Published in the United States of America 2022 * VOLUME 16 « NUMBER 2 AMPHIBIAN & REPTILE CONSERVATION amphibian-reptile-conservation.org ISSN: 1083-446X eISSN: 1525-9153 Front cover: Black-webbed Treefrog (Rhacophorus kio Ohler & Delorme 2006), male, Xuan Nha Nature Reserve, Son La Province, Vietnam. This frog is known to occur in China, Myanmar, Thailand, and Vietnam (Frost 2022). It is categorized as Endangered (EN) by the Vietnam Red Data Book (2007). The Black-webbed Treefrog is a threatened species in Vietnam due to habitat degradation and severe habitat fragmentation. This species was found at night, on trees, and near puddles in the evergreen forests. Photo by Anh Van Pha. Introductory page. Smilisca cyanosticta (Smith, 1953). The Blue-spotted Treefrog occurs on the Atlantic slopes of southern Mexico and northern Central America from Oaxaca and southern Veracruz through northern Chiapas, Mexico, into Guatemala (https://amphibiansoftheworld.amnh.org). These individuals were located at Ejido Villa Guadalupe, in the municipality of Huimanguillo, Tabasco. Wilson et al. (2013b) determined its EVS as 12, placing it in the upper portion of the medium vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, but this species is not listed by SEMARNAT. Photo by Liliana Rios-Rodas. Amphib. Reptile Conserv. 1 August 2022 | Volume 16 | Number 2 | e315 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptile Conservation 16(2) [General Section]: 1-61 (e315). The herpetofauna of Tabasco, Mexico: composition, distribution, and conservation status ‘Ma. del Rosario Barragan-Vazquez, *Liliana Rios-Rodas, *Lydia Allison Fucsko, Louis W. Porras, ‘Vicente Mata-Silva, *Arturo Rocha, ‘Dominic L. DeSantis, ’Eli Garcia-Padilla, Jerry D. Johnson, and ®Larry David Wilson 'Centro de Investigacién para la Conservacion y Aprovechamiento de los Recursos Tropicales, Division Académica de Ciencias Bioldgicas, Universidad Judrez Autonoma de Tabasco, Villahermosa, MEXICO *Facultad Maya de Estudios Agropecuarios, Universidad Aut6noma de Chiapas, Carretera Catazaja-Palenque, Km 4, C.P. 29980, Catazaja, Chiapas, MEXICO +*Department of Humanities and Social Sciences, Swinburne University of Technology, Melbourne, Victoria, AUSTRALIA +7705 Wyatt Earp Avenue, Eagle Mountain, Utah, 84005, USA °Department of Biological Sciences, The University of Texas at El Paso, El Paso, Texas 79968-0500, USA °Department of Biological and Environmental Sciences, Georgia College and State University, Milledgeville, Georgia 31061, USA ‘Oaxaca de Juarez, Oaxaca 68023, MEXICO *Centro Zamorano de Biodiversidad, Escuela Agricola Panamericana Zamorano, Departamento de Francisco Morazan, HONDURAS and 1350 Pelican Court, Homestead, Florida 33035-1031, USA Abstract.—The herpetofauna of Tabasco, Mexico, consists of 170 species, including 39 anurans, five caudates, one caecilian, two crocodylians, 111 squamates, and 12 turtles. We catalogued the distribution of these species among the three physiographic regions we recognize in the state: the Gulf Coastal Plain (88 species), the Sierras Bajas de Petén (93 species), and the Sierra Norte de Chiapas (145 species). The individual species are found in either one, two, or all three regions (mean = 1.9). Approximately 68% of the herpetofauna in Tabasco occupies only one or two of the three regions, which is of important conservation significance. The largest number of single-region species is found in the Sierra Norte de Chiapas (50), followed by the Gulf Coastal Plain (12) and the Sierras Bajas de Peteén (nine). Coefficient of Biogeographic Resemblance (CBR) calculations indicate that the Sierra Norte de Chiapas and the Sierras Bajas de Peten share the greatest number of species (79), followed by 71 species between the Sierra Norte de Chiapas and the Gulf Coastal Plain, and 61 between the Gulf Coastal Plain and the Sierras Bajas de Peten. Fifty-five species occupy all three regions. A similarity dendrogram based on the Unweighted Pair Group Method with Arithmetic Averages (UPGMA) illustrates that the Sierras Bajas de Peteén clusters with the Gulf Coastal Plain at the 0.67 level and the Sierra Norte de Chiapas clusters with the previous pair at the 0.64 level, and overall indicates an intermediate level of similarly. With reference to distributional categories, the greatest number of species is represented by the non-endemic species (146 of 170), followed by the country endemics (20), and the non-natives (five). Of the 146 non-endemic species, the majority (95) are MXCA species (i.e., those found only in Mexico and Central America). The principal environmental threats to the Tabasco herpetofauna are deforestation, agricultural activities, roads, soil contamination and oil extraction, myths and cultural factors (gastronomy), illegal commerce, and forest fires. We evaluated the conservation status of each of the native species by using the SEMARNAT, IUCN, and EVS systems, of which the EVS system provided the most inclusive assessment of the state’s herpetofauna. We also employed the Relative Herpetofaunal Priority (RHP) method to determine the rank order of the three physiographic regions and found the highest values in the Sierra Norte de Chiapas. Most of the protected areas in the state are located in the Gulf Coastal Plain, which is only the second or third most important region from a conservation perspective. Nonetheless, about 95% of the native herpetofauna has been documented within the system of protected areas. Finally, we provide a set of conclusions and recommendations for the future protection of the Tabasco herpetofauna. Keywords. Anurans, caecilians, caudates, conservation status, crocodylians, physiographic regions, protected areas, protection recommendations, squamates, turtles Resumen.—La herpetofauna de Tabasco, Mexico, consta de 170 especies, incluidos 39 anuros, cinco caudados, un cecilido, dos crocodilianos, 111 escamosos y 12 tortugas. Catalogamos la distribuciOn de estas especies entre las tres regiones fisiograficas que reconocemos, incluyendo la Llanura Costera del Golfo (88 especies), las Sierras Bajas del Peten (93 especies) y la Sierra Norte de Chiapas (145 especies). Las especies individuales se Correspondence. rosariobarragan@gmail.com (MRB), aril707@hotmail.com (LR), lydiafucsko@gmail.com (LAF), empub@msn.com (LWP); vmata@utep.edu (VM); arocha3(@miners.utep.edu (AR); dominic.desantis@gcsu.edu (DLD); eligarcia_18@hotmail.com (EG); jjohn- son@utep.edu (JDJ), bufodoc@aol.com (LDW) Amphib. Reptile Conserv. 2 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. encuentran de una a tres regiones (media = 1,9). Aproximadamente el 68% de la herpetofauna de Tabasco ocupa solo una o dos de las tres regiones, lo que es de gran importancia para la conservacion. El mayor numero de especies de una sola region se encuentra en la Sierra Norte de Chiapas (49) seguida por la Llanura Costera del Golfo (12) y las Sierras Bajas del Petén (nueve). Los calculos del Coeficiente de semejanza biogeografica (CBR) demuestran que la Sierra Norte de Chiapas y las Sierras Bajas de Petén comparten el mayor numero de especies (79), seguidas de 71 entre la Sierra Norte de Chiapas y la Llanura Costera del Golfo y 61 entre la Llanura Costera del Golfo y Sierras Bajas del Petéen. Cincuenta y cinco especies ocupan las tres regiones. Un dendrograma de similitud basado en el metodo de grupos de pares no ponderados con promedios aritméticos (UPGMA) ilustra que las Sierras Bajas del Petén se agrupan junto con Llanura Costera del Golfo en el nivel .67 y la Sierra Norte de Chiapas se agrupa con el par anterior en el nivel .64, lo que indica un nivel generalmente intermedio de similitud en general. Con referencia a las categorias de distribucion, el mayor numero de especies es el de las especies no endémicas (146 de 170), seguido de las endémicas del pais (20) y no nativos (cinco). De las 146 especies no endemicas, la mayor parte (95) son especies MXCA. Las principales amenazas ambientales para la herpetofauna de Tabasco son deforestacion, actividades agropecuarias, carreteras, contaminacion del suelo y actividades petroleras, mitos y factores culturales (gastronomia), comercio ilegal, e incendios forestales. El estado de conservacion de cada especie nativa se evaluo mediante el uso de los sistemas SEMARNAT, UICN y EVS, de los cuales el sistema EVS fue de mayor utilidad. También se utilizo el método de Prioridad Relativa de la Herpetofauna (RHP) para determinar el orden de importancia de las tres regiones fisiograficas y los valores mas altos se encontraron en la Sierra Norte de Chiapas. La mayoria de las areas protegidas en el estado estan ubicadas en la Llanura Costera del Golfo, que es solo la segunda o tercera region mas importante desde una perspectiva de conservacion. No obstante, alrededor del 95% de la herpetofauna nativa se ha documentado en el sistema de areas protegidas. Finalmente, se entregan un conjunto de conclusiones y recomendaciones para la futura proteccion de la herpetofauna de Tabasco. Palabras Claves. Anuros, areas protegidas, caudados, cecilidos, crocodilidos, escamosos, estatus de conservacion, recomendaciones de proteccion, regiones fisiograficas, tortugas Citation: Barragan-Vazquez MR, Rios-Rodas L, Fucsko LA, Porras LW, Mata-Silva V, Rocha A, DeSantis DL, Garcia-Padilla E, Johnson JD, Wilson LD. 2022. The herpetofauna of Tabasco, Mexico: composition, distribution, and conservation status. Amphibian & Reptile Conservation 16(2): 1-61 (e315). Copyright: © 2022 Barragan-Vazquez et al. This is an open access article distributed under the terms of the Creative Commons Attribution License [Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are 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: amphibian-reptile-conservation.org. Received: 9 May 2022; Accepted: 21 June 2022; Published: 8 August 2022 “The more we get done ourselves, the easier it will be for our children and their children to move the world back to sustainability. ” Peter H. Raven (2021) Introduction Tabasco is an oddly shaped state in Mexico, in which a western segment and an eastern segment are connected to each other by a slender isthmus. With a total area of 24,731 km”, this state is relatively small (the 24" smallest of the 32 federal entities in Mexico, http://inegi.org; accessed 5 May 2022). The state’s area constitutes only about 1.3% of the country’s area. The coastal region of Tabasco lies adjacent to the southwesternmost corner of the Gulf of Mexico. To the southwest, Tabasco 1s bordered by the state of Veracruz, to the northeast by the state of Campeche, to the south by the state of Chiapas, and to the southeast by a small portion of the northwestern border of Guatemala. To the west, much of the state lies in the Gulf Coastal Plain, where it merges with part of this Amphib. Reptile Conserv. physiographic region in Veracruz, and to the east, this plain merges with the lowlands of the Yucatan Peninsula. The two principal portions of the state are connected by a slender segment of land at least 6 km in width between Campeche and Chiapas, through which passes a portion of the Usumacinta River (Google Earth, https://earth. google.com, accessed: 9 May 2022). The hydrography of Tabasco is dominated by the presence of portions of the first and second largest watersheds in Mexico, those of the Grijalva and the Usumacinta rivers, which arise from divergent points in the central highlands of Chiapas (the Grijalva) and the central highlands of Guatemala (Usumacinta) and join together in a common delta before entering the Gulf of Mexico near the town of Frontera. Tabasco is partitioned into 17 municipalities and its capital is Villahermosa. As of 2020, its population was 2,402,598, which ranks 20" in the country. More recently its density was noted as 97 people/km?’, ranking 12" in the country (http://inegi.org; accessed 5 May 2022). This figure is 1.6 times the average density for Mexico. August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico The southeasternmost portion of the state contains the highest elevation (http://inegi.org; accessed 5 May 2022), between 1,140 and 1,150 m on an unnamed peak located at 17°24’38”N, 92°50’22”W near the border with Chiapas, more or less south of Villahermosa (http:// peakbagger.com; accessed 7 June 2021). As expected, the lowest elevation in the state is sea level, all along the 198.8 km of shoreline (http://inegi.org; accessed 5 May 2022). Materials and Methods Our Taxonomic Position In this paper we follow the same taxonomic position as detailed in previous works on other portions of Mesoamerica (Johnson et al. 2015a,b; Mata-Silva et al. 2015; Teran-Juarez et al. 2016; Woolrich-Pifia et al. 2016, 2017; Nevarez-de los Reyes et al. 2016; Cruz-Saenz et al. 2017; Gonzalez-Sanchez et al. 2017; Lazcano et al. 2019; Ramirez-Bautista et al. 2020; Torres-Hernandez et al. 2021; Cruz-Elizalde et al. 2022). Johnson (2015a) can be consulted for a formal statement of this position, with special reference to the subspecies concept. System for Determining Distributional Status We employed the same system developed by Alvarado- Diaz et al. (2013) for the herpetofauna of Michoacan to ascertain the distributional status of members of the herpetofauna of Tabasco. Subsequently, Mata-Silva et al. (2015), Johnson et al. (2015a), Teran-Juarez et al. (2016), Woolrich-Pifia et al. (2016, 2017), Nevarez-de los Reyes et al. (2016), Cruz-Sanchez et al. (2017), Gonzalez-Sanchez et al. (2017), Lazcano et al. (2019), Ramirez-Bautista et al. (2020), Torres-Hernandez et al. (2021), and Cruz-Elizalde et al. (2022) utilized this system, which consists of four categories: SE = endemic to Tabasco; CE = endemic to Mexico; NE = not endemic to Mexico; and NN = non-native in Mexico. Systems for Determining Conservation Status To assess the conservation status of the herpetofauna of Tabasco, we employed the same three systems (-e., SEMARNAT, IUCN, and EVS) used by Alvarado-Diaz et al. (2013), Mata-Silva et al. (2015), Johnson et al. (2015a), Teran-Juarez et al. (2016), Woolrich-Pifia et al. (2016, 2017), Nevarez-de los Reyes et al. (2016), Cruz- Sanchez et al. (2017), Gonzalez-Sanchez et al. (2017), Lazcano et al. (2019), Ramirez-Bautista et al. (2020), Torres-Hernandez et al. (2021), and Cruz-Elizalde et al. (2022). Detailed descriptions of these three systems appear in the earlier papers of this series, and thus are not repeated here. Amphib. Reptile Conserv. The Mexican Conservation Series The Mexican Conservation Series (MCS) was initiated in 2013, with a study of the herpetofauna of Michoacan (Alvarado-Diaz et al. 2013), as a part of a set of five papers designated as the “Special Mexico Issue” published in Amphibian & Reptile Conservation. The basic format of the entries in the MCS was established in this paper, 1.e., providing an examination of the composition, physiographic distribution, and conservation status of the herpetofauna of a given Mexican state or group of states. Two years later, the MCS resumed with a paper on the herpetofauna of Oaxaca (Mata-Silva et al. 2015). That same year, Johnson et al. (2015a) presented a paper on the herpetofauna of Chiapas. The following year, three entries in the MCS appeared, on Tamaulipas (Teran- Juarez et al. 2016), Nayarit (Woolrich-Pifia et al. 2016), and Nuevo Leon (Nevarez-de los Reyes et al. 2016). Thereafter, three entries were published in 2017, on Jalisco (Cruz-Saenz et al. 2017), the Mexican Yucatan Peninsula (Gonzalez-Sanchez et al. 2017), and Puebla (Woolrich- Pifia et al. 2017), followed by subsequent entries on Coahuila (Lazcano et al. 2019), Hidaldo (Ramirez- Bautista et al. 2020), Veracruz (Torres-Hernandez et al. 2021), and most recently one on Querétaro (Cruz-Elizalde et al. 2022). Therefore, this paper on the herpetofauna of Tabasco is number 14 in this series. Physiography and Climate Physiographic Regions To analyze the distribution of the herpetofauna of Tabasco, we used the classification system of physiographic regions of INEGI (1986 and 2016). According to these studies, two physiographic regions are distinguished, one with two subregions (Fig. 1), which are described here. Gulf Coastal Plain (GCP). This province (Fig. 2) comprises 95.7% of the state’s area. Located in southeastern Mexico, it encompasses the states of Campeche, Chiapas, Oaxaca, Tabasco, and Veracruz; and its average length in each state is between 125 and 150 km. To the north, its limits are defined by the Gulf of Mexico; to the east, by the Yucatan Peninsula and Belize: to the south, by the Central American Cordillera and the Sierras de Chiapas and Oaxaca; and to the west by the Sierra Madre del Sur and Sierra Volcanica Transversal or Eye Neovolcanico. The Gulf Coastal Plain was formed by alluvium carried by the Papaloapan, Coatzacoalcos, Grijalva, and Usumacinta rivers, which cross the province before emptying into the Gulf of Mexico. In the central part of this plain, the lower basins of the Griyalva and Usumacinta rivers (the largest basins in the country) meet and then converge south of the port of Frontera, Tabasco, to exit into the Gulf of Mexico. The Usumacinta August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. 3315000 3400000 3485000 3570000 3655000 Gulf of Mexico SNC Chiapas So Qo ao uw nn w Oaxaca Physiographic regions Mi Gulf Coastal Lowlands ™ Sierras del Norte de Chiapas y Guatemala Physiographic subregions M™) Sierras del Norte de Chiapas (SNC) ™) Sterras Bajas del Petén (SBP) Guatemala 100 km a 3315000 3400000 3485000 3570000 3655000 Fig. 1. Physiographic regions in the state of Tabasco, Mexico, and location of the state in Mexico. and Grijalva rivers contribute about 27% of Mexico’s hydrologic resources (West et al. 1985). Throughout most of this province, relatively young sedimentary rocks form extensive alluvial plains and coastal plains with an almost flat relief at elevations below 100 m. This relief creates extensive flood plains and lagoons, among which La Machona, Mecoacan, Sitio Grande, and El Rosario are the most prominent (INEGI 2006; SAHOP 1980). Two types of climates are evident: (1) warm humid with abundant rainfall in summer, which covers 76.0% of the surface area and is distributed from the coastal zone to the vicinity of the mountains in the southern portion of the state, and (11) warm subhumid with summer rains, which is present toward northeastern Tabasco, in the municipality of Balancan. This region is the least humid in the state, with an average annual temperature of 26.4 °C (INEGI 2016). Grasslands used for grazing livestock have displaced the natural vegetation in this physiographic region, as they cover 30.6% of the area; and additional agricultural areas occupy 25.8% of this region. The third most common type of vegetation is the Tular-popal association, which forms dense patches that cover 26.6% of the swampy areas. In addition, some forests are dominated by a single species (16.2%), such as Cashan (7erminalia amazonia), laurel (Nectandra sp.), Mulato (Bursera simaruba) or Chicozapote (Achras zapota). To a lesser degree, the mangroves (2.8%), which are composed of a group of halophilic plants, are characterized by such dominant species as Red Mangrove (Rhizophora mangle), Black Mangrove (Avicennia germinans), or White Mangrove (Laguncularia racemosa) (INEGI 2016). Sierras Bajas del Petén (SBP). This province (Fig. 3) only covers 4.3% of the area of the state, and includes the mountains that extend from southeastern Mexico to Guatemala. This region is characterized by a parallel Amphib. Reptile Conserv. arrangement of folded mountain ranges with rounded summits, steep flanks, and wide intermontane valleys at its base (Zavala-Cruz and Ortiz-Pérez 2019). To the north, this province is limited by the occurrence of the Gulf Coastal Plain, to the east by Belize, to the south by Guatemala, and to the west it borders the Central American Mountain Cordillera. This province is divided into five physiographic subprovinces, two of which occur in Tabasco: the northern Sierras of Chiapas and the Lower Sierras of Petén. Sierras del Norte de Chiapas (SNC). In Tabasco, this region (Fig. 4) is composed of two small portions to the south that together cover an area of 986.0 km? and comprise parts of the municipalities of Huimanguillo, Macuspana, Tacotalpa, and Teapa (INEGI 1986). In these areas, the highest elevations are the hills of La Pava and La Ventana (at elevations of 880 and 560 m, respectively); and the Madrigal, Tapijulapa, and Poana Mountains (at elevations from 560 to 900 m). The lower hills are La Campana, La Corona, Cocona, Mono Pelado, and El Tortuguero (CONAFOR 2013). Limestone rocks such as dolomites and marls dominate this region, and they alternate with shales and sandstones, but there also are many types of ancient alluvium, igneous rocks formed from volcanic clasts, andesites, and volcanic ash. The lithological diversity gives these mountain ranges a “complex character” (INEGI 1989; Zavala-Cruz and Ortiz-Pérez 2019), and karst features are prominent. Climate Temperature. Here, we present the monthly minimum, mean, and maximum temperatures for a single locality in each of the three recognized physiographic regions in Tabasco (Table 1). The elevations for these three localities range from 10 m at Villahermosa in the Gulf August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Fig. 2. Gulf Coastal Plain. Mangroves in the municipality Paraiso, Tabasco. Photo by José del Carmen Geronimo-Torres. Fig. 4. Sierra Norte de Chiapas. Mountain Cloud Forest fragment in the municipality of Huimanguillo, Tabasco. Photo by Liliana Rios-Rodas. Coastal Plain to 34 m at Huimanguillo in the Sierra Norte de Chiapas. The mean annual temperature (MAT) is highest at Tenosique (elevation 19 m) in the Sierras Bayas del Petén (SBP) at 26.7 °C. The MAT for the other two localities in the Gulf Coastal Plain (GCP) and the Sierra Norte de Chiapas (SNC) differ by only 0.1 °C (26.4 °C for the GCP and 26.3 °C for the SNC). These values are reflective of the limited variation in elevational range in Tabasco. The minimum annual temperatures range from 21.6 °C in the SBP and the SNC to 23 °C in the GCP, which only represents a difference of 1.4 °C (Table 1). The mean minimum monthly temperatures peak in May in the GCP and SBP (at 25.6 °C in the former, and 23.6 °C in the latter) and in June in the SNC (at 23.7 °C). The mean maximum monthly temperatures are highest in May in all three regions, respectively 34.8 °C, 35 °C, and 35.4 °C in the GLC, SNC, and SBP. The monthly maximum temperatures are lowest in January in the GCP (at 26.7 °C) and SBP (at 28.0 °C), and in December and January in the SNC (at 26.9 °C). Precipitation. Naturally, monthly precipitation is lowest during the dry season in February (in the SBP), March (in the GCP), or April (in the SNC), and highest during the rainy season in September in all three regions (Table 2). The data in Table 2 demonstrate that 63.0—76.3% of the Amphib. Reptile Conserv. Fig. 3. Sierras Bajas del Petén. Panoramic view of the Sierras Bajas del Petén, Ejido Nuevo Progreso, municipality of Tenosique, Tabasco, near the border with Guatemala. Photo by Nelly del Carmen Jiménez-Pérez. yearly precipitation falls during the rainy season, from May to October. The annual rainfall ranges from 1,476.0 mm in the SBP to 2,316.8 mm in the SNC (Table 2). Composition of the Herpetofauna Families The members of the native and non-native herpetofauna of Tabasco are arranged among 45 families, including 10 families of anurans, one of salamanders, one of caecilians, one of crocodylians, 24 of squamates, and eight of turtles (Table 3). The total of 45 families includes 72.6% of the 62 families with native, non-native, and introduced/ questionable members represented in Mexico (J. Johnson, unpublished, 26 March 2022). Among the 12 amphibian families, 51.1% of the species (Tables 4 and 5) are classified in the families Craugastoridae (seven species) and Hylidae (16 species). Among the 33 reptile families, 59.5% of the species (Tables 4 and 5) are classified in the families Dactyloidae (14 species), Phrynosomatidae (five), Colubridae (20), Dipsadidae (30), and Viperidae (six). Genera The genera of amphibians and reptiles represented in Tabasco number 104, including 24 genera of anurans, one of salamanders, one of caecilians, one of crocodylians, 67 of squamates, and 10 of turtles. These 104 genera include 48.8% of the 213 recorded for Mexico (J. Johnson, unpublished, 26 March 2022). Among the amphibians (Table 4), the largest numbers of species are classified in the genera Craugastor (seven species) and Bolitoglossa (five); among the reptiles (Table 4), the most speciose genera are Norops (14 species), Sceloporus (five), and Coniophanes (Six). Species The herpetofauna of Tabasco consists of 170 species, including 39 anurans, five salamanders, one caecilian, two crocodylians, 111 squamates, and 12 turtles (Tables 3 August 2022 | Volume 16 | Number 2 | e315 August September October November December Annual July June April May February March January Table 1. Monthly minimum (Min), mean (Mean, in parentheses), maximum (Max), and annual temperature data (in °C) for the three physiographic regions of Tabasco, Mexico. The selected localities for each region and their elevations are as follows: Gulf Coastal Plain—Cardenas (29 m asl), Centla (4 m), Villahermosa (10 m); Sierra Norte de Chiapas—Huimanguillo (29 Physiographic region m), Teapa (41 m), Macuspana (13 m); and Sierras Bajas del Petén—Tenosique (19 m). Data were taken from https://es.climate-data.org and https://smn.conagua.gob.mx/es/climatologia (Accessed: 16 June 2021). Amphib. Reptile Conserv. pee (26.4) 30.4 24.5 23.6 21.9 20.8 (26.9) (25.9) (24.4) (23.5) 30.5 29.1 O17 27.0 25.2 (28.1) 32.1 25.2 25.2 (28.1) (28.4) 32.1 32.5 25.8 (29.5) 34.3 24.4 (28.6) 33.9 22.4 (26.2) 319 21.0 (24.1) 28.4 20.1 (22.8) 26.5 Gulf Coastal Plain 23.1 22.1 20.3 19.1 21.5 (26.8) (25.6) (24.0) (22.8) (25.8) 31.0 29.5 28.0 26.9 30.5 23.3 (27.6) 39 2 23.7 23.3 (28.1) (27.8) 32.8 32.6 23.7 (28.9) 34.3 22.2 (27.7) 33.4 19.0 20.2 (23.4) (25.3) 28.1 30.9 18.3 (22.2) 26.5 Sierra Norte de Chiapas Barragan-Vazquez et al. 21.6 (26.7) 31.9 19.4 (23.9) 28.5 20.7 (25.2) 29.8 22.3 (26.7) BD 22.9 (27.8) 32.7 22.9 (28.2) 33.5 22.9 (28) 33.1 23.5 (28.6) 33.7 23.6 (29.5) 35.4 22.5 (28.6) 34.8 20.4 (26.3) 32.1 19.4 (24.6) 29.7 18.8 (23.4) 28 Sierras Bajas del Petén and 4). The current numbers of native species in Mexico for these six groups are, respectively, 258, 155, 3, 3, 902, and 51 (J. Johnson, unpublished, 26 March 2022). The 165 native species in Tabasco comprise 12.0% of the 1,372 species in the entire native Mexican herpetofauna (J. Johnson, unpublished, 29 May 2021). Three states in Mexico border Tabasco, and all have been evaluated in the Mexican Conservation Series (Chiapas: Johnson et al. 2015a; Campeche: Gonzalez- Sanchez et al. 2017; Veracruz: Torres-Hernandez et al. 2021). Based on these works, the total figures for the native taxa in these states are as follows: Chiapas, 326; Campeche, 125; and Veracruz, 351. The number of native species in Tabasco (165) is closest to that in Campeche, essentially another lowland state in the western portion of the Yucatan Peninsula. As expected, the two larger and more montane states to the north (Veracruz) and south (Chiapas) contain 2.1 and 2.0 times as many species, respectively, as Tabasco. Patterns of Physiographic Distribution We used a system of three regions (Fig. 1) to analyze the physiographic distribution patterns of members of the Tabasco herpetofauna. The results for the 170 species are tabulated in Table 4 and summarized in Table 5. The total number of taxa in each of the three regions we recognize ranges from 88 in the Gulf Coastal Plain to 145 in the Sierra Norte de Chiapas. The total for the remaining area (Sierras Bajas del Petén) is 93. The average of these three values 1s 108.7, or 63.9% of the number for the total herpetofauna (170). The lowest value (88) is 51.8% of the total value (170), and the corresponding percentages for the other two regions in numerical order are 54.7 (93/170) and 85.3 (145/170). These results indicate that the higher elevations in the state, as in the Sierra Norte de Chiapas (see above), exhibit much greater herpetofaunal diversity than the corresponding lower elevations. This situation is consistent with the recognition that herpetofaunal diversity in Mexico is highest in the nearby or bordering states of Oaxaca and Chiapas (Mata-Silva et al. 2015; Johnson et al. 2015a) to the south. Six herpetofaunal groups are represented in Tabasco, i.e., anurans, salamanders, caecilians, crocodylians, squamates, and turtles. As is typical for the state herpetofaunas in Mexico, anurans and squamates contain the largest numbers of species and the caecilians and crocodylians the fewest, while the salamanders and turtles are represented by intermediate numbers. The largest numbers of anurans (36 of 39, or 92.3%), salamanders (four of five, or 80.0%), and squamates (96 of 111, or 86.5%) and of the herpetofauna in general (145 of 170, or 85.4%) occupy the Sierra Norte de Chiapas. Nonetheless, turtles do not follow this pattern, inasmuch as all 12 of the species in Tabasco occur on the Gulf Coastal Plain, with only six of them (50.0%) occurring in the Sierra Norte de Chiapas, and August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico No. 1. Hyalinobatrachium viridissimum (Taylor, 1942). The Northern Glassfrog is a non-endemic species distributed from Guerrero and Veracruz, Mexico, through Guatemala and Belize to northwestern Honduras, and possibly to the departments of Santa Ana and Cabafias in El Salvador (https:// amphibiansoftheworld.amnh.org/). This individual is from Muku Chem, in the municipality of Tacotalpa, Tabasco. Torres- Hernandez et al. (2021) calculated its EVS as 11, placing it in the lower portion of the medium vulnerability category. Its conservation status has not been assessed by either the IUCN or SEMARNAT. Photo by Manuel Herndndez- May. No. 2. Craugastor alfredi (Boulenger, 1898). Alfred’s Rain Frog is distributed from central Veracruz, northern Oaxaca, and southward to the states of Tabasco and Chiapas, Mexico (https:// amphibiansoftheworld.amnh.org/). This individual was located at Muku Chem, in the municipality of Tacotalpa, Tabasco. Wilson et al. (2013b) determined its EVS as 9, placing it at the upper limit of the low vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, but this species is not listed by SEMARNAT. Photo by Liliana Rios-Rodas. No. 3. Craugastor berkenbuschii (Peters, 1870). Berkenbusch’s Robber Frog ranges along the Atlantic slopes of southern San Luis Potosi, Hidalgo, Puebla, Veracruz, Tabasco, and northern Oaxaca, north of the Isthmus of Tehuantepec, Mexico (https:// amphibiansoftheworld.amnh.org/). This individual was located at Muku Chem, in the municipality of Tacotalpa, Tabasco. Wilson et al. (2013b) assessed its EVS as 14, placing it at the lower limit of the high vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN and as a species of Special Protection (Pr) by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. Amphib. Reptile Conserv. No. 4. Duellmanohyla chamulae (Duellman, 1961). The Chamula Mountain Brook Frog is known only from a few localities at elevations above 1,600 m on the northern slopes of the Central Highlands of Chiapas and into adjacent extreme southwestern Tabasco, Mexico (https://amphibiansoftheworld. amnh.org/). This individual was encountered at Ejido Villa Guadalupe, in the municipality of Huimanguillo, Tabasco. Wilson et al. (2013b) determined its EVS to be 13, placing it at the upper limit of the medium vulnerability category. Its conservation status has been evaluated as Endangered (EN) by IUCN, and as a species of Special Protection (Pr) by SEMARNAT. Photo by José del Carmen Geronimo-Torres. August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 2. Monthly and annual precipitation data (in mm) for the three physiographic regions of Tabasco, Mexico. The selected localities for each of the regions, with elevation given in parentheses) are as follows: Gulf Coastal Plain—Cardenas (29 m asl), Centla (4 m), Villahermosa (10 m); Sierra Norte de Chiapas—Huimanguillo (29 m), Teapa (41 m), Macuspana (13 m); Sierras Bajas del Petén—Tenosique (19 m). Data were taken from https://es.climate-data.org and https://smn.conagua.gob.mx/es/climatologia (Accessed: 16 June 2021). The shaded area indicates the months of the rainy season. Physiographic Jan Feb Mar Apr May Jun region GulfCoastal 76g 4g5 0303353 i Plain picwa deh None, “Ase wat Ween TA 120 2573 de Chiapas Sierras Bajas del 62 34 eye 54 95 227 Petén an intermediate number of nine (75.0%) occurs in the Selvas Bayas del Petén. The members of the Tabasco herpetofauna are distributed in either one, two, or three physiographic regions (Table 4), as follows: one (69 of 170 species; 40.6%); two (46; 27.1%); and three (55, 32.4%). The mean regional occupancy figure is 1.9, which is slightly higher than the 1.6 value for Querétaro, another state with three physiographic regions that was assessed in the Mexican Conservation Series (Cruz-Elizalde et al. 2022). A sizable portion of the 170 species in Tabasco (115; 67.6%) occurs in only one or two of the three physiographic regions, which is of considerable conservation significance (see below). The number of species occupying a_ single physiographic region ranges from eight in the Sierras Bajas del Petén (SBP) to 50 in the Sierra Norte de Chiapas (SNC). The 50 single-region species in the SNC (Table 7) are as follows (numbers refer to distributional categories as reported by Wilson et al. [2017], and asterisks indicate the country endemics): Incilius macrocristatus 4 Hyalinobatrachium viridissimum 4 Craugastor berkenbuschii* Craugastor pelorus* Charadrahyla chaneque* Duellmanohyla chamulae* Exerodonta bivocata* Ptychohyla macrotympanum 4 Quilticohyla zoque* Rheohyla miotympanum* Triprion spinosus 4 Gastrophryne elegans 4 Agalychnis moreletii 4 Bolitoglossa platydactyla* Bolitoglossa rufescens 4 Bolitoglossa veracrucis* Norops barkeri* Norops capito 4 Norops compressicauda* Norops laeviventris 4 Amphib. Reptile Conserv. Jul Aug Sept Oct Nov Dec 136.5 193.3 2105 foe «125.8 79.8 L938 247.8 341.5 299 210.3. 178.3 141 206 261 196 104 60 Norops petersi 4 Lepidophyma tuxtlae* Xenosaurus rackhami 4 Dendrophidion vinitor 4 Ficimia publia 4 Phrynonax poecilonotus 6 Senticolis triaspis 7 Stenorrhina freminvillii 4 Tantilla rubra 4 Tantilla schistosa 4 Tantillita lintoni 4 Adelphicos quadrivirgatum 4 Amastridium sapperi 4 Coniophanes piceivittis 4 Dipsas brevifacies 4 Geophis carinosus 4 Geophis laticinctus* Geophis sanniolus 4 Leptodeira maculata 4 Leptodeira septentrionalis 4 Ninia diademata 4 Oxyrhopus petolarius 6 Rhadinaea decorata 6 Sibon dimidiatus 4 Sibon nebulatus 6 Xenodon rabdocephalus 6 Micrurus elegans 4 Scaphiodontophis annulatus 4 Amerotyphlops tenuis 4 Bothriechis schlegelii 6 Order Families Genera Anura 10 24 Caudata 1 1 Gymnophiona 1 1 Subtotal 12 26 Crocodylia l 1 Squamata 24 67 Testudines 8 10 Subtotal 33 78 Total 45 104 Annual 1,489.8 2,316.8 1,476 Table 3. Taxonomic composition of the native and non-native herpetofauna of Tabasco, Mexico. Species 39 5 1 45 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico No. 5. Exerodonta bivocata (Duellman and Hoyt, 1961). The Chiapan Highlands Treefrog is distributed along the Atlantic slopes of extreme southwestern Tabasco, Oaxaca, and Chiapas in southern Mexico (https://amphibiansoftheworld.amnh.org/). This individual was located in Ejido Villa Guadalupe, in the municipality of Huimanguillo, Tabasco. Wilson et al. (2013b) assessed its EVS as 15, placing it in the lower portion of the high vulnerability category. Its conservation status has been judged as Endangered (EN) by the IUCN, but this species is not listed by SEMARNAT. Photo by Liliana Rios-Rodas. No. 6. Ptychohyla macrotympanum (Tanner, 1957). The Pine Forest Stream Frog is distributed in humid montane and pine- oak forest, on the northern slopes of the Chiapan Highlands of Tabasco and Chiapas in Mexico (https://amphibiansoftheworld. amnh.org/). This individual was found in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. Wilson et al. (2013b) assessed its EVS as 11, placing it in the lower portion of the medium vulnerability category. Its conservation status has been considered as Vulnerable (VU) by the IUCN, but this species is not listed by SEMARNAT. Photo by Jenny del Carmen Estrada- Montiel, No. 7. Quilticohyla zoque (Canseco-Marquez, Aguilar-Lopez, Luria-Manzano, Pineda-Arredondo, and Caviedes-Solis, 2017). The Zoque Treefrog is distributed in evergreen tropical forest at three localities in southern Mexico in the Selva Zoque, two in southern Veracruz (Paso del Moral and Arroyo Zarco), one in extreme southwestern Tabasco near the Veracruz and Chiapas borders, and one in northeastern Oaxaca (Chalchijapa) (https:// amphibiansoftheworld.amnh.org/). This individual was located in Ejido Villa Guadalupe, in the municipality of Huimanguillo, Tabasco. Torres-Hernandez et al. (2021) assessed its EVS as 14, placing it at the lower limit of the high vulnerability category. Its conservation status has been judged as Endangered (EN) by the IUCN, but this species is not listed by SEMARNAT. Photo by Liliana Rios-Rodas. Amphib. Reptile Conserv. No. 8. Zriprion spinosus (Steindachner, 1864). The Coronated Treefrog occurs in humid forests, primarily in the premontane zone of eastern Mexico in the states of Tabasco, Veracruz, Puebla, Oaxaca, and Chiapas, and on into Central America south into Panama (https://amphibiansoftheworld.amnh.org/). This individual is from Cerro El Madrigal, in the municipality of Teapa, Tabasco. Wilson et al. (2013b) calculated its EVS as 10, placing it in the lower portion of the medium vulnerability category. Its conservation status has been considered as Near Threatened (NT) by IUCN, but this species is not listed by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 4. Distribution of the herpetofauna of Tabasco, Mexico, by physiographic region. No asterisk = non-endemic; * = country endemic; ** = non-native. Physiographic region Sierras de Chiapas y Guatemala Number of Taxon Gulf Coastal Plain regions (GCP) Sierra del Norte de Sierras Bajas del Chiapas (SNC) Petén (SBP) | Total herpetofauna(170 species) | | AMPHIBIA(4Sspeciesy) | | Anura(39speciesy | | Bufonidae(3species) | | Incilius macrocristaus | | Centrolenidae(I species) | | Hyalinobatrachium viridissimum | | Craugastoridae(7speciesy | | Craugastoralfred’ | | Craugastor berkenbuschit* | | Craugastorlaticeps | | Craugastorloki | | Craugastorpalenque | | Craugastorpelorus* | ECraugasior@hodoy se ee || =e | Eleutherodactylidae(2 species) | | Eleutherodactylusleprus | | Eleutherodactylusplanirostris** | + | Hylidae (16 species) | | Charadrahylachaneque® | | Dendrosophusebraccatus | HP | Dendrosophus microcephalus | HT 8 | Duellmanohyla chamulac* | | Exerodontabivocata® | | Ptychohyla macrotympanum | | Quilticohylazoques | | Rheohyla miotympamum® | Se AAD AE Ss | | Smiliscabaudini | 8 | Smilisca cyanosticta | | Talocohylaloquax | 8 | Talocohylapicta | | Trachycephalus vermiculatus | HT 8 | Triprionpetasatus | | Triprion spinosus | | Leptodactylidae(3 species) | | Engystomopspustulosus | | Leptodactylusfragilis | | Microhylidae(2 species) | | Gastrophyrneelegans | | Hypopachusvariolosus | | Phyllomedusidae(2 species) | | Agalychnismoreletii | | Ranidae(2 species) | AA {i THEFT ITE Amphib. Reptile Conserv. 11 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 4 (continued). Distribution of the herpetofauna of Tabasco, Mexico, by physiographic region. No asterisk = non-endemic; * = country endemic; ** = non-native. Physiographic region Sierras de Chiapas y Guatemala Number of Taxon Gulf Coastal Plain regions (GCP) Sierra del Norte de Sierras Bajas del Chiapas (SNC) Petén (SBP) Lithobates vaillanti Rhinophrynidae (1 species) Caudata (5 species) Plethodontidae (5 species) Bolitoglossa alberchi* Rhinophrynus dorsalis Bolitoglossa mexicana Bolitoglossa rufescens Bolitoglossa veracrucis* Gymnophiona (1 species) Dermophis mexicanus Reptilia (125 species) Squamata (111 species) Norops biporcatus Norops lemurinus Norops petersii O Norops capito Norops compressicauda* Norops sericeus Norops laeviventris Norops sagrei** Norops tropidonotus Norops uniformis Ctenosaura similis + + + + + + + + + + + + + 3 Amphib. Reptile Conserv. 12 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 4 (continued). Distribution of the herpetofauna of Tabasco, Mexico, by physiographic region. No asterisk = non-endemic; * = country endemic; ** = non-native. Physiographic region Sierras de Chiapas y Guatemala Number of Taxon Gulf Coastal Plain regions (GCP) Sierra del Norte de Sierras Bajas del Chiapas (SNC) Petén (SBP) Amphib. Reptile Conserv. 13 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 4 (continued). Distribution of the herpetofauna of Tabasco, Mexico, by physiographic region. No asterisk = non-endemic; * = country endemic; ** = non-native. Physiographic region Sierras de Chiapas y Guatemala Number of Taxon Gulf Coastal Plain regions (GCP) Sierra del Norte de Sierras Bajas del Chiapas (SNC) Petén (SBP) | Stenorrhina degenhardtii__—— | | Stenorrhina freminvillii__— | | Tantillarubra | | Tantillaschistosa | | Tantillitalinton’ | | Dipsadidae(30 species) | | Adelphicos quadrivirgatum | | Amastridium sapperi_ | [Celia seytalina | | Coniophanes bipunctatus | | Coniophanesfissidens | | Coniophanes imperialis | 8 | Coniophanespiceivitis | | Coniophanes quinquevittatus | HT 8 | Coniophanes schmidti | -ConophisHineatus FP Dipsas brevifacies +]+]+ +/+]+]+]+]4+]4]+ N Je PRO JR PRO PN TN IN JR fe Enulius flavitorques Geophis carinosus Geophis laticinctus* _ SSS SSS ea SSS Ee Geophis sanniolus pf Geophis sartorii + NTH ee | (A | Imantodes gemmistraus | CC CU Cid tT Leptodeira renata | Leptodeiramaculaa | —C—“‘iLSCSCC“‘(S+TOTCOUWT CC | Leptodeira septentrionalis | CE CU CS CC Ninia diademata mT HH ttt | ++ Ninia sebae Oxyrhopus petolarius _——S_—— Pliocercus elapoides SSE Se LS a es | ee Rhadinaea decorata SSE ISS _ ee Sibon dimidiatus i | + tHe Sibon nebulatus Vanes | Tretanorhinus nigroluteus | | Xenodonrabdocephalus | | Elapidae(2 species) | | Micrurusdiastema* | | Micruruselegans | | Leptotyphlopidae(I species) | | Epictiaphenops | | Natricidae(3 species) | | Nerodiarhombifera | | Thamnophismarcianus | | Thamnophisproximus | | Sibynophiidae(I species) | | Scaphiodontophisannulaus | | Typhlopidae(2speciesy | | Amerotyphlopstenuis | + ]+ — |r Amphib. Reptile Conserv. 14 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 4 (continued). Distribution of the herpetofauna of Tabasco, Mexico, by physiographic region. No asterisk = non-endemic; * = country endemic; ** = non-native. Number of Physiographic region Sierras de Chiapas y Guatemala Taxon Gulf Coastal Plain (GCP) Sierra del Norte de Sierras Bajas del Chiapas (SNC) Petén (SBP) Viperidae (6 species) Agkistrodon russeolus Virgotyphlops braminus** Cheloniidae (2 species) Thirteen of these 50 species (26.0%) are country endemics and 37 (74.0%) are non-endemics. Thirty of the 37 non-endemics (81.1%) are MXCA species, and Dermatemydidae (1 species) _—_ _) regions Oxybelis fulgidus 6 Stenorrhina freminvillii 4 Enulius flavitorques 6 thus are distributed some distance from Mexico into Epictia phenops 4 Central America. Six of these non-endemics (16.2%) Nerodia rhombifera 3 are MXSA species, and thus range from Mexico through Virgotyphlops braminus** Central America and into South America. Finally, one Chelonia mydas 9 non-endemic (2.7%) is a USCA species, and thus ranges Lepidochelys kempii 9 from the United States to Central America. The 11 single-region species in the GCP (Table 7) are as follows (numbers refer to the distributional categories as designated by Wilson et al. [2017]; one asterisk indicates a country endemic species; and two asterisks a non- native species): Eleutherodactylus planirostris** Aspidoscelis guttatus* Amphib. Reptile Conserv. Dermochelys coriacea 9 Note that only one of these 11 species (9.1%) is a country endemic, two (18.2%) are non-natives, and eight (72.7%) are non-endemics. Of the eight non-endemics, one is a MXUS species (12.5%), the only one in Tabasco that ranges northward from Mexico into the United States. Two of these are MXCA species (25.0%), two are MXSA species (25.0%), and three are OCEA (or oceanic) species (37.5%; the sea turtles). August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico lipped Frog is distributed from the Lower Rio Grande Valley of southern Texas (USA) through eastern and southern Mexico (southeast from Colima), and into Central America through northern and western Colombia (https://amphibiansoftheworld. amnh.org/). This individual was found in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. Wilson et al. (2013b) calculated its EVS as 5, placing it in the lower portion of the low vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, but this species is not listed by SEMARNAT. Photo by José del Carmen Geronimo-Torres. No. 11. Lithobates vaillanti (Brocchi, 1877). Vaillant’s Frog ranges from “low and moderate elevations from north-central Veracruz and northern Oaxaca to the central Rio Magdalena region in Colombia on the Atlantic versant and on the Pacific versant in southeastern Oaxaca and northwestern Chiapas, Mexico, and from northwestern Nicaragua to southwestern Ecuador” (https://amphibiansoftheworld.amnh.org/). This individual was located in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. Wilson et al. (2013b) determined its EVS as 9, placing it at the upper limit of the low vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, but this species is not listed by SEMARNAT. Photo by José del Carmen Geronimo-Torres. Amphib. Reptile Conserv. 16 No. 10. Agalychnis taylori Funkhouser, 1957. Taylor’s Leaf Frog occurs on the Atlantic slopes and lowlands from southern Veracruz and northern Oaxaca in Mexico, through the more humid portions of Tabasco, Campeche, Quintana Roo and Yucatan, and on through Guatemala to west-central Honduras (https://amphibiansoftheworld.amnh.org/). This individual was found in the municipality of Centro, Tabasco. Torres-Hernandez et al. (2021) calculated its EVS as 11, placing it in the lower portion of the medium vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, but this species is not listed by SEMARNAT. Photo by José del Carmen Geronimo-Torres. No. 12. Bolitoglossa mexicana Dumeéril, Bibron, and Dumeéril, 1854. The Mexican Mushroom-tongued Salamander is distributed from the “Atlantic slope from southern Veracruz (Mexico) across the base of the Yucatan Peninsula, with an isolated population in the northern part of Yucatan Peninsula, to Honduras (extending to the Pacific versant in the Ocotepeque) and El Salvador (Departamento de Chalatenango, municipio de La Palma, Cerro La Palma)” (https://amphibiansoftheworld. amnh.org/). This individual was encountered in Villa Luz, in the municipality of Tacotalpa, Tabasco. Wilson et al. (2013b) assessed its EVS as 11, placing it in the lower portion of the medium vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN and it is allocated to the Special Protection (Pr) category by SEMARNAT. Photo by Liliana Rios-Rodas. August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 5. Distribution of herpetofaunal families in Tabasco, Mexico, by physiographic province. See Table 4 header for an explanation of the abbreviations. Distribution among physiographic regions ———<—<$<$<=__— = — | =< — = | ay ee, a a a a a ri |, = a= fe ee inate alt sie dus age aly cass | J a a a CT a CT CT a SS a a a as a waa sss | Subiowl ae 0 Se | YS SY TY "ST re ——$ $1 — a — ee a PDiploglossidae i |?) ae | | in i ee ies ols oe |. te 4 Ga a ee ee ea ee a a Ts Se a CT Siedegiiee = a pF eS 7 so. | oe Si i Re: ia ge l= ome eal ee | St | a | a a TS ce ee. “|| __™e 0) aS al) @e =|) «= a ——— I i a a a | [sibynophiidae | ies a le a nl el | | PS (| TY ST a ee | | | 7 Number of species im iA AHHH Re HTH i Amphib. Reptile Conserv. 17 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 6. Pair-wise comparison matrix of Coefficient of Biogeographic Resemblance (CBR) data of the herpetofaunal relationships for the three physiographic regions in Tabasco, Mexico. Underlined values = number of species in each region; upper triangular matrix values = species in common between two regions; and lower triangular matrix values = CBR values. The formula for this calculation is: CBR = 2C/N, + N, (Duellman, 1990), where C is the number of species common to both regions, N, is the number of species in the first region, and N, is the number of species in the second region. See Table 4 for an explanation of the abbreviations, and Fig. 12 for the UPGMA dendrogram produced from the CBR data. Gulf Coastal Plain Sierra Norte de Chiapas Sierras Bajas del Petén Gulf Coastal Plain 88 71 61 Sierra Norte de Chiapas 0.61 145 79 Sierras Baja de Petén 0.67 0.66 93 The eight single-region species inthe SBP (Table 7) areas 79 between the SNC and the SBP. The average value of follows (numbers refer to the distributional categories as — shared species among all three regions is 70.3. designated by Wilson et al. [2017]; an asterisk indicates The CBR data in Table 6 demonstrate values ranging from country endemics): 0.61 to 0.67 (see below), with a mean value of 0.65. This range of CBR values is limited and the values are relatively Triprion petasatus 4 high, indicating that many of these species are widespread. Engystomops pustulosus 6 We determined the numbers of species inhabiting one, Bolitoglossa alberchi* two, and three of the recognized physiographic regions Scincella gemmingeri* (Table 7). In each of the two smaller herpetofaunas for Coniophanes schmidti 4 the Gulf Coastal Plain and the Sierras Bajas del Petén Leptodeira frenata 4 subregion the numbers of species found in one, two, Agkistrodon russeolus 4 and three regions increase from the lowest to the highest Crotalus tzabcan 4 value. However, in the area with the largest herpetofauna (145 species), the Sierra de Norte de Chiapas, the number Two of these eight species (25.0%) are country endemics __ of single-region species (50) is higher than the number and the remaining six are non-endemics. Of the six non- — of the double-region species (40), and is closer to the endemic species, one is an MXSA species and the other number of species occupying all three regions (55). five are MXCA species. Of the 170 total herpetofaunal species in Tabasco, 101 In summary, of the 69 single-region species distributed (59.4%) are found in two or three physiographic regions, in Tabasco, 51 (73.9%) are non-endemics, 16 (23.2%) leaving 69 (40.6%) with a distribution in only a single are country endemics, and two (2.9%) are non-natives. region (see above). Thus, 50 of these 69 single-region Of the three physiographic regions in Tabasco, the SNC __ species are restricted to the Sierra Norte de Chiapas. is of greatest conservation importance, given that it The highest CBR value (0.67) is that between the supports the largest overall number of species (145), as | GCP and the SBP, and the lowest value (0.61) is between well as the largest numbers of single-region species (50) — the GCP and the SNC. We expected a relatively high and country-endemics (13). level of resemblance among these three areas, since the two higher-elevation regions are adjacent to the lower- We constructed a Coefficient of Biogeographic elevation region, and all three regions contain relatively Resemblance (CBR) matrix for establishing the —lowelevations (see above). herpetofaunal similarity relationships among the three The overall CBR values among the three physiographic physiographic regions in Tabasco (Table 6). The SNC _ regions are as follows, arranged from the highest to supports the highest level of species richness at 145 — lowest value (species numbers in parentheses): species, followed by 93 in the SBP, and 88 in the GCP. The mean species richness for the three regions is 108.7. GCP (88) — 0.61 — SNC (145) The numbers of shared species among all regional SBP (93) — 0.66 — SNC (145) pairs range from 61 between the GCP and the SBP to GCP (88) — 0.67 — SBP (93) Table 7. Counts of the number of species within each of the three physiographic regions in Tabasco, Mexico, which occupy one, two, or three of the physiographic regions. Physiographic Number of regions inhabited region One Two Three Total Gulf Coastal Plain 1] 22 55 88 Sierra del Norte de 50) AO 55 145 Chiapas Sierras Baja del Petén 8 30 55 93 State total 69 46 55 170 Amphib. Reptile Conserv. 18 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. pes ae SF No. 13. Bolitoglossa veracrucis Taylor, 1951. The Veracruz Salamander previously was known only from the type locality (Veracruz, Mexico), at 100 to 1,000 m elevation (https://amphibiansoftheworld.amnh.org/). In 2008, however, a population of this species was recorded for the first time in the state of Tabasco, from Cuevas de Muku Chem, in the municipality of Tacotalpa (Geronimo-Torres et al. 2022). Wilson et al. (2013b) calculated its EVS as 17, placing it in the middle portion of the high vulnerability category. Its conservation status has been considered as Endangered (EN) by IUCN, and as a species of Special Protection (Pr) by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. No. 15. Corytophanes hernandesii (Wiegmann, 1831). Hernandez’s Helmeted Basilisk occurs at low and moderate elevations on the Atlantic versant from southeastern San Luis Potosi, Mexico, to northwestern Honduras (McCranie et al. 2004). This individual was encountered in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. Wilson et al. (2013a) determined its EVS as 13, placing it at the upper limit of the medium vulnerability category. Its conservation status has not been determined by the IUCN, but this species was provided Special Protection (Pr) status by SEMARNAT. Photo by José del Carmen Geronimo-Torres. Amphib. Reptile Conserv. No. 14. Corytophanes cristatus (Merrem, 1820). The Smooth Helmeted Iguana is found at low and intermediate elevations on the Gulf and Caribbean slopes from central Veracruz to Colombia (Lee 1996). This lizard ranges from central Veracruz and the southern part of the Yucatan Peninsula in Mexico, southward on the Atlantic versant and lowlands of Central America through northern Guatemala and Belize to Costa Rica, where it occurs on both the Atlantic and Pacific slopes into northwestern Colombia (Campbell 1998). This individual was found in the municipality of Tacotalpa, Tabasco, in secondary vegetation. Its EVS has been determined as 11, placing it in the middle portion of the medium vulnerability category, and its IUCN status has been assessed as Least Concern (LC). This species was allocated to the Special Protection (Pr) category by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. No. 16. Norops barkeri Schmidt, 1939. Barker’s Anole is a semiaquatic anole endemic to southern Mexico. This species is known from states of Veracruz, Chiapas, Oaxaca, and Tabasco (Powell and Birt 2001). This individual was found in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. This lizard’s EVS has been assessed as 15, placing it in the lower portion of the high vulnerability category (Wilson et al. 2013a). Its IUCN status has been determined as Vulnerable (VU), and it is considered a species of Special Protection (Pr) by SEMARNAT. Photo by Jenny del C.-Estrada- Montiel. August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico SBP GCP SNC 1.00 90 80 10 ss .64 60 0 AO 30 .20 10 OO Fig. 5. UPGMA-generated dendrogram illustrating the simi- larity relationships of species richness among the herpetofau- nal components in the three physiographic regions of Tabasco (based on the data in Table 6; Sokal and Michener 1958). Simi- larity values were calculated using Duellman’s (1990) Coef- ficient of Biogeographic Resemblance (CBR). Based on the data in Table 6, we created a UPGMA dendrogram (Fig. 5) to demonstrate the herpetofaunal resemblance patterns among the three physiographic regions in Tabasco (see map, Fig. 1). The dendrogram indicates that the CBP and GCP cluster at the 0.67 level and the SNC clusters to the previous pair at the 0.64 level. This overall pattern indicates that all three regions are closely aligned together at an intermediate level of resemblance. Distribution Status Categorizations We utilized the same system as Alvarado-Diaz et al. (2013) to examine the distribution status of members of the Tabasco herpetofauna, and this system has been used in all the subsequent entries in the MCS (see above). The categories in this system are non-endemic, country endemic, state endemic (of which none occur in Tabasco), and non-native. These categorizations are listed in Table 8 and summarized in Table 9. The numbers of species in each of the three applicable categories, in decreasing order of size, are as follows: non-endemics, 145 (85.3% of total of 170 species): country endemics, 20 (11.8%); and non-natives, five (2.9%). As with the states of Oaxaca (Mata-Silva et al. 2015), Chiapas (Johnson et al. 2015a), Tamaulipas (Teran-Juarez et al. 2016), Nuevo Leon (Nevarez-de los Reyes et al. 2016), Coahuila (Lazcano et al. 2019), and Veracruz (Torres-Hernandez et al. 2021), as well as the tri- state Yucatan Peninsula (Gonzalez-Sanchez et al. 2017), Amphib. Reptile Conserv. most of the herpetofaunal taxa in Tabasco fall within the non-endemic category. In the other six states evaluated in the Mexican Conservation Series, the largest number falls within the country endemic category: Michoacan (Alvarado-Diaz et al. 2013); Nayarit (Woolrich-Pifia et al. 2016); Jalisco (Cruz-Saenz et al. 2017); Puebla (Woolrich-Pifia et al. 2017); Hidalgo (Ramirez-Bautista et al. 2020); and Querétaro (Cruz-Elizalde et al. 2022). Twenty country endemic species are present in Tabasco, and perhaps this relatively low number was expected because the state lies largely on the Gulf Coastal Plain and adjacent to relatively low-elevation areas, which generally are not known for significant herpetofaunal endemism. No state endemic species occur in Tabasco. In the 13 previous entries in the MCS (including the Oaxaca update; Mata-Silva et al. 2021), the number of state endemic species ranges from one in Nayarit and Nuevo Leon (Woolrich-Pifia et al. 2016; Nevarez-de los Reyes 2016) to 105 in Oaxaca (Mata- Silva et al. 2021). Five non-native species have been recorded from Tabasco, including Eleutherodactylus _ planirostris, Norops sagrei, Hemidactylus frenatus, H. turcicus, and Virgotyphlops braminus. Two of these five species (7. frenatus and V. braminus) are the most widespread of the non-native species recorded in the 13 entries in the MCS (Cruz-Elizalde et al. 2022), and to date they have been reported in 13 states or tri-state regions. Wilson et al. (2017) introduced a system for the distributional categorization of the Mesoamerican herpetofauna. The data for the categories applicable to this work are summarized in Table 10. Previously, we noted that 145 species are non-endemic to Tabasco, and we allocated them to six of the nine categories developed by Wilson et al. (2017), including MXUS, MXCA, MXSA, USCA, USSA, and OCEA. As expected, the greatest number and proportion of species fall into the MKCA category (95, or 65.5%), given the proximity of Tabasco to Central America and since a significant portion of its eastern border is shared with Guatemala. Interestingly, the next largest number and proportion of species are allocated to the MXSA category (34, or 23.4%). Oddly, only a single species (0.7%) is assigned to the MXUS category. By way of comparison, this category contains 29 species, or 17.2%, in the herpetofauna of the adjacent state to the west (1.e., Veracruz; Torres-Hernandez et al. 2021). The remaining 15 species are in the USCA (eight, or 5.5%), USSA (four, or 2.7%), and OCEA (three, or 2.1%) categories. Principal Environmental Threats Deforestation Deforestation in southeastern Mexico is a serious matter that has worsened over time, and the state of Tabasco is no exception (Fig. 6). The continuous loss of vegetational August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. No. 17. Norops compressicauda Smith and Kerster, 1955. The Malposo Scaly Anole is endemic to Mexico. This anole has been reported from the states of Oaxaca, Veracruz, and Chiapas. Here we present the first records of this species from the state of Tabasco, from the municipalities of Teapa and Tacotalpa, in montane areas at elevations from 100 to 700 m (Rios Rodas et al. 2017). Wilson et al. (2013a) calculated its EVS as 15, placing it in the lower portion of the high vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, but this species is not listed by SEMARNAT. Photo by Liliana Rios-Rodas. a Geer ea y No. 19. Ctenosaura similis (Gray, 1831). The Common Spiny-tailed Iguana occurs at low and moderate elevations from southern Veracruz and Oaxaca southward to Panama (Lee 1996). The native range of this species extends along the Atlantic versant from the Isthmus of Tehuantepec southeastward to northeastern Nicaragua, and on the Pacific versant from the Isthmus of Tehuantepec southeastward to Panama (Kohler 2003). This individual was found in rainforest at an elevation of 200 m, in the municipality of Tenosique, Tabasco. Wilson et al. (2013a) calculated its EVS as 8, placing it in the upper portion of the low vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN. This species was allocated to the Threatened (A) category by SEMARNAT. Photo by Maria del Rosario Barragan-Vazquez. Amphib. Reptile Conserv. 21 No. 18. Coleonyx elegans Gray, 1845. The Yucatan Banded Gecko is distributed on the Gulf and Pacific slopes in the states of the southeastern region of Mexico. In Tabasco, this species has been recorded in the municipalities of Tacotalpa, Huimanguillo, and Teapa. This individual is from Muku Chem, in the municipality of Tacotalpa, Tabasco. Its EVS has been determined as 9, placing it at the upper limit of the low vulnerability category (Wilson et al. 2013a). Its IUCN status has been assessed as Least Concern (LC), and as Threatened (A) by SEMARNAT. Photo by Manuel Herndndez-May. — S ee _ No. 20. Sceloporus teapensis Ginther, 1890. The Teapen Rosebellied Lizard occurs at low elevations on the Atlantic slopes from southern Veracruz and Oaxaca, eastward through Chiapas, Tabasco, and Campeche, and through the Petén region of Guatemala to Belize, and south to Coban, Alta Verapaz, Guatemala (Lee 1996). This individual was encountered in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. Wilson et al. (2013a) determined its EVS as 13, placing it at the upper limit of the medium vulnerability category. Its conservation status has been assessed as Least Concern (LC) by the IUCN, but has not been determined by SEMARNAT. Photo by Jenny del Carmen Estrada- Montiel. August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Fig. 6. Deforestation due to road construction in the vicinity of Paraiso, Tabasco. Photo by José del Carmen Geronimo- Torres. cover in tropical forest has been precipitated primarily by a shift in land use for livestock and agricultural activities, lumber extraction, an increasing number of roads, oil production-related activities, and the direct effect of human population growth (Maldonado-Sanchez et al. 2016). To date, Tabasco has lost more than 90% of its original vegetational cover (Zavala-Cruz and Castillo 2003), and more recent data indicate that approximately only 2% of this vegetation remains (Sanchez-Munguia 2005). In the basins of the Grijalva and Usumacinta rivers, the tropical forest cover has been reduced from 36% in 1993 to only 9% in 2007 (Kolb and Galicia 2012). The above numbers reveal the somber panorama that Tabasco currently faces, which directly affects the prospects for conserving biodiversity, including its herpetofauna. The last remnants of forest in Tabasco are distributed mostly in the municipalities of Teapa, Tenosique, Huimanguillo, and Macuspana (Castillo and Zavala 1996); ironically, these are the same geographic entities where various extension records of amphibians and reptiles have been reported in recent years. Species such as the Northern Glass Frog (Hyalinobatrachium viridissimum), the Chiapan Highlands Treefrog (Exerodonta bivocata), the Smooth-headed Helmeted Basilisk, locally known as Turipache (Corytophanes cristatus), Barker’s Anole (Norops barkeri), and the Keeled Earth Snake (Geophis carinosus) are just a few worthy of mention. These records highlight the need for continuous and urgent exploration, especially in areas that still contain tropical forest. Agricultural Activities As mentioned earlier, one of the main drivers of deforestation is farming (Fig. 7) and livestock activities. In this regard, Alejandro-Montiel et al. (2010) stated that these activities are responsible for 94% of the land change that has taken place in Tabasco. Noteworthy agricultural policies for Tabasco were developed in the 1960s and 1970s (Plan Chontalpa and Plan Balancan- Amphib. Reptile Conserv. 22 Fig. 7. Conversion of land use for agricultural purposes in the community of Villa Luz, in the municipality of Tacotalpa, Tabasco. Photo by Liliana Rios-Rodas. Tenosique), and have affected more than 200,000 ha, resulting in the complete elimination of evergreen tropical forest and the desiccation of wetlands for the later development of urban communities (Barkin 1978). For example, immediately after the completion of one of these projects, a subsequent study revealed that only 8% of the forests remained in the municipalities of Balancan and Tenosique, which increased flooding and soil erosion in those areas (Tudela 1989; Torres-Masuera 2021). These programs did not have the promised results, but on the contrary were responsible for the loss of forest and biodiversity that have not recovered thus far. This infamous action was never reported; therefore, there are no actual numbers that can reveal the specific amount of biodiversity affected. Currently, Tabasco dedicates more than 253,000 ha to the cultivation of banana, sugarcane, cocoa, corn, and oil palm. Unfortunately, these large-scale crops are damaging to the remaining natural ecosystems in the state, whose effects are exacerbated by the large amount of associated chemicals. The municipalities of Huimanguillo and Balancan have the largest amount of land used for cultivation, and Huimanguillo also has the largest livestock production (Infografia Agroalimentaria 2017). Atthe same time, the municipality of Huimanguillo contains remnants of evergreen tropical forest where additional species have been reported in recent times, expanding their geographic distributions. The current and historical situation regarding the development of agriculture in the state also indicates the continuous damage inflicted on natural ecosystems and, therefore, all of the species they harbor. Roads Roads represent an important contributor to the intensification of productivity in communities, and simultaneously are an instrumental component for social, economic, and cultural integration. According to INEGI (2009), Tabasco has an extensive system of roads, and is August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. No. 21. Sphaerodactylus continentalis Werner, 1896. The Upper Central American Geckolet occurs at “low and moderate elevations from the Isthmus of Tehuantepec in northern Oaxaca, Mexico, to about the Catacamas, Olancho, region of east-central Honduras;” this species “also occurs on Utila Island in the Honduran Bay Islands and possibly on Cozumel Island, Quintana Roo, Mexico” (McCranie and Hedges 2012). This individual is from Muku Chem, in the municipality of Tacotalpa, Tabasco. Mata-Silva et al. (2021) determined its EVS as 10, placing it at the lower limit of the medium vulnerability category. Its conservation status has been evaluated as Least Concern (LC) by the IUCN, but it has not been assessed by SEMARNAT. Photo by Liliana Rios- Rodas. ee Ab si ares No. 23. Lepidophyma flavimaculatum Dumeril, 1851. The Yellow-spotted Night Lizard is found at low and moderate elevations on the Atlantic slope from Veracruz eastward through northern Guatemala, Belize, and northern Honduras. In the Yucatan Peninsula it is known from northeastern Chiapas, El Petén, Belize, and southern Quintana Roo (Lee 1996). This individual was located in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. Wilson et al. (2013a) assessed its EVS as 8, placing it in the upper portion of the low vulnerability category. Its conservation status has been evaluated as Least Concern (LC) by the IUCN, and this species was placed in the Special Protection (Pr) category by SEMARNAT. Photo by José del Carmen Gerénimo-Torres. Amphib. Reptile Conserv. No. 22. Holcosus stuarti Smith, 1940. The Rainbow Ameiva occurs on the “Atlantic slopes of Mexico from the middle of the Isthmus of Tehuantepec eastward in the lowlands to the southern borders of Laguna de Términos and to Tenosique, Tabasco; southward up the valley of the Rio Grijalva at least as far as Tuxtla Gutiérrez, Chiapas” (Meza-Lazaro and Nieto- Montes de Oca 2015). This individual was located in the Ejido Villa Guadalupe of Huimanguillo, Tabasco. Wilson et al. (2013a) determined its EVS as 7, placing it in the middle limit of the low vulnerability category. Its conservation status has been evaluated as Least Concern (LC) by the IUCN, but this species is not listed by SEMARNAT. Photo by Jenny del Carmen Estrada- Montiel. No. 24. Leptophis mexicanus Duméril, Bibron, and Dumeéril, 1854. The Mexican Parrot Snake is distributed in southeastern Mexico, including Chiapas, Veracruz, Oaxaca, Tabasco, Yucatan, Campeche, San Luis Potosi, Querétaro, Tamaulipas, Puebla, Hidalgo, Nuevo Leon, Guerrero, and Yucatan Peninsula, into Guatemala, Honduras, Belize, El Salvador, Nicaragua, and Costa Rica. In Guatemala it occurs from near sea level to about 1,360 m in elevation (Lee 1996; Campbell 1998). This individual was found in the municipality of Tacotalpa, Tabasco, in secondary vegetation (acahual). Its EVS has been determined as 6, placing it in the middle portion of the low vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN and it is allocated to the Threatened (A) category by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Fig. 8. Roads. A Boa imperator dead on the road in the Pantanos de Centla Biosphere Reserve, in the municipality of the same name, Tabasco. Photo by Coral J. Pacheco-Figueroa. among the three best-served states in the country, with an index of 248 m/km? (Vidal-Garcia and Negrete 2019). The construction and functioning of roads are elements that have drastic and long-term consequences on the natural landscape, as they significantly affect the survival of the native herpetofauna (Trombulak and Frissell 2000; Coffin 2007). The building and maintenance of roads implies the unavoidable removal of native vegetation cover, thus enabling a series of linked processes that ultimately lead to the detriment of adjacent habitats. With respect to the mortality of fauna on roads (Fig. 8), Pozo-Montuy et al. (2019) reported 111 individuals killed by vehicles on the road from Villahermosa to Zacatal; of those, 22.5% were reptiles and 20.7% amphibians. The species affected more frequently were in the families Iguanidae, Boidae, Colubridae, Viperidae, and Geoemydidae; more specifically for lizards they included Green Iguanas ([guana rhinolopha) and Black Iguanas (Ctenosaura similis) (Canales-Delgadillo et al. 2020). Other studies carried out in the state reported the killing of 1. rhinolopha on highway 186 (Villahermosa- Aeropuerto) and the Cane Toad (Rhinella horribilis) on the Tabascan plains. Lastly, a survey conducted at Reserva de la Bidsfera Pantanos de Centla showed that 43% of the road-kills were amphibians, primarily Brown’s Leopard Frog (Lithobates brownorum) (Pacheco-Figueroa 2021). Soil Pollution and Ojil-related Activities In Tabasco, the municipalities with the highest numbers of oil spills that affected numerous hectares of land from 1995 until 2001 were Cardenas, Huimanguillo, Cunduacan, and Comalcalco (Ochoa-Gaona et al. 2011). The long history of oil spills and gas explosions in Tabasco (Fig. 9) has led to serious consequences in many communities, because this activity also resulted in the pollution of soils and vegetation such as grasslands (Zavala-Cruz et al. 2005). Some studies have identified approximately 7,500 ha that are affected, more than 90% of which are located in wetlands (Adams-Schroeder Amphib. Reptile Conserv. 24 Fig. 9. Deforestation due to oil activities in the vicinity of Paraiso, Tabasco. Photo by José del Carmen Geronimo- Torres. 1999; Beltran-Paz 2006). It was estimated that 0.07% of the state was polluted with fossil fuels (Rivera-Cruz and Trujillo-Narcia 2004; Ferrera-Cerrato et al. 2006). All amphibian groups found in Tabasco have been affected by the oil industry, although a study by Reynoso-Rosales (1999) in southeastern Mexico, including Tabasco, determined that the detriment to amphibians ts the result of a combined effect from both farming and the oil industry. With respect to the latter activity, this includes consecutive processes such as exploration, perforation, and production. Among the direct effects from the oil industry are the disturbances caused by permanent light sources at all installations, which likely affect the behavior of species present around these industrial facilities. For instance, toads (Rhinella and Incilius) congregate at light sources to search for food. Myths and Cultural Factors With respect to the herpetofauna, ethnozoological knowledge includes symbolic, spiritual, and social meanings in indigenous societies (Avila-Najera et al. 2018), although few studies have addressed this subject in Tabasco. Among the most frequent uses of native herpetofauna are for food (iguanas, turtles, and crocodiles, Hernandez-Lopez et al. 2012) and magic- religious uses in conjunction with medicinal application. For example, rattlesnakes (Crotalus) are used to treat cancer, diabetes, acne, and skin health issues (GOmez- Alvarez and Pacheco 2010). On the other hand, snakes generally are considered as dangerous, and _ their encounters usually result in their immediate elimination. A similar situation 1s experienced by amphibians, which are considered mostly as undesirable. People in Tabasco have consumed native terrestrial vertebrates for millennia, primarily reptiles, birds, and mammals as food, as well as for skins, pets, and medicinal purposes (Pozo-Montuy et al. 2019). To date, 16 species of reptiles have been identified as traditionally consumed in Tabasco, such as iguanas, turtles, snakes, August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. © Marco A. Torrez Pérez No. 25. Oxybelis potosiensis (Taylor, 1941). The Gulf Coast Vine Snake is distributed from San Luis Potosi and northern Veracruz, southward to Yucatan, Mexico, and Belize (Jadin et al. 2020). This individual was found in the municipality of Huimanguillo, Tabasco. Its EVS has been determined as 5 (Cruz-Elizalde et al. 2022), placing it in the lower portion of the low vulnerability category. Its conservation status has not been evaluated (NE) by the IUCN, and it is considered as having No Status (NS) by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. ae = = aa j ©Maseo A. TorreZ érez No. 26. Coniophanes imperialis (Baird and Girard, 1859). The Black-striped Snake occurs at low and moderate elevations on the Atlantic slope from southern Texas southward on the Atlantic watershed through eastern Mexico, Yucatan, Belize, and northern and eastern Guatemala to Honduras; it also occurs locally on Pacific slopes in Oaxaca, Chiapas, Yucatan, Campeche, and Quintana Roo (Lee 1996; Campbell 1998). This individual was found in the municipality of Huimanguillo, Tabasco. Its EVS has been determined as 8, placing it in the upper portion of the low vulnerability category. Its conservation status has been established as Least Concern (LC) by the IUCN, but has been assigned No Status (NS) by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. No. 27. Enulius flavitorques (Cope, 1868). The Pacific Long- tailed Snake occurs at low and moderate elevations on the Pacific versant from Jalisco, Mexico, to Panama, and on the Atlantic versant in Chiapas, Mexico, Honduras (including Isla Utila in the Islas de la Bahia), Panama, northern Colombia, and northwestern Venezuela (Hernandez- Valadez et al. 2016). This individual was found in a coconut plantation in Playa Chiltepec, in the municipality of Paraiso, Tabasco. Its EVS has been determined as 5, placing it in the lower portion of the low vulnerability category. Its conservation status has been evaluated as Least Concern (LC) by the IUCN, but its status remains undetermined (NS) by SEMARNAT. Photo by Marco Antonio Lopez-Luna. Amphib. Reptile Conserv. No. 28. Jmantodes cenchoa (Linnaeus, 1758). The Neotropical Blunt-headed Treesnake occurs at low and moderate elevations in Mexico, from Chiapas on the Pacific slope and Tamaulipas on the Atlantic slope, southward throughout most of the Petén region in Guatemala and the northeastern portion of Yucatan Peninsula, through the remainder of Central America to Argentina and Paraguay (Lee 1996; Campbell 1998). This individual was found in Muku Chem, in the municipality of Tacotalpa, Tabasco. Wilson et al. (2013a) determined its EVS as 6, placing it in the middle portion of the low vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, and as a species of Special Protection (Pr) by SEMARNAT. Photo by José del Carmen Geronimo-Torres. August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 8. Distributional and conservation status measures for members of the herpetofauna of Tabasco, Mexico. Distributional status: CE = endemic to country of Mexico; NE = not endemic to state or country; and NN = non-native. The numbers suffixed to the “NE” category signify the distributional categories developed by Wilson et al. (2017) and implemented in the taxonomic list at the Mesoamerican Herpetology website (http://mesoamericanherpetology.com), as follows: 3 (species distributed only in Mexico and the USA); 6 (species ranging from Mexico to South America); 7 (species ranging from the USA to Central America); and 8 (species ranging from the USA to South America). Environmental Vulnerability Score (taken from Wilson et al. 2013a,b): low (L) vulnerability species (EVS of 3-9); medium (M) vulnerability species (EVS of 10—13); and high (H) vulnerability species (EVS of 14-20). IUCN Categorization: CR = Critically Endangered; EN = Endangered; VU = Vulnerable; NT = Near Threatened; LC = Least Concern; DD = Data Deficient; NE = Not Evaluated. SEMARNAT Status: A = Threatened; P = Endangered; Pr = Special Protection; and NS = No Status. See Alvarado-Diaz et al. (2013), Johnson et al. (2015a), and Mata-Silva et al. (2015) for explanations of the EVS, IUCN, and SEMARNAT rating systems. Distributional | Environmental IUCN SEMARNAT Taxa Vulnerability ce Status categorization status Category (score) Amphib. Reptile Conserv. 26 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 8 (continued). Distributional and conservation status measures for members of the herpetofauna of Tabasco, Mexico. Distributional status: CE = endemic to country of Mexico; NE = not endemic to state or country; and NN = non-native. The numbers suffixed to the “NE” category signify the distributional categories developed by Wilson et al. (2017) and implemented in the taxonomic list at the Mesoamerican Herpetology website (http://mesoamericanherpetology.com), as follows: 3 (species distributed only in Mexico and the USA); 6 (species ranging from Mexico to South America); 7 (species ranging from the USA to Central America); and 8 (species ranging from the USA to South America). Environmental Vulnerability Score (taken from Wilson et al. 2013a,b): low (L) vulnerability species (EVS of 3—9); medium (M) vulnerability species (EVS of 10-13); and high (H) vulnerability species (EVS of 14-20). IUCN Categorization: CR = Critically Endangered; EN = Endangered; VU = Vulnerable; NT = Near Threatened; LC = Least Concern; DD = Data Deficient; NE = Not Evaluated. SEMARNAT Status: A = Threatened; P = Endangered; Pr = Special Protection; and NS = No Status. See Alvarado-Diaz et al. (2013), Johnson et al. (2015a), and Mata-Silva et al. (2015) for explanations of the EVS, IUCN, and SEMARNAT rating systems. Distributional. | E2vironmental IUCN SEMARNAT Taxa Vulnerability ahee status categorization status Category (score) Amphib. Reptile Conserv. 27 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 8 (continued). Distributional and conservation status measures for members of the herpetofauna of Tabasco, Mexico. Distributional status: CE = endemic to country of Mexico; NE = not endemic to state or country; and NN = non-native. The numbers suffixed to the “NE” category signify the distributional categories developed by Wilson et al. (2017) and implemented in the taxonomic list at the Mesoamerican Herpetology website (http://mesoamericanherpetology.com), as follows: 3 (species distributed only in Mexico and the USA); 6 (species ranging from Mexico to South America); 7 (species ranging from the USA to Central America); and 8 (species ranging from the USA to South America). Environmental Vulnerability Score (taken from Wilson et al. 2013a,b): low (L) vulnerability species (EVS of 3—9); medium (M) vulnerability species (EVS of 10-13); and high (H) vulnerability species (EVS of 14-20). IUCN Categorization: CR = Critically Endangered; EN = Endangered; VU = Vulnerable; NT = Near Threatened; LC = Least Concern; DD = Data Deficient; NE = Not Evaluated. SEMARNAT Status: A = Threatened; P = Endangered; Pr = Special Protection; and NS = No Status. See Alvarado-Diaz et al. (2013), Johnson et al. (2015a), and Mata-Silva et al. (2015) for explanations of the EVS, IUCN, and SEMARNAT rating systems. Distributional. | Environmental IUCN SEMARNAT Taxa Vulnerability aoe status categorization status Category (score) Amphib. Reptile Conserv. 28 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 8 (continued). Distributional and conservation status measures for members of the herpetofauna of Tabasco, Mexico. Distributional status: CE = endemic to country of Mexico; NE = not endemic to state or country; and NN = non-native. The numbers suffixed to the “NE” category signify the distributional categories developed by Wilson et al. (2017) and implemented in the taxonomic list at the Mesoamerican Herpetology website (http://mesoamericanherpetology.com), as follows: 3 (species distributed only in Mexico and the USA); 6 (species ranging from Mexico to South America); 7 (species ranging from the USA to Central America); and 8 (species ranging from the USA to South America). Environmental Vulnerability Score (taken from Wilson et al. 2013a,b): low (L) vulnerability species (EVS of 3—9); medium (M) vulnerability species (EVS of 10-13); and high (H) vulnerability species (EVS of 14-20). IUCN Categorization: CR = Critically Endangered; EN = Endangered; VU = Vulnerable; NT = Near Threatened; LC = Least Concern; DD = Data Deficient; NE = Not Evaluated. SEMARNAT Status: A = Threatened; P = Endangered; Pr = Special Protection; and NS = No Status. See Alvarado-Diaz et al. (2013), Johnson et al. (2015a), and Mata-Silva et al. (2015) for explanations of the EVS, IUCN, and SEMARNAT rating systems. Distributional | Environmental IUCN SEMARNAT Vulnerability Aa status categorization status Category (score) NS [wes mc) COdOCitSCS” NEA NEA NE6 NEA NEA NET Amphib. Reptile Conserv. 29 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico — No. 29. Ninia sebae (Duméril, Bibron, and Duméril, 1854). The Redback Coffee Snake occurs at low and moderate elevations on the Atlantic slope from Veracruz and the Pacific slope from Oaxaca, Mexico, southeastward and eastward through Central America to Costa Rica (Lee 1996). In Panama it has been recorded in Changuinola district (Ponce et al. 2008). In Guatemala it ranges from near sea level to about 2,000 m in elevation (Campbell 1998). This individual was found in a cornfield in the municipality of Tenosique, Tabasco. Its EVS has been determined as 5 (Wilson et al. 2013a), placing it in the lower portion of the low vulnerability category. Its conservation status has been assessed as Least Concern (LC) by the IUCN, but as Not Evaluated (NS) by SEMARNAT. Photo by Maria del Rosario Barragan-Vazquez. No. 31. Micrurus elegans Jan, 1858. The Elegant Coral Snake is distributed from Mexico to southwestern Guatemala. In Mexico it has been reported from the states of Chiapas, Oaxaca, Puebla, Veracruz, and in the municipality of Teapa, Tabasco (Soto-Huerta and Clause 2017). This species ranges from 100 to 1,700 m in elevation. This individual was found in the municipality of Tacotalpa, Tabasco. Its EVS has been determined as 13 (Torres-Hernandez et al. 2021), placing it at the upper limit of the medium vulnerability category. Its conservation status has been considered as Least Concern (LC) by the IUCN, but it is considered a species of Special Protection (Pr) by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. Amphib. Reptile Conserv. a ~ a J ‘ ~~ ae aot wail Be SEF sa No. 30. Sibon dimidiatus (Gunther, 1872). The Slender Snail Sucker occurs at low, moderate, and intermediate elevations on the Pacific slope of Guatemala, and in premontane areas from northern Veracruz, Mexico, southward through Central America to northern Brazil, Colombia, Peru, Venezuela, Guyana, and Ecuador west of the Andes (Lee 1996; Espinal et al. 2021). This individual is from Muku Chem, in the municipality of Tacotalpa, Tabasco. Wilson et al. (2013a) determined its EVS as 10, placing it at the lower limit of the medium vulnerability category. Its conservation status has been assessed as Least Concern by the IUCN, but it has not been assessed by SEMARNAT. Photo by Marco Antonio Torrez-Pérez. © Marco A. Torrez Pérez Cal No. 32. Epictia phenops (Cope, 1875). The distribution of the Slender Threadsnake extends “from southern Mexico to western Honduras” (Wallach 2016: 254). This individual was found in the city of Villahermosa. The EVS of this blindsnake has been calculated as 4 (Mata-Silva et al. 2021), placing it in the lower portion of the low vulnerability category. The conservation status of this species has not been assessed by either the IUCN or SEMARNAT. Photo by Marco Antonio Torrez-Pérez. August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 9. Summary of the distributional status data for herpetofaunal families in Tabasco, Mexico. Distributional status Number of species Non-endemic (NE) oe ae Non-native (NN) fF Bifonidie | as | a eS | CCentrolenidae | CC | —sCrangastoridae | C‘iEC“ M4 8 | _Eleutherodactylidae | 2 ed Hylidae | CSubtotal | 8 8 8 (SBP lcthodont ass S| ee ee | CSubtotal | SC“<C~d 8 | Dermophiidae TE | Subtotal aT eee LOSS SS EE EEE Ee a ae SS) | Crocodylidae PE | Subtotal | Corytophanidae | 4 4 | Dactyloidae a | —SDiploglossidae | CC CE | —sEublepharidae | | Gekkonidae | SSS an se ST EES ee | Mabuyidae Td | __Phrynosomatidae | C“C~*dC“($R USC | Phyllodactylidae | ECE ee Te oSeicidae To a SE ee | _Sphaerodactylidae | Ci | Sphenomorphidae | t—<C~ss Ss a a a a ee) eer |e ExXaniiiidaes ee | Xenosauridae PE Sabicea PE Ee | Sea Ee ese se I = 0c: nD ee ME Echidne UE | CdDipsadidae | Ci CD ee eee Ne Sa | _Leptotyphlopidae | Ce ee aaa | (0: | ——sSSibynophiidae | ——— ih prlopidas —— | | | |__| ho Vipera | CSubtotal |G eT rnd CHeloniidie. car | ros ie ee eee | Chelydridae Te | Ss Dermatemyidae | CE Emydidae __——— | | CSubtotal Te SS a ee: |S | an: | 7a i, | |i: | iil | CC Sum Total] iO eT HS OO te NIN]; eyo pm fr | HPO fHPOPN[ofefutefolryo]—|—|R]aTE]y BSleEfPofelrmolrmo]—e[nfelr ott Leet LO) tl Rot BNO) } Amphib. Reptile Conserv. 31 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico BA ons ~ i Jie ne ~ x ~ on pa ey. : No. 33. Thamnophis marcianus (Baird and Girard, 1853). The Checkered Garter Snake occurs at low and moderate elevations throughout the southwestern United States and northern Mexico, and on the Pacific slope of the Isthmus of Tehuantepec. On the Atlantic slope, it ranges from northern Chiapas and eastern Tabasco through the Yucatan Peninsula and southward to Costa Rica (Lee 1996). This individual was found in the municipality of Huimanguillo, Tabasco. Its EVS has been determined as 10 (Wilson et al. 2013a), placing it at the lower limit of the medium vulnerability category. Its conservation status has been assessed as Least Concern (LC) by the IUCN, and it has been allocated to the Threatened (A) category by SEMARNAT. Photo by Marco Antonio Torrez- Pérez. No. 34. Agkistrodon russeolus Gloyd, 1972. The distribution of the Mexican Moccasin primarily extends along the outer part of the Yucatan Peninsula, from west-central Campeche and the northern portion of Yucatan and Quintana Roo on the Gulf side, and in northern Belize on the Caribbean side, although isolated records are available from extreme southeastern Campeche and central Petén, Guatemala (Porras et al. 2013). This individual was found at Nuevo Pochote, in the municipality of Emiliano Zapata, Tabasco (Charruau et al. 2014). Its EVS has been determined as 15 (Porras et al. 2013), placing it in the lower portion of the high vulnerability category (Gonzalez-Sanchez et al. 2017). Its conservation status has been evaluated as Near Threatened (NT) by the IUCN and it has been allocated to the Special Protection (Pr) category by SEMARNAT. Photo by Marco Antonio Lopez- Luna. No. 35. Crotalus tzabcan Klauber, 1952. The Yucatan Neotropical Rattlesnake occurs in the Yucatan Peninsula, including Campeche, northeastern Chiapas, Quintana Roo, Tabasco, and Yucatan, México, northern Belize and El Petén, Guatemala (Lee 1996; Campbell 1998; Campbell and Lamar 2004). This individual was found in the village of El Triunfo in the municipality of Balancan, Tabasco. Its EVS has been determined as 16 (Gonzalez-Sanchez et al. 2017), placing it in the middle portion of the high vulnerability category. Its conservation status has been designated as Least Concern (LC) by the IUCN, but as No Status (NS) by SEMARNAT. Photo by Marco Antonio Lopez-Luna. Amphib. Reptile Conserv. No. 36. Metlapilcoatlus mexicanus (Dumeéril, Bibron, and Dumeril, 1854). The Central American Jumping Pitviper occurs at low, moderate, and intermediate elevations on the Atlantic slope “from southern Mexico through Central America south to Costa Rica and Panama, where it is also found on the Pacific versant” (Heimes 2016). This individual was found in the municipality of Tacotalpa, Tabasco, in secondary vegetation. Its EVS has been determined as 12 (Wilson et al. 2013a), placing it in the upper portion of the medium vulnerability category. Its conservation status has been assessed as Least Concern (LC) by the IUCN and it is allocated to the Threatened (A) category by SEMARNAT. Photo by Marco Antonio Lopez-Luna. August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 10. Summary of the distributional categories of the herpetofaunal families in Tabasco, Mexico, containing non-endemic species. The categorizations are as follows: MXUS, species distributed only in Mexico and the United States (except perhaps for a few also found in Canada); MXCA (species found only in Mexico and Central America); MXSA (species ranging from Mexico to South America); USCA (species ranging from the United States to Central America (except perhaps for a few also found in the Antilles); USSA (species ranging from the United States to South America); and OCEA (oceanic species). Number of Distributional status Family non-endemic | = MXU MXCA MXSA USCA USSA OCEA species species 3) species (4) | species (6) species (7) | species (8) | species (9) | ° - ° no a el et i ——— | Phyllomedusidae | 2 | Oo | 2 Ranidaes ee | Rhinophrynidae | | CO | Subtotal | 30 | | Plethodontidae | 3 | 8 Subtotal ees ee ee | Dermophiidae | [Subtotal | | Total | HA | Crocodylidae_ | 2 | COT |Subtotal | 2 CO | Corytophanidae |] 4 | |Dactyloidae | | | Diploglossidae | | | Eublepharidae | | |Iguanidae | | CT Re a Ss ee ee | Phrynosomatidae | S| ST | Phyllodactylidae | | |Scincidae | | CC | Sphaerodactylidae | 2 | SO | 2 | Sphenomorphidae_ | 1 | |Teiidae | |Xantusiidae | EE (uimne2 ceo | OTR | NW] ete RINE MIN Xenosauridae Subtotal Boidae Colubridae Dipsadidae Elapidae Leptotyphlopidae Natricidae a Typhlopidae Viperidae |Subtotal | | |Cheloniidae [| 2 CT |Chelydridae | Dermatemyidae |] | | Dermochelyidae | | semyidae S|) eS |e |Geoemydidae | tt | Kinosternidae | 3 | Staurotypidae |] 2 | [Subtotal | | [Total A § | a |Sumtotal | 45 TS Amphib. Reptile Conserv. 33 August 2022 | Volume 16 | Number 2 | e315 ae =a ps — 1 — os a a —S— i — ee ee Ol Pe Pt Oe _ Se ee ee en a | 36 Wye tRelRe|Re|TRe [Nb The herpetofauna of Tabasco, Mexico = J » ae 3 a oe “ oe - ae 2 fe “ Cae - - ie Me “a i % " ' Las Te at Ray ae ‘ = ‘all A isto pes | Sab af chen © = aed Fig. 10. The human consumption of meat from iguanid lizards of the genus Crenosaura documented in the municipality of Paraiso, Tabasco. Photo by Liliana Rios-Rodas. and crocodiles. Among the turtles, the species most affected are the MHicotea (Trachemys_ venusta), Pochitoque Tres Lomos (Kinosternon scorpioides), Pochitoque Jahuactero (Kinosternon acutum), and Chiquiguao (Chelydra rossignonii). Furthermore, the Lagarto or Cocodrilo (Crocodylus moreletii) and the Iguana Verde (/guana rhinolopha) also are in frequent demand (Pozo-Montuy et al. 2019). The consumption of these species varies according to the region and season of the year. For instance, the consumption of freshwater turtles is a tradition for numerous Tabascan families during Lent. Illegal Trade Unfortunately, the illegal trafficking of reptile species in Tabasco is a common activity due to the high demand for meat (iguanas and crocodiles) and turtle eggs (Figs. 10-11). More specifically, many turtles in the state have been part of the Tabascan gastronomy (Guevara-Chumacero et al. 2017). Among _ turtles, people primarily eat Dermatemys mawii due to its size and meat quality, and consequently this consumption has pushed the species to near extinction (Zenteno-Ruiz et al. 2004). Although this species is consumed mostly in local communities, the species also 1s sold outside its distributional range, with prices varying according to the time of year (Guichard-Romero 2006). The crocodile (C. moreletii) is desired for its fat, since local communities use it for treating asthma. Furthermore, all of the species reported above often are purchased by people to keep as pets in tanks within their homes. With regard to amphibians, individuals of the treefrog Agalychnis taylori are sold as pets due to their attractive coloration, and often are advertised on websites by people lacking legal documentation. A similar situation is happening with the Central American Boa (Boa imperator), of which individuals usually are kept as pets, but also are sacrificed for their skin. Amphib. Reptile Conserv. - 4 * a £ vy : =, = —6mw©w ry me = Itty 7 ~ » a Ae _— : a ' | = " = \ - — ¥ e _—c : , 2 a ok. Ww. -3 a ee i } a 4 f ae “ . ~~ : .. : af wt y « ¥ 2 4 Fig. 11. Illegal trafficking of turtle species in the municipality of Centla, Tabasco. Photo by Liliana Rios-Rodas. Fig. 12. Forest fires caused by agricultural activities, Laguna San Isidro, Reserva de la Biosfera Pantanos de Centla, Tabasco. Photo by Marco Antonio Torrez-Pérez. August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. No. 37. Dermatemys mawii (Gray, 1847). The Central American River Turtle occurs in the Caribbean lowlands of southern Mexico from central Veracruz southeastward through the southern portion of the Yucatan Peninsula (Campbell 1998). In Tabasco this species is distributed practically throughout the state; however, wild populations have decreased considerably because it is hunted as a food source and its habitat has been severely modified (Rangel-Mendoza and Weber 2015). This individual belongs to the management unit of the Academic Division of Biological Sciences at the Universidad Juarez Auténoma de Tabasco. Its EVS has been determined as 17 (Wilson et al. 2013a), placing it in the middle portion of the high vulnerability category, and its IUCN status has been assessed as Critically Endangered (CR). It was allocated to the Endangered (P) category by SEMARNAT. Photo by Liliana Rios-Rodas. No. 39. Claudius angustatus Cope, 1865. The Narrow-bridged Musk Turtle occurs at low elevations on the Gulf versant of Mexico from southeastern Veracruz, Tabasco, and Campeche, and it is restricted to the base of the Peninsula de Yucatan, through northern Guatemala and northern Belize (Lee 1996). This individual was located at Division Académica de Ciencias Biologicas of Universidad Juarez Autonoma de Tabasco, in the municipality of Centro. Wilson et al. (2013a) assessed its EVS as 14, placing it at the lower limit of the high vulnerability category. Its conservation status was evaluated as Near Threatened (NT) by the IUCN, and it was placed in the Endangered (P) category by SEMARNAT. Photo by Liliana Rios-Rodas. Amphib. Reptile Conserv. No. 38. Kinosternon leucostomum (Duméril, Bibron, and Dumeril, 1851). The White-lipped Mud Turtle occurs at low elevations from southern Veracruz, Mexico, southeastward through Central America to Colombia and the Pacific lowlands of Ecuador (Lee 1996). This individual was located at Division Académica de Ciencias Biologicas of Universidad Juarez Autonoma de Tabasco, in the municipality of Centro. Wilson et al. (2013a) assessed its EVS as 10, placing it at the lower limit of the medium vulnerability category. Its conservation status has not been evaluated by the IUCN, but was assessed as Special Protection (Pr) by SEMARNAT. Photo by Liliana Rios-Rodas. . wey, 5 a Ce | x No. 40. Staurotypus triporcatus (Wiegmann, 1828). The Mexican Giant Musk Turtle occurs at low elevations on the Atlantic slope from central Veracruz, northern Oaxaca, northern and eastern Chiapas, western Campeche, Mexico, as well as southward and eastward through northern Guatemala and Belize (Lee 1996; Reynoso et al. 2016). This individual was located at Division Académica de Ciencias Bioldgicas of Universidad Juarez Autonoma de Tabasco, in the municipality of Centro. Wilson et al. (2013a) assessed its EVS as 14, placing it at the lower limit of the high vulnerability category. Its conservation status has been evaluated as Near Threatened (NT) by the IUCN, and as Threatened (A) by SEMARNAT. Photo by Liliana Rios-Rodas. August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Wildfires In general, farming activities cause most wild fires in Tabasco. The deliberate burning of grasses before cultivation is a frequent practice by farmers who burn the land to eliminate undesirable plants in order to benefit their grasslands. Furthermore, the resulting ashes from these fires are regarded as valuable fertilizer for their grasses (Camara-Cabrales et al. 2019). Unfortunately, these practices are performed without any regulations, and may end up accidently burning a larger area than originally planned, including entire forest plantations (Camara-Cabrales et al. 2019). In addition, remnant areas of tropical forest also are burned, since many farms are located within this vegetation type. Fires have caused communities such as Villa de Guadalupe in Huimanguillo, Sierra El Madrigal in Teapa, and Sierra de Tenosique, to eradicate large tracts of tropical forest, and consequently many animals are killed, impeded by their slow movement. Another important area that has suffered the consequences of wildfires is Reserva de la Bidsfera Pantanos de Centla, where local villagers traditionally use fires to capture turtles during the dry season; and these fires not only kill the turtles, but also burn their nests and eggs (Beauregard-Solis et al. 2010; Zenteno- Ruiz et al. 2004). Conservation Status We used the same three systems of conservation assessment as in the previous entries in the Mexican Conservation Series (see above), 1.e., SEMARNAT (2010), the IUCN Red List (http://tucnredlist.org), and the EVS (Wilson et al. 2013a, b). We have continued to update the assessments from these three systems as necessary. The SEMARNAT System The Secretaria del Medio Ambiente y Recursos Naturales (SEMARNAT) of the federal government of Mexico developed a system of conservation assessment for the national fauna (SEMARNAT 2010), which is used by many Mexican herpetologists. Three categories are employed in the SEMARNAT system: endangered (P), threatened (A), and under special protection (Pr). We allocated the species remaining unassessed in this system to date into a “No Status” (NS) category. The ratings available for the Tabasco herpetofauna are given in Table 8 and summarized in Table 11. As noted in previous entries in the Mexican Conservation Series (see above), only a small portion of the herpetofauna of Tabasco has been assessed using this system. Of the 165 native species occurring in Tabasco, only 56 species (33.9%) have been provided with SEMARNAT ratings and are placed in the three categories as follows: Endangered (P), four (2.4%); Amphib. Reptile Conserv. Threatened (A), 10 (6.1%); and Special Protection (Pr), 42 (25.5%). The majority of the species native to Tabasco (109, or 66.1%), however, have a No Status (NS) rating by the SEMARNAT system. In our opinion, until and unless all the species occurring in the state are assessed this system will be of little help in understanding the conservation needs of the herpetofauna of Tabasco. Assuming that the SEMARNAT personnel have placed a greater emphasis on assessing endemic species in Mexico, then this should be evident by comparing the distributional category assignments and the SEMARNAT assessments. To ascertain whether such a bias exists, the pertinent data in Table 12 indicate that the majorities of species in Tabasco are non-endemic and have not been assessed (98, or 59.4%), and the evaluated species are also largely non-endemic (47, or 28.5%). Thus, these data indicate no bias toward the assessment of country endemic species. The IUCN System The system of conservation assessment developed and implemented by the International Union for Conservation of Nature is used broadly, but it has been criticized in earlier entries of the Mexican Conservation Series for several reasons, as discussed in Johnson et al. (2015b). Still, the assessments available for the Tabasco herpetofauna are collated in Table 8 and summarized in Table 13. Of the 165 native herpetofaunal species in Tabasco, 114 (69.1%) have been evaluated using the IUCN system (Table 13). Of these 114 species, 19 have been allocated to the three threat categories of CR (four, or 3.5%); EN (four, or 3.5%); and VU (11, or 9.6%). The four CR species are the anurans Ptychohyla macrotympanum and Agalychnis moreletii and the turtles Lepidochelys kempii and Dermatemys mawii;, and all four are non-endemic. The four EN species are the anurans Charadrahyla chaneque and Duellmanohyla chamulae, the salamander Bolitoglossa veracrucis, and the turtle Chelonia mydas, and the anurans and the salamander are country endemics while the turtle is non-endemic. The 11 VU species are the anurans Incilius macrocristatus, Craugastor alfredi, C. rhodopis, and Eleutherodactylus leprus, the salamander Bolitoglossa alberchi, the caecilian Dermophis mexicanus, the crocodylian Crocodylus acutus, the lizard Norops barkeri, and the turtles Chelydra rossignonii, Dermochelys coriacea, and Trachemys venusta. The three anurans are non-endemic, except for C. rhodopis, which is a country endemic, the salamander is a country endemic, the caecilian and crocodylian are non-endemic, the lizard is a country endemic, and the three turtles are non-endemic. The remaining 95 species are placed in the “lower risk” categories of NT (10, or 6.1% of the total of 165 species) and LC (85, or 51.5%). The 10 NT species are the anurans Craugastor berkenbuschii, C. laticeps, Rheohyla August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 11. SEMARNAT categorizations for the herpetofaunal species in Tabasco, Mexico, arranged by families. Non-native species are excluded. Number SEMARNAT categorizations Family of Speci pecial No status species adangered)(F). .|"Ehreavened'(’) “| irotection< Py) (NS) i el 3 ee — eet | Hylidae Leptodactylidae Microhylidae Phyllomedusidae Ranidae Rhinophrynidae Subtotal Plethodontidae Sarr EEE SSE _ EES EE | Dermophiidac TP Sibit) a E SSS Gr eS Ee ee ae |Crocodylidae TS USS iAD5E05 a | 2 || LN | = SN | NI? | L=—~ | Corytophanidae | A 8 pDaciviodact i | Ss eS eS CoB i i a a eee eee eee eee CESS | | [ee eee Viguatidsd ss (| 22 | Sa ts) ee | Lab uyides <a | Phrynosomatidae | S| SS | Phyllodactylidae |] St | CO [SCE A ets [fs Sel | lc) neal | = fos |e tll ee ee | Sphaerodactylidae | 2 | Sphenomorphidae | 2 Ci i, i ce i Saar ae ee ee eee pXantusiidae J Sf CXensaitdde nl | Oe. Olim on | ke | | Subtotal a 8 ea, a a a ee eee eee |Colubridae | 2 CC ZS |Dipsadidae | 8 cele | | | | Leptotyphlopidae | tS Notice | ee eee |S a eel psibynophiidaess |S er Sle CU C—O ee ee ee ee ee DS SS ESE ESS SSS ee Eee |Subtotal | SS ecidomidasis ih Be ee Ee ee Chel ditdie | | Pees eS Se |Dermatemyidae | dt Isbensochelyidace i) eS u | Kinosternidae | 3 BO lAPmofuolrm]—|nf—frmolel_]|7 elole[p Amphib. Reptile Conserv. 37 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 12. Comparison of the SEMARNAT and distributional categorizations for the Tabasco herpetofauna. Non-native species are excluded. Distributional categories Endangered (P) Non-endemic species (NE) 4 Country-endemic species (CE) — Total 4 miotympanum, and Smilisca cyanosticta (two country endemics and two non-endemics), the salamander Bolitoglossa platydactyla (a country endemic), the lizard Celestus rozellae (a non-endemic), and the turtles Rhinoclemmys areolata, Kinosternon acutum, Claudius angustatus, and Staurotypus triporcatus (all non-endemics). The 85 LC species comprise the largest group of the native species (Table 13), but whether this large portion of the native species are in reality of “Least Concern” is a question we examine below. Of the remaining 51 species in the herpetofauna, four are allocated to the DD category (2.4% of the total of 165 species) and 47 are in the NE category (28.5%). In the next section, we examine the status of these 51 species using the EVS system. To determine the relationship between the application of the IUCN categories and the distribution categories, the data on these correlations are assembled in Table 14. These data indicate that of the 20 country endemic species, six (30.0%) are allocated to the “threat categories.” None of these six species is placed in the CR category, thus three species are in the EN category, including the anurans Charadrahyla chaneque and Duellmanohyla chamulae, and _ the salamander Bolitoglossa veracrucis. The other three species are consigned to the VU category, including the anuran Craugastor rhodopis, the salamander Bolitoglossa alberchi, and the anole Norops barkeri. The remaining CE species, numbering 14, are distributed rather uniformly among the other IUCN categories, with the highest number (five) placed in the LC category. As expected, the majority of the 145 non-endemic species (80, or 55.2%) are also allocated to the LC category. The next largest number (44, or 30.3%) was placed in the Not Evaluated (NE) category. The remaining non- endemic species (21, or 14.5%) are distributed among the remaining IUCN categories, with 13 placed in the “threat categories” (CR, EN, and VU). Based on the data in Table 14, no correlation 1s evident between the placements of the country endemic or non-endemic species among the IUCN’s “threat categories.” More to the point, as commonly found in earlier entries of the Mexican Conservation Series, most species of either distribution category (country endemic or non-endemic) are placed either in the LC category or are not assessed using the IUCN system. In the case of Tabasco, these species amount to 124 of the 145 non-endemic species (85.5%) and eight of the 20 country-endemic species Amphib. Reptile Conserv. SEMARNAT categories Special No Status Threatened (A) Protection (Pr) (NS) Total 9 34 98 145 9 10 20 10 43 108 165 38 (40.0%). In total, 132 of the 165 total native herpetofaunal species in Tabasco (80.0%) are placed either in the Least Concern or NE categories using the IUCN system. At this juncture, the IUCN assessment system has demonstrated that the majority of the evaluated herpetofauna is either of little concern (1.e., is in reasonably good shape from a conservation perspective) or simply has been ignored (i.e., non-evaluated). We further examine these LC and NE species in the next section. The EVS System The EVS (Environmental Vulnerability Score) system of conservation evaluation initially was created as a means for assessing the conservation status of the amphibians and reptiles of Honduras (Wilson and McCranie 2004). Subsequently, it has been used for the same purpose with other segments of the Mexican and Central American herpetofaunas (e.g., Townsend and Wilson 2010, 2013a,b; Johnson et al. 2015b, 2017; Mata-Silva et al. 2015, 2019; and all entries in the MCS [see above]). In this study, the EVS values for all 162 native non-marine Species occurring in Tabasco are given in Table 8 and summarized in Table 15. The EVS values range from 3 to 19, one less than the entire theoretical range of 3—20. The most frequent values (applied to 10 or more species) are 6 (13 species), 7 (12), 8 (13), 9 (16), 10 (21), 11 (16), 12 (15), 13 (17), 14 (11), and 15 (10). These 10 values are applied to 144 native non-marine species (88.9% of the total of 162 species). The lowest possible score of 3 was established for two anuran species (Rhinella horribilis and Smilisca baudini) and the highest score of 19 for one turtle (7rachemys venusta). As with previous MCS reports, herein the EVS scores are aggregated into three categories of low (EVS of 3-9), medium (10-13), and high (14-19) vulnerability. On the basis of this categorization, the species counts increase slightly from low (66) to medium (69) and then decrease markedly to high (27). This sort of pattern is emblematic of herpetofaunas that contain more non-endemic species (145 in the case of Tabasco) than endemic species (20), as was previously determined in Chiapas (Johnson et al. 2015a), Tamaulipas (Teran-Juarez et al. 2016), Nuevo Leon (Nevarez-de los Reyes et al. 2016), Jalisco (Cruz-Saenz et al. 2017), the Mexican Yucatan Peninsula (Gonzalez- Sanchez et al. 2017), and Coahuila (Lazcano et al. 2019). August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 13. IUCN Red List categorizations for the herpetofaunal families in Tabasco, Mexico. Non-native species are excluded. The shaded columns to the left are the “threat categories,” and those to the right are the categories for which too little information on conservation status exists to allow the taxa to be placed in any other IUCN category, or they have not been evaluated. IUCN Red List categorization Number 7 Family of Critically Eendanoered! | Walners bis Near Least Data Not species Endangered 8 Threatened | Concern | Deficient | Evaluated (NE) 1 3 Dactyloidae — — Xantusiidae Xenosauridae N —) | Colubridae | 2 | Dipsadidae | 3 | Total | 12 1 6 N Subtotal 4 Corytophanidae 1 _ al 2 l 0 0 2 6 8 Sum total 5 Category total 165 19 95 51 Amphib. Reptile Conserv. 39 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 14. Comparison of IUCN and distributional categorizations for the Tabasco herpetofauna. Non-native species are excluded. IUCN category a LP Critically Near Least Data Not ee eae Endangered ee aaa Threatened Concern Deficient Evaluated Total (CR) (NT) (LC) (DD) (NE) Non-endemic species 4 1 8 Ti 80 1 44 145 (NE) Country-endemic species (CE) _ 4 ; : 2 2 A ae Total 4 4 11 10 85 4 47 165 The results of applying the IUCN categories to the Tabasco herpetofauna are compared to those obtained from using the EVS system in Table 16. This comparison demonstrates that only 16 of the 27 high vulnerability species (59.3%) are placed in the three IUCN “threat categories.” These 16 species are the anurans /ncilius macrocristatus (VU 11), Craugastor alfredi (VU 11), C. rhodopis (VU 14), Eleutherodactylus leprus (VU 12), Charadrahyla chaneque (EN 13), Duellmanohyla chamulae (EN 13), Ptychohyla macrotympanum (CR 11), Agalychnis moreletii (CR 7), the salamanders Bolitoglossa alberchi* (VU 15) and _ Bolitoglossa veracrucis* (EN 17), the caecilian Dermophis mexicanus (VU 11), the crocodylian Crocodylus acutus (VU 14), the anole Norops barkeri* (VU 15), and the turtles Chelydra rossignonii (VU 17), Dermatemys mawii (CR 17), and Trachemys venusta (VU 19). At the other extreme, the 65 low vulnerability species constitute 75.6% of the 86 LC species (Table 16). As demonstrated in the other MCS studies, there is a general lack of correspondence between the application of the IUCN and EVS assessment systems. Only four of the 162 native non-marine species in the Tabasco herpetofauna are allocated to the DD category (Table 17), which are the anurans Craugastor palenque, C. pelorus*, and Exerodonta bivocata*, and the night lizard Lepidophyma tuxtlae*. Based on similar arguments presented in previous MCS studies (e.g., Torres-Hernandez et al. 2021), we suggest that the three anurans, each with an EVS of 15, would be better served by being placed in the EN category and the lizard, with an EVS of 11, in the NT category. Forty-seven species still remain to be evaluated using the IUCN system, and thus we allocated them to the NE category (Tables 8 and 18). Only three of these species are country endemics (the anuran Quilticohyla zoque and the lizards Holcosus amphigrammus and H. stuarti). The remaining 44 species are all non-endemics. The EVS values range from 3—15, which allocates a certain number of species to each of the three summary categories (Table 8). Twenty-four species have a low EVS score, 19 have medium scores, and four have high scores. When these species are assessed by the IUCN, we suggest that the four high vulnerability species (Quilticohyla zoque, Oxybelis potosiensis, Oxyrhopus petolarius, and Amphib. Reptile Conserv. 40 Agkistrodon russeolus), with an EVS of 14 or 15, should be placed in one of the three “threat categories.” The 10 species with an EVS of 11, 12, or 13 should be allocated to the NT category. The remaining 33 species, with an EVS of 3-10, can be placed in the LC category. As with all the previous entries in the Mexican Conservation Series, in this entry we ascertained that IUCN has placed a rather large segment of the Tabasco herpetofauna in the Least Concern category (Table 19). This includes 85 species, or 52.5% of the total of 162 native non-marine species. Since over half of the species in Tabasco have been judged by IUCN to be of Least Concern, one might conclude that the conservation status of this herpetofauna is in reasonably good shape. To examine whether this is the case, the determinations of the EVS values for these 85 species are shown in Table 19. Given that the majority of the Tabasco herpetofauna is comprised of non-endemic species, one might expect that a large portion of these species should be assigned to the LC category, which proves to be the case. Only five (5.9%) of these LC species are country endemics. The EVS values for the 85 LC species range from 3 to 17, or only three fewer than the entire theoretical range for the EVS (1.e., 3-20). This range is two fewer than the entire range for Tabasco (3-19). Allocation of the EVS values for the 85 LC species into the three summary categories indicates the following: low (3-9), 40 species; medium (10-13), 38 species; and high (14-20), 7 species. On the basis of these allocations, we suggest that a more realistic evaluation would position the seven high vulnerability species in one of the three threat categories, as: CR (Micrurus diastema), EN (Crotalus tzabcan), and VU (Triprion spinosus, Norops compressicauda, Sceloporus lundelli, Dipsas brevifacies, and Porthidium nasutum). The 40 medium vulnerability species most logically should be allocated to the NT category, and the 40 low vulnerability species should be retained in the LC category, at least until more up-to-date, targeted conservation status surveys can be completed. Relative Herpetofaunal Priority Johnson et al. (2015a) developed the concept of Relative Herpetofaunal Priority (RHP) in the third entry of the MCS. This device is a simple means for measuring the August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 15. Environmental Vulnerability Scores (EVS) for the herpetofaunal species in Tabasco, Mexico, arranged by family. The and the one to the right indicates the high vulnerability scores. Non- P. shaded area to the left encompasses low vulnerability scores native species are excluded. ) pe S 3 N >> = ra) % ze 5) = = > S —_ =| o = = S = > —| a Number of lossidae Eublepharidae astoridae Iguanidae Eleutherodactylidae Sphenomorphidae Sibynophiidae Typhlopidae Viperidae Subtotal Chelydridae Geoemydidae Kinosternidae Subtotal Sum total Subtotal Corytophanidae Phyllomedusidae Dactyloidae Phrynosomatidae Sphaerodactylidae Phyllodactylidae Scincidae Leptotyphlopidae Natricidae Dermatemyidae Plethodontidae Dermophiidae Subtotal Crocodylidae Xenosauridae Subtotal Emydidae Rhinophrynidae Subtotal Bufonidae Centrolenidae Leptodactylidae Microhylidae Subtotal Mabuyidae Xantusiidae Colubridae Dipsadidae Elapidae Crau Diplo Category total August 2022 | Volume 16 | Number 2 | e315 41 Amphib. Reptile Conserv. The herpetofauna of Tabasco, Mexico Table 16. Comparison of Environmental Vulnerability Scores (EVS) and IUCN categorizations for members of the herpetofauna of Tabasco, Mexico. Non-native species and marine species are excluded. The shaded area at the top encompasses low vulnerability category scores, and the one at the bottom includes the high vulnerability category scores. IUCN category Least Data Not Total Concern Deficient | Evaluated Critically Endangered Vulnerable Endangered ne en ee ee eee ee —_ is) relative importance of the herpetofaunal components of the physiographic regions in any given geographic entity, such as states in Mexico in the case of the MCS. Ascertaining the RHP is accomplished by using two metrics, 1.e., (1) the proportions of state and country endemics (only country endemics in the case of Tabasco) among the physiographic regional herpetofaunas, and (2) the absolute quantity of high vulnerability category species in each physiographic regional herpetofauna. The data resulting from these calculations are presented in Tables 20 and 21, respectively. The data in Table 20 are based on the relative number of country endemics (since there are no state endemic species in Tabasco). These data demonstrate that the first rank is occupied by the SNC with 17 species of a total of 145 species (11.7%). The second rank is held by the GCP with five country endemics among a total of 89 species (5.6%), and the third rank is the SBP with three country endemics among a total of 93 species (3.2%). The data in Table 21 show the relative numbers of high vulnerability species, but the rankings differ Threatened Near somewhat from those seen in Table 20. The first rank is the same in both instances, 1.e., the Sierra Norte de Chiapas, with 23 high vulnerability species among a total of 142 species (16.2%). The second rank relative to the high vulnerability species, however, 1s held by the Sierras Bajas de Petén with 15 such species among a total of 91 (16.5%), although it holds rank number three with respect to country endemics. The third rank in Table 21 is for the Gulf Coastal Plain, with 10 high vulnerability species among a total of 79 (12.7%), while this region’s status is rank two relative to country endemics. Based on the results of the RHP analyses, the physiographic region with the highest priority is clearly the SNC, since it supports the highest numbers of both country endemics (Table 20) and high vulnerability species (Table 21). The 17 country endemics, as indicated by the asterisks in Table 4, include eight anurans (Craugastor berkenbuschii, C. pelorus, C. rhodopis, Charadrahyla chaneque, Duellmanohyla chamulae, Exerodonta bivocata, Quilticohyla zoque, and Rheohyla miotympanum), two salamanders (Bolitoglossa Table 17. Environmental Vulnerability Scores (EVS) for members of the herpetofauna of Tabasco, Mexico, allocated to the IUCN Data Deficient category. * = country endemic. Environmental Vulnerability Score (EVS) Species Geographic Ecological Reproductive mode/Degree Total distribution distribution of persecution score Craugastor palenque i a ae a | ae | Craugastor pelorus* lo el | Exerodontabivocata* | OTS Lepidophyma tuxtlae* Amphib. Reptile Conserv. 42 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 18. Environmental Vulnerability Scores (EVS) for members of the herpetofauna of Tabasco, Mexico, currently Not Evaluated (NE) by the IUCN. Non-native species are excluded. * = country endemic. Environmental Vulnerability Score (EVS) Species ‘ : Reproductive Geographic Ecological distribution distribution more Degree Ol persecution |Rhinellahorribilis | 8 ee a cS | Noropsbiporcats | Et E Ce | Noropstemurims 88 Norops rodriguezii 10 ce | Norops tropidonoms | 8 |Noropsuniformis, TA a ‘iguinavkinotopha ———=—~S~SC‘iaSSSCdYS Cd Cd | Marisoralineola | 8 8 | Sceloporusvariabilis | eC 10 |Scincellacherrieg | 8 8 | Holcosus amphigrammus® | SL [Hotcomssuri® ———=S=~=~“‘is;CSC*‘( UBT dUATCOCdSOC“‘SUSSOOOUOCO#*d;SCO*~‘<(‘ia]R |Xenosaurusrackhami | 8 |Boaimperator 8 | Drymobius margaritiferus | Lampropeltis polyzona | Onybelis potosiensis | SS a a | Coniophanesfissidens | 8 TT ee Cimantodes cenchea—=—=~—CSC‘ SSCS) COC‘ | Imantodes gemmisrams | 8 | Leptodeira septenrionalis, | | [Onrhopus peters ———=S~C~wSC“‘ TUSSCdTSOOUéCUdOOC~<“ RSSCC‘*dSC“‘éNM# [Rhadinaea decoraa ——=~=~sC~SCiS| Cd Cd |Sibonnebulas Epictia phenops Wlul]ey_nftre Aa WW} ute |e ies) ios) t ios) N ies) i t + HEE er AA N | GW HA Ww | Go] Go | Go | tHE HHH HHH Bothrops asper Kinosternon leucostomum Kinosternon scorpioides Amphib. Reptile Conserv. 43 August 2022 | Volume 16 | Number 2 | e315 — The herpetofauna of Tabasco, Mexico Table 19. Environmental Vulnerability Scores (EVS) for members of the herpetofauna of Tabasco, Mexico, assigned to the IUCN Least Concern (LC) category. Non-native species are excluded. * = country endemic. Environmental Vulnerability Score (EVS) ppecics Geographic Ecological Reproductive mode/Degree Total distribution distribution of persecution score 3 2 Mesoscincus schwartzei ne Kee) — os) — — — — — N — — ie) N — N — ies) — — — — — — tee Amphib. Reptile Conserv. 44 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 19 (continued). Environmental Vulnerability Scores (EVS) for members of the herpetofauna of Tabasco, Mexico, assigned to the IUCN Least Concern (LC) category. Non-native species are excluded. * = country endemic. Environmental Vulnerability Score (EVS) ppecics Geographic Ecological Reproductive mode/Degree Total distribution distribution of persecution score 2 ] — N 1 — ies) N ne Kak Kom) — a a a a ae i a rr a EE i Lee eee ee a a a a SSS sss SSIES SSS a sss Se SS ee ee ae ee eee eee ee —— ol 2S 1 es ss Se ______ GC —— 1 a es | es ee es (Se SS. e2= — | ____F_)fsLe a eel eee a pee == ee eS Er a a a a ee EE a SSS rae rr Se a ee ee a 12 Crotalus tzabcan 16 12 14 platydactyla and B. veracrucis), five lizards (Norops Bolitoglossa veracrucis* barkeri, N. compressicauda, Holcosus amphigrammus, Crocodylus acutus H.. stuarti, and Lepidophyma tuxtlae), and two snakes Norops barkeri* (Geophis laticinctus and Micrurus diastema). The 23 Norops compressicauda* high vulnerability species found in the SNC are identified Sceloporus lundelli in Table 8 and are listed here for emphasis: Oxybelis potosiensis Dipsas brevifacies Craugastor berkenbuschii* Oxyrhopus petolarius Craugastor palenque Micrurus diastema* Craugastor pelorus* Porthidium nasutum Craugastor rhodopis* Chelydra rossignonii Exerodonta bivocata* Dermatemys mawii Quilticohyla zoque* Trachemys venusta Triprion spinosus Kinosternon acutum Bolitoglossa platydactyla* Staurotypus triporcatus Amphib. Reptile Conserv. 45 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 20. Number of herpetofaunal species in the three distributional status categories among the three physiographic regions of Tabasco, Mexico. Rank order is based on the number of country endemics. Distributional categories Physiographic region : Country Non-endemics ‘ Endemics Gulf Coastal Plain 78 5 Sierra Norte de Chiapas 125 17 Sierras Baja del Petén 88 3 Of these 23 species, 10 are country endemics (*) with EVS values ranging from 14 to 19. The GCP includes five country endemics: the anuran Craugastor rhodopis, the lizards Aspidoscelis guttatus, Holcosus amphigrammus, and H. stuarti, and the snake Micrurus diastema. The GCP also harbors 10 high vulnerability species, which are indicated in Table 8 and listed here for emphasis: Craugastor rhodopis* Crocodylus acutus Oxybelis potosiensis Micrurus diastema* Chelydra rossignonii Dermatemys mawii Trachemys venusta Kinosternon acutum Claudius angustatus Staurotypus triporcatus Only two of these 10 species are country endemics (*), but the EVS values for all ten range from 14 to 19. Finally, the SBP contains only three country endemics: the anuran Craugastor rhodopis, the salamander Bolitoglossa alberchi, and the lizard Scincella gemmingeri. This region, however, supports 15 high vulnerability species that are listed in Table 8 and here for emphasis: Craugastor palenque Craugastor rhodopis* Bolitoglossa alberchi* Crocodylus acutus Sceloporus lundelli Oxybelis potosiensis Agkistrodon russeolus Crotalus tzabcan Porthidium nasutum Total Rank order Non-natives 88 2 145 93 3 Chelydra rossignonii Dermatemys mawii Trachemys venusta Kinosternon acutum Claudius angustatus Staurotypus triporcatus Only two of these 15 species are country endemics (*), but the EVS values for all 15 species range from 14 to 19. In each of the three physiographic regions we recognize in Tabasco, the largest distributional group, as expected, 1s comprised of the non-endemic species. Similarly, the high vulnerability species in each region are non-endemic species. As a result, unlike the many states surveyed thus far in the MCS, the group of principal conservation concern in Tabasco is the non- endemic segment. Consequently, this group of species is examined more closely below in an effort to protect the herpetofauna of Tabasco. Natural Protected Areas in Tabasco The ostensible purpose for the establishment of natural protected areas in any location is to protect key portions of the ecosystems contained within them from the depredations of societal elements outside them for perpetuity. Basically, there are two types of issues, i.e., agriculturalization and urbanization. To be maximally effective, such protected areas should include functionally capable segments of the ecosystems originally present in a given entity (e.g., a state), whose size and extent is sufficient to support viable populations of all the organisms found within the designated protected area. Most often, however, such areas are established without the completion of the requisite work to demonstrate the existence of survivable populations of anything more than a handful of the resident creatures. When Table 21. Number of herpetofaunal species in the three EVS categories among the three physiographic regions of Tabasco, Mexico. Rank order is determined by the relative number of high EVS species. Non-native and marine species are excluded. Physiographic province Gulf Coastal Plain 39 Sierra Norte de Chiapas 60 Sierras Baja del Petén 38 Amphib. Reptile Conserv. Low Medium High Total Rank order 31 10 79 3 59 23 142 1 38 15 91 2 August 2022 | Volume 16 | Number 2 | e315 an-Vazquez et al. oO °o °o Z ie} Barrag ° Z ° TS CFO TEP aqeAlld epuss Ll PIOZ/LO/OI Adv soqueny BIOSOD BINUL]'T ING OFJOH jap osteleg URdBODIJ| sedeiyd op ING OFJOD ep IN OFJOD [ep oe aI edyeiooe] ZLS 0002/90/01 yeoreg-ny, SHON [OP SBIOIS ie} a BU0D0D OLIAD Jap seynH vAeld OY BULIOJOY ° Z. o Baa ING OF[OH Jap Pare) el B1olsor) BIN a1e1g seuopie) $66 1/70/80 ap voIsQ[OIT PAIOSOY INg OJ[OD Jap INS O}[OD Jap ugrewed sederyD op . SMION [ap SexIOIS pms edvay, “edjejoov], 8861/Z0/P7 INg OJ[OD Jap udleg JUDWIUIOAOS anbisous eyuTovUNns(), [op sefeg seuoig | [esopoay ueorxopy ; L 800¢/60/C¢ dddV [op uoueD INS OF[OD Jap JUSUWIUIOAOS euedsnoeyy ‘ eIOJSOIg P] epusag eIoJsOD einuey’T [elopoy UvITXO|Y pue eynuor “epuag LOL COE C66 1/80/90 Op BAIOSOY op souvjueg (AKAA/UU k i leas Sraeiieas ued carded LAE TBAG i ae UOTLIIPsLNne soredrorunyy (vy) Bory /PP) 99.190q K1083938) dIAINS [BVUNLJOJIC.IIA JUSWIOSvURI] &q patdnddQ SIIB] aydeasorsdyd sie ia a yo aq ‘1Z07 Aln¢ OF :passad0¥_ “ooseqry} -Sepisojo1d-soyeinyeu-seose/XUW' gos ooseqe}//:sdyy ‘sepIsojO1g SO[BININ SVOIV OP [eILIS| CUIDSIS ‘SePISI}OIg SO[VINIVN SBOIE OP OPRISTT G6 1OT (OOSAS) ooneunyyD orquiey A pepryiqeyuajsns ‘Ie|SOUDT OP BIILIAINIS (IT) puv SOdTX9YI Op OpeisY “eONTOL WAVN CUROTXO| BIISOTOJAdIOE Pepsts0s ‘sousunsoy A vUIvISOIg ‘VISO[OJOdIOH Op [eUOTOBN UOTUNDY [KX ‘OoSeqR], “PSOULIOYP] IIA ua eyuny GNYV [ep afeso op epesioduis} ey ud vorjstunejojodioy peprunwios ev] op vinjonyjsa A UOIdISOdWIOD ‘OIOT MV BIOATY-zZoulyRe A VS zodoT-eistneg “yV O1ezeT-so[eAy (1) JOM Ppol[NsuOd SddINOS vJeP OM} SY], UOTOBVAIOSUOD LI B BJUSUILLILJUNTOA SsepeUT|sop seoTY =OACV pure ‘yeInjeN oJyUOUINUO|Y =NIW ‘Soyeyejsa seAIOSOY =AY “ezoyeinyeN PL] op ojuoTUTD0U0Z A UOLDVISIU] OP OUD =NOIO “[eyeisq onbieg =qd ‘eune, A vO] Ip UOIIID}O1g Op volV <add V ‘BIOJSOL VI IP CAIASOY =Gy :SUOTeIADIQge AIOSI}CD ‘SIOUSIA JOJ SONI[IO’J = A pure ‘shemuyed JO waysks = § ‘sprens yred = yY SSOOIAJOS SATILIISIUTLIPe = VW :SMOTIOF SB oe ayqeyreae SOIPIO’] Ul SUOTILIADIQGY ‘OdTXd|[ “OOSeQLI, UI SVOIY Pd}99}01g [BINJLN JO SONSIIDOvIVYD °7Z IQeVL NA da qa da da da da oO Zz iS) ooseqe |, ap BLIIS PT dd dd dd dd ie} © August 2022 | Volume 16 | Number 2 | e315 47 Amphib. Reptile Conserv. The herpetofauna of Tabasco, Mexico Table 23. Distribution of herpetofaunal species in the Natural Protected Areas of Tabasco, Mexico, based on herpetofaunal surveys. Abbreviations are as follows: * = species endemic to Mexico and ** = non-native species. The numbers signifying the 14 Natural Protected Areas in Tabasco are as follows: 1 = Pantanos de Centla; 2 = Cafion del Usumacinta; 3 = Agua Blanca; 4 = La Sierra de Tabasco; 5 = Laguna del Camaron; 6 = CICN Yumka; 7 = Reserva Ecoldgico de la Chontalpa; 8 = Laguna de las Ilusiones; 9 = Yu-Balcah; 10 = Cascadas de Reforma; 11 = Rio Playa; 12 = Grutas del Cerro Cocona; 13 = Laguna Mecoacan; and 14 = Guaritec. Pita Ts [sts [s [ofa [> Pola fetafie CEE a GG PBufonidaeGspets) | | | | |. | |. 1117) 1— ‘inciins macrocrisans | | | {+} | | | | | | [+/ | _ [Centrotenidae specs) | | || ilar [Avainobatrachium virdssimum | + | | [+] [Craugastoridae species) |_| | Be Craugastor laticeps Craugastor loki Craugastor palenque [sl Craugastor pelorus* ard Craugastor rhodopis* = [Eleutherodactylidae (2 species) | || || [Eleutherodactylusteprus | TE + | +L +L [Eleutherodactylus planirostris*® ||| TL Hylidae(15 species) | TT [Charadrahylachaneque* | TT +H | LDendrosophus ebraceams | | | TL [Dendrosophus microcephalus | + | + | | + | || ial Duellmanohyla chamulae* Exerodonta bivocata* Ptychohyla macrotympanum QOuilticohyla zoque* Rheohyla miotympanum* Scinax staufferi Smilisca baudinii a Smilisca cyanosticta aa Tlalocohyla loquax Tlalocohyla picta el elelet Trachycephalus vermiculatus a5 rr Triprion spinosus (ele [Leptodactylidae @speciesy | |_| Eneystomops pustiosus | [+ | Leptodacryius fagiis | + [+ | [Microhylidae(2 species) || || [Gasrophyrne elegans | | | [+ [ial eee Es ean Amphib. Reptile Conserv. 48 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 23 (continued). Distribution of herpetofaunal species in the Natural Protected Areas of Tabasco, Mexico, based on herpetofaunal surveys. Abbreviations are as follows: * = species endemic to Mexico and ** = non-native species. The numbers signifying the 14 Natural Protected Areas in Tabasco are as follows: 1 = Pantanos de Centla; 2 = Cafion del Usumacinta; 3 = Agua Blanca; 4 = La Sierra de Tabasco; 5 = Laguna del Camaron; 6 = CICN Yumka; 7 = Reserva Ecoldgico de la Chontalpa; 8 = Laguna de las Ilusiones; 9 = Yu-Balcah; 10 = Cascadas de Reforma; 11 = Rio Playa; 12 = Grutas del Cerro Cocona; 13 = Laguna Mecoacan; and 14 = Guaritec. iT? Ts [+ ts[s [7 [e [> Pola fetafie [popacmasverotous _|+{+| | | 111111111 PPaylomedusione@speey | | | | ||. |. |} )) 11 [Agalychnismoreens | [| | | | | — [niobate brownorm «t+ | | t+le| PRhinophrynidae (Tepeciesy) |__| || [Rhinophymus dorsais—_———~st + || Caudata (5 species) = Plethodontidae (5 species) = Bolitoglossa alberchi* [airy Bolitoglossa mexicana | Bolitoglossa platydactyla* =| = Bolitoglossa rufescens = [Bolitoglossaveracrucis* | | LT [Gymnophiona (I species) | || TT [Dermophiidae (I species) ||| TT [Dermophis mexicanus | + | LTT Sree S| Crocodylia (2 species) Crocodylidae (2 species) Crocodylus acutus Crocodylus moreletii Squamata (110 species) Corytophanidae (4 species) Basiliscus vittatus Corytophanes cristatus Corytophanes hernandezii [iaemanctustongives | +|+|_ TDactyloidae 4 specs) | |__| _ [Norops barter «did [Norops becterh |_| [+ Novbpstipersans—||[[ ops aio Norops capito Norops compressicauda* — [Norops laevis i [Norapsredriguecii [| __ Amphib. Reptile Conserv. 49 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 23 (continued). Distribution of herpetofaunal species in the Natural Protected Areas of Tabasco, Mexico, based on herpetofaunal surveys. Abbreviations are as follows: * = species endemic to Mexico and ** = non-native species. The numbers signifying the 14 Natural Protected Areas in Tabasco are as follows: 1 = Pantanos de Centla; 2 = Cafion del Usumacinta; 3 = Agua Blanca; 4 = La Sierra de Tabasco; 5 = Laguna del Camar6én; 6 = CICN Yumka; 7 = Reserva Ecoldgico de la Chontalpa; 8 = Laguna de las Ilusiones; 9 = Yu-Balcah; 10 = Cascadas de Reforma; 11 = Rio Playa; 12 = Grutas del Cerro Cocona; 13 = Laguna Mecoacan; and 14 = Guaritec. Pet [s[*ts[s [oe [> Pola fetafie [Noropstropidonoms | oT +L +] +] oT | | t+] | t+] ft [Noropspetersi | TT dT + TT CT CT ET [Noropsuniformis | CT +L +] +] Tt tT TT eT [Noropsunilobams | tT +L TT UT CT CE ET Diploglossidae(Ispeciesy |_| | | | TT Celestus rozellae a | [Eublepharidae(1speciesy |_| || Coleonyx elegans Gekkonidae (2 species) Hemidactylus frenatus** Hemidactylus turcicus** Iguanidae (2 species) Ctenosaura similis Iguana rhinolopha Mabuyidae (1 species) Marisora lineola + /Phrynosomatidae(Sspecies) ||| | [Sceloporustundelli | [Sceloporusserrifer | [Sceloporusteapensis || | + Sceloporus variabilis afr | + | Phyllodactylidae (1 species) [aa] Thecadactylus rapicauda Scincidae (2 species) (| - = Sphaerodactylus glaucus Sphaerodactyius milepuncianes |__| + [| [Sphenomorphidae @ species) | | [| er Treiidae Sepecisy | | | | [Aspidosels depot =i + | [+ [Aspidoscelisgutans® ++ + Mesoscincus schwartzei Plestiodon sumichrasti Sphaerodactylidae (2 species) Amphib. Reptile Conserv. 50 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 23 (continued). Distribution of herpetofaunal species in the Natural Protected Areas of Tabasco, Mexico, based on herpetofaunal surveys. Abbreviations are as follows: * = species endemic to Mexico and ** = non-native species. The numbers signifying the 14 Natural Protected Areas in Tabasco are as follows: 1 = Pantanos de Centla; 2 = Cafion del Usumacinta; 3 = Agua Blanca; 4 = La Sierra de Tabasco; 5 = Laguna del Camarén; 6 = CICN Yumka; 7 = Reserva Ecologico de la Chontalpa; 8 = Laguna de las Ilusiones; 9 = Yu-Balcah; 10 = Cascadas de Reforma; 11 = Rio Playa; 12 = Grutas del Cerro Cocona; 13 = Laguna Mecoacan; and 14 = Guaritec. Pita Ts [*ts[s [7 [a] Pola feta fie SNR a GC Lepidophyma fvinacuam | [+ [*|*|| | ||| | [+] | _ [Lepidophyma nates | | | | |_| PXenossuridae speci) | | | ||| _ Taottaeigpcisy —-| | | | | | | ee = 1 [Pseudelaphefiairgs |_| | [+*| | | | | | |_ [Senicoisriaps [+ {*{ f+; | | | | [+] _ amis II ale) Saleen Stenorrhina degenharatii ==) Stenorrhina freminvillii Tantilla schistosa Tantilla rubra Tantillita lintoni Dipsadidae (27 species) Adelphicos visoninum Clelia scytalina rN S 3 a & = S iS S ies) sf [Coniophanes quinguevinatus | + | + | | + | [Coniophanes schmidti | TE + | TL [Conophistineams LH T+ LT Dipsasbrevifacies | TH LT [Geophiscarinous | TEL | [Geophislaticinctus® | EE + TE CE Amphib. Reptile Conserv. 51 August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico Table 23 (continued). Distribution of herpetofaunal species in the Natural Protected Areas of Tabasco, Mexico, based on herpetofaunal surveys. Abbreviations are as follows: * = species endemic to Mexico and ** = non-native species. The numbers signifying the 14 Natural Protected Areas in Tabasco are as follows: 1 = Pantanos de Centla; 2 = Cafion del Usumacinta; 3 = Agua Blanca; 4 = La Sierra de Tabasco; 5 = Laguna del Camaron; 6 = CICN Yumka; 7 = Reserva Ecologico de la Chontalpa; 8 = Laguna de las Ilusiones; 9 = Yu-Balcah; 10 = Cascadas de Reforma; 11 = Rio Playa; 12 = Grutas del Cerro Cocona; 13 = Laguna Mecoacan; and 14 = Guaritec. PieT2 Ts [*ts[s [7 [a] Pola fetafie |Geophissartorit | Ct + TT +] OT UT +] Imantodescenchoa | oT Cd HT +] oT LT LImantodes gemmistratus | + | | oT +] TL [Leptodeirafrenata | CE +H TL +P +] oT UL PT UT HT OT eT [Leptodeiramacuata | TL E+] UT UT UE [Leptodeira seprentrionalis | + | =o +] +] +] +t+} +) P+] f+] | [Niniadiademaa | Et TT + TT UT Niniaseboe dt + TP + t+] oT PT oT TP UT TT eT [Oxyrhopus perolarius | + | Hc el [Pliocercus elapoides | * || -Rhadinaeadecoraa | + | + Sibon dimidiatus Micrurus diastema* a [Micruruselegans | TL | [Leptotyphlopidae (I species) ||| TT Epictiagoudo’ LH TL + | | Thamnophismarciams | *# | LT + LT Thamnophisproximus | *# TL TL [Sibynophiidae(I species) ||| = [Scaphiodontophisannulatus |_| + | + | + io = Lt ie Typhlopidae (2 species) | Amerotyphlops tenuis Indotyphlops braminus** = ar Viperidae (7 species) Li Agkistrodon russeolus | + | Bothriechis schlegelii + Chelonia mydas [iepidecheystenpa |_| | | [| | [chetyaritacaspeas) | | | 1.111...) ?P?f?t. I Amphib. Reptile Conserv. 52 August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. Table 23 (continued). Distribution of herpetofaunal species in the Natural Protected Areas of Tabasco, Mexico, based on herpetofaunal surveys. Abbreviations are as follows: * = species endemic to Mexico and ** = non-native species. The numbers signifying the 14 Natural Protected Areas in Tabasco are as follows: 1 = Pantanos de Centla; 2 = Cafion del Usumacinta; 3 = Agua Blanca; 4 = La Sierra de Tabasco; 5 = Laguna del Camaron; 6 = CICN Yumka; 7 = Reserva Ecoldégico de la Chontalpa; 8 = Laguna de las Ilusiones; 9 = Yu-Balcah; 10 = Cascadas de Reforma; 11 = Rio Playa; 12 = Grutas del Cerro Cocona; 13 = Laguna Mecoacan; and 14 = Guaritec. Taxon Natural Protected Area put 2tsitat stot 7} sf 9 | wo] nm | 2] a3 | 1 | [Chelydrarossignonii | + TL TE + | TT Dermatemydidae (I species) ||| TTT LDermatemysmawii | # T+] TL TE Ee PDermocteyidae tspeces) | | | | | | | | | Dermochelys coriacea SSS SaaS aaa Emydidae (1 species) Trachemys venusta Geoemydidae (1 species) Rhinoclemmys areolata Kinosternidae (3 species) Kinosternon acutum Kinosternon leucostomum Kinosternon scorpioides Staurotypidae (2 species) Claudius angustatus Staurotypus triporcatus Total + + such information is available, generally it is assembled in a sufficiently detailed management plan that, in the best-case scenario, is used to justify the recognition of a given natural protected area. Oftentimes, however, the management plan is drawn up after the official designation of the protected area, or does not exist at the time of the designation. This scenario often is the case with herpetofaunal surveys. In order to assess the extent to which the natural protected areas of Tabasco are able to protect the state herpetofauna, we collected a variety of data on these areas (Table 22). The number of these areas in Tabasco is rather substantial, amounting to 14 entities, which is the same number as seen in the state of Puebla (Woolrich- Pifia et al. 2017). The Mexican Federal government administers two of these 14 areas, 11 are administered at the state level, and one 1s a private reserve. The 14 areas range in size from seven to 302,707 ha. Their total area is 743,808.5 ha or 7,438.1 km, which is 30.1% of the total area of the state and close to three times the proportion occupied by the 14 areas located in Puebla (Woolrich et al. 2017). In Tabasco, these areas were established relatively recently, during the 33-year period from 1988 to 2019. The representation of these areas among the physiographic regions of Tabasco is skewed toward the Llanura Costera de Golfo Sur or Gulf Coastal Lowlands, with 10 of the 14 located there. Three areas are found Amphib. Reptile Conserv. + = within the Sierras del Norte de Chiapas and only one in the Sierras Baja del Petén. With respect to the range of facilities available in these 14 protected areas, eight have the full range (Table 22), and the remaining six have fewer. A major concern for the stability of the state’s protected areas is that, to some degree, landowners occupy nine of the 14 (64.3%) areas. Unfortunately, the nine occupied areas include all of the largest ones, and the largest area not occupied by landowners encompasses only 572 hectares (5.72 km”). Also unfortunate is that only five of the 14 areas have had management plans developed for them. Fewer than half (six) of the 14 areas have had herpetofaunal surveys conducted for them. Below we examine the impact of this situation on the protection of the state’s herpetofauna. Of the 165 native species known from Tabasco, all but seven (158, or 95.8%) have been recorded from one or more of the state’s natural protected areas (Table 23). In addition, all five non-native species have been recorded from one or more of these areas (Table 24). The number of species recorded from these 14 areas ranges from 18 in PE Laguna del Camarén and ADVC Guaritec to 112 in PE La Sierra de Tabasco (Table 23). The seven species that are not represented in any of the 14 areas are: Rheohyla miotympanum*, Bolitoglossa veracrucis* ; Lepidophyma tuxtlae*; Xenosaurus rackhami;, Chelonia mydas, Lepidochelys kempii;, and Dermochelys coriacea. August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico These seven species include three country endemics, and all three of the sea turtles known from the state. Unlike the situation commonly encountered in the other states surveyed in the MCS, a high percentage of the known herpetofauna in Tabasco has been documented in the 14 natural protected areas in the state (Table 24). To date, of the 162 species thus far recorded from these areas, most (141, or 87.0%) are non-endemic species, which is a similar percentage (85.3%) for the representation of non- endemic species in the herpetofauna as a whole (Table 9). In Tabasco, 16 of the 162 (9.9%) species known from these areas are country endemics, again similar to the percentage (11.8%) for the state as a whole (Table 9). All five of the non-native species (100%) have been shown to occur in the natural protected areas in the state, which is not desirable, as these species have been recorded in from one to all 14 of these areas. Nonetheless, the goal of complete representation of the native herpetofauna in the established natural protected areas is within reach, as only seven species need to be added. As noted above, however, four of these seven species are country endemics and three are sea turtles. All but one of these species have been recorded from only a single physiographic region (Table 4), with the four terrestrial species documented from the Sierra del Norte de Chiapas and the three marine species from the Gulf Coastal Plain. Apparently, a special effort must be undertaken to incorporate all seven species within the existing system of natural protected areas. Conclusions and Recommendations Conclusions A. The herpetofauna of Tabasco presently consists of 165 native species, including 38 anurans, five salamanders, one caecilian, two crocodylians, 107 squamates, and 12 turtles. In addition, five non-native species have been recorded from the state, including one anuran and four squamates. B. We recognize three physiographic regions in Tabasco: the Gulf Coastal Plain (GCP), the Sierras Bajas del Petén (SBP), and the Sierra Norte de Chiapas (SNC). C. The three physiographic regions we recognize in Tabasco support from 88 species in the Gulf Coastal Plain (GCP) to 145 in the Sierra del Norte de Chiapas (SNC), with an intermediate number of 93 in the Sierra Bajas del Petén (SBP). D. The numbers of species shared among. the physiographic regions range from 61 between the GCP and the SBP to 79 between the SNC and the SBP. The Coefficient of Biogeographic Resemblance (CBR) values range from 0.61 between the GCP and the SNC to 0.67 between the GCP and the SBP. The UPGMA dendrogram (Fig. 5) indicates that the SBP and GCP cluster at the Amphib. Reptile Conserv. 54 0.67 level, while the SNC clusters to the previous pair at the 0.64 level. This pattern indicates that all three regions are closely aligned at a relatively intermediate level of overall resemblance. E. The level of herpetofaunal endemism in Tabasco is relatively low. Of the 165 recorded native species, only 20 are country endemics (12.1%), including eight anurans, three salamanders, and nine squamates. No state endemics are known from this state. F. The distribution status of the 170 species comprising the Tabasco herpetofauna is as follows (in decreasing order of species numbers): non-endemics (145, 85.3%); country endemics (20, 11.8%); and non-natives (5, 2.9%). Of the 145 non-endemic species, their allocation among six of the nine distributional categories are as follows: MXCA (95, 65.5%); MXSA (34, 23.4%); USCA (eight, 5.5%); USSA (four, 2.7%); OCEA (three, 2.1%); and MXUS (one, 0.7%). G. The principal environmental threats to the herpetofauna of Tabasco are deforestation, agricultural activities, roads, soil contamination and oil extraction, myths and cultural factors (gastronomy), illegal commerce, and forest fires. H. The conservation status of the Tabasco herpetofauna was evaluated by using the SEMARNAT, IUCN, and EVS systems. As in previous MCS entries, the SEMARNAT system was determined to be of limited value, given that of 165 native species distributed in Tabasco, only 56 (33.9%) have been assessed using this system. A comparison of the SEMARNAT and _ distributional categorizations demonstrates that the majority of the species in Tabasco that have not been evaluated (98, 59.4%) are non-endemic species. Otherwise, the species that have been assessed also are primarily non-endemic species (47 or 28.5%), indicating no bias toward the consideration of country endemic species. I. The results of the application of the IUCN system (by category and proportion) are: CR (four, 2.4% of 165 native species); EN (four, 2.4%); VU (11, 6.7%); NT (10, 6.1%); LC (85, 51.5%); DD (four, 2.4%); and NE (47, 28.5%). J. A comparison of the IUCN and _ distributional categorizations illustrates that most of the 165 native species (132, 80.0%) are either allocated to the LC category (85, 51.5%) or Not Evaluated (NE; 47, 28.5%). K. The application of the EVS system of conservation assessment to the 162 native non-marine species of Tabasco demonstrates that the categorical values increase slightly from low vulnerability (66, 40.7% of 162 native non-marine species) to medium vulnerability (69, 42.6%), and then decrease markedly at high vulnerability (27, 16.7%). August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. L. A comparison of the IUCN and EVS conservation status categorizations demonstrates that only 16 of the 27 high vulnerability species (59.3%) are placed in the three “threat categories” (CR, EN, or VU), while 66 low vulnerability species or 77.6% are among the 85 species in the IUCN LC category. As found in previous MCS studies, these two conservation systems lacked correspondence when applied to the Tabasco herpetofauna. M. An examination of the conservation status of the species allocated to the IUCN DD, NE, and LC categories indicates that many of these 136 species (82.4% of the 165 native species) have not been assessed adequately compared to their respective EVS values. Thus, we strongly recommend that these species be reassessed to better demonstrate their propects for future survival. N. The Relative Herpetofaunal Priority (RHP) measure was utilized to determine the conservation significance of the three regional herpetofaunas in Tabasco. This analysis demonstrates that the herpetofauna of the Sierra del Norte de Chiapas is the most significant among the three regions, inasmuch as it supports the greatest numbers of country endemic species and high vulnerability species. The two other areas differ in their rankings (i.e., the rankings are reversed) based on these two RHP measures. O. The number of protected areas in Tabasco 1s 14, of which the Mexican Federal Government administers two, while 11 are administered at the state level, and one is a private reserve. These 14 areas have been established relatively recently, from 1988 to 2019. Collectively, these areas comprise 30.1% of the total area of the state. Most of these areas (10 of the 14) are located in the Gulf Coastal Plain, while three are found in the Sierra del Norte de Chiapas, and only one ts in the Sierras Baya del Petén. Landowners occupy nine (64.3%) of the 14 areas, an undesirable situation with respect to the protection of the included herpetofaunal species. Unfortunately, only five of the 14 areas have developed management plans. In addition, only six of the 14 have completed herpetofaunal surveys. P. One highly desirable aspect, however, is that 158 (95.8%) of the 165 native species from the state have been recorded from one or more of the 14 areas. On the other hand, however, all five non-native species known from the state also are found in one or more of these areas. Of the 158 native species, 141 are non-endemics and 17 are country endemics. Q. Future conservation efforts should be directed toward either locating sustainable populations of the seven unrecorded species within existing natural protected areas or establishing new areas, or perhaps enlarging Amphib. Reptile Conserv. existing areas to encompass these species. In addition, herpetofaunal surveys need to be prepared for the eight areas presently lacking them. Recommendations A. Our principal interest in preparing this 14" entry in the MCS is to document the composition, physiographic distribution, and conservation status of the 165 native Species constituting the herpetofauna of Tabasco. The use of the EVS conservation system demonstrates that the categorical values increase only slightly from low vulnerability (66 species) to medium vulnerability (69 species), and then decrease markedly at high vulnerability (27). The Relative Herpetofaunal Priority measure indicates that the herpetofauna of the Sierra del Norte de Chiapas is the most significant among the three physiographic regions in Tabasco, because it supports the highest numbers of country endemic species and high vulnerability species. B. The most important conservation challenge in Tabasco is to conduct the herpetofaunal surveys for eight of the 14 protected areas, with the hope that populations of the Species not known to be represented within this system can be found in one or more of the areas located in the Gulf Coastal Plain and Sierra del Norte de Chiapas. C. Once the presence of the entire native herpetofauna has been ascertained in the system of natural protected areas, then the next step will be to establish monitoring programs for all native species in order to guarantee their long-term survival. We submit that these steps need to be taken with the greatest speed, given that Tabasco is the 20" most populous state in Mexico and the 12" most densely populated. “Living wild species are like a library of books still unread. Our heedless destruction of them is akin to burning the library without ever having read its books.” John D. Dingell (1991) Acknowledgments.—We are grateful to the following individuals for providing their photographs for this article: Jenny del Carmen Estrada-Montiel, Coral J. Pacheco-Figueroa, José del Carmen Geronimo-Torres, Manuel Alberto Hernandez-May, Marco Antonio Torrez- Pérez, Manuel Hernandez-May, and Marco Antonio Lopez-Luna. We also are very thankful to Guillermo Woolrich-Pifia for his perceptive and helpful review of the manuscript. Literature Cited Adams-Schroeder RH, Dominguez Rodriguez VI, Garcia Hernandez L. 1999. Potencial de la biorremediacion August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico de suelo y agua impactados por petroleo en el tropico mexicano. Terra Latinoamericana 17(2): 159-174. Alvarado-Diaz J, Suazo-Ortufio I, Wilson LD, Medina- Aguilar O. 2013. Patterns of physiographic distribution and conservation status of the herpetofauna of Michoacan, Mexico. Contribution to Special Mexico Issue. Amphibian & Reptile Conservation 7. 128-170 (e7.): Avila-Najera DM, Mendoza GD, Villarreal O, Serna- Lagunes R. 2018. Uso y valor cultural de la herpetofauna en México: una revision de las ultimas dos décadas (1997-2017). Acta Zoolégica Mexicana Nueva Serie 34: 1-15. Barkin D. 1978. Desarrollo Regional y Reorganizacion Campesina: La Chontalpa como Reflejo del Problema Agropecuario Mexicano. Editorial Nueva Imagen, México, DF, México. 173 p. Beauregard-Solis G, Zenteno-Ruiz CE, Armijo-Torres R, Guzman-Juarez E. 2010. Las tortugas de agua dulce: patrimonio zooldgico y cultural de Tabasco. Kuxulkab’ 17: 5-19. Beltran-Paz OI, Vela-Correa G. 2006. Suelos Contaminados con Hidrocarburos y su Efecto en la Formacion de Agregados del Suelo en La Venta, Tabasco. Universidad Autonoma Metropolitana- Xochimilco, México, DF, México. 6 p. Camara-Cabrales LS, Rullan-Silva CD, Galindo- Alcantara A. 2019. Estudio de caso: Incendios Forestales en la Region de los Rios. La biodiversidad en Tabasco. Estudio de Estado Ill: 1-64. Campbell JA. 1998. Amphibians and Reptiles of Northern Guatemala, the Yucatan, and Belize. University of Oklahoma Press, Norman, Oklahoma, USA. xix + 380 p. Campbell JA, Lamar WW. 2004. The Venomous Reptiles of the Western Hemisphere. 2 Volumes. Comstock Publishing Associates, Cornell University Press, Ithaca, New York, USA. 870 p. Canales-Delgadillo J, Pérez-Ceballos R, Zaldivar- Jiménez A, GOmez-Ponce M, Vazquez-Pérez N, De la Rosa M, Potenciano-Morales L. 2020. Muertes por trafico sobre la carretera costera del golfo de México: ,cuantas y cuales especies de fauna silvestre se estan perdiendo? Revista Mexicana de Biodiversidad 91: 1-12. Castillo AO, Zavala CJ. 1996. Fisiografia, recursos vegetales y alternativas de manejo en el Parque Estatal Agua Blanca Tabasco. Universidad y Ciencia 12: 63-70. Coffin AW. 2007. From roadkill to road ecology: a review of the ecological effects of roads. Journal of Transportation Geography 15: 396-406. CONAFOR. 2013. Inventario Estatal Forestal y de Suelos-Tabasco. SEMARNAT, México, DF, México. 125 p. Cruz-Elizalde R, Ramirez-Bautista A, Pineda-Lopez R, Mata-Silva V, DeSantis DL, Garcia-Padilla E, Amphib. Reptile Conserv. Johnson JD, Fucsko LA, Wilson LD. 2022. The herpetofauna of Querétaro, Mexico: composition, distribution, and conservation status. Amphibian & Reptile Conservation 16(1) [General Section]: 148- 192 (e308). Cruz-Saenz D, Mufioz-Nolasco FJ, Mata-Silva_ V, Johnson JD, Garcia-Padilla E, Wilson LD. 2017. The herpetofauna of Jalisco, Mexico: composition, distribution, and conservation status. Mesoamerican Herpetology 4: 22-118. Dingall JD. 1991. The Endangered Species Act: legislative perspectives on a living law. Pp. 25-30 In: Balancing on the Brink of Extinction: Endangered Species Act and Lessons for the Future. Editor, Kohm KA. Island Press, Washington, DC, USA. 329 p. Duellman WE. 1990. Herpetofauna in neotropical rainforests: comparative composition, history, and resource use. Pp. 455-505 In: Four Neotropical Rainforests. Editor, Gentry AH. Yale University Press, New Haven, Connecticut, USA. 627 p. Espinal MR, Solis JM, Mora JM. 2021. First record of Sibon dimidiatus (Gunther, 1872) for the department of La Paz, south-western Honduras. Herpetology Notes 14: 99-102. Ferrera-Cerrato R, Rojas-Avelizapa NG, Poggi- Varaldo HM, Alarcon A, Cafizares-Villanueva RO. 2006. Procesos de biorremediacién de suelo y agua contaminados por hidrocarburos del petrdleo y otros compuestos. Revista Latinoamericana de Microbiologia 48: 179-187. Geronimo-Torres JC, Barragan- Vazquez MR, Rios-Rodas L. 2022. Incorporando la distintividad taxonémica en estudios de diversidad: Anfibios del Parque Estatal de la Sierra de Tabasco, México. Ecosistemas 31: 2294. Gomez-Alvarez G, Pacheco N. 2010. Uso medicinal de serpientes comercializadas en dos mercados de la ciudad de Mexico. Etnobiologia 8: 51-58. Gonzalez-Sanchez VH, Johnson JD, Garcia-Padilla E, Mata-Silva V, DeSantis DL, Wilson LD. 2017. The herpetofauna of the Mexican Yucatan Peninsula: composition, distribution, and — conservation. Mesoamerican Herpetology 4: 263-380. Guevara-Chumacero M, Pichardo-Fragoso A, Martinez Cornelio M. 2017. La tortuga en Tabasco: comida, identidad y representacion. Estudios de Cultura Maya 49: 97-122. Guichard-Romero CA. 2006. Situacién actual de las poblaciones de tortuga blanca (Dermatemys mawii) en el sureste de México. Instituto de Historia Natural y Ecologia. Informe final SNIB-CONABIO proyecto No. AS003. CONABIO, México, DF, México. 32 p. Heimes P. 2016. Snakes of Mexico. Herpetofauna Mexicana. Volumen I. Edition Chimaira, Frankfurt am Main, Germany, and ECO Publishing, Rodeo, New Mexico, USA. 572 p. Hernandez-Lopez A, Lopez-Alamilla E, Rodriguez Ramirez A, Aquino-Bravata V. 2012. Diagnostico del August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. uso de la fauna silvestre, en el area de Protecci6n de Flora y Fauna “Cafion del Usumacinta,” Tenosique, Tabasco. Ra Ximhai Numero Especial 1, 9: 1-13. Infografia Agroalimentaria. 2017. Tabasco. Servicio de Informacion Agroalimentaria y Pesquera (SIAP), México, DF, México. 51 p. INEGI (Instituto Nacional de Estadistica, Geografia e Informatica). 1986. Sintesis Geogrdafica, Nomenclator y Anexo Cartografico del Estado de Tabasco. Instituto Nacional de Estadistica, Geografia e Informatica, México, DF, México. INEGI (Instituto Nacional de Estadistica, Geografia e Informatica). 1989. Censos Economicos. Instituto Nacional de Estadistica, Geografia e Informatica, México, DF, México. INEGI (Instituto Nacional de Estadistica, Geografia e Informatica). 2016. Anuario Estadistico y Geogrdfico de Tabasco. Instituto Nacional de _ Estadistica, Geografia e Informatica, México, DF, México. INEGI (Instituto Nacional de Estadistica, Geografia e Informatica). 2009. Anuario Estadistico del Estado de Tabasco. Instituto Nacional de Estadistica, Geografia e Informatica, México, DF, México. Johnson JD, Mata-Silva V, Garcia-Padilla E, Wilson LD. 2015a. The herpetofauna of Chiapas, Mexico: composition, distribution, and conservation. Mesoamerican Herpetology 2: 271-329. Johnson JD, Mata-Silva V, Wilson LD. 2015b. A conservation reasssessment of the Central American herpetofauna based on the EVS measure. Amphibian & Reptile Conservation 9 [General Section]: 1-94 (e100). Kolb M, Galicia L. 2012. Challenging the linear forestation narrative in the Neo-tropic: regional patterns and processes of deforestation and regeneration in southern Mexico. The Geographical Journal 178: 147-161. Lazcano D, Nevarez-de los Reyes M, Garcia-Padilla E, Johnson JD, Mata-Silva V, DeSantis DL, Wilson LD. 2019. The herpetofauna of Coahuila, Mexico: composition, distribution, and conservation status. Amphibian & Reptile Conservation 13(2) [General Section]: 31-94 (e189). Lee JC. 1996. The Amphibians and Reptiles of the Yucatan Peninsula. Comstock Publishing Associates, Cornell University Press, Ithaca, New York, USA. xii + 500 p. Maldonado-Sanchez EA, Ochoa-Gaona S, Ramos R, Guadarrama-Olivera MA, Gonzalez-Valdivia N, De Jong BHJ. 2016. La selva inundable de canacoite en Tabasco, México, una comunidad vegetal amenazada. Acta Botanica Mexicana 115: 75-101. Mata-Silva V, Johnson JD, Wilson LD, Garcia-Padilla E. 2015. The herpetofauna of Oaxaca, Mexico: composition, physiographic distribution, and conservation. Mesoamerican Herpetology 2: 5—62. Mata-Silva V, DeSantis DL, Garcia-Padilla E, Johnson Amphib. Reptile Conserv. JD, Wilson LD. 2019. The endemic herpetofauna of Central America: a casualty of anthropocentrism. Amphibian & Reptile Conservation 13(1) [General Section]: 1-64 (e168). Mata-Silva V, Garcia-Padilla E, Rocha A, DeSantis DL, Johnson JD, Ramirez-Bautista A, Wilson LD. 2021. A reexamination of the herpetofauna of Oaxaca, Mexico: composition update, physiographic distribution, and conservation commentary. Zootaxa 4996: 201-252. McCranie JR, Townsend JH, Wilson LD. 2004. Corytophanes hernandesii. Catalogue of American Amphibians and Reptiles 790.1—790.6. McCranie JR, Hedges SB. 2012. Two new species of geckos from Honduras and resurrection of Sphaerodactylus. continentalis Werner from the synonymy of Sphaerodactylus millepunctatus Hallowell (Reptilia, Squamata, Gekkonoidea, Sphaerodactylidae). Zootaxa 3492: 65-76. Meza-Lazaro RN, Nieto-Montes de Oca A. 2015. Long forsaken species diversity in the Middle American lizard Holcosus undulatus (Teiidae). Zoological Journal of the Linnaean Society 175(1): 1-22. Nevarez de los Reyes M, Lazcano D, Garcia-Padilla E, Mata-Silva V, Johnson JD, Wilson LD. 2016. The herpetofauna of Nuevo Leon, Mexico: composition, distribution, and conservation. Mesoamerican Herpetology 3: 557-638. Ochoa-Gaona S, Pérez-Hernandez I, Frias-Hernandez JA, Jarquin-Sanchez A, Méndez-Valencia A. 2011. Estudio Prospectivo de Especies Arboreas Promisorias para la Fitorremediacion de Suelos Contaminados por Hidrocarburos. Gobierno del Estado de Tabasco/Secretaria de Recursos Naturales y Proteccién Ambiental/El Colegio de la Frontera Sur/Petroleos Mexicanos, Villahermosa, Tabasco, Mexico. 145 p. Pacheco-Figueroa C, Luna-Ruiz RC, Valdez-Leal JD, Saenz JC, Gordillo-Chavez EJ, Moguel-Ordofiez E, Gama-Campillo LM, Mata-Zayas EE, Rangel- Ruiz LJ, Santiago-Plata VM. 2021. Puntos de alta siniestralidad de vertebrados en la carretera costera en Tabasco. Pp. 348-367 In: Impacto de las vias de comunicacion sobre la fauna silvestre en dreas protegidas. Estudios de caso para el sureste de Mexico. Editors, Benitez J. A, Escalona-Segura G. El Colegio de la Frontera Sur, Chetumal, Quintana Roo, México. 732 p. Ponce M, Solis FA, Ibafiez R, Jaramillo CA. 2008. Geographic distribution. Ninia sebae. Herpetological Review 39: 372. Porras LW, Wilson LD, Schuett GW, Reiserer RS. 2013. A taxonomic reevaluation and conservation assessment of the common cantil, Agkistrodon bilineatus (Squamata: Viperidae): a race against time. Amphibians & Reptile Conservation 7(1): 48-73 (e63). Pozo-Montuy G, De la Cruz-Félix HK, Guzman- Aguirre CC. 2019. Uso tradicional de los vertebrados August 2022 | Volume 16 | Number 2 | e315 The herpetofauna of Tabasco, Mexico terrestres. Pp. 309-317 In: La Biodiversidad de Tabasco. Estudio de Estado. Tomo 1. CONABIO, México, DF, México. 393 p. Ramirez-BautistaA, Hernandez-Salinas U, Cruz-Elizalde R, Berriozabal-Islas C, Moreno-Lara I, DeSantis DL, Johnson JD, Garcia-Padilla E, Mata-Silva V, Wilson LD. 2020. The herpetofauna of Hidalgo, Mexico: composition, distribution, and conservation status. Amphibian & Reptile Conservation 14(1) [General Section]: 63-118 (e224). Rangel-Mendoza JA, Weber M. 2015. Evaluacion del estado fisico de la tortuga blanca, Dermatemys mawii, bajo condiciones de cautiverio en Tabasco, México. Agrociencia 49: 499-511. Raven PH. 2021. Driven by Nature: a Personal Journey from Shanghai to Botany and Global Sustainability. Missouri Botanical Garden Press, Saint Louis, Missouri, USA. 359 p. Reynoso-Rosales WH, Vazquez-Cruz ML, Rivera- Arroyo RC. 2016. Estado de Conservacion, Uso, Gestion, Comercio y Cumplimiento de los Criterios de Inclusion a los Apéndices de la CITES para las Especies Claudius angustatus y Staurotypus triporcatus. Instituto de Biologia. Informe final SNIB- CONABIO, Ciudad de México, México. 202 p. Rivera-Cruz MC, Trujillo-Narcia A. 2004. Estudio de toxicidad vegetal en suelos con petrdleos nuevo e intemperizado. Interciencia 29: 369-376. Rios-Rodas L, Barragan-Vazquez M del R, Torres Pérez MA, Triana Ramirez DI. 2017. Ampliacion de distribucién de Anolis compressicauda Smith and Kerster, 1955 (Squamata: Dactyloidae) en el estado de Tabasco, México. Acta Zoolégica Mexicana Nueva Serie 33: 120-122. Sanchez-Munguia A. 2005. Uso del Suelo Agropecuario y Deforestacion en Tabasco 1950-2000. Division Académica de Ciencias Biologicas, Universidad Juarez Autonoma de Tabasco, Villahermosa, Tabasco, México. 123 p. SEMARNAT (Secretaria de Medio Ambiente y Recursos Naturales). 2010. Norma Oficial Mexicana Nom- 059-Semarnat-2010, Proteccién Ambiental-Especies Nativas de México de Flora y Fauna Silvestres- Categorias de Riesgo y Especificaciones para su Inclusion, Exclusién o Cambio-Lista de Especies en Riesgo. Diario Oficial de la Federacion, 30 de diciembre de 2010. SEMARNAT, México, DF, Mexico. Sokal RR, Michener CD. 1958. A statistical method for evaluating systematic relationships. University of Kansas Science Bulletin 38: 1,409-1,438. Soto-Huerta KA, Clause AG. 2017. Distribution and range extension of the Elegant Coralsnake, Micrurus elegans. The Southwestern Naturalist 62: 303-308. Teran-Juarez SA, Garcia-Padilla E, Mata-Silva_ V, Johnson JD, Wilson LD. 2016. The herpetofauna of Tamaulipas, Mexico: composition, distribution, and Amphib. Reptile Conserv. conservation. Mesoamerican Herpetology 3: 42-113. Torres-Hernandez LA, Ramirez-Bautista A, Cruz- Elizalde R, Hernandez-Salinas U, Berriozabal-Islas C, DeSantis DL, Johnson JD, Rocha A, Garcia- Padilla E, Mata-Silva V, et al. 2021. The herpetofauna of Veracruz, Mexico: composition, distribution, and conservation status. Amphibian & Reptile Conservation 15(2) [General Section]: 72—155 (e285). Torres-Masuera G. 2021. La “modernizacion forzada” del sureste mexicano: déja vu de la polarizacion socioambiental en nombre del desarrollo. DOSSIER: Socioenvironmental Destruction, Extractivism, and Dependency 52(4): 28-33. Townsend JH, Wilson LD. 2010. Conservation of the Honduran herpetofauna: Issues and imperatives. Pp. 460-487 In: Conservation of Mesoamerican Amphibians and Reptiles. Editors, Wilson LD, Townsend JH, Johnson JD. Eagle Mountain Publishing, LC, Eagle Mountain, Utah, USA. i-xvili + 1-812 p. Trombulak SC, Frissell CA. 2000. Review of ecological effects of roads on terrestrial and aquatic communities. Conservation Biology 14: 18-30. Tudela F. 1989. La Modernizacion Forzada del Tropico: el Caso de Tabasco. El Colegio de México, Ciudad de México, México, México. Vidal-Garcia M, Negrete MMM. 2019. Infraestructura. En: La Biodiversidad en Tabasco. Estudio de Estado. Volumen 1. CONABIO, México, DF, México. 117 p. West R, Psuty NP, Thom BG. 1985. Las Tierras Bajas de Tabasco: en el Sureste de México. 2™ Edition. Gobierno del Estado de Tabasco, Villahermosa, Tabasco, Mexico. 409 p. Wilson LD, Mata-Silva V, Johnson JD. 2013a. A conservation reassessment of the reptiles of Mexico based on the EVS measure. Contribution to Special Mexico Issue. Amphibian & Reptile Conservation 7(1): 1-47 (e61). Wilson LD, Johnson JD, Mata-Silva V. 2013b. A conservation reassessment of the amphibians of Mexico based on the EVS measure. Contribution to Special Mexico Issue. Amphibian & Reptile Conservation 7(1): 97-127 (e69). Wilson LD, Johnson JD, Porras LW, Mata-Silva V, Garcia-Padilla E. 2017. A system for categorizing the distribution of the Mesoamerican herpetofauna. Mesoamerican Herpetology 4: 901-913. Wilson LD, McCranie JR. 2004. The conservation status of the herpetofauna of Honduras. Amphibian & Reptile Conservation 3(1): 6—33 (e12). Woolrich-Pifia GA, Ramirez-Silva JP, Loc-Barragan J, Ponce Campos P, Mata-Silva V, Johnson JD, Garcia- Padilla E, Wilson LD. 2016. The herpetofauna of Nayarit, Mexico: composition, distribution, and conservation status. Mesoamerican Herpetology 3: 375-448. Woolrich-Pifia GA, Garcia-Padilla E, DeSantis DL, Johnson JD, Mata-Silva V, Wilson LD. 2017. The August 2022 | Volume 16 | Number 2 | e315 Barragan-Vazquez et al. herpetofauna of Puebla, Mexico: composition, distribution, and conservation status. Mesoamerican Herpetology 4: 790-884. Zavala-Cruz J, Castillo AO. 2003. Uso del suelo y manejo en los cordones litorales de la Reserva de la Bidsfera Pantanos de Centla, Tabasco. Universidad y Ciencia 19: 17-33. Zavala-Cruz J, Gavi-Reyes F, Adams-Schroeder RH, Ferrera-Cerrato R, Palma-Lopez DJ, Vaquera-Huerta H, Dominguez-Ezquivel JM. 2005. Derrames de petroleo en suelos y adaptacion de pastos tropicales en el activo cinco presidentes, Tabasco, México. Jerra Zavala-Cruz J, Ortiz-Pérez MA. 2019. Geomorfologia. Pp. 29-36 In: La Biodiversidad en Tabasco. Estudio de Estado. Volumen 1. CONABIO, México, DF, México. 117 p. Zenteno-Ruiz CE, Zamora LF, Cabrera S, Marsella D. 2004. Captura y uso de fuego en el aprovechamiento de las tortugas dulceacuicolas en la Reserva de la Biosfera Pantanos de Centla. Pp. 106-116 In: Etnopaisaje, Trabajo Comunitario, Manejo y Conservacion de Recursos Naturales. Editors, Gama L, Ochoa-Gaona S, Chiappy C. Universidad Juarez Autonoma de ‘Tabasco, Villahermosa, Tabasco, Latinoamericana 23. 293-302. Amphib. Reptile Conserv. México. 116 p. Ma. Del Rosario Barragan-Vazquez graduated with a Master’s degree in Environmental Sciences from the Universidad Juarez Autonoma de Tabasco (UJAT) in Villahermosa, Tabasco, Mexico. She is a full-time Research Professor at UJAT in the Biology and Environmental Management degree programs. She is interested in the study of amphibians and reptiles at the community level, and from ecological, taxonomic, and management and utilization points of view. She has undertaken academic appointments at CINVESTAV-Merida, Faculty of Sciences, UNAM, and at the Universidad Veracruzana, with work on amphibian cultivation, taxonomy, and behavior. She has participated in research projects on the herpetofauna in the municipalities of the Sierra de Tabasco, and has authored or co-authored various articles, notes, and book chapters, primarily on the community level biodiversity and population genetics in turtles. She is in charge of the Coleccion de Anfibios y Reptiles de Tabasco (CART), curated within the Division Académica de Ciencias Bioldgicas de la UJAT. Liliana Rios-Rodas has a degree in Biology from the Universidad Juarez Autonoma de Tabasco (UJAT), a Master’s degree in Agricultural Sciences and Natural Resources from the Universidad Autonoma de México, and a Ph.D. in Ecology and Management of Tropical Systems from UJAT. Her main research topics involve the ecology of communities and populations of amphibians and reptiles of Tabasco, focusing on riparian ecosystems of tropical environments. She has worked on the ecomorphology of the genus Sce/oporus, geometric morphometry in Dryophytes plicatus, and trophic ecology in Craugastor berkenbuschii. Liliana has participated in genetic conservation projects for Dermatemys mawii and Trachemys venusta, and in updating the Tabasco Amphibian and Reptile Collection (CART). She is the author or co-author of several articles on the distribution, diversity, and conservation of the herpetofauna of Tabasco. Lydia Allison Fucsko, who resides in Melbourne, Australia, is an environmental activist and amphibian conservationist. As a photographer with international publications, she has taken countless amphibian photographs, including photo galleries of frogs mostly from southeastern Australia. Dr. Fucsko has a Bachelor of Humanities from La Trobe University (Bundoora, Victoria, Australia) and a Diploma in Education from the University of Melbourne (Parkville, Victoria, Australia). She has postgraduate diplomas in computer education and in vocational education and training from the University of Melbourne (Parkville). Additionally, Dr. Fucsko has a Master’s degree in Counseling from Monash University (Clayton, Victoria, Australia). She received her Ph.D. in Environmental Education, which promoted habitat conservation, species perpetuation, and global sustainable management, from Swinburne University of Technology (Hawthorn, Victoria, Australia), while being mentored by the late Australian herpetologist and scholar Dr. Michael James Tyler (Order of Australia recipient). As a sought-after educational consultant, Dr. Fucsko has academic interests that include: clinical psychology, focusing on psychopathology; neuroscience and empathy; environmental education for sustainable development; sentient ecology; academic writing; and creative writing, which includes poetry and creative non-fiction books for children and young adults. Dr. Fucsko also is the senior author (with Boria Sax) of a chapter in the 2019 Springer Encyclopedia of Sustainability in Higher Education entitled “Learning activities for environmental education for sustainable development.” Recently, Dr. Fucsko has co-authored an obituary of Jaime D. Villa, a study of the introduced Mesoamerican herpetofauna, a treatment of the conservation prospects of the Mesoamerican salamander fauna, papers on the herpetofauna of Veracruz and Querétaro, Mexico, a review of the book Advances in Coralsnake Biology, and a study on the biological and cultural diversity of Oaxaca, Mexico, among several other academic papers. In 2020, the species Zantilla lydia, with the suggested common name of Lydia’s Little Snake, was named in her honor. 59 August 2022 | Volume 16 | Number 2 | e315 Amphib. Reptile Conserv. The herpetofauna of Tabasco, Mexico Louis W. Porras graduated in 1971 with a degree in Biology from what today is known as Miami- Dade College in Miami, Florida, USA. Over his career he has authored or co-authored over 60 academic publications, including the descriptions of two new species, and two taxa have been named in his honor. Louis developed an interest in herpetology at an early age in his native Costa Rica. His passion for the field led him to travel to many remote areas, including throughout the Bahamas, the United States, Mesoamerica, and parts of South America. In 1968, he worked at the Houston Zoological Gardens, and from 1982 to 1984 at Utah’s Hogle Zoo. In 1976, he attended the inaugural meeting of the International Herpetological Symposium (IHS), and later served the group as Vice-President and President. In 1993, along with Gordon W. Schuett, he helped launch the journal Herpetological Natural History, and for IHS’ 20" anniversary, in recognition of his contributions, three former Presidents dedicated the book Advances in Herpetoculture in his honor. Louis’ career in publishing began in 1995, when as a member of Canyonlands Publishing Group he helped publish Fauna magazine. In 2002 he founded Eagle Mountain Publishing, LC, which has published such herpetological titles as Biology of the Vipers (2002), Biology of the Boas and Pythons (2007), Amphibians, Reptiles, and Turtles in Kansas (2010), Conservation of Mesoamerican Amphibians and Reptiles (2010), and Amphibians and Reptiles of San Luis Potosi (2013). From 2014 to 2018 he was the Publisher and Managing Editor of the journal Mesoamerican Herpetology, and recently he was the Publisher and Co-editor of the book Advances in Coralsnake Biology: with an Emphasis on South America. Vicente Mata-Silva is a herpetologist originally from Rio Grande, Oaxaca, Mexico. His interests include ecology, conservation, natural history, and biogeography of the herpetofaunas of Mexico, Central America, and the southwestern United States. He received a B.S. degree from the Universidad Nacional Autonoma de México (UNAM), and MLS. and Ph.D. degrees from the University of Texas at El Paso (UTEP). Vicente is an Assistant Professor of Biological Sciences at UTEP, in the Ecology and Evolutionary Biology Program, and Co-Director of UTEP’s Indio Mountains Research Station, located in the Chihuahuan Desert of Trans-Pecos, Texas, USA. To date, Vicente has authored or co- authored over 100 peer-reviewed scientific publications. He also was the Distribution Notes Section Editor for the journal Mesoamerican Herpetology, and is currently Associate Editor for the journal Herpetological Review. Arturo Rocha is a Ph.D. student in the Ecology and Evolutionary Biology program at the University of Texas at El Paso. His interests include the study of the biogeography, physiology, and ecology of amphibians and reptiles in the southwestern United States and Mexico. A graduate of the University of Texas at El Paso, his thesis centered on the spatial ecology of the Trans-Pecos Rat Snake (Bogertophis subocularis) in the northern Chihuahuan Desert. To date, he has authored or co-authored over 20 peer-reviewed scientific publications. Dominic L. DeSantis is an Assistant Professor of Biology at Georgia College and State University, Milledgeville, Georgia, USA, in the Department of Biological and Environmental Sciences. Dominic’s research interests broadly include the behavioral ecology, conservation biology, and natural history of herpetofauna. In addition to ongoing collaborative projects associated with the “| Mesoamerican Research Group, much of Dominic’s current research focuses on using novel animal- borne sensor technologies to study the behavior of snakes in the field. While completing his Ph.D. at the University of Texas at El Paso, Dominic accompanied Vicente Mata-Silva, Eli Garcia-Padilla, and Larry David Wilson on survey and collecting expeditions to Oaxaca in 2015, 2016, and 2017, and is a co-author on numerous natural history publications produced from those visits, including an invited book chapter on the conservation outlook for herpetofauna in the Sierra Madre del Sur of Oaxaca. 60 August 2022 | Volume 16 | Number 2 | e315 Amphib. Reptile Conserv. Barragan-Vazquez et al. Eli Garcia-Padilla is a Social Biologist and Professional Photographer with more than 12 years of experience in the formal study and photo documentation of the biological and cultural diversity of Mexico. He has published one book, entitled Mexican Biodiversity: the Snake, the Jaguar and the Quetzal, and more than 100 formal contributions on knowledge, the communication of science and the conservation of Mesoamerican biodiversity. Since 2006, he has been exploring Oaxaca and Chiapas, which are the most biodiverse and multicultural states in Mexico. In 2017, he began to enter the mythical region of Los Chimalapas in the Isthmus of Tehuantepec, which is the most biologically rich region in all of Mexico, under a community social conservation scheme. Eli has published his photographic work in prestigious magazines such as National Geographic in Spanish and Cuartoscuro. In 2020, he co-founded the Mesoamerican Biodiversity initiative with the aim of creating a community around the dissemination of the most important wealth of Mexico, which is its biodiversity and its culture. His writings are published regularly in Oaxaca Media, the Jornada Ecologica and the Ojarasca Supplement of La Jornada. Jerry D. Johnson is Professor of Biological Sciences at The University of Texas at El Paso, and has extensive experience studying the herpetofauna of Mesoamerica, especially that of southern Mexico. Jerry is the Director of the 40,000-acre Indio Mountains Research Station, and was a co- editor on the book Conservation of Mesoamerican Amphibians and Reptiles and co-author of four of its chapters. He was also the senior author of the recent paper “A conservation reassessment of the Central American herpetofauna based on the EVS measure” and is Mesoamerica/Caribbean editor for the Geographic Distribution section of Herpetological Review. Jerry has authored or co-authored over 130 peer-reviewed papers, including two 2010 articles, “Geographic distribution and conservation of the herpetofauna of southeastern Mexico” and “Distributional patterns of the herpetofauna of Mesoamerica, a Biodiversity Hotspot.” One species, Zantilla johnsoni, has been named in his honor. Presently, he is an Associate Editor and Co-chair of the Taxonomic Board for the journal Mesoamerican Herpetology. Larry David Wilson is a herpetologist with lengthy experience in Mesoamerica. He was born in Taylorville, Illinois, USA, and received his university education at the University of Illinois at Champaign-Urbana (B.S. degree) and at Louisiana State University in Baton Rouge (M.S. and Ph.D. degrees). He has authored or co-authored more than 460 peer-reviewed papers and books on herpetology. Larry was the senior editor of Conservation of Mesoamerican Amphibians and Reptiles and a co-author of seven of its chapters. His other books include The Snakes of Honduras, Middle American Herpetology, The Amphibians of Honduras, Amphibians & Reptiles of the Bay Islands and Cayos Cochinos, Honduras, The Amphibians and Reptiles of the Honduran Mosquitia, and Guide to the Amphibians & Reptiles of Cusuco National Park, Honduras. To date, he has authored or co-authored the descriptions of 75 currently-recognized herpetofaunal species, and seven species have been named in his honor, including the anuran Craugastor lauraster, the lizard Norops wilsoni, and the snakes Oxybelis wilsoni, Myriopholis wilsoni, and Cerrophidion wilsoni. In 2005, he was designated a Distinguished Scholar in the Field of Herpetology at the Kendall Campus of Miami- Dade College in Miami, Florida, USA. Currently, Larry is a Co-chair of the Taxonomic Board for the website Mesoamerican Herpetology. 61 August 2022 | Volume 16 | Number 2 | e315 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptile Conservation 16(2) [General Section]: 62—68 (e316). Reproductive characteristics of the Burmese Narrow-headed Softshell Turtle, Chitra vandijki, in captivity ‘Hong Xiaoyou, *2*Zhu Xinping, ‘Chen Chen, ‘Cai Xiaodan, *Li Yongming, and ‘Li Xinping 'Key Laboratory of Tropical and Subtropical Fishery Resources Application and Cultivation, Ministry of Agriculture, Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences, Guangzhou, CHINA *College of Life Science and Technology, Shanghai Ocean University, Shanghai, CHINA 3Xishuangbanna Indigenous Fish Research and Breeding Center, Dai Autonomous Prefecture of Xishuangbanna, CHINA +Xishuangbanna Dai Autonomous Prefecture Fishery Technology Extension Station, Dai Autonomous Prefecture of Xishuangbanna, CHINA Abstract.—The purpose of this study was to provide basic data for the breeding biology of Chitra vandijki and to contribute to the conservation of this species. The Burmese Narrow-headed Softshell Turtle, Chitra vandijki, is a CITES Appendix I-listed species, and biological information on wild and captive C. vandijki is relatively scarce. In 2019, we studied the reproductive biology of two C. vandijki specimens (a female and a male) that had been in captivity for approximately 25 years. The oviposition period of the domesticated female C. vandijki was from June to August. The female laid eggs at night, and no egg protection behavior was observed. The female C. vandijki laid five clutches of eggs in a year representing 564 eggs in total, with 100-131 eggs/clutch, and the interval between successive clutches was 9-28 d. The fertilization rate of C. vandijki was 90.4%, and the hatching rate was 38.6%. The eggs were spherical and rigid, with an average mass of 15.04 + 0.65 g and an average diameter of 2.96 + 0.22 cm. The average hatching period of C. vandijki was 65.3 d at 28.0—29.0 °C, and the average accumulated incubation temperature was 44,688.6 °C-h. The average mass of newly hatched neonates was 10.51 + 0.57 g, and the average mass of juvenile C. vandijki reached 150.37 + 53.86 g after one year of feeding live fry in a greenhouse. Keywords. Burmese Narrow-headed Softshell Turtle; captive-breeding; Chitra vandijki, conservation; egg laying; juvenile; threatened species Citation: Hong X, Zhu X, Chen C, Cai X, Li Y, Li X. 2022. Reproductive characteristics of the Burmese Narrow-headed Softshell Turtle, Chitra vandijki, in captivity. Amphibian & Reptile Conservation 16(2) [General Section]: 62-68 (e316). Copyright: © 2022 Hong et al. This is an open access article distributed under the terms of the Creative Commons Attribution License [Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are 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: amphibian-reptile-conservation.org. Accepted: 22 February 2022; Published: 1 September 2022 Introduction The Burmese Narrow-headed Softshell Turtle, Chitra vandijki, is a large turtle with a straight-line carapace length of up to 1 m. It is listed on CITES Appendix I and classified as Critically Endangered (CR) on the JUCN Red List of Threatened Species (Rhodin et al. 2018; Platt et al. 2021). It is mainly distributed in rivers in Myanmar and Thailand (Platt et al. 2014). The abundance and distribution of C. vandijki have been sharply reduced because of human hunting and habitat destruction (Kuchling et al. 2004; Platt et al. 2005, 2014). Because little 1s known about the turtle’s ecological habits, successful cases of artificial breeding are very few and knowledge of its breeding biology is extremely lacking (Platt et al. 2018, 2020). Because the external appearance of C. vandijki 1s similar to the Asian Giant Softshell Turtle, Pelochelys cantorii, and given the demands of the Chinese wild animal market, C. vandijki has been illegally traded to China as food or for rearing in the last century. Although P. cantorii in China is Critically Endangered (Gong et al. 2017; Hong et al. 2019; Wu et al. 2020), we have successfully carried out artificial breeding of six P. cantorii (three females and three males) turtles since 2014 (Zhu et al. 2015; Hong 2020). At present, we have bred more than 800 P. cantorii between 1 and 6 years old (Ministry of Agriculture and Rural Affairs of People’s Republic China 2020). Based on our successful experience in the artificial breeding of captive P. cantorii, we carried out research on the reproductive biology of two captive C. vandijki, and the mitochondrial genomes of the individual hatched offspring confirmed their identity as Burmese Small- headed Turtles (Chen et al. 2021). The findings of this study enrich the basic biological data of C. vandijki and provide a theoretical basis for its conservation biology. Correspondence. hongxiaoyoul216@163.com (HX); zhuxinping_1964@163.com (ZX), chenchen3729@163.com (CC), cxd430@163.com (CX); 1258313402@qq.com (LY); 13988165908@163.com (LX) Amphib. Reptile Conserv. September 2022 | Volume 16 | Number 2 | e316 Hong et al. Materials and Methods Captive Care, Conditions, and Management Two C. vandijki, one male and one female, were accidentally captured and rescued from the waters of the Mekong River at the border between Myanmar, Laos, and Guanli Town, Mengla County, Xishuangbanna Prefecture, Yunnan Province, China in 1995. Their body weights at the time of capture were approximately 3 kg and 7 kg, respectively. The two C. vandijki were raised in an outdoor fish pond at an elevation of 570 m asl (21°35’40.84’"N 101°14’5.02”E) in Xishuangbanna Prefecture, Yunnan Province, China. This region has a north tropical and south tropical humid monsoon climate, which includes a long summer without winter. The annual average temperature 1s between 18.6 and 21.9 °C, and the annual average precipitation 1s between 1,200 and 1,700 mm. The dimensions of the pond were 30 m =< 25 m, the water depth was 1.2 m, and the bottom mud was 30-40 cm thick (Fig. 1A). In 2012, a 25 m x 2 m nesting sand pond with a depth of 60 cm was built on the side of the main pond, and there was a shed above the sand pool for shade. Tiles were used to build an incline of about 30° so the turtles could climb up from the water to the sand pond (Fig. 1A—B). In the pond, Tilapia Oreochromis niloticus, Carp Cyprinus carpio, and Crucian Carp Carassius auratus were cultured together, and the C. vandijki lived by feeding on these fish. box, (D) rearing facilities. Amphib. Reptile Conserv. Fig. 1. Artificial rearing facility of Burmese Narrow-headed Softshell Turtles: (A) breeding pond, (B) nesting area, (C) incubation In December 2018, the fish pound was cleared, and the large fish were removed. From February to May 2019, 200 kg of live fish fry, including Carp, Crucian Carp, and Mud Carp (Cirrhinus molitorella) with body lengths of 3—5 cm, were regularly added to the pond to serve as the food for improving the cultivation of C. vandijki. Collection and Hatching of Eggs In April 2019, the stones and plants in the spawning sand pool were cleared, and the sand was raked loose and sieved. From May to July, water was sprayed irregularly into the spawning sand pond to ensure that the sand remained damp. A surveillance camera was installed above the spawning pool to observe the oviposition activity of C. vandijki. For the first clutch, the eggs were incubated in situ for 25 days, the clutch was dug manually, and artificial incubation was continued. For the other clutches, within 16—24 h after the turtle had laid the eggs, they were collected by excavating the nest. The numbers of eggs and fertilized eggs were counted and recorded. The diameter of each egg was measured with a Vernier caliper (+ 0.01 cm), and the egg weight was measured using an electronic balance (+ 0.01 g). The incubators for fertilized eggs were plastic boxes with dimensions of 57 x 41 x 36 cm. The medium was sieved fine river sand, and the moisture content of the sand was 8—10% (weight ratio, Fig. 1C). The thickness of the sand pile was approximately 15 cm. The fertilized eggs in a given clutch were arranged on the sand pile in >) — September 2022 | Volume 16 | Number 2 | e316 Reproductive characteristics of the Burmese Narrow-headed Softshell Turtle the incubator and covered with 2 cm of fine sand with the same dampness. The incubator was then covered. The temperature of the incubator was controlled by an indoor air conditioner, and maintained at 28.0—29.0 °C. Water was sprayed regularly onto the sand to control the humidity. Cultivation of Hatchlings After emergence, the hatchlings were observed and photographed. The body mass of each hatchling was obtained using an electronic balance (+ 0.01 g). The length and width of the carapace and the length and width of the snout of each juvenile were measured using a caliper (+ 0.01 cm). Hatchlings were reared according to the rearing method of Asian Giant Softshell Turtle (P. cantorii) hatchlings (Hong et al. 2018), and cultured in six custom-designed round buckets with a diameter of 1.2 m and a water depth of 0.5 m. The cultivation density was 25-30 individuals/m’, the bottom of the bucket contained 15 cm of fine sand, and the water was filtered through circulation (Fig. 1D). The pH of the water was measured regularly and adjusted to 7.0—7.5 with quicklime. The neonates were fed live Mosquito Fish, Gambusia affinis. The temperature was controlled by air conditioning and the water temperature was maintained at 26.0—31.0 °C. In July 2020, five juvenile a C. vandijki were randomly selected from each barrel. The body mass and length and width of the carapace of 30 hatchlings were measured. Statistics The data shown below and labeled as “this study” are expressed as the mean + SD, and were compared and analyzed using ANOVA. Statistical analysis was conducted using IBM SPSS 23.0 software. All statistical tests were two-tailed, and the significance level was set as P<0.05. Results and Discussion Morphology of the Parents In 2012, the body mass of the female C. vandijki was 38.0 kg. On 6 December 2018, the body masses of the female and male parental C. vandijki were 59.2 kg and 40.0 kg, respectively. There were irregular, slightly fuzzy yellow stripes on the adult carapace (Fig. 2A). The longitudinal stripes on the neck and back merged behind the head, and the neck stripes were more obvious than the stripes on the back (Fig. 2B). The neck was not obviously separated from the anterior edge of the carapace. The neck of the Fig. 2. Characteristics of Burmese Narrow-headed Softshell Turtles: (A) back; (B) head and neck, close-up; (C) male, ventral view; (D) female, ventral view. Amphib. Reptile Conserv. September 2022 | Volume 16 | Number 2 | e316 Hong et al. Table 1. Egg laying and hatching data of Burmese Narrow-headed Softshell Turtles in 2019. Number of Date of egg laying Clutch size fertilized eggs/ clutch 3 June 2019 101 69 1 July 2019 131 117 18 July 2019 110 105 27 July 2019 122 120 9 August 2019 100 99 Total 564 510 adult C. vandijki was slightly short and could not turn to the middle or rear of the carapace. The front edge of the carapace was flat, without folds or warts. The head of the adult C. vandijki was small, and the snout was short (Fig. 2A—B). The main morphological difference between the males and females was the tail. The tail of male C. vandijki was thick and long, extending out from the edge of carapace, while the tail of female C. vandijki was thin and short, extending no longer than the edge of the carapace (Fig. 2C-D). Egg Laying In 2004 and 2006, 10 and 20 eggs were laid, respectively, in the culture pond water of the parent C. vandijki. After the construction of the spawning sand pond, 98 C. vandijki hatchlings were collected from the fish pond in September 2018. However, they all died within 30 d of captive feeding following the cultivation method of the Chinese Softshell Turtle Pe/odiscus sinensis (Zhao et al. 1997). Beginning in June 2019, the female turtle was observed climbing up the sand dunes at night, looking for nesting sites. On the nights of 3 June to 9 August, the female turtle laid five clutches of eggs, for a total of 564 eges. The four intervals between the five clutches were 28 d, 17 d, 9 d, and 13 d, respectively (Table 1), and the average interval was 16.75 + 8.18 d. The C. vandijki eggs were nearly round and rigid, and the calcareous layer of the eggshell was thin. The number of eggs in each clutch varied from 100 to 131 (Table 1), with an average of 112.8 + 13.5 eggs/clutch. We randomly selected 40 eggs from the first clutch and another 40 eggs from the second clutch for measurements. Fertilization rate Number of hatchings/ Hatching rate (%) clutch (%) 69.3 43 62.3 85.4 80 80.0 95.5 54 68.4 98.4 20 9.1 99.0 90.4 197 38.6 The egg masses were 13.37—16.47 g (15.04 + 0.65 g) and egg diameters were 2.76—3.15 cm (2.96 + 0.22 cm). According to the average egg weight, the total weight of the five clutches of eggs could be estimated as 8,482.56 g, accounting for 14.335% of the maternal body weight. Hatching and Characteristics of Hatchlings The five clutches included 510 fertilized eggs, and the fertilization rate was 90.4%. In total, 197 hatchlings emerged, for a hatching rate of 38.6% (Table 1). The average incubation period of the fertilized eges was 65.3 + 5.4 d, and the average accumulated incubation temperature was 44,688.6 °C-h at a room temperature of 28.0—29.0 °C (28.51 °C on average) based on the hatching data of the second clutch. Under artificial conditions, the hatching rate for the last four clutches of eggs was 34.9%. The carapace of the newly hatched neonate C. vandijki was approximately round, with obvious yellow stripes on the neonate’s back, neck, and limbs. The carapace was covered with small protuberances and the posterior edge was yellow without stripes (Fig. 3). The newly hatched neonate C. vandijki weighed 9.44—11.75 g (10.51 + 0.57 g, n = 60, 30 neonates in the first clutch and another 30 neonates in the second clutch). The length of the neonate C. vandijki carapace was 4.18—4.70 cm (4.41 + 0.13 cm), and the width of the carapace was 3.75—4.24 cm (4.01 + 0.10 cm). The length of the snout of neonate C. vandijki was 0.14—0.22 cm (0.16 + 0.02 cm) and the width of the snout was 0.15—0.24 cm (0.19 + 0.02 cm). The reproductive biology data for four species of softshell turtles bred in captivity are shown in Table 2. C. vandijki, P. cantorii, and the Siamese Narrow-headed Softshell Turtle (Chitra chitra) are all large Trionychidae Table 2. Comparison of the reproductive biology of four species of softshell turtles in captivity. Burmese Narrow- Siamese Narrow- Asian Giant Softshell Chinese Softshell Turtle headed Softshell headed Softshell Turtle Turtle (Hong et al. 2018; (Yang et al. 1999; Zhou Turtle (this study) (Kitimasak et al. 2003) Hong 2020) 2004) Parent sample size (9, 3) Vol 2,4 Io >800, >100 Number of clutches/year 5 3-4 46 5-7 Clutch size 100-133 40-88 32-55 8-25 Egg diameter (cm) 2.96 + 0.22 3.32+40.15 3.10+0.18 2.00-2.40 Egg mass (g) 15.04 + 0.65 19.00 + 1.67 16.82 + 1.99 3.55-6.77 Mass of neonate (g) 10.51 + 0.57 13.10 + 1.03 13.60 + 0.85 2.33-4.83 HCC UnUar Mn UpaLlon 44,688.60 Not reported 44,886.50 36,000 temperature (°C-h) Amphib. Reptile Conserv. 65 September 2022 | Volume 16 | Number 2 | e316 Reproductive characteristics of the Burmese Narrow-headed Softshell Turtle iW BN) ut Wi i{ (Ct AAV HAN (NM iil : AC (i | quae «6 Fig. 3. Hatchling of Burmese Narrow-headed Softshell Turtle. animals with similar breeding biology, but they are very different from the Chinese Softshell Turtle, Pelodiscus sinensis. There are very limited breeding data for wild C. vandijki. Platt et al. (2020) reported that the numbers of eggs in four collected clutches were 58, 76, 89, and 102; and the diameter of the eggs was 2.01—3.66 cm (2.60—2.95 cm on average). The length of the carapace of the hatchlings was 2.73-4.10 cm (3.52 + 0.35 cm), and the width of the carapace was 2.75—3.86 cm (3.39 + 0.26 cm). However, data for the parent C. vandijki were not reported in that study. In general, in our study, the clutch size, diameter of eggs, and body size of neonate C. vandijki were all higher than those reported by Platt et al. (2020), suggesting that the maternal size in our study might be larger and/or that the nutritional status of C. vandijki in captivity is better than that in the field (Gibbons et al. 1990; Litzgus et al. 2008; Hong et al. 2018). Based on the egg laying and hatching data, we found that the number of eggs in each clutch was relatively constant and the fertilization rate remained at a high level. However, the hatching rates of the last two clutches were relatively low (Table 1). The total mass of the five clutches of eggs accounted for approximately 14% of the body mass of the female C. vandijki. We considered that the low hatching rate of the last two clutches may be due to the influence of oviposition frequency and the availability of reproductive resources for the female C. vandijki (Jackson and Prange 1979; Ferguson et al. 1982). We speculate that too much of the energy of the female C. vandijki had been consumed by the late stage of oviposition, and the spawning intervals between the last two clutches were short (9 d and 13 d), which may have resulted in an insufficient energy supply for the development of the eggs, leading to improper development of the embryos. This phenomenon has been reported for P. cantorii (Hong et al. 2018). Growth of Juveniles By July 2020, 180 juvenile C. vandijki survived, for a survival rate of 91.4%. The body weight of juvenile C. Amphib. Reptile Conserv. vandijki was 49.80-311.10 g (150.37 + 53.86 g, n = 30), the length of the juvenile carapace was 7.54—13.19 cm (11.05 + 1.66 cm), the width of the carapace was 7.56— 12.87 cm (10.84 + 1.53 cm), the length of the juvenile snout was 0.22—0.40 cm (0.29 + 0.05 cm), and the width of the snout was 0.23—-0.43 cm (0.31 + 0.06 cm). The living habits of P. cantorii and C. vandijki are similar; thus, the juvenile breeding method of P. cantorii is also suitable for C. vandijki and 1s ideal in terms of survival rate and growth rate. P. cantorii is endangered due to overhunting and habitat destruction, and China has increased its artificial conservation efforts to gradually restore wild resources, which has achieved initial results (Ministry of Agriculture and Rural Affairs of People’s Republic China 2020). Currently, the C. vandijki population has been greatly reduced, and so it is also in a critical condition state. For this reason, referring to the Chinese protection strategy for P. cantorii, we can use the limited captive population of C. vandijki to carry out conservation biological research in order to achieve the artificial conservation of this species, and subsequently release the captive turtles into the wild to restore the wild population. Acknowledgments.—Thanks are due to the staff of Xishuangbanna Dai Autonomous Prefecture Fishery Technology Extension Station for the breeding turtles for more than 25 years. We also thank all those who have contributed to the conservation of Pelochelys cantorii and Chitra vandijki. The financial support for this work was provided by the Foshan City Financial Special Fund-2020, the Guangdong Agricultural Science and Technology Demonstration City Project, and the Central Public-interest Scientific Institution Basal Research Fund, CAFS (#2019ZD04, #2020TD35 and #2020ZJTDO1). Literature Cited Chen C, Hong X, Li W, Yu L, Zhu X. 2021. Complete mitochondrial genome and the phylogenetic position of the Burmese Narrow-headed Softshell Turtle Chitra vandijki (Testudines: Trionychidae). Mitochondrial DNA Part B 6(3): 1,216-1,218. Ferguson GW, Brown KL, DeMarco VG. 1982. Selective basis for the evolution of variable egg and hatchling size in some iguanid lizards. Herpetologica 38(1).: 178-188. Gibbons JW, Greene JL. 1990. Reproduction in the Slider and other species of turtles. Pp. 124-134 In: Life History and Ecology of the Slider Turtle. Editor, Gibbons JW. Smithsonian Institution Press, Washington, DC, USA. 368 p. Gong S, Shi H, Jiang A, Jonathan JF, Daniel G, Wang J. 2017. Disappearance of Endangered turtles within China’s nature reserves. Current Biology 27(5): 170— 171. Hong X. 2020. Conservation biology research of the Asian Giant Softshell Turtle, Pelochelys cantorii. Ph.D. Dissertation, Shanghai Ocean University, Shanghai, China. [In Chinese with English abstract]. Hong X, Cai X, Chen C, Liu X, Zhao J, Qiu Q, Zhu X. 2019. Conservation status of the Asian Giant Softshell September 2022 | Volume 16 | Number 2 | e316 Hong et al. Turtle (Pelochelys cantorii) in China. Chelonian Conservation and Biology 18(1): 68—74. Hong X, Zhu X, Chen C, Zhao J, Ye Z, Qiu Q. 2018. Reproductive traits of captive Asian Giant Softshell Turtles, Pelochelys cantorii. Acta Hydrobiologica Sinica 42: 794-799. [In Chinese with English abstract]. Jackson DC, Prange HD. 1979. Ventilation and gas exchange during rest and exercise in adult Green Sea Turtles. Journal of Comparative Physiology B 134(A4): 315-319. Kitimasak W, Thirakhupt K, Moll DL. 2003. Captive breeding of the Siamese Narrow-headed Softshell Turtle, Chitra chitra Nutphand, 1986 (Testudines: Trionychidae). Thai Journal of Agricultural Sciences 36: 141-154. Kuchling G, Win KK, Lwin T, Min SA, Myo KM, Khaing TT, Mar WM, Khaing TT. 2004. The softshell turtles and their exploitation at the upper Chindwin River, Myanmar: range extensions for Amyda cartilaginea, Chitra vandijki, and Nilssonia formosa. Salamandra 40: 281-296. Litzgus JD, Bolton F, Schulte-Hostedde AI. 2008. Reproductive output depends on body condition in Spotted Turtles (Clemmys guttata). Copeia 2008(1): 86-92. Ministry of Agriculture and Rural Affairs of People’s Republic China. 2020. Asian Giant Softshell Turtles, Pelochelys cantorii: Adaptive Protection Activities held in Foshan, Guangdong. Available: http://www.cjyzbgs.moa. gov.cn/gzdt/202009/ t20200928 6353394.htm [Accessed: 26 October 2020]. [In Chinese]. Platt SG, Win MM, Platt K, Haislip NA, Rainwater TR. 2020. Aspects of the reproductive biology of Chitra vandijki with a description of neonates. Salamandra 56: 66-70. Platt SG, Win KK, Khaing LL, Myo KM, Rainwater TR. 2005. Noteworthy records and exploitation of chelonians from the Ayeyarwady, Chindwin, and Dokhtawady rivers, Myanmar. Chelonian Conservation and Biology 4(4): 942-948. Amphib. Reptile Conserv. 67 Platt SG, Platt K, Win KK, Rainwater TR. 2014. Chitra vandijki McCord and Pritchard, 2003, Burmese Narrow-Headed Softshell Turtle. Pp. 074.1-074.7 In: Conservation Biology of Freshwater Turtles and Tortoises: a Compilation Project of the IUCN/ SSC Tortoise and Freshwater Turtle Specialist Group. Chelonian Research Monographs Number 5. Editors, Rhodin AGJ, Pritchard PCH, van Dijk PP, Saumure RA, BuhImann KA, Iverson JB, Mittermeier RA. Chelonian Research Foundation, Lunenburg, Massachusetts, USA. Platt SG, Zug GR, Platt K, Win KK, Myo KM, Soe MM, Lwin T, Win MM, Aung SHN, Kyaw NW, et al. 2018. Field records of turtles, snakes, and lizards in Myanmar (2009-2017), with natural history observations and notes on folk herpetological knowledge. Natural History Bulletin of the Siam Society 63: 67-114. Platt K, Praschag P, Horne BD. 2021. Chitra vandijki. The IUCN Red List of Threatened Species 2021: e.T170525A1316195 Rhodin AJ, Stanford CB, van Dijk PP, Eisemberg C, Luiselli L, Mittermeier RA, Hudson R, Horne BD, Goode EV, Kuchling G, et al. 2018. Global conservation status of turtles and tortoises (order Testudines). Chelonian Conservation and Biology 17: 135-161. Wu J, Wu Y, Rao D, Zhou T, Gong S. 2020. China’s wild turtles at risk of extinction. Science 368(6493): 838. Yang Z, Niu C, Sun R. 1999. Advances in studies of the Chinese Soft-Shelled Turtle 7rionyx sinensis. Biology. Journal of Zoology 34(6): 41-44. [In Chinese]. Zhao C, Huang T, Zhou J. 1997. Turtle pool renovation. Pp. 12-35 In: Artificial Breeding of Chinese Soft- Shelled Turtle and New Technology for Disease Control. Editor, Pan Q. Rural Reading Publishing House, Beijing, China. 173 p. Zhou G. 2004. Technique of reproduction and juvenile rearing of Trionyx sinensis. Shangdong Fisheries 21(2): 1+. [In Chinese with English abstract]. Zhu X, Hong X, Zhao J, Liang J, Feng Z. 2015. Reproduction of captive Asian Giant Softshell Turtles, Pelochelys cantorii. Chelonian Conservation and Biology 14(2): 143-147. September 2022 | Volume 16 | Number 2 | e316 Reproductive characteristics of the Burmese Narrow-headed Softshell Turtle Amphib. Reptile Conserv. Hong Xiaoyou completed his B.A. degree at Huazhong Agriculture University, People’s Republic of China, and his M.S.A. and Ph.D. degrees in Conservation Biology from Shanghai Ocean University, People’s Republic of China. In 2011, he joined the Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences in Guangzhou, where his research is focused on the exploitation, conservation of genetic resources, artificial breeding, and genetic improvement of turtles. In particular, Dr. Hong has successfully bred Pe/ochelys cantorii in captivity for the first time in China. Zhu Xinping received his Ph.D. degree in Genetics from the Institute of Hydrobiology, Chinese Academy of Sciences, People’s Republic of China, in 2004. He has worked at the Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences in Guangzhou since 1988. Prof. Zhu is now the Deputy Director of the Research Institute. The research scope of the team he leads focuses on conservation and breeding of Pelochelys cantorii, the developmental mechanisms of turtle germ cells and transplantation technology, breeding of high-fecundity Mauremys mutica, and early propagation and breeding of fast-growing Pelodiscus sinensis. Chen Chen completed his B.A. degree at Ludong University, People’s Republic of China, and his M.S. and Ph.D. degrees at the Ocean University of China, People’s Republic of China. After graduation, Dr. Chen joined the Laboratory of Aquatic Germplasm Resources and Genetic Breeding at the Chinese Academy of Fishery Sciences, Pearl River Fisheries Research Institute in Guangzhou. Recently, his research direction is in turtle population genetics, focusing on the allele distributions of turtle populations and metapopulation dynamics. Cai Xiaodan received her M.S. degree in Fishery Resources at Nanjing Agricultural University, People’s Republic of China. She joined the Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences in Guangzhou in 2007, where her work is focused on river fisheries management and aquatic life protection, with a special interest in Endangered aquatic wildlife protection. Li Yongming graduated in 1988 from the Yunnan Agricultural School, People’s Republic of China, majoring in Freshwater Aquaculture, and has been engaged in the domestication and breeding of indigenous fishes in the Lancang River. As a government employee, Mr. Li conducts fishery supervision and management as well as aquatic wildlife protection work. Li Xinping is a farmer who cultivates aquatic seedlings, with a special interest in the protection and breeding of turtles. Mr. Li has been rearing the two captive Chitra vandijki specimens used in this study for more than 25 years. 68 September 2022 | Volume 16 | Number 2 | e316 Amphibian & Reptile Conservation 16(2) [General Section]: 69-87 (e317). Official journal website: amphibian-reptile-conservation.org Applying population genetics to define the units for conservation management in the European Tree Frog, Hyla arborea ‘Astrid Krug, ‘?Jana Auffarth, and '*Heike Prohl ‘Institute of Zoology, University of Veterinary Medicine, Biinteweg 17, 30559 Hannover GERMANY *AquaEcology GmbH & Co. KG, Steinkamp 19, 26125 Oldenburg, GERMANY Abstract.—Population genetic analyses are a powerful tool for obtaining information about cryptic genetic lineages, population structure, and the distribution of intra- and interpopulation genetic diversity across the landscape. This knowledge is crucial for establishing units for the conservation management of endangered species. Species with limited dispersal capacities, such as amphibians, are particularly affected by habitat fragmentation and reductions in gene flow among isolated populations. The European Tree Frog, Hyla arborea, has suffered from dramatic population declines in the last decades and is categorized as Vulnerable to Critically Endangered in its north-western distribution range. In Lower Saxony (Germany), the current distribution of the tree frog is fragmented. In this study, we aimed to assess the population structure, genetic diversity, gene flow, and migration rates in order to define the units for conservation management. Across a distribution area of 250 km’, frogs were sampled at 14 localities and genotyped at seven microsatellite loci, and the mtDNA cytochrome b gene was sequenced for a subsample. Whereas microsatellite pairwise D,,,and F,, values showed genetic differentiation among nearly all sampled populations, Bayesian analyses assigned the 14 localities to two distinct genetic clusters including seven subclusters. Together with a slight correlation between geographic and genetic distance, the population structure indicates ongoing fragmentation. The cytochrome b haplotype distribution does not indicate divergence into mtlineages, but highlights the former connection of populations along the river Elbe. The results of this study suggest that the intense anthropogenic pressures in this area over the last decades have had negative genetic consequences for this species. The fragmented population structure calls for reconnection of the isolated occurrences by the implementation of conservation measures. Keywords. Amphibian, Bayesian assignment, conservation genetics, genetic diversity, population fragmentation, population structure Citation: Krug A, Auffarth J, Préhl H. 2022. Applying population genetics to define the units for conservation management in the European Tree Frog, Hyla arborea. Amphibian & Reptile Conservation 16(2) [General Section]: 69-87 (e317). Copyright: © 2022 Krug et al. This is an open access article distributed under the terms of the Creative Commons Attribution License [Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are 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: amphibian-reptile-conservation.org. Accepted: 14 September 2022; Published: 24 October 2022 When reconnection of the habitats of endangered Introduction Genetic diversity and connectivity mediated by migrating individuals between populations are critical for the maintenance of many threatened species and can be evaluated by population and landscape genetic analyses (Shaffer et al. 2015). Loss of connectivity disrupts gene flow between formerly connected habitats and leads to the isolation of populations. Isolation in turn imposes a more rapid erosion of genetic diversity, exacerbating the effects of genetic drift and inbreeding on local gene pools (Andersen et al. 2004; Crnokrak and Roff 1999; Hedrick and Kalinowski 2000; Luquet et al. 2011). Species 1S necessary, it 1s essential to determine the genetic structures and migration patterns for effective conservation management. This information can be used to delineate conservation units (e.g., Palsboll et al. 2007), even though the concepts that are applied to define them are somewhat uneven among studies and taxa (Shaffer et al. 2015). While Evolutionary Significant Units (ESUs) are used to delineate entities which possess a long (evolutionary) history (Crandall et al. 2000; Moritz 1994), management efforts are often restricted to a more recent and regional space. In such cases, population boundaries need to be identified among which gene flow is limited. To achieve this, both mitochondrial Correspondence. “heike.proehl@tiho-hannover.de (HP, corresponding author); krug_astrid@gmx.de (AK); auffarth.jana@gmail.com (JA) Amphib. Reptile Conserv. October 2022 | Volume 16 | Number 2 | e317 Conservation management units of Hyla arborea and nuclear markers are informative. Mitochondrial DNA has been widely used to analyze the phylogenetic relationships of amphibian populations (Dufresnes et al. 2013; Stock et al. 2012), while nuclear markers like microsatellites are well suited for detecting fine-scale structuring of populations and recent loss of genetic variation (Selkoe and Toonen 2006). Population genetic approaches such as Bayesian assignment tests use population allele frequencies to group individuals into genetic clusters. Together with information on genetic divergence between genetic clusters, this approach can be used to denote conservation units (Olsen et al. 2014; Rowe and Beebee 2007). In Europe, habitat loss, fragmentation, and degradation — mostly due to anthropogenic pressure (Cushman 2006; Pimm and Raven 2000) — are the most significant threats to endangered wildlife populations (Fahrig and Merriam 1994; Sih et al. 2000; Stuart et al. 2004). Amphibian populations are especially vulnerable to fragmentation and loss of genetic variation due to their low dispersal capabilities (as reviewed in Smith and Green 2006). For safeguarding vulnerable species of this most endangered vertebrate group (Stuart et al. 2004), it 1s necessary to counteract genetic depletion by maintaining the exchange of individuals among populations. The European Tree Frog has shown long-term decline in much of its Western European distribution, mainly caused by habitat fragmentation (Andersen et al. 2004; Dubey et al. 2009; Krug and Prohl 2013). The highest genetic diversity of this species has accumulated in South-eastern Europe, where it survived in refugia during glaciations. After late-Pleistocene diversification on the Balkan Peninsula, one of several major genetic groups recolonized North and Western Europe. Postglacial expansions resulted in decreasing genetic diversity across the range and therefore increased the vulnerability of populations towards North-Western Europe (Dufresnes et al. 2013; Stock et al. 2012). Indeed, the tree frog is not categorized as threatened in South-Eastern Europe, while it is reported to have declined and is now classified as Vulnerable to Critically Endangered in different areas in the north-west (see review in Dufresnes et al. 2013, Table S1). In Lower Saxony in Germany, where the current distribution is patchy with some main occurrences in the lowlands (Fig. 1), the conservation status of the tree frog 1s Endangered (see the Red List at http://www.amphibienschutz.de). Although the species was widespread in the past, severe declines have been observed mainly in the second half of the last century (Manzke and Podloucky 1995). In some places, measures for conservation management have been successful (Brandt and Liiers 2017; Buschmann et al. 2006; Richter and Mugge 2012). For supporting further conservation activities, analyses of the genetic structure are required for assessing the genetic clusters as a way to delineate the units for conservation management (Rowe and Beebee Amphib. Reptile Conserv. 2007). So far, several studies have measured the genetic structure and diversity in more or less fragmented metapopulation systems (Andersen et al. 2004; Angelone and Holderegger 2009; Arens et al. 2006; Dubey et al. 2009; Edenhamn et al. 2000; Krug and Prohl 2013). The aim of this study was to perform a conservation genetic survey of the European Tree Frog across its distribution in Lower Saxony and adjacent areas. The specific intention was to assess significant genetic differentiation in order to define those conservation units among which dispersal is restricted. The obtained information was then used to identify population management goals and to provide specific recommendations about conservation priorities to ensure the long-term survival of the tree frog in this region. In this study, we tested hypotheses regarding population genetic structure, differentiation, and diversity in the Endangered European Tree Frog by analyzing mitochondrial sequence and nuclear microsatellite data with a series of statistical techniques. First, we tested for the existence of diverged genetic lineages. We further predicted that past population expansion and recent habitat fragmentation 1) reduced the migration among localities; 2) reduced genetic diversity as well as genetic population size within localities; and 3) resulted in significant genetic structure among the remaining tree frog localities. Therefore, 4) we expected a small to moderate effect of geographic distance on genetic differentiation as a result of the ongoing population disconnection. This work reveals ongoing population fragmentation with moderate genetic diversity for the populations of the European Tree Frog in Lower Saxony. Materials and Methods Sample Collection and Preparation Fourteen sites were sampled across the tree frog distribution in Lower Saxony and adjacent distributions in North Rhine Westphalia and Saxony-Anhalt, all in Germany. We chose one sample site within each main occurrence of the tree frog in this region (Fig. 1). In the occurrence near Hannover, however, we sampled four sites: two in the west of Hannover (KZ, KO, see Table 1 for site definitions) and two in the east of Hannover (KH, BH) for a comparison of smaller scaled spatial distances. In total, 237 individuals were sampled with 5—22 individuals per sample site (Table 1). Genetic material was collected from the tips of tadpole tails and buccal swabs of adult frogs. The adults were collected from the choruses during the breeding seasons in spring 2007 and 2008. Tadpoles were sampled in summer 2007 at three localities. In this year, the climatic conditions for breeding where unfavorable and adult catch rates were low at these sites. To avoid bias in the results from tadpole samples representing offspring from only one breeding pair, tadpoles were sampled in different breeding ponds. October 2022 | Volume 16 | Number 2 | e317 Krug et al. heel \ 0 10 20 30 40 ad Sa er aS oe Fig. 1. Current distribution of the European Tree Frog, and the distribution of cyt b haplotypes in Lower Saxony and adjacent areas on the basis of TK25-quadrants (grey squares) during 1994—2010 in Lower Saxony (NLWKN 2011), 1993-2006 in North Rhine Westphalia (LANUV 2011), and 1990-2000 in Saxony Anhalt (Meyer et al. 2004). Dashed lines denote state borders, dots denote sample sites. Inset in upper left corner: Haplotype network of 11 distinct haplotypes of cyt 6 of H. arborea (901 bp) in Lower Saxony and adjacent areas. Each haplotype is represented by one circle and color. The size of a circle corresponds to the haplotype frequency. Lines between haplotypes denote mutational steps between sequences. The genetic diversity for these localities was similar to other locations, suggesting that relatedness among samples did not bias the results (Table 1). DNA from the tail clips was fixed in 99% ethanol and extracted using a proteinase K digestion followed by a Phenol-Chloroform protocol (Sambrook et al. 1989), and then stored at -20 °C. DNA was extracted from the buccal swabs with an Invisorb Spin Swab Kit (Invitek) following the manufacturer’s protocol, and stored at -20 °C. Another study confirmed that buccal swabbing is a very efficient method for obtaining DNA of adequate quality for microsatellite amplification (Broquet et al. 2007). A total of seven polymorphic microsatellite loci (WHA1-9, WHA1-20, WHA1-25, WHA1-67, WHAI1- 103, WHA1-104, and WHA1-140) previously isolated by Arens et al. (2000) were amplified following the author’s protocol, except that the annealing temperature for WHA1-20 was changed to 64.6 °C. The PCR products were genotyped using the capillary sequencer MegaBace 1000 (Amersham Bioscience). Allele scoring was performed using the software Genetic Profiler v. Amphib. Reptile Conserv. 71 2.2. The genotyping results can be found in the file that accompanies this article (Supplementary file 1). Because earlier analyses (Stock et al. 2012) included only four samples of mt DNA from Germany, we also sequenced cytochrome 5 (cyt b) fragments of 901 bp for 5—20 individuals from each sample site, excluding KO and BH from the Hannover population. The cyt 5 fragment was amplified via PCR using the primers MVZ 15-L (5'- GAACTAAT GGCCCA CACWWTACGNAA -3') and cyt b AR-H (TAWAAGGGTCTTCTACTGGTTG) from Moritz et al. (1992) and Goebel et al. (1999). The PCR reaction (25 ul) consisted of 20-100 ng DNA, | ul of each primer (10 uM), 0.8 ul dNTPs (10 mM, 5PRIME), 2.5 ul 10x advanced Buffer (SPRIME), 1.25 U Taq DNA Polymerase (SPRIME), and 17.45 ul H,O. The PCR conditions were as follows: an initial denaturation at 94 °C for 3 min; 35 cycles at 94 °C for 45 s, annealing temperature of 50 °C for 45 s, and extension at 65 °C for 1 min. The PCR products were sent to the Macrogen Company (Seoul, South Korea) for purification and sequencing with an ABI3730XL genetic analyzer (Applied Biosystems). October 2022 | Volume 16 | Number 2 | e317 Conservation management units of Hyla arborea Table 1. Overview of data from the various sample sites. *: Samples from adult frogs, '' samples from tadpoles, H,: observed heterozygosity, H,: expected heterozygosity, SD: standard deviation, F’,: inbreeding coefficient, with bold values for significant differences after 1,000 permutations, R: mean allelic richness over all loci, 4: haplotype diversity, 2: nucleotide diversity, NV: number of sampled individuals, N,: mean number of alleles over all loci. ID Sample site mean H, + SD mean H, + SD Fs IN’. R h m [%| N QU Quakenbriick * 0.786 + 0.248 0.741 0.143 -0.070 4.00 ~ 0.00 0.00 5 WK Westerkappeln* 0.661 + 0.187 0.579 + 0.158 -0.154 3.43 - 0.54 0.06 8 EK Espelkamp* 0.796 + 0.169 0.754 + 0.087 -0.058 S29 5.18 0.00 0.00 12 KZ Kananohe Zentrum * 0.667 + 0.169 0.666 + 0.134 -0.001 5.14 4.65 0.00 0.00 20 KO Kananohe Ost? 0.701 + 0.208 0.684 + 0.089 -0.027 4.57 4.53 - - 11 KH Kolshorn# 0.754 + 0.131 0.713 + 0.094 -0.059 6.29 3.37 0:53 0.13 20 BH Beinhorn* 0.693 + 0.089 0.693 + 0.093 -0.001 S57 4.84 - - 20 BA Bassum ' 0.771 + 0.099 0.748 + 0.092 -0.032 5.43 4.98 0.41 0.05 20 RU Ruschwedel * 0.731 + 0.064 0.721 + 0.050 -0.015 5.00 4.50 0.68 0.14 18 WG Wolfsburg-Gifhorn * 0.790 + 0.158 0.799 + 0.080 -0.011 7.71 6.65 0.61 0.08 20 ST Strothe 0:735,+ 0.153 0.708 + 0.126 -0.039 6.29 5.34 OAQ 0.02 21 AN Amt Neuhaus # 0.708 + 0.111 0.750 + 0.090 0.057 6.43 5.50 0.66 0.15 22 SW Salzwedel ' 0.600 + 0.227 0.687 + 0.181 0.130 6.00 5.01 0.57 0.11 20 PW Pevestorfer Wiesen* 0.793 + 0.110 0.764 + 0.091 -0.039 6.43 5.55 0.42 0.05 20 Statistical Analysis Analysis of mtDNA Both directions of the cyt b sequences were assembled using the computer software SeqMan™ I (DNASTAR, Inc., Konstanz, Germany). Multiple sequence alignments were performed in MEGA 4 (Tamura et al. 2007) using the Muscle algorithm (Edgar 2004), and all variable sites were confirmed by visual inspection of the chromatograms. The EMBL-EBI sequence analytical tool (Madeira et al. 2022) was used to convert the sequences to the corresponding amino acid sequences in order to assure that nuclear copies were not sequenced. The program MEGA was applied to calculate p-distances between sample sites (Tamura et al. 2004). Haplotype diversity (4) and nucleotide diversity (ma) (Nei 1987) were determined with ARLEQUIN ver. 3.11 (Excoffier et al. 2005). A haplotype network of the cyt 6 data set was constructed via the statistical parsimony analysis of the program TCS 1.21 (Clement et al. 2000) using the default settings. Analysis of Microsatellites Microsatellite data were checked for null alleles, stuttering, and allelic dropout using Micro-CHECKER (Van Oosterhout et al. 2004). The program Fstar v. 2.9.3 (Goudet 1995) was used to test for genotypic disequilibrium of all pairs of loci in each sample and to calculate average allelic richness per population. For the calculation of average allelic richness, sample sites with less than ten individuals (QU and WK) were excluded. For each sample site and locus, the observed and Amphib. Reptile Conserv. 72 expected heterozygosity (Nei 1987) and deviation from Hardy-Weinberg equilibrium (HWE) (Guo and Thompson 1992) were determined with ARLEQUIN Ver. 3.11 (Excoffier et al. 2005). Genepop ver. 4.1 (Rousset 2008) was used to test for a global deviation from HWE in each sample site. The inbreeding coefficient fF’. per sample site (Weir and Cockerham 1984) was calculated using Genetix ver. 4.05 (Belkhir et al. 2004) and the significance was tested with a permutations test (1,000 permutations). Genetic differentiation between the sample sites was calculated as global F’,.and pairwise F’,. values (Weir and Cockerham 1984) in ARLEQuUIN (Excoffier et al. 2005). In addition, pairwise D_., (Jost 2008), a substitute measure of genetic differentiation, was calculated using the R package DEMEtics (Gerlach et al. 2010). Significance was calculated by 10,000 bootstraps. The data were also tested for Isolation By Distance in sampled populations (IBD; Storfer et al. 2010; Wright 1943). IBD occurs when gene flow occurs but declines with increasing distances between pairs of populations, and is typical for the genetic population structure of many animal species (Hitchings and Beebee 1997; Spear et al. 2005; Vergara et al. 2015). To test for IBD, a Mantel test for correlation between pairwise genetic distances (F,, and D,.) and pairwise geographic distances was conducted, implemented in IBDWS 3.23 (Jensen et al. 2005). As proposed by Rousset (1997) for populations in two-dimensional habitats, geographical distance was log-transformed and genetic distance was expressed as F. (1 — F,,), and D,,, (1 — D,,). Significance for tr > 0 was assessed via 10,000 bootstraps. The linear geographic distances among sample sites were calculated in ArcView GIS 3.3 using the Distance Matrix extension (Jenness 2005). October 2022 | Volume 16 | Number 2 | e317 Krug et al. Even though the Mantel test is widely used in landscape genetic studies, an evaluation of different methods revealed that Mantel tests exhibit high type- 1 error rates (Balkenhol et al. 2009). Those authors recommended applying a combination of statistical methods to avoid inaccurate conclusions derived from only one method. Therefore, two additional hierarchical Bayesian methods, GESTE (Foll and Gaggiotti 2006) and BIMr (Faubet and Gaggiotti 2008), were applied here for evaluating the effect of distance by means of generalized linear models. Both BIMr and GESTE perform well for moderate samples sizes and limited numbers of loci, as in our study (Balkenhol et al. 2009). GESTE estimates the genetic distance (F.,. values) for each local population pair from multilocus genotypes and correlates them to environmental factors. Posterior probabilities associated with each factor allow the identification of factors with the highest effect on genetic structure. The regression coefficient estimate (Alpha) indicates whether a factor reduces or enhances genetic differentiation. The estimation of model parameters 1s performed by using a combination of Markov Chain Monte Carlo (MCMC) and Reversible-Jump MCMC (RJMCMC) (Green 1995). As environmental factors, we included latitude (G1) and longitude (G2; geographic coordinates in GK3 format) as approximations of the effect of distance among population pairs. The software BIMr 1.1 estimates contemporary gene flow and assesses the influence of genetic distance on gene flow. This program quantifies the gametic disequilibrium from multilocus genotypes (here, microsatellite alleles) generated by the progeny of recent migrants to calculate the proportion of the population that immigrated during the last generation (Faubet and Gaggiotti 2008). Five replicates (= runs) were run with a total of 1,020,000 iterations (burn-in: 1,000,000, sample size: 20,000) and a thinning interval of 50 iterations. For each replicate, first 20 short pilot MCMC runs of 1,000 iterations were conducted, and the run with the lowest Bayesian deviance (De) and the highest posterior probability was selected to extract the parameter estimates (Faubet et al. 2007; Faubet and Gaggiotti 2008). Two models were calculated: model 0 did not include environmental factors and model 1 included factor G1 which is the geographic distance between pairs of populations. As an alternative, the BayesAss software (Rannala 2007; http://www.rannala.org/software/) was also used to infer contemporary migrations rates. The software was run with 10,000,000 iterations (i), a burn-in (b) of 1,000,000 repetitions, and the interval between samples (n) was set to 1,000. The default values were used for all other parameters at first. Then, we adjusted the mixing parameters for migration rate (m), allele frequencies (a), and inbreeding coefficients (f) to maintain their acceptance rates between 20% and 60% as recommended in the Manual. Amphib. Reptile Conserv. To infer genetic clusters, individual assignments to populations were conducted by means of a combination of non-spatial and spatial Bayesian algorithms with STRUCTURE Version 2.3.3 (Pritchard et al. 2000) and TESS version 2.3 (Chen et al. 2007; Francois et al. 2006). Simulation data suggested the combination of TESS and STRUCTURE as a reliable approach for deducing the spatial population structure (Chen et al. 2007), outperforming other Bayesian clustering programs. All StrucTURE runs used 500,000 iterations after a burn-in period of 100,000. An admixture ancestry model and correlated allele frequencies were used between populations. STRUCTURE was run both without and with information about the sampling location (prior population information) and the results were compared as recommended by Pritchard etal. (2000; see also Dufresnes et al. 2013; Olsen et al. 2014). Hierarchical analyses were performed by repeating the STRUCTURE runs with each of the major clusters. Twenty runs were conducted for each K. The range of possible Ks tested spanned from | to 14, according to the number of sampled breeding sites. The average log likelihood Pr(X|K) (given by the estimated Ln Prob of data = Ln P(D) in the software result output, see Table 4) was calculated for each K across all runs. Since detecting the true number of K is not always straightforward, we included the AK statistics proposed by Evanno et al. (2005), using STRUCTURE HARVESTER V.0.6.8 (Earl and von Holdt 2012). TESS uses a Bayesian method to detect population structure, but it considers the spatial information (geographical coordinates) of the individuals. After assessing the preliminary runs as recommended in the software manual, the maximum number of allowed genetic clusters (K,.) was varied from 2 to 10. One hundred independent runs for each K_ were conducted under the admixture model, with 50,000 sweeps and a burn-in period of 10,000 sweeps for each run. Finally, the NeEstimator v2 software was used to estimate the contemporary effective population sizes at all sample sites (Do et al. 2014). Three single sample estimators were implemented: the linkage disequilibrium method, the heterozygote-excess method, and the molecular coancestry method. The lowest allele frequencies (P....) were set to 0.05, 0.02, 0.01, and 0+. Crit Results Genetic Diversity Genetic diversity was estimated for each population based on the microsatellite alleles and the cyt 6 haplotypes (Table 1). While genetic diversity indices based on microsatellites (H,, H,, N,, and R) are moderate to high across the range, they are always highest in WG (except for H.); while the indices based on cyt b (h, 2) tend to increase from west (WK/QU) to east (PW) (Table 1; Fig. 1). October 2022 | Volume 16 | Number 2 | e317 Conservation management units of Hyla arborea Table 2. Pairwise D est Table 1 for sample site acronym definitions. values (lower matrix) and pairwise F’,. values (upper matrix) between sample sites; ns = not significant. See QU WK EK KZ KO KH QU 0 0.120 0.066 0.111 0.113 0.110 WK 0.208 0 0.114 0.064 0.107 0.117 EK 0.195 0.286 0 0.041 0.031"° 0.069 KZ 0.284 0.145 0.102 0 0.002" 0.090 KO 0.324 0.243 0.097 0.001" 0 0.092 KH 0.412 0.301 0.222 0.279 0.290 0 BH 0.387 0.372 0.230 0.284 0.311 0.059 BA 0.189 0.230 0.220 0.199 0.189 0.172 RU 0.424 0.449 0.285 0.382 0.386 0.173 WG 0.139 0.309 0.296 0.230 0.306 0.237 ST 0.299 0.208 0.329 0.281 0.308 0.175 AN 0.265 0.315 0.230 0.248 0.263 0.245 SW 0.238 0.391 0.289 0.233 0.228 0.368 PW 0.213 0.365 0.278 0.278 0.322 0.350 BH BA RU WG ST AN SW PW 0.103 0.039 0.117 0.043 0.086 0.086 0.084 0.052 0.154 0.097 0.191 0.107 0.083 0.119 0.162 0.147 0.076 0.060 0.090 0.061 0.095 0.065 0.097 0.071 0.099 0.066 0.139 0.081 0.092 0.081 0.095 0.105 0.104 0.059 0.130 0.081 0.092 0.074 0.091 0.106 0.025 0.053 0.062 0.064 0.056 0.074 0.110 0.094 0 0.074 0.092 0.082 0.076 0.102 0.118 0.100 0.225 0 0.074 0.037 0.072 0.070 0.083 0.066 0.252 0.235 0 0.068 0.079 0.072 0.110 0.073 0.283 0.142 0.252 0 0.056 0.071 0.086 0.061 0.238 0.239 0.222 0.201 0 0.057 0.110 0.070 0.316 0.242 0.238 0.290 0.211 0 0.097 0.065 0.367 0.272 0.347 0.287 0.369 0.343 0 0.078 0.337 0.253 0.267 0.249 0.226 0.241 0.253 0 Mitochondrial Sequence Analysis The analysis revealed 11 haplotypes of the cytochrome b fragment which differed by ten variable sites and nine parsimony informative sites (Fig. 1). There was no evidence for any diverged haplotype groups that would correspond to different genetic lineages. Most haplotypes were closely related but unique to one sample site, except for haplotypes Hy-1, Hy-2, and Hy-5. While Hy-1 (blue) and Hy-5 (red) showed a broad distribution over almost the complete sampling area, Haplotype Hy-2 (green) was restricted to five sample sites in the northeast (Fig. 1). Eight haplotypes were found at only a single locality: Hy-3 (white) in AN, Hy-4 (orange) in BA, Hy-6 (light blue) in EK, Hy-7 (yellow) and HY-8 (brown) in KH, Hy-9 (dark blue) in RU, and Hy-10 (grey) and Hy-11 (dark grey) in WG. In WG and RU, four different haplotypes were detected, while in QU, EK, and KZ (all in the west of Hannover) only one haplotype was found. The p-distances among localities were low, varying between O and 0.4 % (Supplementary Table S1). The GenBank accession numbers can be found in Supplementary Table S3. Microsatellite Analysis The seven microsatellite markers examined were polymorphic with seven to 16 alleles per locus. The analysis with Micro-CHECKER uncovered signs of null alleles for locus WHA1-67 in sample site KO and for locus WHA1-140 in sample site SW. As null alleles for these two loci were found at only a single sample site, we did not adjust for null alleles. Furthermore, this analysis revealed no evidence for large allele dropout or scoring errors due to stuttering. Deviation from Hardy-Weinberg-Equilibrium (HWE) was found for WHA1-104, with a significant excess of heterozygotes in sample site KH. The global test for Amphib. Reptile Conserv. HWE over all loci in each population resulted in no significant deviation from HWE. Significance values for the inbreeding coefficient F, were obtained for the sample sites SW (F,, = 0.130) and WK (F,, = -0.154, Table 1). No linkage (genetic) disequilibrium was found between any pair of loci. The global F’,. value across all localities was 0.083 and highly significant (P > 0.0001). Genetic differentiation calculated as pairwise D,_ and pairwise F’,. values were significant in all cases except between the two sample sites in the West of Hannover (KZ and KO), as well as EK and KO regarding the F’,. values (Table 2). The Mantel test for IBD showed a significant but low correlation between the genetic and geographic distances (Fig. 2; Der =Q:28eP O01 Eh. F029: 20,0 1S see also Supplementary Table S2), indicating that genetic differentiation is only partially explained by geographic distances among the sites. GESTE calculated five different models (Table 3). The probability of a model was not improved by including either latitude (G1) or longitude (G2) without interaction. The model with the highest posterior probability was model 4, which included the constant, latitude, and longitude as well as their interaction. The Alpha values were low for the effects of both factors, while the Alpha value of the inter- action indicates a significant effect on genetic differentiation. All five replicates of the BIMr analysis showed a De of 0.0. The highest posterior probability for the null model was 0.79 (run 1), and the lowest was 0.55 (run 4). The posterior probabilities for model 1 (including G1) were lower than the posterior probabilities of the null model (Table 4); i.e., the geographic distance did not seem to affect recent gene flow or migration among sample sites. Mean migration rates were extremely low and varied from 2.88e"" to 1.1le°; while the highest mean migration rate was observed among PW and WG, and the lowest was among AN and QU. Also, the October 2022 | Volume 16 | Number 2 | e317 Krug et al. Genetic Distance [D,,/(1-D...)1 Genetic Distance [F,,/(1-F;,)] “0.0 0.6 4:2 1.8 2.4 3.0 =<010 0.6 1.2 1.8 2.4 3.0 Geographic Distance (Log km) Geographic Distance (Log km) Fig. 2. Isolation By Distance plots. (a) D,, /(1 — D.,,) versus log geographic distance; and (b) F’,./(1 — F,,,) versus log geographic est est distance. The lines are the RMA (Reduced Major Axis) regressions. migration rates between two close population pairs inthe by TESS (see below). Within each cluster, a fine surroundings of Hannover were very low. For example, — substructuring could be detected, and the two approaches the mean migration rates between KH to BH and vice — (with/without prior population information) provided versa were only 1.66e!° and 2.34e!°, and those between __ slightly different clustering solutions. In those runs where KZ and KO were 1.92e'° and 2.24e'° even though the __ the sampling location was used as prior information, the distances between these pairs were only 3.94 km and 2.6 — western/central cluster was split along Hannover with km, respectively. The analysis with BayesAss provided some admixed populations in the western part of the very similar results for the runs with default and adjusted region (K =3, Fig. 3B). The Structure runs without prior parameters. The migration rates were higher than those — population information supported the existence of two calculated with BIMr, most of which fluctuated around genetic subclusters within this region (Supplementary 0.01. Interestingly, the migration rates between EK and Fig. S1B). Inthe northern/eastern cluster, both approaches KZ, KO and KZ, as well as KH and BH were considerably provided similar results (Fig. 3C and Supplementary Fig. higher (0.14, 0.16, and 0.20, respectively) and more SIC). RU was differentiated from WG and ST, which in consistent with the genetic population structure (Fig. 3) — turn differed from the populations lying 1n the eastern as well as the above-mentioned geographic distances. part of the study area in Saxony Anhalt. More detailed Bayesian assignments conducted by Structure _ results of log likelihood Pr(X|K) values and AK statistics suggested two major groups (K = 2), separating the — are provided in the Supplementary figures (Figs. S2 and western/central populations (WK-BA-EK-KZ-KO-BH- __ S3, respectively). KH) and the northern/eastern populations (RU-WG-ST- The TESS analysis supported the two major genetic AN-PW-SW, Figs. 3A and 4, see also Supplementary —§ groups separated into a western/central cluster and a Fig. SIA). The population QU appeared to be admixed, __ northern/eastern cluster (Fig. 5). Increasing the number but was unambiguously assigned to the western group _ of K,. resulted in only a slight decrease in the Deviance Table 3. Posterior probabilities for five possible models calculated with GESTE explaining the genetic differentiation of European Tree Frogs as a function of the environmental factors latitude (G1) and longitude (G2). Constant is the intercept of the regression model. The regression coefficients (Alpha) for different environmental factors used in the models are given in the right side of the table. Model Factors included Posterior probability Factors Regression coefficients ModelO — Constant 0.19 Constant Alpha 0 -3.69 Model 1 Constant, G1 0.19 Gl Alpha 1 0.06 Model2 — Constant, G2 0.18 G2 Alpha 2 -0.07 Model 3 Constant, G1, G2 0.17 G1*G2 Alpha 3 -2.14 Model 4 Constant, G1, G2, G1*G2 0.26 Amphib. Reptile Conserv. 75 October 2022 | Volume 16 | Number 2 | e317 Conservation management units of Hyla arborea A es § Y & F F F F&F YF F FF F F Fig. 3. Estimation of the number of Hy/a arborea populations using the program Structure ver 2.3.1 (Pritchard et al. 2000) for the admixture model with prior population information; QU, WK, EK, etc. = sample sites, separated by fine black lines. Each individual is represented by a single vertical line broken into colored segments, with lengths proportional to the corresponding clusters. (A) Plot for K = 2 in the analysis of the entire data set, (B) plot for K = 3, and (C) K = 4 for hierarchical analysis on each of the two main clusters. Information Criterion (DIC) of the models while the resulting population structure was not consistent among runs for each K, or in comparison with the STRUCTURE results (data not shown). STRUCTURE as Well as TESS revealed that frogs from WG in the eastern group (Fig. 3A) are of admixed origin from both genetic groups as the assignment results from the two were similar. Effective Population Sizes Mean effective population sizes (N.) varied among the sample sites and statistical methods applied. The results also differed between P.,.. = 0.05 and the other P.,.. but were the same for P,,,, = 0.02, 0.01, and 0+ (Table 5). The large confidence intervals indicate that the results might not be very reliable. However, most calculated N, values were small (100 individuals or less). Only three sample sites (KO, AN, and PW) showed consistently high values for N (500 or higher, or infinite) for most methods. Discussion The analyses presented here provide valuable information for the conservation management of the Endangered European Tree Frog species, Hyla arborea, which suffers from population isolation in its northern distribution range. In Lower Saxony, a weak correlation between genetic and geographic distances suggests a low level of recent gene flow among localities, and further analyses indicate a lack of current migration at least during the last generation. Two major genetic clusters, one in the east and one in the west, were found with some admixture in a central population. Both main clusters were further subdivided into several distinct regional clusters. The substantial population structure, verified by significant genetic distances among localities, suggests that the populations are currently isolated to a large extent. Consequently, conservation management is needed to ensure the long-term persistence of this species in Lower Saxony with suitable effective population sizes and high levels of genetic diversity that are necessary to counteract the reductions in fitness and adaptive potential (Andersen et al. 2004; Frankham 2005; Allentoft and O’ Brien 2010; Angelone 2010). Isolation by Geographic and Genetic Distances In addition to the Mantel test, the landscape genetic analysis in GESTE provided some insight into the role of geographic distance on genetic differentiation. While the models including latitude and longitude alone did not offer a better explanation for genetic differentiation than the null model, the most complex model including the interactions between latitude and longitude did. We interpret this as the effect of the geographic distance Table 4. Results of BIMr analysis (means of posterior probability and Alpha for run 1) for estimating migration rates among localities with European Tree Frogs in Lower Saxony. The factor G1 is the geographic distance. Alpha 0 and Alpha | represent estimates of the constant term and factor G1, respectively. Model Factor Posterior probability Alpha 0 Alpha 1 Model 0 0.79 1.64 Model 1 G1 = geographic distance 0.21 1.48 0.14 Amphib. Reptile Conserv. 76 October 2022 | Volume 16 | Number 2 | e317 Krug et al. A Mean of estimated Ln probability of data ~~ Delta K oOo NM fF OO 2 4 .6 -8 -10 12. 44 K Fig. 4. Mean values of estimated Ln probability of data (LnPD) for each K (a) and delta K (b) when prior population information was implemented. on genetic dissimilarity between localities. Overall, however, the posterior probability and alpha values illustrate that the distance effect 1s not very large. Habitat fragmentation might play a more important role in shaping the genetic structure of the tree frogs in this area. According to Podloucky and Fischer (2013), habitat fragmentation in Lower Saxony is mainly caused by the loss of summer habitat, breeding ponds, and corridors suitable for migration. This is in accordance with the disconnected distribution (Fig. 1) and significant genetic distances (F,.and D, ) among most tree frog populations. A significant population structure as a result of limited dispersal between isolated populations 1s typically accompanied by a slight to moderate effect of isolation by distance. Isolation by distance was detected in some European amphibian species (Rana dalmatina, Sarasola-Puente et al. 2012; Bombina_ variegata: Weihmann et al. 2009; Hantzschmann et al. 2020), but not others (Bufo calamita: Allentoft et al. 2009; Bombina bombina: Dolgener et al. 2012). For the tree frogs in Lower Saxony, the small positive correlation (7 ~ 0.28) between genetic and geographic distances suggests a very low level of recent gene flow among localities. This finding is in accordance with earlier studies on tree frogs, which reported small to moderate correlation coefficients between both distances and significant population structure (Andersen et al. 2004; Angelone and Holderegger 2009; Arens et al. 2006). In all these studies, limited gene flow was explained by habitat fragmentation, particularly the loss of breeding ponds. In contrast, in those frog species which occur in more continuous habitats or that have higher dispersal capacities, the Fig. 5. Map of geographic-genetic cluster membership for A, = 2 as inferred by TESS. Amphib. Reptile Conserv. 77 October 2022 | Volume 16 | Number 2 | e317 Table 5. Results for effective populations sizes (NV) calculated with the NV Estimator for three different methods and different P Conservation management units of Hyla arborea Crit values. N. values were the same for P.,,, = 0.02, 0.01, and 0+. Only 0+ is used in the Molecular Coancestry method. Mean values (N.) and Confidence Intervals (CIs) are given for each sample site, method, and P Crit The three methods option. Linkage disequilibrium Heterozygote excess Molecular coancestry Sample site Pre 0.05 0.02/0.01/0+ 0.05 0.02/0.01/0+ 0+ QU N, D5:7. 55.7 78 78 9.5 Cls 1.4 — inf. 1.4 — inf. 3.7 — inf. 3.7 — inf. 5.1-—15.2 Wk N, 1.3 13 5.1 5/1 2h Cls 0.7 —2.7 0.7-2.7 2.5 — inf. 2.5 — inf. 2.0 —3.5 EK N, 16.3 33:5 9.9 9.9 8.2 ClIs 5.6 — inf. 8.7 — inf. 4.5 — inf. 4.5 — inf. 2.7—-16.9 KZ N, 46.3 28.2 34.7 34.7 5.6 CIs 15.8 — inf. 12.1 —319.2 7.5 — inf. 7.5 — inf. 2.3-—10.5 KO N, inf. 611.3 2515 2515 inf. CIs 11.4 — inf. 10.4 — inf. 4.3 — inf. 4.3 — inf. inf. — inf. KH N, 3193 159.7 inf. 443 59 CIs 30.5 — inf. 30.2 — inf. 5.8 — inf. 6.0 — inf. 3.2-8.9 BH N, 83.5 735 PORT inf. inf. Cls 21.0 — inf. 21.3 — inf. 7.2 — inf. 7.7 — inf. inf. — inf. BA N, W772 188.0 322.0 322.0 10.8 Cls 26.5 — inf. 26.8. inf. 6.5 — inf. 6.5 — inf. 1.8 —27.7 RU N, 19.3 27:5 67.7 81.9 inf. Cls 8.4 — 106.4 10.8 — inf. 6.8 — inf. 7.1 — inf. inf. WG N, 31.1 65.5 inf. inf. 9.1 CIs 16.0 — 113.5 26.0 — inf. 8.1 — inf. 9.3 — inf. 3.8 — 16.6 ST N, 34.7 TLS 39.7 51.0 inf. Cls 15.7-315.7 25.4 — inf. 6.4 — inf. 7.2 — inf. inf. — inf. AN Ny inf. inf. inf. inf. inf. Gis 41.7 — inf. 74.0 — inf. 30.9 — inf. 57.8 — inf. inf. — inf. SW N, StS 36.0 inf. inf. 43.4 CIs 15.2 — inf. 16.1 — inf. inf. — inf. 280 — inf. O-—217.7 PW N, 631.9 inf. inf. inf. inf. CIs 36.2 — inf. 73.1 — inf. 6.1 — inf. 6.6 — inf. inf. — inf. correlation between geographic and genetic distance 1s higher (e.g., Prohl et al. 2006) or disappears in case of panmixia (Leblois et al. 2000). The BIMr analysis revealed that contemporary migration (1.e., during the last generation) was practically absent among the sample sites of tree frogs in Lower Saxony. This is in contrast to the migration rates calculated with BayesAss and the results of the SrRuCTURE analysis (see below), which imply that individuals of mixed ancestry exist in the different genetic clusters. However, gene flow over the last few tree frog generations may have suffered due to the expanding habitat fragmentation, while some decades ago far more tree frog localities were reported for lower Saxony, thus, the connectivity among them was much better (Manzke and Podloucky 1995). Amphib. Reptile Conserv. Therefore, the signals for gene flow are still apparent in the results of some analyses (StrucTURE, BAYESAss) but not in those where the calculations are restricted to the most recent years (BIMr). Genetic Diversity The expected microsatellite heterozygosity (H,) has been measured in a number of previous population genetic studies of the European Tree Frog. Interestingly, H, was higher in the current study area (H,: 0.60 — 0.79) than in most other northern areas where the genetic situation of the tree frog was investigated (e.g., H, values in Denmark: 0.35—0.54, Andersen et al. 2004; Switzerland: 0.27—0.71, Angelone and Holderegger 2009; and the Netherlands: October 2022 | Volume 16 | Number 2 | e317 Krug et al. 0.39-0.59, Arens et al. 2006). As expected, the more peripheral populations, such as those in Denmark and the Netherlands, show lower genetic diversity values. In comparison, the mthaplotypes diversity (mean h= 0.38, n = 14 populations) is lower in the Lower Saxony area than in the southern part of the distribution range (Greece, Albania, Croatia, Serbia, and Romania) were the average h amounts to 0.7 (n = 20 populations, calculated from Dufresnes et al. 2013, Table S2). From all cumulative data within the framework of this study, we can conclude that in central and northern Europe, human induced fragmentation processes involving habitat destruction in a previously widely distributed frog species are contributing to the depletion of genetic diversity. Genetic Structure and Conservation Units The Bayesian cluster analyses conducted with STRUCTURE and TESS support the division of the tree frog populations into two major geographic-genetic clusters, one in the west and one in the east. One population in the south- east of the area (WG) shows admixture between the two groups and therefore a relatively high genetic diversity. Thus, it seems that both groups were previously connected by migrating animals when habitat fragmentation was less severe. This result is consistent with the moderate correlation between geographic distance and genetic distances. Both analyses also provide evidence for further fragmentation within both groups. In contrast, the haplotype network does not indicate any older, distinct (e.g., postglacial) lineages supporting the finding of Dufresnes et al. (2013), that only one evolutionary unit is present in this area. Microsatellite pairwise D, and F.,. values all showed significant genetic differentiation except for the two closest sites (KZ and KO) in the West of Hannover. However, mthaplotype distribution and Bayesian analyses of the microsatellites suggest distinct relationships among the currently fragmented localities. In the Northeast of Lower Saxony, the distribution of the mthaplotypes indicates a former connection of the populations along the river Elbe. Interestingly, the easternmost occurrences at AN, PW, and SW in the current and former distributions display a relatively well-connected area, nonetheless the presently distinct genetic sub-clusters in the Bayesian analyses and pairwise F’. values are relatively high. The significant F’, value found for SW indicates that the separation of this site may have resulted in inbreeding in an isolated population. Altogether, the available microsatellite data point to recent fragmentation of tree frog populations in this area, while similar mthaplotypes provide some evidence for a former connection. Interestingly, there is a significant genetic divergence between the sample sites KZ and KO in the West and the sample sites KH and BH in the East of Hannover. One possible explanation is that recently constructed motorways in combination with genetic drift contributed Amphib. Reptile Conserv. 79 to population differentiations, which are also apparent in the mthaplotype frequencies. Roads have been identified as barriers to gene flow in some other amphibians (Arens et al. 2007; Lesbarreéres et al. 2006). One central question is whether these relatively young barriers (motorways expanding in the 1960s, and dense urban areas) are the only reason for the differentiation of these formerly linked tree frog localities (Manzke and Podloucky 1995). The low haplotype diversity in KZ, EK, and QU points to a loss of genetic diversity as a consequence of increased genetic drift in isolated occurrences. In summary, the genetic analyses point to a highly structured population, as was observed in other surveys of European amphibians (Dolgener et al. 2012; Rowe and Beebee 2007; Sarasola-Puente et al. 2012; Hantzmann et al. 2020). The risks of fragmentation include population reduction, loss of genetic diversity and declining fitness, and finally extinction (Hitchings and Beebee 1997; Cushman 2006). To reverse such negative processes, conservation management that takes the genetic population structure into account is important (Allentoft et al. 2009; Olsen et al. 2014). For this endangered and fragmented frog species, we suggest delineating the two major genetic clusters as conservation units; and then within those, intense reconnection efforts should be undertaken by creating suitable habitats for migrating frogs. Moreover, there is clear evidence of admixture in WG, and gene flow along this route could be reestablished between both clusters. Conservation Measures In our opinion, future conservation management should be directed towards two aims. The first aim is to maintain high genetic diversity in large and stable populations within each conservation unit. In this context, it has been argued that an effective breeding size (N_) of at least 50 animals is necessary to avoid inbreeding in the short term and that an N, of 500 is necessary to maintain the evolutionary potential that would allow adaptations to environmental changes and assure long term viability (Jamieson and Allendorf 2012). Most of our isolated sample sites (e.g., BA, Fig. 3B—C) or sub-clusters (e.g., KH-BH) do not reach these effective population sizes. For tree frogs, the ratio of effective breeding size (N_) to census size (N) is ~ 0.5 (Broquet et al. 2009). Therefore, we recommend the monitoring of population sizes and maintaining population sizes of at least 100 breeding frogs in isolated populations, 1.e., each sub-cluster, for short term conservation goals, but increasing the population sizes to 1,000 or more embedded in each of several meta- population systems in every conservation unit (see also Andersen etal. 2004; Frankham etal. 2014). Incases where populations within a conservation unit are genetically and geographically separated, genetic rescue can be attained by establishing corridors to stimulate dispersal (e.g., in AN-SW-PW). Dense networks of suitable spawning October 2022 | Volume 16 | Number 2 | e317 Conservation management units of Hyla arborea ponds have been destroyed by habitat conversion but are of great importance for the maintenance of large tree frog populations and the connection of subpopulations. There are several reports that tree frogs not only respond well to new suitable water bodies, but also depend on them for migrations exceeding several km and often colonize them in subsequent breeding seasons (e.g., Angelone and Holderegger 2009; Brandt and Ltiers 2017; Hansen 2004; Schwartze 2002; Zumbach 2004). The second aim is to maintain overall genetic diversity among the genetic clusters within the species and to protect local co-adapted gene complexes (Savolainen et al. 2013). To achieve this second goal, we recommend the re-establishment of gene flow between genetic clusters where possible, but at a lower level than within them. This particularly applies to the western and eastern clusters between which (former) gene flow is evidenced by the Structure analysis (Fig. 3). Habitat reconnection between these areas would allow a few frogs to travel between the breeding ponds of different clusters, thereby refreshing genetic diversity and counterbalancing the loss of genetic diversity through drift, while diverse selection pressures would sustain local adaptation. Levels of genetic diversity inferred from neutral markers are not necessarily correlated with variation in locally adapted traits. In this context, more research is necessary to understand which traits are locally adapted and how their variation affects the fitness of a population. Very isolated and small populations might benefit from translocations, 1.e., the introduction of individuals from other populations. In such cases, translocations should be restricted to within the conservation unit to avoid causing outbreeding depressions that have sometimes been observed between distantly related populations (Sagvik et al. 2005). Conclusions Populations of the European Tree Frog in Lower Saxony are highly fragmented geographically and genetically, and therefore endangered. We identified two major genetic clusters and recommend that they should be considered as local conservation units. Conservation efforts should entail areconnection of the populations within these conservation units, and to a lesser degree between them. Moreover, the maintenance of large and stable meta-populations within genetic sub-clusters (mostly consisting of isolated populations) needs to be achieved for long-term survival. For translocations of individuals to recovering very small and inbred populations or for reintroduction, we suggest a mixing of individuals from different populations within the same conservation unit to increase genetic diversity and enhance the adaptive capacity regarding changing environmental conditions. This study offers one example of how population genetic studies can help to delineate conservation units, and our recommendations might apply just as well to other endangered species where declines are connected to increasing habitat fragmentation. Amphib. Reptile Conserv. Acknowledgements.—This research was supported by grants from the German Federal Environmental Foundation (DBU), Heidehof-Stiftung, and MHans-Schiemenz- Fonds - Deutsche Gesellschaft fiir Herp-etologie und Terrarienkunde (DGHT). We thank the following nature conservation authorities for permission to collect the tree frogs: Kreis Minden Ltibbecke, Kreis Steinfurt, Landkreis Diepholz, Landkreis Gifhorn, Landkreis Ltneburg, Landkreis Osnabriick, Landkreis Stade, Landkreis Uelzen, Region Hannover, and Stadt Wolfsburg. We are especially grateful to the many people who helped during field work, and R. Podloucky for providing distribution maps. Finally, we thank our technician Sabine Sippel for her assistance in the molecular lab, Sonke von den Berg for help by creating the figures, Tjard Bergmann for submitting the Cyt b sequences to Genbank, and Oscar Gaggiotti for providing advice on BIMr. Literature Cited Allentoft ME, O’Brien J. 2010. Global amphibian declines, loss of genetic diversity, and fitness: a review. Diversity 2: 47-71. Allentoft ME, Siegismund HR, Briggs L, Andersen LW. 2009. Microsatellite analysis of the Natterjack Toad (Bufo calamita) in Denmark: populations are islands in a fragmented landscape. Conservervation Genetics 10: 15-28. Andersen LW, Fog K, Damgaard C. 2004. Habitat fragmentation causes bottlenecks and inbreeding in the European Tree Frog (Hy/a arborea). Proceedings of the Royal Society of London B-Biological Sciences 271: 1,293-1,302. Angelone S. 2010. Are differences in fitness traits related to genetic cluster? An empirical test on the European Tree Frog. Biological Conservation 143: 471-478. Angelone S, Holderegger R. 2009. Population genetics suggests effectiveness of habitat connectivity measures for the European Tree Frog in Switzerland. Journal of Applied Ecology 46: 879-887. Arens P, van’t Westende W, Bugter R, Smulders MJM, Vosman B. 2000. Microsatellite markers for the European Tree Frog, Hyla arborea. Molecular Ecology 9: 1,944—1,946. Arens P, Bugter R, van’t Westende W, Zollinger R, Stronks J, Vos CC, Smulders MJM. 2006. Microsatellite variation and population structure of a recovering Tree Frog (Hyla arborea L.) metapopulation. Conservation Genetics 7: 825-835. Arens P, van der Sluis T, van’t Westende W, Vosman B, Vos CC, Smulders MJM. 2007. Genetic population differentiation and connectivity among fragmented Moor Frog (Rana arvalis) populations in The Netherlands. Landscape Ecology 22: 1,489-1,500. Balkenhol N, Waits LP, Dezzani RJ. 2009. Statistical approaches in landscape genetics: an evaluation of methods for linking landscape and genetic data. October 2022 | Volume 16 | Number 2 | e317 Krug et al. Ecography 32: 818-830. Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F. 2004. Genetix 4.05, population genetics software for Windows™. Université de Montpellier II, Montpellier, France. Brandt T, Lters E. 2017. Die erfolgreiche Wiederansiedlung von Europdaischen Laubfroschen (Hyla arborea) in der Sachsenhager Niederung, Landkreis Schaumburg, Niedersachsen. Zeitschrift fiir Feldherpetologie 24: 209-222. Broquet T, Berset-Braendli L, Emaresi G, Fumagalli L. 2007. Buccal swabs allow efficient and reliable microsatellite genotyping in amphibians. Conservervation Genetics 8: 509-511. Broquet T, Jaquiery J, Perrin N. 2009. Opportunity for sexual selection and effective population size in the lek-breeding European Treefrog (Hyla arborea). Evolution 63: 674-683. Buschmann H, Scheel B, Brandt T. 2006. Amphibien und Reptilien im Schaumburger Land und am Steinhuder Meer. Natur und Text, Rangsdorf, Germany. 184 p. Chen C, Durand E, Forbes F, Francois O. 2007. Bayesian clustering algorithms ascertaining spatial population structure: a new computer program and a comparison study. Molecular Ecology Notes 7: 747-756. Clement M, Posada D, Crandall KA. 2000. TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1,657—1,659. Crandall KA, Bininda-Emonds OR, Mace GM, Wayne RK. 2000. Considering evolutionary processes in conservation biology. Trends in Ecology and Evolution 15: 290-295. Crnokrak P, Roff DA. 1999. Inbreeding depression in the wild. Heredity 83: 260-270. Cushman SA. 2006. Effects of habitat loss and fragmentation on amphibians: a review and prospectus. Biological Conservation 128: 231-240. Dolgener N, Schroder C, Schneeweiss N, Tiedemann R. 2012. Genetic population structure of the Fire-bellied Toad, Bombina bombina, in an area of high population density: implications for conservation. Hydrobiologia 689: 111-120. Do C, Waples RS, Peel D, Macbeth GM, Tillett BJ, Ovenden JR. 2014. NeEstimator V2: re- implementation of software for the estimation of contemporary effective population size (N,) from genetic data. Molecular Ecology Resources 14: 209- 214. Dubey S, Ursenbacher S, Pellet J, Fumagalli L. 2009. Genetic differentiation in two European Tree Frog (Hyla arborea) metapopulations in contrasted landscapes of western Switzerland. Amphibia-Reptilia 30: 127-133. Dufresnes C, Wassef J, Ghali K, Brelsford A, Stock M, Lymberakis P, Crnobrnja-Isailovic J, Perrin N. 2013. Conservation phylogeography: Does historical diversity contribute to regional vulnerability in Amphib. Reptile Conserv. 81 European Tree Frogs (Hyla arborea)? Molecular Ecology 22: 5,669—5,684. Earl DA, von Holdt BM. 2012. StrucrURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4: 359-361. Edenhamn P, Hoggren M, Carlson A. 2000. Genetic diversity and fitness in peripheral and central populations of the European Tree Frog, Hy/a arborea. Hereditas 133: 115-122. Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32: 1,792-1,797. Evanno G, Regnaut S, Goudet J. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14: 2,611-2,620. Excoffier L, Laval G, Schneider S. 2005. ARLEQUIN (version 3.0): an integrated software package for population genetics data analysis. Evolutionary Bioinformatics 1: 47-50. Fahrig L, Merriam G. 1994. Conservation of fragmented populations. Conservation Biology 8: 50-59. Faubet P, Waples RS, Gaggiotti OE. 2007. Evaluating the performance of a multilocus Bayesian method for the estimation of migration rates. Molecular Ecology 16: 1,149-1,166. Faubet P, Gaggiotti OE. 2008. A new Bayesian method to identify the environmental factors that influence recent migration. Genetics 178: 1,491—1,504. Foll M, Gaggiotti O. 2006. Identifying the environmental factors that determine the genetic structure of populations. Genetics 174: 875-891. Frangois O, Ancelet S, Guillot G. 2006. Bayesian clustering using hidden Markov random fields in spatial population genetics. Genetics 174: 805-816. Frankham R. 2005. Genetics and extinction. Biological Conservation 126: 131-140. Frankham R, Bradshaw CJA, Brook BW. 2014. Genetics in conservation management: revised recommendations for the 50/500 rules, Red List criteria, and population viability analyses. Biological Conservation 170: 56-63. Gerlach G, Jueterbock A, Kraemer P, Deppermann J, Harmand P. 2010. Calculations of population differentiation based on G(ST) and D: forget G(ST) but not all of statistics! Molecular Ecology 19: 3,845— 3/852: Goebel AM, Donnelly JM, Atz ME. 1999. PCR primers and amplification methods for 12S ribosomal DNA, the control region, cytochrome oxidase I, and cytochrome b in bufonids and other frogs, and an overview of PCR primers which have amplified DNA in amphibians successfully. Molecular Phylogenetics and Evolution 11: 163-199. Goudet J. 1995. Fstrar (version 1.2): a computer program to calculate F-statistics. Journal of Heredity 86: 485— October 2022 | Volume 16 | Number 2 | e317 Conservation management units of Hyla arborea 486. Green PJ. 1995. Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika 82: 711-732. Guo SW, Thompson EA. 1992. Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics 48: 361-372. Hansen F. 2004. Verbreitung und Gefahrdung des Laubfrosches (Hyla arborea) auf Bornholm (Dane- mark) und Mafinahmen zur Lebensraumoptimierung. Zeitschrift ftir Feldherpetologie (Supplement 5): 133— 143. Hantzschmann AL, Sinsch U, Gottlicher C, Prohl H. 2020. Conservation genetics of Yellow-bellied Toads (Bombina variegata): a matter of geographical scale and isolation. Conservation Genetics 22: 83-96. Hedrick PW, Kalinowski ST. 2000. Inbreeding depression in conservation biology. Annual Review of Ecology and Systematics 25(31): 139-162. Hitchings SP, Beebee TJ. 1997. Genetic substructuring as a result of barriers to gene flow in urban Rana temporaria (Common Frog) populations: implications for biodiversity conservation. Heredity 79: 117-127. Jamieson IG, Allendorf FW. 2012. How does the 50/500 rule apply to MVPs? Trends in Ecology and Evolution 27: 578-584. Jenness J. 2005. Distance Matrix (dist_mat_jen.avx) extension for ArcView 3.x, v. 2. Jenness Enterprises, Flagstaff, Arizona, USA. Available: http://www. jennessent.com/arcview/dist_matrix.htm [Accessed: 1 August 2011]. Jensen JL, Bohonak AJ, Kelley ST. 2005. Isolation by distance, web service. BMC Genetics 6: 6. Jost L. 2008. G(ST) and its relatives do not measure differentiation. Molecular Ecology 17: 4,015—4,026. Krug A, Prohl H. 2013. Population genetics in a fragmented population of the European Tree Frog (Hyla arborea). Amphibia-Reptilia 34: 95-107. LANUV. 2011. Laubfrosch-Hy/la arborea. Landesamt fiir Natur, Umwelt und Verbraucherschutz Nordrhein- Westfalen, Recklinghausen, German. Available: http:// www.lanuv.nrw.de [Accessed: 1 December 2011]. Leblois R, Rousset F, Tikel D, Moritz C, Estoup A. 2000. Absence of evidence for isolation by distance in an expanding Cane Toad (Bufo marinus) population: an individual-based analysis of microsatellite genotypes. Molecular Ecology 9: 1,905—1,909. Lesbarréres D, Primmer CR, Lode T, Merila J. 2006. The effects of 20 years of highway presence on the genetic structure of Rana dalmatina populations. Ecoscience 13: 531-538. Luquet E, David P, Lena JP, Joly P, Konecny L, Dufresnes C, Perrin N, Plénet S. 2011. Heterozygosity-fitness correlations among wild populations of European Tree Frog (Hyla arborea) detect fixation load. Molecular Ecology 20: 1,877—1,887. Madeira F, Pearce M, Tivey ARN, Basutkar P, Lee J, Amphib. Reptile Conserv. 82 Edbali O, Madhusoodanan N, Kolesnikov A, Lopez R. 2022. Search and sequence analysis tools services from EMBL-EBI in 2022. Nucleic Acids Research 50: W276-W 279. Manzke U, Podloucky R. 1995. Der Laubfrosch Hyla arborea in Niedersachsen und Bremen—Verbreitung, Lebensraum, Bestandsituation. Pp. 57-72 In: Der Laubfrosch (Hyla arborea), Okologie und Artenschutz. Editor, Geiger A. Mertensiella, Bonn, Germany. 196 p. Meyer F, Buschendorf J, Zuppke U, Braumann F, Schadler M, Grosse WR. 2004. Die Lurche und Kriechtiere Sachsen-Anhalts. Verbreitung, Okologie, Gefahrdung, und Schutz. Laurenti- Verlag, Bielefeld, Germany. 240 p. Moritz C, Schneider CJ, Wake DB. 1992. Evolutionary relationships within the Ensatina_ eschscholtzii complex confirm the ring species interpretation. Systematic Biology 41: 273-291. Moritz C. 1994. Defining evolutionarily significant units for conservation. Trends in Ecology and Evolution 9: 373-374. Nei M. 1987. Molecular Evolutionary Genetics. Columbia University Press, New York, New York, USA. 512 p. NLWKN. 2011. Vollzugshinweise zum Schutz von Amphibien und Reptilienarten in Niedersachsen Hannover. Niedersachsischer Landesbetrieb fur Wasserwirtschaft, Kustenund Naturschutz, Hannover, Germany. 33 p. Olsen MT, Andersen LW, Dietz R, Teilmann J, Harkonen T, Siegismund HR. 2014. Integrating genetic data and population viability analyses for the identification of Harbour Seal (Phoca vitulina) populations and management units. Molecular Ecology 23: 815-831. Palsboll PJ, Berube M, Allendorf FW. 2007. Identification of management units using population genetic data. Trends in Ecology and Evolution 22: 11-16. Pimm S, Raven P. 2000. Biodiversity: extinction by numbers. Nature 403: 843-845. Podloucky R, Fischer C. 2013. Rote Listen und Gesamtlisten der Amphibien und _ Reptilien in Niedersachsen und Bremen. Informationsdienst Naturschutz Niedersachsen, Band 4. Niedersachsischer Landesbetrieb fiir Wasserwirtschaft, Kustenund Naturschutz, Hannover, Germany. 167 p. Pritchard JK, Stephens M, Donnelly P. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945-959, Prohl H, Koshy RA, Mueller U, Rand AS, Ryan MJ. 2006. Geographic variation of genetic and behavioral traits in northern and southern Tungara Frogs. Evolution 60: 1,669-1,679. Rannala B. 2007. BayesAss Edition 3.0 User’s Manual, Update 2015. University of California Davis, Davis, California, USA. 12 p. Richter M, Migge G. 2012. Bestandsentwicklung October 2022 | Volume 16 | Number 2 | e317 Krug et al. und Schutz des Laubfroschs (Hyla arborea) im Landkreis Diepholz/Niedersachsen. Zeitschrift fiir Feldherpetologie 19: 79-90. Rousset F. 1997. Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance. Genetics 145: 1,219-1,228. Rousset F. 2008. GENEPOP’007: a complete re- implementation of the GENEPOP software for Windows and Linux. Molecular Ecology Resources 8: 103-106. Rowe G, Beebee T. 2007. Defining population boundaries: use of three Bayesian approaches with microsatellite data from British Natterjack Toads (Bufo calamita). Molecular Ecology 16: 785-796. Sagvik J, Uller T, Olsson M. 2005. Outbreeding depression in the Common Frog, Rana temporaria. Conservation Genetics 6: 205-211. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular Cloning. Cold Spring Harbor Laboratory Press, New York, New York, USA. 1,659 p. Sarasola-Puente V, Madeira MJ, Gosa A, Lizana, M, Gomez-Moliner B. 2012. Population structure and genetic diversity of Rana dalmatina in the Iberian Peninsula. Conservation Genetics 13: 197-209. Savolainen O, Lascoux M, Merila J. 2013. Ecological genomics of local adaptation. Nature Reviews Genetics 14: 807-820. Schwartze M. 2002. Neuanlage und Verbesserungen von Kleingewassern fir den Laubfrosch und andere Amphibien: eine Untersuchung im 6stlichen Munsterland (NRW). Zeitschrift fiir Feldherpetologie 9: 61-73. Selkoe KA, Toonen RJ. 2006. Microsatellites for ecologists: a practical guide to using and evaluating microsatellite markers. Ecology Letters 9: 615-629. Shaffer HB, Gidis M, McCartney-Melstad E, Neal KM, Oyamaguchi HM, Tellez M, Toffelmier EM. 2015. Conservation genetics and genomics of amphibians and reptiles. Annual Reviews of Animal Biosciences 3: 113-138. Sih A, Jonsson BG, Luikar G. 2000. Habitat loss: ecological, evolutionary, and genetic consequences. Trends in Ecology and Evolution 15: 132-134. Smith, MA, Green DM. 2006. Sex, isolation, and fidelity: unbiased long-distance dispersal in a terrestrial amphibian. Ecography 29: 649-658. Spear SF, Peterson CR, Matocq MD, Storfer A. 2005. Amphib. Reptile Conserv. Landscape genetics of the Blotched Tiger Salamander (Ambystoma tigrinum melanostictum). Molecular Ecology 14: 2,553-2,564. Stock M, Dufresnes C, Litvinchuk SN, Lymberakis P, Biollay S, Berroneau M, Borzée A, Ghali K, Ogielska M, Perrin N. 2012. Cryptic diversity among Western Palearctic tree frogs: postglacial range expansion, range limits, and secondary contacts of three European Tree Frog lineages (Hyla arborea group). Molecular Phylogenetics and Evolution 65: \-9. Storfer A, Murphy MA, Spear SF, Holderegger R, Waits LP. 2010. Landscape genetics: Where are we now? Molecular Ecology 19: 3,496—3,514. Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues AS, Fischman DL, Waller RW. 2004. Status and trends of amphibian declines and extinctions worldwide. Science 306: 1,783—1,786. Tamura K, Nei M, Kumar S. 2004. Prospects for inferring very large phylogenies by using the neighbor-joining method. Proceedings of the National Academy of Sciences of the United States of America 101: 11,030— 11,035. Tamura K, Dudley J, Nei M, Kumar S. 2007. MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution 24: 1,596-1,599. Van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P. 2004. Micro-cHECKER: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4: 535-538. Vergara M, Basto MP, Madeira MJ, Gomez-Moliner BJ, Santos-Reis M, Fernandes C, Ruiz-Gonzalez A. 2015. Inferring population genetic structure in widely and continuously distributed carnivores: the Stone Marten (Martes foina) as a case study. PLoS One 10: 00134257. Weihmann F, Podloucky R, Hauswaldt S, Prohl H. 2009. Naturschutzgenetische Untersuchungen von Populationen der Gelbbauchunke (Bombina _ v. variegata) im stidlichen Niedersachsen. Zeitschrift fiir Feldherpetologie 16: 183-200. Weir BS, Cockerham CC. 1984. Estimating F-statistics for the analysis of population-structure. Evolution 38: 1,358—1,370. Wright S. 1943. Isolation by distance. Genetics 28: 114. Zumbach S. 2004. Die Laubfrosche (Hyla arborea und Hyla intermedia) in der Schweiz — Verbreitung, Gefahrdung, und Schutz. Zeitschrift fiir Feld- herpetologie (Supplement 5): 183-192. October 2022 | Volume 16 | Number 2 | e317 Amphib. Reptile Conserv. Conservation management units of Hyla arborea Astrid Krug has a Ph.D. in Biology from the University of Veterinary Medicine Hannover in Germany. During the time of her Diploma and Ph.D. theses, Astrid worked on several herpetological projects with a focus on molecular genetic analyses in the European Tree Frog. Since then, working in the field of clinical research, her unabated and keen interest in herpetology remains. Jana Auffarth holds a Ph.D. in Veterinary Research and Animal Biology from University of Veterinary Medicine Hannover in Germany. At the time of this study, she was a Postdoctoral Research Assistant at the Institute of Zoology, University of Veterinary Medicine Hannover, where her research centered around amphibian population ecology, conservation, and management on a molecular level. Currently, she is working on assessments of aquatic systems in line with the EU water framework directive and national regulations. Heike Préhl discovered her interest in studying frogs in the wild while she studied Biology at the University of Hannover, and while spending a year studying tropical biology at the Universidad de Costa Rica. During that appointment, she benefitted from a wide range of field courses, including herpetology. After sampling field data on Neotropical frog behavior and ecology for her Diploma and Doctoral theses, as well as her Postdoctoral project, she started to work as a Junior Professor at the Veterinary University of Hannover (TiHo) in Germany. Heike is currently an apl. Professor for Zoology and teaches courses related to Zoology, Ecology, and Evolution to biology and veterinary students. Her research focuses on the behavior, ecology, and conservation of Neotropical and European amphibians. 84 October 2022 | Volume 16 | Number 2 | e317 Krug et al. e@ g¢ ¢ Ff ¢ €& C HT War e g 6 > ¢ @ Supplementary Fig. S1. Estimation of the number of Hyla arborea populations using the program STRUCTURE ver. 2.3.1 (Pritchard et al. 2000) without prior population information; QU, WK, EK, etc. = sample sites, separated by fine black lines. Each individual is represented by a single vertical line broken into K-colored segments, with lengths proportional to each of the K-inferred clusters. (A) Plot for K = 2 in the analysis of the entire data set, (B) plot for K = 4, and (C) K = 4 for hierarchical analysis on each of the two main clusters. ‘@) Mean of estimated Ln probability of data" Mean of estimated Ln probability of data~—~ 2 3 4 o K K Supplementary Fig. S2. Mean values of estimated Ln probability of data (LnPD) for each K (A, C) and delta K (B, D) calculated from STRUCTURE runs with STRUCTURE HARVESTER (20 replicates per K) in those analyses where prior population information was implemented. (A—B), graphs for hierarchical analysis of the red cluster; (C—D), corresponding graphs for the green cluster (compare to Fig. 3 in the main text). Amphib. Reptile Conserv. 85 October 2022 | Volume 16 | Number 2 | e317 Conservation management units of Hyla arborea B) -5,600 -5,800 a 20 6,000} . 15 & e ’ i) -6,200 : 2140 -6,400 \ ©) Mean of estimated Ln probability of data Mean of estimated Ln probability of data~—~ 2 3 4 5 6 Fé E) £ i) <°) Ke) £ Fe} s iS jos c —! 1?) D o E 3 is) c o co) = 2 3 4 9 K K Supplementary Fig. S3. Mean values of estimated Ln probability of data (LnPD) for each K (A, C, E) and delta K (B, D, F) without prior population information. (A—B), results for the entire data set. (C—D), graphs for hierarchical analysis of the red cluster. (E—F), corresponding graphs for the green cluster. Amphib. Reptile Conserv. 86 October 2022 | Volume 16 | Number 2 | e317 Krug et al. Supplementary Table S1. Estimates of evolutionary divergence over cyt b sequence pairs between sample sites (p-distances). QU WK EK KZ KH BA RU WG ST AN SW PW QU 0 WK 0.001 0 EK 0.001 0.002 O KZ 0.000 0.001 0.001 0 KH 0.003 0.002 0.004 0.003 O BA 0.001 0.001 0.003 0.001 0.002 O RU 0.001 0.001 0.002 0.001 0.003 0.002 O WG 0.001 0.001 0.002 0.001 0.002 0.001 0.001 0 St 0.001 0.001 0.002 0.001 0.002 0.000 0.001 0.001 0 AN 0.001 0.001 0.002 0.001 0.003 0.001 0.002 0.001 0.001 0 SW 0.001 0.001 0.002 0.001 0.003 0.001 0.002 0.001 0.001 0.001 0 PW 0.001 0.001 0.002 0.001 0.002 0.001 0.001 0.001 0.000 0.001 0.001 0 Supplementary Table S2. Geographic distances (km) among sample sites. QU WK EK KZ KO KH BH BA RU WG ST AN SW PW QU 0 24.89 59.36 118.51 121.07 138.45 138.18 59.60 145.74 186.98 194.94 225.27 21432 244.29 WK 24.89 0 55.80 121.60 124.18 140.04 140.47 77.42 165.05 189.68 206.53 236.57 221.51 254.05 EK 59.36 55.80 0 66.97 69.52 84.52 85.18 57.94 135.51 134.27 158.58 187.71 168.42 203.12 KZ 118.51 121.60 66.97 0 2.59 20.80 19.80 79.20 107.70 68.55 98.15 125.14 101.84 138.06 KO 121.07 12418 69.52 2.59 0 18.53 17.33 81.32 107.77 65.97 96.27 123.05 99.39 = 135.75 KH 138.45 140.04 84.52 20.80 18.53 0 3.94 99.81 118.84 49.76 93.08 117.07 88.33 126.83 BH 138.18 140.47 85.18 19.80 17.33 3.94 0 98.08 115.08 49.22 89.62 114.01 86.17 124.28 BA 59.60 77.42 57.94 79.20 81.32 99.81 98.08 0 87.65 142.67 137.64 167.94 161.85 188.28 RU 145.74 165.05 135.51 107.70 107.77 118.84 115.08 87.65 QO 134.01 84.08 108.84 12469 133.14 WG 186.98 189.68 134.27 68.55 65.97 49.76 49.22 14267 134.01 0 73.62 86.47 47.60 88.04 ST 194.94 206.53 158.58 98.15 96.27 93.08 89.62 137.64 84.08 73.62 0 30.35 43.49 51.77 AN 225.27 236.57 187.71 125.14 123.05 117.07 114.01 167.94 108.84 86.47 30.35 0 42.06 24.53 SW 214.32 221.51 168.42 101.84 99.39 88.33 86.17 161.85 124.69 47.60 43.49 42.06 0 40.48 PW 24429 25405 203.12 138.06 135.75 126.83 12428 188.28 133.14 88.04 51.77 24.53 40.48 0 Supplementary Table S3. GenBank accession numbers for Hyla arborea CytB haplotypes Hy 1 to Hy 11. Sequence ID GenBank accession number BankIt2634361 Seql OP690610 BankIt2634361 Seq2 OP690611 BankIt2634361 Seq3 OP690612 BankIt2634361 Seq4 OP690613 BankIt2634361 Seq5 OP690614 BankIt2634361 Seq6 OP690615 BankIt2634361 Seq7 OP690616 BankIt2634361 Seq8 OP690617 BankIt2634361 Seq9 OP690618 BankIt2634361 Seq10 OP690619 BankIt2634361 Seq11 OP690620 Amphib. Reptile Conserv. 87 October 2022 | Volume 16 | Number 2 | e317 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptile Conservation 16(2) [General Section]: 88-103 (e318). New biological data for two rare reedfrog species, Hyperolius nimbae Laurent, 1958, and H. chlorosteus (Boulenger, 1915) (Anura: Hyperoliidae) ‘Basseu Aude-Inés Gongomin, ‘*N’Goran Germain Kouameé, ‘Konan Laurent Agoh, ‘Kouassi Philippe Kanga, and *Mark-Oliver Rodel ‘Université Jean Lorougnon Guédé, UFR Environnement, Laboratoire de Biodiversité et Ecologie Tropicale (BioEcoTrop), Daloa, BP 150, COTE D'IVOIRE *Museum fiir Naturkunde - Leibniz Institute for Evolution and Biodiversity Science, Invalidenstr. 43, 10115 Berlin, GERMANY Abstract.—Our recent surveys have generated new biological data on Hyperolius nimbae and H. chlorosteus, two little-known West African reedfrog species of conservation concern. During fieldwork at the eastern foothills of the Ivorian part of the Nimba Mountains, H. nimbae and H. chlorosteus individuals were found by acoustic and visual encounter surveys. In addition, various natural history data were recorded, including clutch size and egg-deposition sites. Hyperolius nimbae males were found at the edges of a permanent swamp. The area was dominated by cocoa and coffee plantations, and used in small-scale subsistence farming, i.e., growing plantains, cassava, and rice. A new locality is reported here for the species, but our surveys failed to confirm its presence at some previously known sites. The entire range of the species is now confined to a small, non-protected, human-impacted area. Under these conditions, this species is facing high extinction risk. Hyperolius chlorosteus was found in a patch of dense, broadleaf and evergreen primary forest at a mid- elevation along a torrent stream, and in a degraded lowland forest edging a large stream. The habitats of both species are suffering from forest degradation and deforestation along streams, so we urgently recommend the strict protection of the habitats of both species. Given these concerns, we suggest that the IUCN threat status of H. nimbae should be updated to Critically Endangered. Considering the large range of H. chlorosteus, the current IUCN categorization of this species as Least Concern seems to be correct. However, based on the fact that in Ivory Coast it only occurs in the westernmost parts of the country, several of the few known national populations have been lost, and the remaining forest habitats are declining, so we feel that H. chlorosteus should be regarded as Endangered nationally. Keywords. Biodiversity hotspot, endemic, habitat protection, Mount Nimba Reedfrog, Threatened species, Upper Guinean forest area Citation: Gongomin BA-I, Kouamé NG, Agoh KL, Kanga KP, Rédel M-O. 2022. New biological data for two rare reedfrog species, Hyperolius nimbae Laurent, 1958, and H. chlorosteus (Boulenger, 1915) (Anura: Hyperoliidae). Amphibian & Reptile Conservation 16(2) [General Section]: 88-103 (e318). Copyright: © 2022 Gongomin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License [Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are 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: amphibian-reptile-conservation.org. Accepted: 23 August 2022; Published: 24 October 2022 Introduction Ivorian montane areas, such as Mount Péko, Mount Sangbé, and Mount Nimba, are small in extent but comprise a high diversity of amphibian species (e.g., Rodel 2003; Rodel and Ernst 2003; Kanga et al. 2021). Mount Nimba, in particular, hosts the richest amphibian fauna of West Africa including various rare and endemic species (Kanga et al. 2021; Rodel et al. 2021). For instance, this mountain is the only known region where the matrotrophic Nimba Toad, Nimbaphrynoides occidentalis (Angel, 1943), and the Mount Nimba Reedfrog, Hyperolius nimbae Laurent, 1958, are found. Correspondence. *ngoran_kouame@yahoo.fr Amphib. Reptile Conserv. The biology of N. occidentalis has been investigated in detail (Lamotte 1959; Lamotte and Sanchez-Lamotte 1999; Xavier 2009; Sandberger et al. 2010; Sandberger- Loua et al. 2017, 2018), and this toad is considered to be the flagship species for the conservation of the Nimba mountains (Hillers et al. 2008a; Sandberger-Loua et al. 2016). Much less attention has been directed toward various other little-known amphibian species which are, nevertheless, of high conservation concern in Ivory Coast. These include, two notable reedfrogs, the Mount Nimba Reedfrog Hyperolius nimbae and the Large Green Reedfrog H. chlorosteus (Boulenger, 1915), for which October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. we know very little concerning their life histories. Hyperolius nimbae is currently classified as Endangered on The IUCN Red List of Threatened Species (IUCN 2021). It is endemic to the lowlands of the eastern slopes of the Ivorian part of Mounts Nimba. After Schiotz (1967) last collected this frog on 28 July 1963, the species was only recently rediscovered by Kouamé et al. (2016). This species is rare throughout its limited range in some swamp forests near Danané (Schiotz 1967; Kouamé et al. 2016; Kanga et al. 2021). By contrast, H. chlorosteus inhabits rainforests in Sierra Leone, Guinea, Liberia, and western Ivory Coast (Rédel et al. 2004; Channing and Rodel 2019) and is categorized as Least Concern on the IUCN Red List (IUCN 2021). Despite a broad distribution, H. chlorosteus is confined to the edges of rivers in primary lowland forests (Schiotz 1967; Rodel and Glos 2019), and thus must be regarded as threatened in Ivory Coast due to the decline of those habitats. A previously known population in Lakota (south-central Ivory Coast; Schietz 1967) no longer exists, and more may have been lost due to deforestation. For instance, Ivorian records of H. chlorosteus have been published from the classified forests (managed forests) of Cavally and Haute Dodo (Rodel and Branch 2002), and the national parks of Mount Sangbé (Rédel 2003), Mount Péko (Rodel and Ernst 2003), and Tai (e.g., Ernst and Rodel 2008; Kpan et al. 2021), as well as from Mount Nimba (Kanga et al. 2021), which are all situated in the south-western part of the country. However, during a decade of conflict, the gallery forests in Mount Péko as well as both classified forests have been destroyed (NG Kouameé, pers. obs.). These losses restrict the remaining known Ivorian range of H. chlorosteus to the southern part of Mount Sangbé, the lowland forests of Mount Nimba, and the Tai National Park. The aim of this study was to search for the persisting populations of H. nimbae and H. chlorosteus at Mount Nimba, and to collect additional data on the morphology, call characteristics, breeding sites, and potential threats for these poorly-known species. Material and Methods Study site. Fieldwork was conducted in the Mount Nimba Integrated Nature Reserve (MNINR: 07°25’-07°45’N, 008°20’—008°35’W; Fig. 1) during both the rainy and dry seasons, and was carried out on 84 days between 2 August 2019 and 8 July 2020. The MNINR covers 5,000 ha of various habitats ranging from altered and former forests in the lowlands, through dense, broadleaf and evergreen forests stretching from the lower to mid-elevations, and montane grasslands at the highest elevations (Lamotte et al. 2003a,b; Lauginie 2007). Mean annual temperatures vary from 22—27 °C on the mountain bases, and 16-21 °C on the mountain ridge. The rainy season extends from eight to nine months and is only interrupted by a short dry season from November/December to February/ Amphib. Reptile Conserv. March. The precipitation is highest on the mountain top, where it may reach up to 3,500 mm. During the dry season, the humidity is usually below 30%, but exceeds 80% in the rainy season (Lauginie 2007). The dry season is characterized by a warm, dry, and dusty wind known as Harmattan. For a detailed description of MNINR’s amphibian fauna see Kanga et al. (2021). Sampling. As previous research has shown that these two reedfrog species are only active at night (Kanga et al. 2021), searches took place from 1800-2200 h GMT. All searches were conducted by three people, totalling a search-effort of 1,008 person-hours. Search techniques focused on visual scanning of the terrain, supplemented by acoustic surveying to find the males, and examination of potential calling and breeding sites. Particular attention was given to searching the vegetation around and/or along forest streams and swamps. We investigated the area seven times per month and counted only individuals that were captured. Hyperolius nimbae was found at one site at Yéalé, while H. chlorosteus was collected at one site at MNINR and another site at Yéalé. We visited each site 56 times in the rainy season and 28 times in the dry season (Table 1; Appendices 1 and 2). Frogs were photographed, measured, and sexed. To avoid re-counting frogs, individuals that were not collected were marked as vouchers by toe clipping following the recommendations of Grafe et al. (2011), and released at their respective sites of capture. Areas close to four villages (Dagbonpleu, Danipleu, Kouan-Houlé, and Zéalé), within the formerly known range of H. nimbae, were re-investigated seven times. These visits were in the rainy season, which is the presumed reproductive period of the species. We recorded potential threats to H. nimbae and H. chlorosteus, defined here as any anthropogenic activity that may negatively impact the two species and/or their habitats. Although the panzootic chytrid fungus Batracho- chytrium dendrobatidis (Bd) seems to be absent from West Africa, west of the Dahomey Gap (Penner et al. 2013), newly sterilized equipment was always used at each site. Geographical coordinates were recorded with a GPS (datum: WGS84). A few vouchers were euthanized in a_ 1,1,1-trichloro-2-methyl-2-propanol hemihydrate (MS222) solution and thereafter preserved in 80% ethanol. These frogs are deposited in the research collection of NGK at the Université Jean Lorougnon Guédé, Daloa, Ivory Coast, and will serve as the bases of a national reference collection and for research and teaching purposes. Morphological characteristics and advertisement calls. Measurements of morphological features (in mm) were taken by one person (BAIG) with a dial calliper (accuracy + 0.1 mm), and are given as means (X) with standard deviations (SD). They comprise: snout-urostyle length, head width and length both at the level of jaw October 2022 | Volume 16 | Number 2 | e318 Biology and conservation of Hyperolius nimbae and H. chlorosteus 10°W o°w GUINEA LIBERIA City Village Road Stream State boundaries 0123 4km Lit jj 8°W 8°N OAST ouan-Houlé Danané Fig. 1. Geographical location of the Mount Nimba Integrated Nature Reserve within the westernmost extension of Ivory Coast at the border crossing point with Guinea and Liberia. The altered forest area where Hyperolius nimbae and H. chlorosteus were found at Yéalé is represented by a purple star; the habitats of H. chlorosteus ranged from dense, broadleaf and evergreen forests from lower to mid-elevations (yellow star). The inset figure indicates the location of Ivory Coast (green patch) on the African continent. articulation, interorbital space, distance between eye and nostril, distance from nostril to tip of snout, eye diameter, internarial distance, tympanum diameter, crus length, thigh length, and foot length including the tarsus and the longest toe. Other characters recorded included color pattern variation following the definitions by Schiotz (1967) and Channing and Rodel (2019). To ensure that clutches of H. nimbae and H. chlorosteus could not be confused with those of other sympatric reedfrogs, couples were collected and kept separately in a plastic terrarium (25 x Amphib. Reptile Conserv. 15 cm, 16 cm depth) with water (8.5 cm) and submerged and floating vegetation until eggs were laid. Egg numbers and diameters (accuracy + 0.1 mm) were determined for each clutch. For comparison, further H. chlorosteus clutch data collected between 1999 and 2001 in Tai National Park by MOR are also reported here. Hyperolius nimbae was not included in a phylogeny of the genus (Portik et al. 2019), but is assumed to be a member of the H. viridiflavus complex (Schietz 1999; Channing 2022). Very recently, this was confirmed by genetic data, and it was shown to be the sister species to October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. Table 1. Numbers of daily searches for Hyperolius nimbae and H. chlorosteus, during the 84-night period from 2 August 2019 to 8 July 2020, in Mount Nimba Integrated Nature Reserve and Yéaleé. ; if HAyperolius nimbae HAyperolius chlorosteus Sire oh eantite Yéalé (07°31.928’N, MNINR (07°34.652’N, Yéalé (07°31.932’N, P 008°25.401’W; 425 m asl) 008°24.966’ W; 716 m asl) 008°25.508’ W; 387 m asl) season season the Central African H. tuberculatus (Kouamé et al.2022). (Fig. 3). The calling males were mostly well concealed, Nine sequential advertisement calls were collected from —§ perching between thick branches and leaves of cocoa one male, in order to compare these data with those and coffee trees at ~1.80—2.10 m above the ground. presented by Schietz (1967). Five advertisement calls | Througout the study, a total of 305 frogs were captured were also recorded from a H. chlorosteus male. (Fig. 3; Appendix 1). They comprised 277 individuals, All calls were recorded with a Huawei recorder (44.1 including one couple, caught in the rainy season and 28 kHz sample ratio, 16 bits of resolution, FFT length=256) males caught in the dry season. Generally, additional and analyzed with the software Soundruler 0.9.6 (Gridi- —_ calling males were inaccessible, calling from high up in Papp 2007; Kohler et al. 2017; Emmrich et al. 2020). For tall trees close to the deep pond. each advertisement call, the following measurements Sympatric amphibian species in Yéalé were the White- were recorded: call duration (s), dominant frequency — lipped Frogs Amnirana sp. “albolabris west”, Afrixalus (Hz), fundamental frequency (Hz), minimum frequency dorsalis, A. fulvovittatus, Hyperolius concolor, H. (Hz), maximum frequency (Hz), duration intervals fusciventris fusciventris, H. picturatus, Phrynobatrachus between calls (s), duration intervals between notes (s), gutturosus, and P. latifrons, thus representing a typical and overall frequency bandwidth (Hz). composition of a farmbush anuran community (Schietz 1967). Results Morphology of H. nimbae. We retained 13 males and Observations on H. nimbae one female H. nimbae as vouchers. Snout-urostyle lengths of the voucher males ranged from 28.8—35.0 Habitat, population size, and activity of H. nimbae. mm (32.3 + 1.8 mm), thus slightly exceeding the known The Mount Nimba Reedfrog was found only in the range of H. nimbae (Schiotz 1967). The head was slightly Yéalé village (07°31.928’N, 008°25.401’W; 425 masl) — longer (12.0 + 1.1 mm; range: 10.0—13.5 mm) than broad at the periphery of MNINR. Our efforts to confirm the (10.8+0.8 mm; range: 9.9-12.5 mm); interorbital space species around village forests at Dagbonpleu, Danipleu, ranged from 5.0—7.5 mm (6.4 + 0.7 mm); the distance Kouan-Houlé, and Zéalé were unsuccessful. The forest between eye and nostril (2.1 + 0.3 mm; range: 2.0—3.0 areas formerly at these sites were destroyed and replaced =mm) approximated the distance from nostril to tip of by roads and new settlements. At Yéalé, the habitat snout (2.2 + 0.3; range: 1.5—2.0 mm); eye diameter (4.6 comprised altered forest bordered by patches of bamboo, +0.4 mm; range: 4.0—-5.1 mm) was larger than internarial intact forests, degraded forests with large clearings, space (3.2 + 0.3 mm; range: 3.0-4.0 mm) and tympanum and thick grassy and shrubby vegetation around the diameter (2.3 + 0.3 mm; range: 2.0—3.0 mm); crus length settlement. Large parts of the village’s surroundings were (16.5+1.0 mm; range: 15.1—18.9 mm) slightly exceeded dominated by small-scale subsistence farming, mainly _ thigh length (15.0 +0.8 mm; range: 13.5—16.1 mm); and plantings of corn, cassava, plantains, cocoa, and coffee, foot length including the longest toe ranged from 21.0- as well as rice in swamps. During the entire study, the 26.9 mm (23.1 + 1.6 mm). vegetation around swamps was checked at seven sites The female measured 34.0 mm SUL (Fig. 4A). Its without detecting any sign of H. nimbae’s presence. head was slightly longer (14.0 mm) than broad (12.1 Specimens of H. nimbae were found only ina mixed mm); the interorbital space reached 8.0 mm; the cocoa and coffee plantation (including plantains) edging —_ distance between eye and nostril (3.4 mm) was longer a large and deep pond (greater than 100 x 70 m; Fig. 2), than the distance from nostril to tip of snout (1.8 mm); where males started calling at dusk, usually around 1810 — eye diameter (5.0 mm) was larger than internarial space h GMT. Calling males were active in each month of our = (3.0 mm) and tympanum diameter (2.9 mm); crus length study and congregated in large choruses, particularly (18.0 mm) slightly exceeded thigh length (17.1 mm); during the rainy season. Calling activity never stopped, and foot length including the tarsus and the longest toe but was low during the dry season, as indicated by the —_ reached 24.5 mm. much lower numbers of frogs caught during that time Mount Nimba reedfrogs have a brief and truncated Amphib. Reptile Conserv. 91 October 2022 | Volume 16 | Number 2 | e318 Biology and conservation of Hyperolius nimbae and H. chlorosteus Fig. 2. Habitats of Hyperolius nimbae at the foo in heavily degraded habitats at the edge of large swamps. snout and rough to warty dorsal skin with a varied color pattern. The basic dorsal pattern varied from silver-grey to chocolate brown. The dark pattern on the back varied in intensity, but the general pattern was always similar and recognizable (Fig. 4). The female coloration was within the variation of the males (Fig. 4), but the discs on toes and fingers were more reddish than in most males (Fig. 4A). Likewise, the ventral color differed slightly between individuals, but the female was not different from the males (Fig. 5). However, instead of a dark Amphib. Reptile Conserv. hills of the Ivoria 92 n part of the Nimba Mountains. In Yéalé, the species was foun i - Babee vocal sac and yellowish gular gland, the female had a white throat, with the edges beset by orange and black speckles. The armpits and inner parts of limbs including webbing were blood red (Fig. 5C). Some males had similarly reddish skin parts (Fig. 5A), areas which are used for water uptake, see Rddel (2000), whereas others had blue-grey skin instead (Fig. 5B). Reproduction of H. nimbae. At Yéalé, a couple of H. nimbae in amplexus was captured on a Raphia Palm in October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. w ~ eal (op) (oe) jo) fo) jo) Number of captured frogs N oO Aug Sep Oct Nov Dec — Jan Feb Mar Apr Months May Jun Jul Fig. 3. Number of Hyperolius nimbae individuals recorded from 2 August 2019 to 8 July 2020 at Yéalé. deeper water (> 1 m). The couple was placed into a terrarium, and the following morning, 227 eggs were found attached to a leaf above the water surface. The eggs had a dark and light green pole that was surrounded by a thin transparent jelly (Fig. 5D). Egg diameter varied from 1.6—2.5 mm, with a mean value of 1.99 mm (SD: + 0.18 mm). Vocalization of H. nimbae. The acoustic properties of nine advertisement calls of one H. nimbae male were analyzed (Fig. 6). Hyperolius nimbae call in a long, fast series of pulse groups (notes), thus confirming the call descriptions provided by Schiotz (1967). The call duration averaged 2.77 + 0.56 s (range: 1.89-3.25 s). The duration intervals between calls averaged 0.44 + 0.23 S (range: 0.22-0.87 s, N = 8). Intervals between notes within each call were 0.15 + 0.03 s (range: 0.09-3.48 s, NV = 131). The mean fundamental frequency was 1,359.37 Hz (N = 9), while the dominant frequency reached 2,718.75 Hz (N = 9). The maximum frequency intensity was at 2,906.25 Hz (N = 9), and the minimum frequency intensity averaged 2,718.75 Hz (N=9). Threats to H. nimbae. We failed to confirm the species presence at all sites where we had recorded it only a few years before (see above). In addition, we observed various signs of habitat degradation in the habitats of H. nimbae at Yéalé. Aside from the close proximity to a human settlement, the frogs’ habitats were dominated by cocoa and coffee plants, and used for small-scale subsistence farming, i.e., plantains and cassava. Large parts of the breeding area of H. nimbae were used to establish rice paddies. Bamboo patches were steadily harvested by the local human population as construction materials. The rainforest edges were impacted by logging (Fig. 2). Observations on H. chlorosteus Habitat, population size, and activity of H. chlorosteus. Specimens of H. chlorosteus were found in MNINR ina patch of dense, broadleaf and evergreen primary forest at a mid-elevation along a torrent stream (07°34.652’N, 008°24.966’W; 716 m asl; Fig. 7A), as well as in a Amphib. Reptile Conserv. degraded forest edging a large stream at Yéalé village (07°31.932’N, 008°25.508°W; 387 m asl; Fig. 7B). From April—November, the rainy season, a total of 341 H. chlorosteus were captured in the two sites, always at night along the streams (Fig. 8; Appendix 2). In the dry season, from December to March, no signs of the Species’ presence were observed. Males started calling at around 1800 h GMT, most often after heavy rainfall. In a patch of primary forest, 327 calling males perched on leaves and branches of small shrubs at ~1.5—2.0 m height above the ground (e.g., Fig. 7C) were captured. The other recorded males were heard calling from very high up in tall trees along the forest stream. A few calling males (V = 11) were also found in a degraded forest, edging a large stream at Yéalé village (07°31.932’N, 008°25.508’W; 387 m asl; Fig. 7B). There, they called on shrubs, with the calling sites ranging from 1.80 m above the ground to much higher. Sympatric frogs were Leptopelis macrotis, Hyperolius picturatus, and White-lipped Frogs Amnirana sp. “albolabris west.” Morphology of H. chlorosteus. The snout-urostyle lengths of 15 males, retained as vouchers, ranged from 31.5-35.0 mm (33.8 + 1.3 mm), thus occurring within the known range of H. chlorosteus (Schiotz 1967). The head was slightly longer (12.0 + 0.1 mm; range: 12.0- 12.2 mm) than broad (11.5 + 0.7 mm; range: 10.5—13.0 mm); the interorbital space ranged from 5.9—7.2 mm (6.5 + 0.4 mm); the distance between eye and nostril (3.0 + 0.2 mm; range: 2.5—3.5 mm) was approximately twice the distance from nostril to tip of snout (1.4 + 0.6; range: 0.6—-2.4 mm); the eye diamter (5.1 + 0.3 mm; range: 4 4-5.5 mm) was larger than internarial space (2.6 + 0.4 mm; range: 2.1—3.0 mm) and tympanum diameter (2.8 + 0.2 mm; range: 2.2-3.0 mm); the crus length (17.0 + 0.7 mm; range: 15.4—-18.1 mm) slightly exceeded the thigh length (15.9 + 0.8 mm; range: 14.9—-16.9 mm); and foot length including the longest toe ranged from 21.5—23.5 mm (22.2 + 0.8 mm). Two females measured 36.8 and 37.5 mm in SUL. As in the males, their heads were slightly longer (13.5 mm) than broad (13.2 mm); interorbital spaces were 7.0 and 7.2 mm; the distance between eye and nostril was longer (3.3 + 0.3 mm; range: 3.1—3.5 mm) than the distance from nostril to tip of snout (1.9 + 0.1 mm; range: 1.9-2.0 mm); the eye diamter (5.0 + 0.1 mm; range: 5.0—5.1 mm) exceeded the internarial space (2.9 + 0.1 mm; range: 2.9-3.0 mm) and tympanum diameter (2.9 + 0.1 mm; range: 2.8-3.0 mm); the crus length (17.7 + 0.3 mm; range: 17.5—-17.9 mm) slightly exceeded thigh length (16.9 1.5 mm; range: 15.9-18.0 mm); and foot length including the tarsus and longest toe ranged from 23.5—23.8 mm (23.6 + 0.2 mm). The color pattern of H. chlorosteus showed some variations (see Figs. 7 and 9), however, all were within the range known for the species (compare Schiotz 1967; Rodel 2003; Channing and Rodel 2019). October 2022 | Volume 16 | Number 2 | e318 Biology and conservation of Hyperolius nimbae and H. chlorosteus Fig. 4. Benoa views of the Tapers nimbae fantale (A) ra males (B—H) from Yéalé, western oe Coast. Note the variation in coloration. Reproduction of H. chlorosteus. One female perched on a broad leaf of a shrub overhanging a forest stream, at ~1.5 m height, was captured. Eggs were visible through its ventral skin (Figs. 7D and 9F). An amplectant couple was seen sitting on low vegetation in this site (Fig. 7E). This couple was placed in a terrarium and a clutch of 62 eggs was present the next morning. The eggs were attached to the glass above the water. They had beige and dark poles surrounded by a thin transparent capsule (Fig. Amphib. Reptile Conserv. 7F). Egg diameters varied from 2.0—2.5 mm, mean 2.2 mm (SD: + 0.25 mm). The jelly of H. chlorosteus eggs is slightly opaque. Six additional H. chlorosteus clutches observed in Tai National Park, south-west of Ivory Coast, ranged from 54-122 eggs (86.2 + 28.8 eggs; Table 2). Vocalization of H. chlorosteus. Five advertisement calls were recorded from one male (Fig. 10). The calls were repeated 2—7 times (Fig. 10B). The acoustic impression October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. Table 2. Data for Hyperolius chlorosteus clutches from Tai National Park, Ivory Coast. Clutch number | Date | Number of eggs Location of clutch Remarks 16 March 1999 pene canaen Clutch looks like a blackberry above stream above stream Egg size 3-6 mm, nucleus 1.9—2 eo ene ae Bourecolded mm; clutch looks like a blackberry 4 February 71 Clutch on stone covered Egg size 3-6 mm, nucleus 2.2 mm; 1999 with mosses above water looks like a blackberry Fig. 5. Ventral views of Hyperolius nimbae males (A and B) and fnate (C) adults, and the eggs (D) from Yéalé, western Ivory Coast. of the call is a low-pitched, far ranging metallic click | Hz, N = 5). The maximum frequency intensity was at with a mean duration of 0.87 + 0.34 s (range: 0.51-1.41 2,643.75 + 167.70 Hz (range: 2,531.25—2,906.25 Hz, N s). The duration intervals between calls averaged 2.90+ =5), and the minimum frequency intensity averaged at 1.27 s (range: 1.36-4.0 s, N=4). The mean fundamental 2,531.25 + 229.64 Hz (range: 2,343.75—2,906.25 Hz, N frequency was 1,303.12 + 83.85 Hz (range: 1,265.62—. _=5). 1,453.12 Hz, N = 5), while the dominant frequency reached 2,606.3 + 167.70 Hz (range: 2,531.25—2,906.25 Threats to H. chlorosteus. Major threats to H. Amphib. Reptile Conserv. 95 October 2022 | Volume 16 | Number 2 | e318 Biology and conservation of Hyperolius nimbae and H. chlorosteus Ampl (Pa) oO 2 4 6 Time (s) Frequency (kHz) O-NWAH OH T eT ey a aD DT Ea RI RSI SU PI A UT Dy DS 0D Ds 0 | 0.25 Amplitude 2 4 6 Time (s) 1 2 3 4 ce) Frequency (kHz) Fig. 6. (A) Hyperolius nimbae usually calls perched between branches and leaves of trees. (B) Waveform (above), spectogram (center), and powerspectrogram (below) of the advertisement calls of a H. nimbae male from Yéalé, western Ivory Coast. chlorosteus were observed at Yéalé, where some forest pockets along streams had been cleared and converted to cocoa and coffee plantations, as well as rice cultivation. Discussion This study gathered new data on the biology and distribution of two rare reedfrog species, Hyperolius nimbae and H. chlorosteus, living in different forest habitats in the western Upper Guinea forest zone (Channing and Rodel 2019). Hyperolius nimbae has been documented only from a few individuals and populations on the Ivorian part of the Nimba Mountains (Schiotz 1967; Kouamé et al. 2016; Kanga et al. 2021), and little was known about its phenotypic variation, biology, and habitat requirements prior to this study. A recent study based on 16S rRNA sequences of members of the H. viridiflavus clade revealed that H. nimbae is a member of the H. viridiflavus clade and sister species to the Central African H. tuberculatus (Kouameé et al. 2022). At Yéalé, a relatively large population of H. nimbae was found in patches of heavily degraded forest area, partly neighboring human settlements. This habitat fit well with the earlier habitat description, 1.e., from the villages at Dagbonpleu, Danipleu, Kouan-Houleé, and Zéalé (Kouamé et al. 2016). However, an alarming result of this study is that due to ongoing development of the area, all formerly known sites have been destroyed, restricting the currently known range of H. nimbae to Yéalé alone. Our morphological assessments of a larger number of individuals confirmed former descriptions, but slightly increased our knowledge about color variability (Schiotz 1967; Kouamé et al. 2016; Figs. 3 and 4). Hyperolius nimbae apparently is another reedfrog species that lacks Amphib. Reptile Conserv. sex specific color dimorphism (compare Schigtz 1967; Veith et al. 2009; Portik et al. 2019). The new data herein also confirms previous descriptions of the advertisement calls of H. nimbae (Schigetz 1967). In contrast to most other Hyperolius (Schigetz 1967, 1999), H. nimbae males always produce a succession of continuous clicks and thus much longer calls. Most Hyperolius species are “prolonged breeders” (Schiotz 1967, 1999: Lotters et al. 2004; Rodel et al. 2006; Kouamé et al. 2015). Hyperolius nimbae 1s no exception to that rule, with males calling during all months of the rainy season, usually at the edges of large and permanent water bodies. We documented the first known clutch for this species (Fig. 5D). The eggs were attached to a leaf above the water surface, a feature which is common to forest species within Hyperolius. Savannah dwelling Hyperolius species deposit eggs under water, most likely as a response to the higher desiccation risk (Rodel 2000). We take the egg deposition above water as a further hint that this species depends on forest, at least to some extent. It remains a mystery why H. nimbae apparently tolerates degraded forests but only occurs in the Ivorian foothills of Mount Nimba, as comparable habitats are available beyond its range (Schiotz 1967). Most likely this species was trapped at these sites when the Nimba mountains were surrounded by savannah during drier periods, representing a refugium for the survival of forest species during unfavorable Pleistocene periods (Maley 1996). The new Hyperolius chlorosteus data reported herein confirmed previous descriptions of the morphology and advertisement call (Schiotz 1967, 1999; Channing and Rodel 2019). This species is morphologically similar to October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. = ae, é gi es Fig. 7. (A—B) Typical forest streams in the Mount Nimba Integrated Nature Reserve where Hyperolius chlorosteus breeds. A calling be al f a male (C), a female (D), a couple in amplexus (E), and their clutch (F) which was deposited on a leaf above a forest stream. H. laurenti (from the eastern parts of the Upper Guinea forest zone; Schiotz 1967, 1999: Channing and Rodel 2019), with which it also shares habitat requirements, i.e., perching on branches above small to medium-sized streams in primary or only slightly degraded forests, often at considerable height (Rodel et al. 2005; Kouamé et al. 2014). The breeding period of H. chlorosteus is confined to the rainy season, and clutches are attached at various heights on leaves, branches, and stones above flowing water. The habitat characteristics of H. chlorosteus Amphib. Reptile Conserv. 97 encountered in MNINR were similar to those reported from previous sites, 1.e., dense forests with streams or small rivers (Schiotz 1967; Rodel 2003; Ernst and Rodel 2008). Unfortunately, some of the known Ivorian sites for H. chlorosteus no longer exist, such as Lakota (e.g., Schiotz 1967), the Mount Péko National Park (Rédel and Ernst 2003), and the Cavally and Haute Dodo lowland forests in the western part of the country (R6édel and Branch 2002). In MNINR, the species was still found in abundance, but this study also registered the increasing and continuous degradation of its forest habitat. October 2022 | Volume 16 | Number 2 | e318 Biology and conservation of Hyperolius nimbae and H. chlorosteus wow fF u DBD oO CO O Oo Number of captured frogs N io) Jan Feb Jul Months Mar Apr May = Jun Fig. 8. Number of Hyperolius chlorosteus captured from 2 August 2019 to 8 July 2020 along lowland forest streams at Mount Nimba. Sa 4) =e xs “~< % * , = a ¥ Fig. 9. Variation of color pattern o Amphib. Reptile Conserv. a } Te of . ii Nee | Hae 4 4 an f Hyperolius chlorosteus from Mount Nimba Integrat 98 Conclusions and Threat Status This study revealed the most viable and_ largest population of H. nimbae known thus far. However, it also revealed that formely known sites are most likely gone, and the one reported herein might be the last site where the species is thriving. Amphibians currently have no official protection status in Ivory Coast, and no dedicated conservation policies apply to them. The only existing conservation measures for their protection are the designations of protected areas (Rodel et al. 2021). Unfortunately, the small pocket of forest zone where H. nimbae was found falls entirely within a non-protected area, and consequently it is exposed to increasing and : wy sees ed Nature Reserve, western Ivory Coast. October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. Ci = ron 0 E xt -1| Time (s) ® SP z 4} B 3p , Cc g 2r Si} Ww OE | | 5 10 Time (s) 1 o Tc = | = 0.57 acai ee xt 0 1 0 2 3 4 5 Frequency (kHz) Fig. 10. Calling male of Hyperolius chlorosteus (A) with (B) waveform (above), spectogram (centre) and powerspectrogram (below), of the species’ advertisement call, from Mount Nimba Integrated Nature Reserve. Note that the data for five calls are shown. ongoing human influence. This species is thus facing an alarming risk of extinction, and the long-term persistence of its population in the forthcoming years is unlikely. The most important conservation action would be the immediate protection of its habitat at Yéalé, possibly accompanied by the establishment of a captive rescue population. Currently, H. nimbae is classified as Endangered (IUCN 2017). However, based on JUCN Red List of Threatened Species criteria Al(a) (population decline directly observed, only one known population remaining), B1 (extent of occurrence less than 100 km/7), and B2 (area of occupancy less than 10 km7?, a: only one population known, b: population globally declining), we believe that the threat status of this species should be updated to Critically Endangered. The second study species, H. chlorosteus, is similarly exposed to habitat loss due to intensive deforestation. The few remaining primary forests on which it depends do not enjoy sufficient protection or sustainable management (e.g., Chatelain et al. 1996; Wood 2003; Bakarr et al. 2004). The MNINR, Mount Sangbé, and the lowland rainforest of Tai National Park remain the three known Ivorian sites of H. chlorosteus at present (R6del 2003; Ernst and Rodel 2008; Kanga et al. 2021). The recent IUCN Red List (2020) classifies H. chlorosteus as Least Concern. Given its large range, that category appears to be correct. However, based on the facts that in Ivory Coast the species only occurs in the westernmost parts of the country, several of the few national populations are gone, and the remaining forest habitats are declining, we think that H. chlorosteus should be regarded as Endangered nationally. In summary, we urgently recommend the strict protection of the habitats of both species. Amphib. Reptile Conserv. Acknowledgments.—We are greatly indebted to the Ministere de |’Environnement et du Développement Durable, and the Office Ivoirien des Parcs et Réserves, Direction de Zone Ouest, Céte d’Ivoire, for permitting the research. We would like to acknowledge the local populations of Danané villages for their hospitality, and are especially grateful for the support and collaboration from Paul Seu, elder of Kouan-Houlé. We thank Nicolas Granier, a conservationist of the Biotope Foundation France, for facilitating the relationships between the local people of Yéalé and our team. We are thankful to our local assistants and field guides Droh David Gueu and Zoda Alphonse Tokpa, for their invaluable help during the field expedition. Alan Channing and Lucinda Lawson improved our manuscript greatly with their constructive comments. We are grateful to Michael L. Grieneisen for his valuable comments. Literature Cited Bakarr M, Oates JF, Fahr J, Parren M, Rodel MO, Demey R. 2004. Guinean forests of West Africa. Pp. 123-130 In: Hotspots Revisited: Earth's Biologically Richest and Most Endangered Terrestrial Ecoregions. Editors, Mittermeier RA, Gil PR, Hoffmann M, Pilgrim J, Brooks T, Mittermeier CG, Lamoreux J, da Fonseca GAB. Conservation International and CEMEX, Washington, DC, USA. 390 p. Channing A. 2022. Color patterns to sequences: a perspective on the systematics of the Hyperolius viridiflavus group (Anura: Hyperoliidae) using mitochondrial DNA. Zootaxa 5134: 301-354. Channing A, Rodel MO. 2019. Field Guide to the Frogs and other Amphibians of Africa. Struik Nature, Cape October 2022 | Volume 16 | Number 2 | e318 Biology and conservation of Hyperolius nimbae and H. chlorosteus Town, South Africa. 407 p. Chatelain C, Gautier L, Spichiger R. 1996. A recent history of forest fragmentation in south-western Ivory Coast. Biodiversity and Conservation 5. 37-53. Emmrich M, Vences M, Ernst R, Kohler J, Bare; MF, Glaw F, Jansen M, Rodel MO. 2020. A guild classification system proposed for anuran advertisement calls. Zoosystematics and Evolution 96: 515-525. Ernst R, R6del MO. 2008. Patterns of community composition in two tropical tree frog assemblages: separating spatial structure and environmental effects in disturbed and undisturbed forests. Journal of Tropical Ecology 24: 111-120. Grafe TU, Stewart MM, Lampert KT, R6del MO. 2011. Putting toe clipping into perspective: a viable method for marking anurans. Journal of Herpetology 45: 28— 35. Gridi-Papp M. (Editor). 2003-2007. SoundRuler: acoustic analysis for research and teaching. Available: http://soundruler.sourceforge.net [Accessed: 22 October 2013]. Hillers A, Loua NS, Rodel MO. 2008a. Assessment of the distribution and conservation status of the viviparous toad Nimbaphrynoides occidentalis on Monts Nimba, Guinea. Endangered Species Research 5: 13-19. Hillers A, Veith M, Rodel MO. 2008b. Effects of forest fragmentation and habitat degradation on West African leaf litter frogs. Conservation Biology 22: 762-772. IUCN SSC Amphibian Specialist Group. 2017. Hyperolius nimbae. The IUCN Red List of Threatened Species 2017: e.T56170A 16926587. IUCN SSC Amphibian Specialist Group. 2020. Hyperolius chlorosteus (amended version of 2019 assessment). The IUCN Red List of Threatened Species 2020: T56123A176499160. IUCN. 2021. The IUCN Red List of Threatened Species. Version 2021-3. International Union for the Conservation of Nature, Gland, Switzerland. Available: https://www.iucnredlist.org [Accessed: 4 June 2022]. Kanga KP, Kouamé NG, Zogbassé P, Gongomin BAI, Agoh KL, Kouamé AM, Konan JCBYN, Adepo-Gourene AB, Gourene G, Rodel MO. 2021. Amphibian diversity of a West African biodiversity hotspot: an assessment and commented checklist of the batrachofauna of the Ivorian part of the Nimba Mountains. Amphibian & Reptile Conservation 15(1) [General Section]: 71—107 (e275). Kouamé NG, Adepo-Gourene AB, Konan JCBYN, Emmrich M, Penner J, Rodel MO. 2014. Second record of Ayperolius laurentii Schiotz, 1967 (Anura: Hyperoltidae) in south-eastern Ivory Coast, with observations on the species’ variability, call characteristics and habitat. Herpetology Notes 7: 59— 65. Kouamé NG, Assemian NE, Tohé B, Adeba PJ, Gouréne G, Rodel MO. 2016. Rediscovery of the Mount Amphib. Reptile Conserv. 100 Nimba Reedfrog, Hyperolius nimbae Laurent, 1958, in western Ivory Coast (Anura: Hyperoltidae). Herpetozoa 29: 3-13. Kouamé NG, Gongomin BAI, Rédel MO, Channing A. 2022. The taxonomic status of Hyperolius nimbae Laurent, 1958 (Amphibia: Anura: Hyperoltidae). Zootaxa 5174: 596-599. Kouamé AM, Kouamé NG, Konan JCBYN, Adepo- Gourene AB, Rodel MO. 2015. Contributions to the reproductive biology and behaviour of the Dotted Reedfrog, Hyperolius guttulatus, in southern-central Ivory Coast, West Africa. Herpetology Notes 8: 633-— 641. Kohler J, Jansen M, Rodriguez A, Kok PJR, Toledo LF, Emmrich M, Glaw F, Haddad CFB, Rédel MO, Vences M. 2017. The use of bioacoustics in anuran taxonomy: theory, terminology, methods, and recommendations for best practice. Zootaxa 4251: 1-124. Kpan TF, Ernst R, Rodel MO. 2021. Follow the forest: slow resilience of West African rainforest frog assemblages after selective logging. Forest Ecology and Management 497: 119489. Lamotte M. 1959. Observations écologiques sur les populations naturelles de Nectophrynoides occidentalis (Family Bufonidae). Bulletin Biologique 4: 355-413. Lamotte M, Rougerie G, Roy R, Schnell R. 2003a. Le Nimba et ses principaux biotopes. Mémoire du Muséum National d’Histoire Naturelle 190: 29-50. Lamotte M, Roy R, Xavier F. 2003b. Les premiers temps de |’étude scientifique et de la protection du Nimba (1942-1978). Mémoires du Muséum National d’Histoire Naturelle 190: 11-27. Lamotte M, Sanchez-Lamotte C. 1999. Adaptation aux particularités climatiques du cycle biologique d’un anoure tropical, Nectophrynoides occidentalis Angel, 1943 (Bufonidae). Alytes 16(3-4): 111-122. Lauginie F. 2007. Conservation de la Nature et Aires Protégées en Cote d’Ivoire. Editions CEDA/NEI et Afrique Nature, Abidjan, Ivory Coast. 668 p. Lotters S, Schick S, Scheelke K, Teege P, Kosuch J, Rotich D, Veith M. 2004. Bio-sketches and partitioning of sympatric reedfrogs, genus Hyperolius (Amphibia; Hyperoliidae), in two humid tropical African forest regions. Journal of Natural History 38: 1,969-1,997. Maley J. 1996. The African rain forest: main character- istics of changes in vegetation and climate from the Upper Cretaceous to the Quaternary. Proceedings of the Royal Society of Edinburgh 104B: 31-73. Penner J, Adum GB, McElroy MT, Doherty-Bone T, Hirschfeld M, Sandberger L, Weldon C, Cunningham AA, Ohst T, Wombwell E, et al. 2013. West Africa - a safe haven for frogs? A sub-continental assessment of the chytrid fungus (Batrachochytrium dendrobatidis). PLoS ONE 8: e56236. Portik DM, Bell RC, Blackburn DC, Bauer AM, Barratt CD, Branch WR, Burger M, Channing A, Colston TJ, October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. Conradie W, et al. 2019. Sexual dichromatism drives diversification within a major radiation of African amphibians. Systematic Biology 68: 859-875. Rodel MO. 2000. Herpetofauna of West Africa. 1. Amphibians of the West African Savanna. Edition Chimaira, Frankfurt am Main, Germany. 335 p. Rodel MO. 2003. The amphibians of Mont Sangbé National Park, Ivory Coast. Salamandra 39: 91-112. Rodel MO, Adum GB, Aruna E, Assemian NE, Bare} MF, Bell RC, Burger M, Demare G, Doherty- Bone T, Doumbia J, et al. 2021. Diversity, threats, and conservation of western and central African amphibians (Senegal, The Gambia, Guinea Bissau, Mali, Guinea, Sierra Leone, Liberia, Ivory Coast, Burkina Faso, Ghana, Togo, Benin, Nigeria, Niger, Cameroon, Gabon, SA0 Tome and Principe, Equatorial Guinea, Central African Republic, Chad, Republic of the Congo, Democratic Republic of the Congo, northern Angola). Pp. 11-101 In: Status and Threats of Afrotropical Amphibians — Sub-Saharan Africa, Madagascar, Western Indian Ocean Islands. Amphibian Biology, Volume 11, Part 7. Editors, Heatwole H, Rodel MO. Edition Chimaira, Frankfurt am Main, Germany. 243 p. Rodel MO, Bangoura MA, Bohme W. 2004. The amphibians of south-eastern Republic of Guinea (Amphibia: Gymnophiona, Anura). Herpetozoa 17: 99-118. Rodel MO, Branch WR. 2002. Herpetological survey of the Haute Dodo and Cavally forests, western Ivory Coast, Part I: Amphibians. Salamandra 38: 213-232. Rodel MO, Ernst R. 2003. The amphibians of Marahoué and Mont Péko National Parks, Ivory Coast. Herpetozoa 16: 23-39. Rodel MO, Gil M, Agyei AC, Leaché AD, Diaz RE, Fujita MK, Ernst R. 2005. The amphibians of the forested parts of south-western Ghana. Salamandra 41: 107-127. Rodel MO, Glos J. 2019. Herpetological surveys in two proposed protected areas in Liberia, West Africa. Amphib. Reptile Conserv. 101 Zoosystematics and Evolution 95: 15-35. Rodel MO, Lampert, KP, Linsenmair KE. 2006. Reproductive biology of the West African savannah frog Hyperolius nasutus Gunther, 1864 (Amphibia: Anura: Hyperoliidae). Herpetozoa 19: 3-12. Sandberger L, Hillers A, Doumbia J, Loua NS, Brede C, Rodel MO. 2010. Rediscovery of the Liberian Nimba Toad, Nimbaphrynoides liberiensis (Xavier, 1978) (Amphibia: Anura: Bufonidae), and reassessment of its taxonomic status. Zootaxa 2355: 56-68. Sandberger-Loua L, Doumbia J, Rodel MO. 2016. Conserving the unique to save the diverse. Identifying key environmental determinants for the persistence of the viviparous Nimba toad in a West African World Heritage Site. Biological Conservation 198: 15—21. Sandberger-Loua L, Miller H, Rodel MO. 2017. A review of the reproductive biology of the only known matrotrophic viviparous anuran, the West African Nimba Toad, Nimbaphrynoides occidentalis. Zoosystematics and Evolution 93(1): 105-133. Sandberger-Loua L, R6del MO, Feldhaar H. 2018. Gene- flow in the clouds: landscape genetics of a viviparous, montane grassland toad in the tropics. Conservation Genetics 19: 169-180. Schiotz A. 1967. The treefrogs (Rhacophoridae) of West Africa. Spolia Zoologica Musei Haunienses 25: 1-346. Schiotz A. 1999. The Treefrogs of Africa. Edition Chimaira, Frankfurt am Main, Germany. 351 p. Veith M, Kosuch J, Rodel MO, Hillers A, Schmitz A, Burger M, Lotters S. 2009. Multiple evolution of sexual dichromatism in African reed frogs. Molecular Phylogenetics and Evolution 51: 388-393. Woods D. 2003. The tragedy of the cocoa pod: rent- seeking, land, and ethnic conflic in Ivory Coast. The Journal of Modern African Studies 41(4): 641-655. Xavier F. 2009. La belle histoire du petit crapaud vivipare du Mont Nimba. Bulletin de la Société Zoologique de France 134: 13-21. October 2022 | Volume 16 | Number 2 | e318 Biology and conservation of Hyperolius nimbae and H. chlorosteus ee | CHASSE, P Amphib. Reptile Conserv. Basseu Aude-Inés Gongomin is a Ph.D. candidate at the Universté Jean Lorougnon Guédé (Daloa, Ivory Coast). Her current research topic focuses on taxonomy, and the population ecology of endemic amphibians of the Ivorian part of the Nimba Mountains. N’Goran Germain Kouamé is an Ivorian herpetologist and biologist. He is a Senior Lecturer at the Jean Lorougnon University (Daloa, Ivory Coast), and is also the current Chair of the West African region of the IUCN SSC Amphibian Specialist Group (ASG). He holds a Diploma and a Ph.D. in natural sciences from the Université d’Abobo-Adjamé (actually Université Nangui Abrogoua), Abidjan, Ivory Coast, where he used leaf-litter frogs (Phrynobatrachus spp.) as models to determine the conservation status of the Banco National Park, one of the rare remaining primary forests situated in the midst of a West African mega-city. His current research interests mainly focus on the taxonomy, ecology, distribution, and conservation of rare, threatened, and new amphibian species in Ivory Coast. Konan Laurent Agoh is a postgraduate student at the Université Jean Lorougnon Guédé (Daloa, Ivory Coast) with an M.Sc. in Herpetology. His research interest focuses on the diversity and ecology of amphibian assemblages in forest fragments of the valley of the Bandama River, an area which lies between Lamto Faunal Reserve (central Ivory Coast) and Azagny rainforest (southern Ivory Coast). Kouassi Philippe Kanga is a recent graduate of the Université Jean Lorougnon Guédé (Daloa, Ivory Coast) with a Ph.D. in Herpetology. His work emaphsized amphibian diversity and its spatio- temporal dynamics in the Ivorian part of Mount Nimba. His research was under the direction of Dr. N’Goran Germain Kouamé, and aimed at linking amphibian assemblages with species function and assessing how species react in response to environmental changes in Mount Nimba. Mark-Oliver Rédel works with his team at the Museum fir Naturkunde, Berlin, Germany, where he is the Curator of Herpetology. He has studied the systematics, taxonomy, biogeography, and ecology of African amphibians for almost 30 years. With his group, comprising students from around the world, he currently runs projects in Germany, Ecuador, Guinea, Ivory Coast, Cameroon, and Mozambique. His special interest is on how species and ecological communities react to environmental changes. He has authored or co-authored about 400 scientific and popular publications, including several books. 102 October 2022 | Volume 16 | Number 2 | e318 Gongomin et al. Appendix 1. Counts of Hyperolius nimbae individuals captured ina heavily degraded site (07°31.928’N, 008°25.401’W; A425 m asl) at Yéalé from 2 August 2019 to 8 July 2020. Appendix 2. Counts of Hyperolius chlorosteus individuals captured along forest streams at Mount Nimba Integrated Nature Reserve (MNINR: 07°34.652’N, 008°24.966’W; 716 m asl) and Yéalé (07°31.932’N, 008°25.508’W; 387 m asl) from 2 August 2019 to 8 July 2020. SS |) 0") POF eS oS [a ole [ss lie ee Amphib. Reptile Conserv. 103 October 2022 | Volume 16 | Number 2 | e318 Amphibian & Reptile Conservation 16(2) [General Section]: 104-119 (e319). Official journal website: amphibian-reptile-conservation.org Goliath Frog (Conraua goliath) abundance in relation to frog age, habitat, and human activity 1*Nono LeGrand Gonwouo, ?Marvin Schafer, *Sedrick Junior Tsekané, 24Mareike Hirschfeld, 'F. Arnaud M. Tchassem, and 24Mark-Oliver Rodel ‘Laboratory of Zoology, Faculty of Science, University of Yaoundé I, P.O. Box 812, Yaoundé, CAMEROON *Museum fiir Naturkunde — Leibniz Institute for Evolution and Biodiversity Science, InvalidenstraBe 43, D-10115 Berlin, GERMANY *Department of Biology and Animal Physiology, University of Douala, P.O. Box 24157 Douala, CAMEROON *Frogs & Friends e.V., GroBbeerenstrape 84, 10963 Berlin, GERMANY Abstract.—Habitat change and overexploitation are major factors driving species population declines worldwide, and they often act in union. The Goliath Frog, Conraua goliath, is an iconic species that is known to be extensively exploited by humans. However, Goliath Frog populations have not yet been assessed quantitatively in relation to their proximity to human settlements, nor has the loss of terrestrial habitat adjacent to the frogs’ riverine habitat been investigated. In this study, populations of the Goliath Frog were assessed across its range in Cameroon during nocturnal, time-constrained, visual encounter surveys. Goliath Frogs showed a patchy distribution along torrent rivers in three main habitat types: primary forest, secondary forest, and agroforestry plantations. There were no significant differences in the encounter rates among the three habitat types. However, we noted higher frog abundances, including larger sized adults, with increasing distance from human settlements, an observation confirmed by local frog hunters. Our observations revealed strong segregation in microhabitats with respect to age classes, as juvenile frogs were frequently found along river beds with rock pools/rock crevices, while sub-adults were mostly encountered around exposed rocks at river rapids, and adults were mostly recorded near cascades and waterfalls. The adults predominately perched on rocks around waterfalls and rapids, with distances of about 3-5 m between them, suggesting both territoriality and site fidelity. Adults were observed foraging at night, beyond 10 m from the river bank. During the day, adults were seen basking on rocks along the river bank. The lower abundance and size of Goliath Frogs near human settlements indicates the effects of hunting pressure, with terrestrial habitat showing less of an effect on this species. Monitoring of the remaining Goliath Frog populations, raising local awareness on the effects of hunting and habitat preservation, as well as law enforcement, are suggested as further efforts to conserve the world’s largest frog species. Keywords. Amphibia, Anura, Cameroon, conservation, exploitation, habitat choice, monitoring, threatened species Citation: Gonwouo NL, Schafer M, Tsekané SJ, Hirschfeld M, Tchassem FAM, Rédel M-O. 2022. Goliath Frog (Conraua goliath) abundance in relation to frog age, habitat, and human activity. Amphibian & Reptile Conservation 16(2) [General Section]: 104-119 (e319). Copyright: © 2022 Gonwouo et al. This is an open access article distributed under the terms of the Creative Commons Attribution License [Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are 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: amphibian-reptile-conservation.org. Accepted: 29 August 2022; Published: 15 November 2022 Introduction The world’s biodiversity crisis is almost exclusively due to human activities, most notably the conversion and destruction of natural habitats. However, the overexploitation of many species, such as for food, is an increasingly serious threat as well. Frogs are no exception to this trend (Mohneke et al. 2010; Altherr et al. 2022), and both of these threats may also affect the world’s largest frog, Conraua goliath (Boulenger 1906). This species is restricted to southwestern Cameroon and northern Equatorial Guinea, where it occurs in lowland to mid-altitude rainforests below 1,000 m asl (Lamotte Correspondence. */gonwouo@gmail.com Amphib. Reptile Conserv. and Perret 1968; Sabater-Pi 1985; Wild et al. 2004; Stuart et al. 2008; Channing and Rodel 2019). This frog is associated with fast flowing rivers and larger streams with rocky outcrops, rapids, and waterfalls (Perret 1957; Amiet 1975; Sabater-Pi 1985; Gewalt 1996; Herrmann et al. 2005). These natural habitats are becoming progressively altered through various human activities, such as conversion to farmland, construction of roads and hydroelectric dams, and exploitation for artisanal and commercial timber resources. The combination of logging and conversion of the remaining forests to agroforestry plantations has tremendous negative consequences on biodiversity, including amphibians. The November 2022 | Volume 16 | Number 2 | e319 Gonwouo et al. progressive fragmentation of original forest landscapes leads to modified microclimates with obvious adverse effects on amphibian communities (e.g., Ernst et al. 2006; Stuart et al. 2004, 2008; Ernst and Rodel 2008; Hillers et al. 2008; Ofori-Boateng et al. 2013). Previous studies suggested that fragmentation and destruction of habitat has led to a reduction in Goliath Frog populations at various Cameroonian sites (Amiet 2004; Herrmann et al. 2005). If this is true, it is likely that these processes would cause general population decline over the entire range of the species. In addition, frog meat may be an important (or at least a much valued) protein source for many indigenous people throughout the Goliath Frogs’ range (Gonwouo and Rodel 2008). Hence, the increasing human population, expansion of settlements and increased efficiency of hunting tools will intensify the pressure on this species. This is especially evident in the professionalized hunting methods (traps, hooks, spears, throwing nets, etc.) that have been developed specifically to collect Goliath Frogs (Amiet 2004; Gonwouo and Rodel 2008; Schafer et al. 2019). The hunt for subsidiary consumption, as well as for local bush meat markets, might be one of the main factors driving the population decline of Goliath Frogs. The commercial harvesting of Hoplobatrachus occipitalis has contributed to the population decline of this species in northern Bénin (Mohneke et al. 2010). Based on its rarity and much larger size (assuming a longer time until frogs become mature), similar or even more severe consequences may be assumed for the Goliath Frog. As a result of these pressures, this species 1s currently listed as Endangered by the IUCN and Class A under Cameroonian law (IUCN Amphibian Specialist Group 2019a; NLG et al., unpub. data). Previous studies on the Goliath Frog have focused mostly on its distribution (Perret 1957, 1960; Sabater- Pi 1962, 1967; Amiet and Perret 1969; Gewalt 1977), taxonomy, and phylogeny (Lamotte and Perret 1968: Neguiffo et al. 2019; Blackburn et al. 2020). Some investigations have also investigated various aspects of life-history, including larval development, parasites, and reproduction (Lamotte et al. 1959; Perret 1957, 1960; Sabater-P1 1985; Nguiffo et al. 2015). Parental care has recently been documented in the species (Schafer et al. 2019), and additional studies by the authors of this paper are in progress to further improve our knowledge of the life-history of this species. However, very little is known concerning the population trends and habitat preferences of this species, or the specific threats that this species is facing. In the absence of research on the habitat requirements and responses to the various threats facing the Goliath Frog, the development and implementation of appropriate conservation measures are difficult. To remedy this general lack of knowledge, this study examines the impact of land use and proximity to human settlements on the relative abundance, demographics, and body size of Goliath Frogs. The data presented in Amphib. Reptile Conserv. this paper are based on six years of investigations on the Goliath Frogs in Cameroon, and allow us to examine the correlations between this species and human-caused forest alteration and to propose the directions and goals of future research and conservation strategies. Material and Methods Study Area Fieldwork was carried out from November 2014 to December 2019 during both rainy and dry seasons (Table 1), although the dry seasons (November to February) were emphasized as the rivers were more accessible. Fieldwork focused on the areas around three main localities in south-western Cameroon: Moungo, Sanaga maritime, and Nkam division (Littoral region); Nyong- Ekele (Central region); and Bipindi in Ocean division (South region) (Fig. 1). In total, 13 rivers (Table 1) were surveyed, including the Nkam and Sanaga rivers. Investigations were carried out from near sea level around Kribi (Ocean division), up to the foothills of Mount Manengouba near Nkongsamba (Moungo division). The latter locality hosts the northernmost population of the Goliath Frog and is characterized by several large rivers and streams. Overall, the landscapes of our sites constituted mostly low to medium elevation habitats, with the elevations of our frog observations ranging from 39 m asl along Lobe River (Bifa, Ocean division) to about 677 m asl along Nkam River (Nkoungsou, Moungo division). Other than the Moungo area, which is characterized by a heterogeneous, mountainous landscape (Mts. Kupe, Nlonako, Manengouba), the remaining sample localities have a low-rise or flat relief, only rarely interrupted by hills. The study areas comprised a mixture of several large forest patches of Guinea-Congolian lowland rainforest (both pristine and logged), agroforestry plantations, and small-scale subsidiary agricultural sites. Especially in the area around Yabassi (Nkam division), logging companies commercially exploit timber for exportation, despite the remoteness and difficulty in accessing the area. The entire study region has a tropical climate, with the wet season extending from March to October and the dry season from November to February. Rainfall peaks in August and September, and the driest period extends from late December to the end of February. The annual precipitation ranges from 2,000 to 3,000 mm (Amiet 1975). Surveys and Data Acquisition The visual encounter survey (VES) method (Heyer et al. 1994; Rodel and Ernst 2004) was used in suitable habitats to systematically survey for Goliath Frogs. The VES consisted of counting all Goliath Frogs encountered in every major habitat type, and provided an encounter rate per person-hour. Three major habitat types were November 2022 | Volume 16 | Number 2 | e319 Goliath Frog populations in Cameroon Table 1. River sites investigated for Goliath Frogs in three surveyed habitat types in western Cameroon. The information provided includes river name, total length of sampled trails, geographic position, and a short habitat description (including length of river investigated and the total sampling effort per site in person-hours). Locality (river name, coordinates, elevation [m asl]) Habitat characteristics (including approximate length surveyed for each section and person-hours of effort for each site) Section one: Composed of logged secondary forest (SF) with open to semi- Poe PAD Scene closed canopy forest which covered about 70% of the surveyed trail, ~250 m. 2014 02°36'47.23”N, 10°01°05.51”E: 39 masl Section two: Small-scale cocoa plantation with few native trees interspersed in the plantation and a relatively open canopy, ~150 m (9.77 person-hours). Section one: River bank bordered by primary forest with a closed canopy of native trees that were about 25-30 m high. Open understory including leaf litter on floor. Steep flanks at some sites that were difficult to access, with no rs signs of human activity, ~1,500 m 14 February e . MALY 2 ' - 2 a NY ce Section two: Selectively logged secondary forest along a path with constant 9°58’06.18”E; 231 m asl human signs. Trees about 25 m high with relatively open to closed canopy. Secondary growth trees observed where logging had been carried out, ~400 m (13.74 person-hours). Section one: Composed of logged forest along steep portions of the river where farming activities are impossible. Many footpaths present and rampant Ekomtolo wood extraction for local furniture and domestic fuel, ~300 m. 12 February 4°47°32.57°N, 2015 Section two: Composed of large- to small-scale cocoa plantations with 9°53’11.60”E; 332 m asl few native trees spaced all over the area and constantly managed by the community. Chemicals used to sustain the crops are processed in the nearby river with possible contamination, ~500 m (4.25 person-hours). Forest composed of a mosaic of primary forest, secondary forest, and agroforestry plantation interspersed all through the surveyed trail. Section one: Mainly closed canopy of native trees, about 25—30 m tall, along difficult-to-access terrain, no previous logging had occurred. Forest floor with Dibombe ; about 70% leaf litter cover, ~900 m. 15 February 9°4 Bs eee — 74 acl a Section two: Old selectively logged forest, easy access due to the many i footpaths present, empty cartridges left behind by hunters, ~400 m. Section three: Small-scale plantation, mainly composed of cocoa and banana plants which covered the flat sections along the river on both flanks, ~300 m (14.26 person-hours). Sta MNna Mipein Section one: Primary forest on slopes along the river where movement and St eee 7-12 July 2016 | tree exploitation is difficult, ~1,000 m. 10° a So Leet Section two: Included portions where access was easier with several human ° foot paths present, signs of forest exploitation, ~650 m (10.55 person-hours). Section one: Composed of small patches of native large trees around difficult- Sanaga (Tributary River) 13-15 July 2016 to-access points of the river, bordered by very large rocks, ~700 m. 4°03’23.19’N, Section two: Composed of recently logged forest patches, with several hunting 10°37? 12.88”E; 297 m asl paths. Frequent use of the forest to collect non-timber forest products, ~600 m (14.55 person-hours). Section one: Consisted of closed canopy trees of about 25 m in height on both meinke sides of the river. Footpaths present within the forest seem to be regularly used 2-3 March 2017 | by fishermen and Goliath Frog hunters, ~900 m. 3 -4 2 me is NM Section two: Recently logged forest with open canopy and bushy understory. 10°04’56.75”E; 48 m asl Regular use of this forest section evident, with many footpaths present, ~200 m (11.72 person-hours). Amphib. Reptile Conserv. 106 November 2022 | Volume 16 | Number 2 | e319 Gonwouo et al. Table 1 (continued). River sites investigated for Goliath Frogs in three surveyed habitat types in western Cameroon. The information provided includes river name, total length of sampled trails, geographic position, and a short habitat description (including length of river investigated and the total sampling effort per site in person-hours). Locality (river name, coordinates, elevation [m asl]) Magatnba 6 April 2016 4°45'19.03"N, 9°52'16.68"E; 308 m asl Habitat characteristics (including approximate length surveyed for each section and person-hours of effort for each site) Trail bordered by agroforestry plantations as well as subsistence plantations on both banks. Fallow land present along river bordered by degraded forest with very dense vegetation, edges with shrubs and only a few trees present, ~600 m (6 person-hours). Section one: Consisted of agroforestry plantations, mainly cash crops Nkam 29 October —2 November 2018 5°08'17.43'N, 9°59'43.17"E; 677 m asl including coffee, cocoa, and palm oil trees growing on the river bank, ~800 m. Section two: Mainly composed of small relic forest patches on steep valleys along the river, access was difficult. No possibility of farming at this site, but forest patches appeared to have been logged with several foot paths found, ~350 m (4.96 person-hours). Section one: Vegetation composed of a mosaic of secondary forest and agroforestry plantations on both river banks. Secondary forest composed of Mbo fallow land, more than 10 years old, and several cocoa plants and large palm 6 April 2016 4°49°39.16"N, 9°47 18.58”E; 465 m asl trees could still be found. Section two: Permanently cultivated plantation with young cocoa and palm trees, intensively managed with signs of constant human presence, ~250 m (14.97 person-hours). Mpoula 27 February — 5 4°38'15"N, 9°43'07"E; 200 May 2018 m asl Njuma 27 August —17 September 2019 4°20°53.1°N, 10°13’56.3”E; 304 m asl Bisoue 27 August —17 September 2019 4°21°38.3°N, 10°12’30.4”E; 152 m asl identified along the 13 rivers: primary or pristine forest (PF; Fig. 2A), selectively logged or secondary forest (SF; Fig. 2B), and agroforestry plantations (AP; Fig. 2C, also see Table 1 and below for detailed descriptions of the habitat types). Transects along the rivers could comprise either a single vegetation type, or a mosaic of different vegetation types or segments (e.g., PF, SF, and AP) that could vary considerably in short succession. This was especially true when human settlements were nearby. To measure the portion of a certain vegetation type in a single transect, we passed the respective segment and measured the covered distance with a GPS unit. The vegetation type was identified on both sides of the river by assessing the canopy cover (visual estimation to 25% accuracy), estimated height (4+ 5 m), and measured diameter at breast height (DBH) (+ 0.5 cm) of the trees, as well as any obvious human impacts, such as selective logging, hunting, or extraction of non-timber forest products (e.g., leaves, tree bark, fruits, resins, or roots). Amphib. Reptile Conserv. Surveyed trail composed of a mosaic of secondary forest (~200 m) and agroforestry plantation (~300 m) on both sides of the river, constant human activities noted along the trail (10.8 person-hours). River bordered by primary forest (~1,200 m) on both sides with little or no signs of human activity, though foot paths where present and have been used by poachers and seasonally by Goliath Frog hunters (5.31 person-hours). River bordered by primary forest (~ 400 m) on both sides with little or no human impact. Access was difficult at some points due to the dense undergrowth along old footpaths (15 person-hours). Primary Forest (PF) consisted of closed canopy forest with 75—-100% canopy cover. This forest type was dominated by large, native trees of about 25—30 m in height, although the largest exceeded 50 cm in DBH. No evidence of recent logging was present in PF. Although hunting or fishing paths were regularly found along the rivers (especially close to human settlements), revealing some degree of natural resource exploitation, this forest type was still considered mature and relatively undisturbed. Primary Forest comprised seven segment portions of the 13 rivers surveyed, and their lengths and brief descriptions are provided in Table 1. Selectively logged Secondary Forest (SF) included all vegetation formations with a relatively closed canopy (50-75% canopy cover) and medium to large trees (10- 25 m in height). Here, the tree composition included both native and non-native tree species with DBH usually exceeding 40 cm, including many secondary growth trees. These forests had been previously logged for commercial timber exportation, and/or by the local November 2022 | Volume 16 | Number 2 | e319 Goliath Frog populations in Cameroon 10°0'0.000°E 5°0'0.000"N Mpo Ekomtolo Magamb#a ¢. ene, Nkebe Mpoul? ibomie Littoral 4°0'0.000"N 2 [=] o = Ss So ° om pe obe 2°0'0.000"N 10°0'0.000"E e kam AS einke 11°0'0.000°E 5°0'0.000"N goisoue Njuma 4°0'0.000"N 3°0'0.000"N “=polowa Legend © Study sites Hl Rivers —— Main roads GB South and Littoral region Cameroon {__] Other borders 2°0'0,000"N 11°0'0,000°E Fig. 1. Map of Cameroon indicating the locations of the study sites. population for house construction or for local trade. Many footpaths, as well as the remains of abandoned logs, snare traps, and rifle cartridges, were found within these forests, indicating ongoing and constant use for hunting and timber exploitation. Secondary Forest comprised ten segment portions of the 13 rivers surveyed. Agroforestry Plantations (AP) included — all plantations, ranging from small-scale subsidiary crops and/or cash crops cultivated by the local people to intensively farmed, large-scale monocultures created by international commercial agro-companies. This habitat type, especially when cultivated by local farmers, could include native trees with more or less open canopies (< 50%), the largest stem diameters exceeded 40 cm, and they were spaced all throughout the cropped species. Most of the cultivated plants were introduced species, including manioc (Manihot esculenta), papaya (Carica papaya), pineapple (Ananas comosus), mango (Mangifera indica), cocoa (Theobroma cacao), palm oil (Elaeis guineensis), avocado (Persea americana), and bananas (Eumusa spp.). Native large trees were found mostly in cocoa plantations as shade-trees, whereas banana tended to be cultivated in large monocultures. The latter is cultivated at a commercial scale in the Njombe-Penja area for exportation. Agroforestry Plantations comprised seven Amphib. Reptile Conserv. segment portions of the 13 rivers surveyed. To enhance crop production, herbicide and pesticide mixtures are prepared in nearby rivers and streams with likely runoff, polluting the rivers. Sampling took place between 0700 and 1200 h, and included various microhabitats, such as waterfalls, rocky rapids, rock pools, and riverbanks, as well as forest strips up to 20 m from the rivers. Gaining access to the rivers and their banks demanded careful clearance of trails, e.g., removing of some lianas or dead wood along a narrow trail. Trails were set-up at least 24 h before surveying the respective area. To maximize the probability of documenting all individual frogs, teams of two or three researchers walked along the rivers at a slow, steady rate of 0.2 m per second, avoiding any jerky movements that could disturb the frogs. Headlamps and handheld flashlights were used to detect the frogs, particularly by picking up eye-shine. All spatial data were recorded with a Garmin GPS (60 cx; accuracy of 5-10 m). For every frog, the perch site, date, time, posture, size, and distance from the riverbed were recorded. Size categories were classified as follows: adults (= 19 cm snout-vent length) (Fig. 3A); subadults (approx. 10-18 cm) (Fig. 3B); and juveniles (< 9 cm) (Fig. 3C). Daytime habitat assessments preceded the nighttime November 2022 | Volume 16 | Number 2 | e319 Gonwouo et al. ee > ie £ Fig. 2. Examples of the three different forest types investigated in this study: (A) primary forest (River Dibombe), (B) secondary forest (River Nkam), and (C) agroforestry plantations (River Mpoula). surveys, with notes taken on habitat features, such as dominant vegetation, as well as notable, anthropogenic influences on the sites. Notes on the microhabitats of sun-basking frogs were made accordingly. Data for the daytime searches are not generally comparable to the nighttime searches, given that the frogs are mostly active during the night. However, the locations of daytime frog encounters regularly coincided with the presence of similarly-sized individuals during the night. Local Ecological Knowledge (LEK) was obtained from local frog hunters around the surveyed areas who could reliably identify Goliath Frogs and regularly hunt them. This information was collected through informal interviews and discussions, and it greatly contributed to our assessments. To prevent any biasing toward certain answers, we asked all of the respondents the following six questions: Which are the rivers where Goliath Frogs are present? How far are they from the village? What was your biggest catch ever, and when was that? What was your biggest catch in 2019? How often do you hunt for the frogs? What is your perception about the Goliath Frog population around the village? Data Analysis The sampling effort was recorded only for the nocturnal surveys. Daytime searches were not time constrained as they were mostly meant to identify nocturnal survey sites. Amphib. Reptile Conserv. Sampling effort is given in person-hours, 1.e., the number of hours spent surveying multiplied by the number of observers for any given river and habitat type (Table 2). The relative abundance of frogs was calculated as the number of individuals observed per time unit, divided by the number of sightings through the total sampling effort for each river or habitat type, and given as frogs per person-hour. As the relative abundances of Goliath Frogs were not normally distributed, Pearson’s Correlation Coefficient was applied to compare the abundance of frogs and age groups per habitat type. In order to examine frog abundance in relation to human presence, the GPS coordinates of the study sites were used and a 10 km buffer around each point was drawn in a geo-information system (QGIS Development Team 2021). All streets and settlements within the buffer zone were extracted from the Open Street Map database (https://planet.openstreetmap.org), and the total length of all roads (motorways, interregional and regional highways, urban as well as agricultural roads) was determined, as well as the number and type of settlements. No recent, fine-scale census data are available for the study area, thus the human population within each buffer zone was estimated by assigning fixed values to each of the different settlement types. Hamlets, the smallest type of settlement, accounted for 200 people, villages for 1,500, towns for 25,000, and cities for 100,000 inhabitants. Note that these values were November 2022 | Volume 16 | Number 2 | e319 Goliath Frog populations in Cameroon Ps ea NC ad wee t Fig. 3. The three different age sizes of Goliath Frogs considered in this study: (A) adult (= 19 cm), (B) subadult (10-18 cm), and (C) juvenile (< 9 cm). based on the open street maps criteria. Human population density was calculated by dividing total population by the area within the 10 km radius buffer zone (314 km/?). The distance of each sampling site to the nearest road and settlement boundary were determined. Subsequently, the data were screened for any linear dependencies of the GIS extracted values and the number of observed frogs (iph) in a regression analysis. Distance measures were log-transformed before the analysis. To analyze the LEK data, a Welch-test comparing the hunter’s perceived frog weights in different years was performed. A linear regression model was used to determine if there was a correlation between frog weight measured by local hunters and distance to the nearest settlement. All analyses were conducted using R v.4.0.0 (R Development Core Team 2014). Results Encounter Rates and Distances to Settlements and Roads A total of a 100 person-hours were spent surveying for Goliath Frogs at 13 rivers across the entire range of the species in southwestern Cameroon. During the study, 490 frogs were observed along 26.7 km of riverine habitats, including 13.1 km in PF, 7.0 km in SF, and 6.6 km in AP. The number of person-hours spent on each Amphib. Reptile Conserv. habitat type varied, given the differences in river size and habitat accessibility. The encounter rates of frogs varied with respect to habitat types, rivers, and with grade of anthropogenic influence. For the entire study period, the average encounter rate was five frogs per person-hour (5 iph). Within the three habitat types, the highest mean iph was 8.2 recorded in SF, followed by 7.1 in PF, and 4.6 in AP. However, these values were not statistically different (Fig. 4). The two individual study sites with the highest encounter rates (17.0 iph) were both in SF, along the Sanaga and Keinke rivers. At these sites, the forest had been legally and commercially logged. These localities are far from human settlements and the frogs seemed to show lower flight-distances when approached compared to the frogs at other sites (however, we did not collected data to support this general impression). The lowest encounter rates were along the Ekomtolo and Mpoula Rivers (1.0 iph), and both localities are close to human settlements. Although disturbed, the surveyed portion of Mbo River, which is bordered by about 50% AP, revealed very high encounter rates (8.4 iph). Human densities and the levels of anthropogenic disturbance varied considerably around the study sites. Total road length within a 10 km radius buffer around the sites ranged between 9 km and 168 km, with a mean of 84.5 km. The number of settlements ranged from 1 to 20 within the buffers, with a mean value of 10.1. The estimated human population densities ranged from November 2022 | Volume 16 | Number 2 | e319 Gonwouo et al. Table 2. Encounter rates of Goliath Frogs (in person-hours of searching effort) during time-constrained visual encounter surveys in the three different habitat types: PF, primary forest; SF, secondary forest; and AP, agroforestry plantation. Data are given for each of three different age sizes: a, adult; s, subadult; and j, juvenile. Habitat type PF Age size a S j Dy Lobe Nkebe 2.07 133 0.93 4.33 Ekomtolo Dibombe 3.50 2.33 1.33 7.17 Sanaga Ngo Mpem 240 1.05 1.05 4.50 Sanaga (Tributary River) 6.43 429 1.71 = 12.43 Keinke 3.67 5.67 1.33 10.67 Magamba Nkam Mbo Mpoula Njuma 163 1.25 0.00 2.88 Bisoue 3.38 450 0.00 7.88 Mean 3.29 2.92 0.91 7.12 1,000 to 99,200 inhabitants in the 314 km? of the buffer zones (12.7—1,263.7 persons per km?), with a mean value of 32,592.3 (415.2 persons per km’). Road length and population density were negatively, but not significantly, associated with higher frog numbers (Table 3). The distances between study sites and the nearest settlement ranged from 89 to 9,114 m (median = 2,741 m), and the distances between study sites and the nearest road ranged from 176 to 8,653 m (median = 701 m). While both measures (after logarithmic transformation) indicated a positive association between distance and frog numbers, only the nearest settlement showed a robust and significant linear dependency (Table 3, Fig. 5). In other words, more frogs were found when the nearest settlement was farther away. Only the Njuma River violated this rule, as it was the most remote site but only provided a small number of frogs (Table 3). Age Categories and Microhabitats Of the 490 Goliath Frog observations, 243 (49%) were adults, 170 (35%) were subadults, and 77 (16%) were juveniles. Of all the adults, 48% where from PF, 36% from SF, and 16% from AP. The encounter rates of the three age sizes varied among the rivers, as well as both between and within the different habitat types. A Chi- square test of the three age groups showed that there was no difference in the population structures between the three different habitat types (v7 = 3.48, df= 2, p = 0.48). The highest encounter rates among the three age sizes were for adults (9.0 iph) recorded in SF; while the lowest was in juveniles (1.0 iph) recorded in all three habitats (PF, SF, and AP). The recorded encounter rates Amphib. Reptile Conserv. SF AP a S j x a S j x 7.20 240 0.00 9.60 4.00 600 0.00 10.00 245- 225 050 5.50 100 0.00 0.00 1.00 180 080 0.20 2.80 3.00 1.88 413 9.00 200 250 0.00 4.50 2.77 2.08 1.62 6.46 9.00 650 1.50 17.00 9.00 7.50 0.00 16.50 0.89 0.78 0.33 2.00 429 257 0.00 686 2.63 150 0.00 4.13 420 1.20 1.20 660 240 420 1.80 8.40 2.17 O33 1.00 3.50 044 0.00 0.22 0.67 454 2.67 0.99 8.20 2.02 2.25 0.37 4.64 of the frogs, sorted by age sizes within the three surveyed habitat types, are summarized in Table 4. Goliath Frogs showed a patchy distribution across the study sites and were mostly restricted to particular microhabitats, which included moderate to fast flowing rivers with cascading, turbulent rocky sections (Fig. 6), or waterfalls with mostly sandy soils. The species was recorded from Bifa, Babong, Ekomtolo, Magamba, Manengotang, Nko-Olong, Ngo-Mpen, Nkongsou, and Sole. The altitudinal range of the inhabited sites spanned from 39 m asl at Lobe River around Bifa and near the coast, to about 677 m asl at Nkam River. Goliath Frogs were recorded in 75% of the surveyed rivers, but only when these included suitable microhabitats. The inhabited river sections surveyed varied from 50 m to more than 300 m. The torrent, rocky sections were inhabited while the in-between sections, slow moving, meandering river parts with no rocks, revealed no frogs. Goliath Frogs were completely absent from rivers lacking fast flowing sections and rocks. For example, the five rivers south of the newly constructed deep-sea port at Kribi that were investigated all lacked the above- mentioned microhabitats, and yielded no Goliath Frogs despite being within the range of the species. Discussions with local ethnic groups, including the Bagyli/Bakola people who have lived in this forest for many generations, confirmed that Goliath Frogs never occurred in this area. Age dependent differences were noted in microhabitat use. When Goliath Frogs came out at night to perch on rocks, the adults used the areas around large cascading waterfalls, while sub-adults were more often present on rocks in the rapids, and juveniles inhabited rock pools and crevices. Adult frogs were abundant around waterfalls November 2022 | Volume 16 | Number 2 | e319 Goliath Frog populations in Cameroon 15 10 Individuals per person-hour (ibh) PF SF AP Habitat type Fig. 4. Number of individual Goliath Frogs recorded per person- hour in each of the three surveyed habitat types: PF = primary forest, SF = secondary forest; AP = agroforestry plantation. Note that the encounter rates of frogs were not statistically different between any of the three habitat types. but maintained some distance from one another. The closest distance between two adult frogs was 3 m, observed at Dibombe River. A maximum of seven adult individuals were observed at a single waterfall along Nkebe River. On Mpoula River, adult frogs were seen sitting on large branches within jumping distance of the stream (~5 m in adult frogs), at more than 2 m above the ground. At night, adult frogs that had fled from the surveyors by diving into the stream returned to the same perching rocks shortly (10 to 15 min) after disturbance, indicating fidelity to particular perching sites. Adults and subadults (7 = 24) were frequently observed at night on low bushes and trees more than 10 m away from the riverbed. During the daytime, our observations revealed extensive sun-basking behavior, 1.e., more than 20 adults were seen sun-basking throughout the study period. Adults often leaped and dove into the rivers upon our approach (flight distance 4 to 10 m). One individual was observed basking on the same rock on three consecutive days, along a relatively calm portion of Ekomtolo River. When disturbed, this frog jumped into the river and returned to its perch site after about 30 to 45 min. Presence at Sites Impacted by Pollution and Agricultural Activities The data revealed that Goliath Frogs persist in forest fragments, plantations, and rivers, even when surrounded by human settlements. Observations from the localities of Magamba and Manengotang indicate that small populations can be present at about 200 m from human settlements. Here, the habitat was patchy and comprised a mosaic of small forest remnants and small subsidiary plantations. The Goliath Frog occurrence at these places Amphib. Reptile Conserv. Distance to settlement 7 Individuals per person-hour Fig. 5. Scatterplot of Goliath frog abundance (as individuals per person-hour) and log transformed distance to the nearest settlement. The red line is the trendline of the fitted linear model and the blue dotted lines demarcate the 95% confidence interval. Note that the top outlier point refers to Njuma River, which was the most remote site sampled, however, it also accounted for one of the lowest numbers of individuals. Fig. 6. Typical forested and rocky-sandy riverbed characteristics for Goliath Frog habitat (Dibombe River). was also confirmed by Goliath Frog hunters, and frog hunting at these sites was perpetual. We commonly observed habitat pollution by the dumping of household waste into the rivers and adjacent forests. Unfortunately, there is no quantitative data regarding how long the habitats had been impacted (by logging and/or pollution); for how long and with what intensity the frogs had been/ are being hunted: or how large the populations had been previously and how they had developed. Thus, there is no way to estimate how long these populations may prevail despite the small numbers of individuals and altered habitats. The vast majority of the local people (~70%) around the study sites live on subsidiary and cash crop agriculture. In the study area, many forests along large rivers and streams have been transformed into cocoa and palm oil plantations, with larger portions converted where populations are high. To maintain these plantations and to improve production, fertilizers, herbicides, fungicides, and insecticides are used extensively. The preparation of those chemicals generally happens along the nearby rivers and streams, inevitably contaminating the water (NLG, pers. obs.). The scale of these potential threats November 2022 | Volume 16 | Number 2 | e319 Gonwouo et al. Table 3. Sample sites, numbers of frogs, and extracted values for number of roads, total length of roads, number of settlements, derived population estimate within the 10-km buffer zone, as well as measured distances to nearest settlement and road. Correlation coefficient and P-values from correlation analysis are given below each of the respective measures. Total length of — Distance to River IPH roads nearest road Lobe 971 167.9 650.9 Keinke 172 86.6 2,741.1 Dibombe 14.26 78.2 3,508.5 Bisoue 15.00 23.8 5,624.6 Mpoula 10.80 131.7 2,464.6 Nkebe 13.74 31.8 3,634.7 Sanaga (N) 10.55 69.4 2,199.6 Sanaga (T) 14.55 82.5 4,175.9 Njuma 5.31 8.9 8,653.5 Magamba 6.00 80.5 358.8 Ekomtolo 4.25 94.6 175.8 Mbo 14.97 82.2 3,843.3 Nkam 4.96 160.1 229.9 R? -0.257 0.688 P 0.3972 0.0093 to the frogs varied through the study sites and time. Along the river at Magamba for instance, large quantities (estimated at around 70%) of the original forest were transformed into plantations within our six-year survey period. Nevertheless, the Goliath Frog populations persisted at these sites. Frog Size and Distance to Settlements Frog size, estimated by the authors and the surveyed hunters, was positively correlated with distance to settlements, with smaller frogs living closer to the settlements (Fig. 7). Discussions with 11 frog hunters revealed that they have to move increasing distances of 300 to 4,000 m from the settlement in order to find large frogs, and that the adult and subadult/juvenile Goliath Frogs, which were once common around waterfalls and rock pools, are now less abundant. Based on weight estimates from the hunters, they agreed that the largest Goliath Frog they ever encountered, estimated to weigh about 5 kg, was caught before the year 2010. There was a significant (Welch-test: t= 6.14, df= 15.87, p < 0.0001) drop in the perceived frog weight when compared to the largest frog caught in 2019 (Fig. 8). Discussion This study assessed Goliath Frog habitats and relative abundance over large parts of the known range of this species in Cameroon, in order to understand the influenc- es of land use and vicinity to human settlements. Goliath Frogs were found to occur within all of the three main habitat types surveyed, 1.e., primary forest, secondary Amphib. Reptile Conserv. Distance to nearest Number of Population settlement settlements estimate 701.26 5 1,000 2,941.09 6 9,000 525:02 6 9,000 6,126.63 1 1,500 505.07 10 85,500 2372 22 ci 10,500 2,236.13 5 31,000 5,603.30 19 28,500 9,113.55 1 1,500 149.07 7. 25,500 89.25 19 75,500 541.14 15 46,000 477.44 20 99,200 0.431 -0.304 -0.415 0.1410 0.3130 0.1590 forest, and agroforestry plantations. The frogs utilize a combination of particular microhabitats that are stratified by different age sizes. Torrent water and rocks seem to be requirements for the presence of all ages. Based on our data, Goliath Frogs seem to be able to deal with some degree of habitat alteration, and the population decline is mostly due to hunting. Impacts of Anthropogenic Factors Due to the lack of previous (quantitative) data, it is difficult to reliably assess whether and to what extent the Goliath Frog populations have changed. Therefore, we had to base our assessment on indirect evidence, e.g., the comparisons of frog occurrences in pristine versus altered habitats, the severity of different threats, and interviews with local frog hunters. One exception is the previously published data on the abundance of Goliath Frogs along the Sanaga River, described by Perret (1957, 1960) and Amiet and Perret (1969). These populations still persist today, despite the high degree of selective logging in the area. However, this positive finding does not preclude the fact that the species seems to be threatened by human activities in general, and the situation for many local populations is not very promising. We base this conclusion on our observation that especially large adult frogs are rare around human settlements and increase progressively farther away, a basic finding confirmed by the surveyed frog hunters. Two indirect measures of anthropogenic impact, roads and population densities, indicated a trend of increasing frog numbers with the remoteness of habitats, although only frog abundance and distance to settlement was robustly, positively correlated. November 2022 | Volume 16 | Number 2 | e319 Goliath Frog populations in Cameroon oO aE 2) i) Or KS w) aa EET o- root S =x Cc - a2)) te oO = fe) 2am fap) 1?) oO oO Te) “ A GY 1000 2000 3000 4000 Distance to settlement [m] Fig. 7. Scatterplot of frog weight and distance to the nearest settlement. Note that the weight of the Goliath Frogs was increasing with distance from human settlements. This analysis has some limitations. First, there are limitations in the dataset itself. For instance, we are certain that not all small settlements and roads were consistently recorded, and thus the human impact may be generally larger. Our population estimates include some errors and inaccuracies as well, e.g., the population for the category ‘village’ in OpenStreetMap ranges from 500—5,000. Clearly not every village will have a population of about 1,500 inhabitants. Nonetheless, our data generally indicated coherent patterns. The Goliath Frog population of Lobe, for instance, is situated right next to a vast banana plantation and gave the lowest jo) To) “) + go — wT = = 12 = 0s jo) ro) i) N Before 2010 2019 Year of catch Fig. 8. Changes in weight of Goliath Frogs from around 1990 to 2019, as estimated by the surveyed frog hunters. frog population estimate. On the other hand, several Goliath Frog populations were quite large despite being in close proximity to major roads. Generally, the distance to the nearest settlement was a better predictor for frog abundance. The impact of distance to the nearest settlement was especially notable for populations that are exploited for food and trade, and also comprise anthropogenic impact- ed habitats. For instance, the Nkongsamba area has un- dergone a drastic change in vegetation structure over the past decades, as large-scale agro-industrial plantations and an increasing number of subsistence plantations have Table 4. Counts of the Goliath Frogs observed in the 13 rivers for the three habitat types (PF, primary forest; SF, secondary forest; AP, agroforestry plantation) across its range in Cameroon. Count data are provided with respect to the three different age sizes: a, adult; s, subadult; and j, juvenile. Count data (numbers of observed individuals) River PF a S j x Lobe Nkebe 31 20 14 65 Ekomtolo Dibombe 21 14 8 43 Sanaga Ngo Mpem 16 7 Zz 30 Sanaga (Tributary River) 15 10 4 29 Keinke 1] 17 4 32 Magamba Nkam Mbo Mpoula Njuma 13 10 23 Bisoue 9 12 21 Total 116 90 37 243 Amphib. Reptile Conserv. SF AP Total S j a S j x 2 8 3 5 13 11 2 22 87 3 3 4 1 14 M7 8 11 24 4 9 76 12 9 ip 28 58 18 13 34 63 6 5 11 43 8 a) 18 18 5 3 8 of 4 11 19 7d 11 4 14 25 13 21 4 2 6 27. 23 ZI 89 50 31 170 38 30 9 77 490 114 November 2022 | Volume 16 | Number 2 | e319 Gonwouo et al. been established (NLG, pers. obs.). In this area, Goliath Frogs are also intensively hunted for the food market. This is the area where we recorded the lowest encounter rates throughout the entire study period. Habitat conver- sion and degradation usually went hand in hand with hunting pressure, with both being more pronounced close to settlements. Around the localities of the Ebo forest, an- thropogenic pressure was almost absent, and here we en- countered the highest numbers of frogs per survey effort. Importance of Forest Habitats In contrast to the human impact, vegetation type was not a useful predictor, as Goliath Frogs were present in all forest types from semi-open to close-canopy forests (as long as cascading rocky river sections were pres- ent). Goliath Frogs apparently need some forest, but not necessarily pristine forest. They live in cold water and regularly sun-bask, presumably to regulate body temper- ature. Thus, the opening of forest habitats and resulting raising of temperatures may not negatively impact them, as long as the water temperature remains “low enough” (although, unfortunately, the temperature preferences of the species are unknown) and the habitat surround- ing the rivers can still provide enough food and shelter. Other African frog species with similar life-histories and inhabiting forested, torrent rivers, e.g., Conraua alleni and Odontobatrachus spp., also usually occur in cooler streams in forest, but may persist in areas with little riv- erine forest surrounded by savanna (Rodel 2003; Rodel and Bangoura 2004). In this survey, when cascading, turbulent water with rocks and some forest patches was present, Goliath Frogs were reliably recorded. Thus, Goliath Frogs may be able to tolerate forest degradation to a surprising extent. This is in line with predictions by Hirschfeld and Rodel (2017) that large frogs in particular, with large clutch sizes and aquatic larvae, may be more resilient to forest degrada- tion than species with other trait combinations. We would like to stress, however, that our observations should not be interpreted as indicating that riverine vegetation 1s not important for maintaining Goliath Frog populations. Riverine forest may be important for Goliath Frogs during their nocturnal foraging activity. It is likely that all individuals observed along the river banks were adopting a ‘sit and wait’ foraging strategy. A study on stomach and intestinal contents revealed that the diet of C. goliath consists of approximately 60% arthropods, 20% crustaceans, 10% amphibians, and 10% indeterminate food items, the latter comprising ingested stones as well as pieces of wood and leaves (Sabater-Pi 1985). The majority of the arthropods were terrestrial taxa. The presence of leaves, wood, and stones also suggests a mainly terrestrial foraging mode. If the quality of the riparian forests impacts prey quality and quantity, this would likely impact the Goliath Frog populations as well. Our observations on one particular frog highlight the Amphib. Reptile Conserv. Goliath Frog’s use of riverine forest habitats. In prima- ry forest along Nkebe River, a large adult was found in the forest leaf litter at about 14 m from the river. When disturbed, the frog covered this distance with three long jumps back to the river (also see Herrmann and Edwards 2006). In about 1 m water depth, the frog could then be spotted in the slow flowing, clear water. Further distur- bance (with the torch beam) triggered the frog to bury itself deep in the sandy and leaf-littered river bottom un- til it could not be seen anymore, a behavior also known from its smaller congener, C. crassipes (Knoepffler 1985). Goliath Frogs are less active during the day, and when encountered, they were usually found sun-basking. When disturbed, the behavior was the same as during the night, with the frog seeking shelter in the water or be- neath the rocks it was sitting on. Use of Microhabitats by the Different Age Classes Our observations revealed that Goliath Frogs partition microhabitats by age-class. Therefore, a range of dif- ferent riverine habitat features is likely crucial for sup- porting the full complement of life stages of this species. Large adults predominately perched on rocks around wa- terfalls and rapids, with individual separated by a con- siderable distance (minimum 3-5 m), thereby providing evidence for territorialism and site fidelity, as already suggested by Sabater-Pi (1985). In contrast, subadults were rarely found around waterfalls. They appeared fre- quently on exposed, mid-stream rocks in the vicinity of cascades and waterfalls. Finally, metamorphosing and ju- venile frogs most often used rook pools along the river- beds where the current was slower. There they could find refuge in rock crevices when disturbed (Fig. 3C). Such sections also comprise the breeding sites of the species (Sabater-Pi 1985; Schafer et al. 2019). The reason for this microhabitat partitioning is unclear. It may be a linked to thermoregulation, with small juveniles avoiding colder water, and/or predation pressure, including cannibalism. Habitat segregation has been reported from juvenile/sub- adult and adult European water frogs, Pelophylax spp., in order to escape cannibalism (Gunther 1990). Based on daytime observations, Sabater-Pi (1985) estimated the territory sizes of 20 to 40 m?’ for Goliath Frogs along the Mbia River. Our observations suggest that Goliath Frogs use small core areas or territories for sun-basking and shelter (200 m7), and larger areas (> 1,000 m7) for foraging. However, quantitative research on this issue 1s lacking and ideally should be based on radio-tracked individuals (e.g., Spieler 1997). One surprising finding of this study was that the Goli- ath Frogs showed no apparent impact from the contami- nation with agrochemicals from neighboring plantations (at least to the extent that it is represented in our study sites; although, unfortunately, the composition and quan- tity of agrochemical run-off in the streams is unknown). On the Mpoula River, a large banana plantation that is November 2022 | Volume 16 | Number 2 | e319 Goliath Frog populations in Cameroon regularly sprayed by airplane occurs just upstream from where the Goliath Frogs were observed. In most of the study areas, small to medium-sized plantations along the rivers are run by local farmers. They typically use small spraying pumps and regularly process chemicals in the rivers, and the Goliath Frogs could still be found at these sites. Our observations, as well as the information from the frog hunters, revealed that populations around human settlements are smaller and adult frogs are rarer. As habi- tat degradation seemed to be of little influence, the Goli- ath Frogs might be mainly impacted by targeted hunting for food (locally) and trade. Conservation Needs of the Goliath Frog In order to detect potential declines, populations should be monitored on a regularly basis. However, no national monitoring program for Goliath Frog populations has been implemented thus far. Long-term data collection from specific sites across the Goliath Frog range would be essential for detecting changes in the distribution and local abundances of this species. Standard guidelines and techniques for monitoring amphibian populations and habitats are well-established (e.g., Heyer et al. 1994). Potential monitoring methods for Goliath Frogs should include time-area constrained searches in order to estab- lish baseline data against which population changes with time could be judged. The data presented here may serve as a baseline for future studies. Based on our findings, regular surveys carried out at night by walking along pre-defined river routes would probably be the most effi- cient method. The Nkongsamba area is especially critical for monitoring because of the particularly intensive frog collection for food, and the severe habitat degradation overall. The Campo-Ma’an National Park and the Ebo Forest National Park should likewise be considered for monitoring, as they consist of areas with limited human impact. Monitoring in different parts of the species range will allow comparisons of the population trends within and outside protected areas. This will also potentially allow differentiation between the different threats such as collection, habitat degradation and pollution, climate change, and disease. Given that most amphibian popula- tions naturally fluctuate (Pechmann et al. 1991), it would be ideal to start an initial monitoring program for at least five years. Several key aspects of Goliath Frog ecology remain to be investigated in order to better understand the biology of this species (1.e., larval and juvenile survival rates, growth rates, age at maturity, and the longevity of adults). Such information may ultimately help in setting up a conservation action and management plan for this species. In parallel, additional short-term surveys (detecting as many populations as possible) and long- term monitoring data (to follow population trends) are needed to fully interpret the Goliath Frog’s occurrences and threat status. This study has shown that conservation Amphib. Reptile Conserv. efforts for the Goliath Frog do not need to be prioritized for terrestrial habitat loss, but that hunting is clearly a prominent factor affecting the persistence of robust populations. Acknowledgements.—We are very grateful to the many people and organizations who made this work possible. We are particularly grateful to Frogs & Friends e.V., Germany. The Cameroonian Ministry of Forestry and Wildlife (MINFOF) provided permission for the study. Our thanks are also extended to several organizations which commissioned part of the study (Rainbow Consult, Sylvatrop Consult, and Diggi Wels), and permitted the use of the data. We would also like to thank the local communities for permission to access their land. The people whose assistance in the field was invaluable include: Ekane Pierre, Andre, Blaise, Motto, Alex from the Nkongsamba region, Patrick, Olivier from Akom II and Blaise from around Ngo-Pem and Moukol, Ottou, and Nyanga boy and Black Joe from the Yabassi area. We thank the reviewers, in particular Thomas Doherty-Bone, for their constructive critiques on previous drafts of the manuscript. Literature Cited Amiet JL. 1975. Ecologie et distribution des amphibiens anoures de la région de Nkongsamba (Cameroun). Annales de la Faculté de Sciences de Yaoundé 20: 33-107. Amiet JL. 2004. Conraua goliath. The IUCN Red List of Threatened Species 2004: e.T5263A11121365. Amiet JL, Perret JL. 1969. Contribution a la faune de la région Yaoundé (Cameroun). IT. Amphibiens, anoures. Annales de la Faculté de Sciences du Cameroun 13: 117-137. Altherr S, Auliya M, Nithart C. 2022. Deadly Dish — Role and Responsibility of the European Union in the International Frogs’ Legs Trade. Pro Wildlife and Robin des Bois, Munich, Germany and Paris, France. 28 p. Blackburn DC, Nielsen SV, Bare} MF, Doumbia J, Hirschfeld M, Kouamé NG, Lawson D, Loader S, Ofori-Boateng C, Stanley EL, et al. 2020. Evolution of the African Slippery Frogs (Anura: Conraua), including the world’s largest living frog. Zoologica Scripta 49: 684-696. Boulenger GA. 1906. Descriptions of new batrachians discovered by Mr. G.L. Bates in South Cameroun. Annals and Magazine of Natural History 7(17): 317- 323. Channing A, Rodel M-O. 2019. Field Guide to the Frogs and other Amphibians of Africa. Struik Nature, Cape Town, South Africa. 408 p. Ernst R, Linsenmair KE, Rodel M-O. 2006. Diversity erosion beyond the species level: dramatic loss of functional diversity after selective logging in November 2022 | Volume 16 | Number 2 | e319 Gonwouo et al. two tropical amphibian communities. Biological Conservation 133: 143-155. Ernst R, Rodel M-O. 2008. Patterns of community composition in two tropical tree frog assemblages: separating spatial structure and environmental effects in disturbed and undisturbed forests. Journal of Tropical Ecology 24: 111-120. Gewalt W. 1977. Einige Bemerkungen tber Fang, Trans- port und Haltung des Goliathfrosches (Conraua go- liath Boulenger). Zoologische Garten 47: 161-192. Gewalt W. 1996. Vom Goliathfrosch in Stidkamerun. 77 Magazin 28(129): 44-47. Gonwouo NL, Rodel M-O. 2008. The importance of frogs to the livelihood of the Bakossi people around Mount Manengouba, Cameroon, with special consid- eration of the Hairy Frog, Trichobatrachus robustus. Salamandra 44: 23-34. Gunther R. 1990. Die Wasserfrésche Europas. Die Neue Brehm-Bucherei, Volume 600. A. Ziemsen Verlag, Wittenberg Lutherstadt, Germany. 288 p. Herrmann H-W, Bohme W, Herrmann PA, Plath M, Schmitz A, Solbach M. 2005. African biodiversity hotspots: the amphibians of Mt. Nlonako, Cameroon. Salamandra 41: 61-81. Herrmann H-W, Edwards T. 2006. Conraua goliath (Go- liath Frog) skittering locomotion. Herpetological Re- view 37: 202-203. Heyer WR, Donnelly MA, McDiarmid RW, Hayek LAC, Foster MS. 1994. Measuring and Monitoring Biological Diversity: Standard Methods for Amphibians. Smithsonian Institution _ Press, Washington, DC, USA. 364 p. Hillers A, Veith M, Ro6del M-O. 2008. Effects of forest fragmentation and habitat degradation on West African leaf-litter frogs. Conservation Biology 22: 762-772. Hirschfeld M, Rodel M-O. 2017. What makes a successful species? Traits facilitating survival in altered tropical forests. BMC Ecology 17: 25. Knoepffler L-P. 1985. Le comportement fouisseur de Conraua crassipes (Amphibien anoure) et son mode de chasse. Biologia Gabonica 1(3): 239-245. Lamotte M, Perret J-L. 1968. Révision du genre Conraua Nieden. Bulletin de I’Institut Fondamental d'Afrique Noire Série A 30: 1,603—1,644. Lamotte M, Perret J-L, Dzieduszycka S. 1959. Contribution a l’étude des batraciens de l’Ouest Africain IX. Les formes larvaires de Petropedetes palmipes, Conraua goliath et Acanthixalus spinosus. Bulletin de I’Institut Fondamental d’Afrique Noire Série A 21: 762-776. Mohneke M, Onadeko AB, Hirschfeld M, Rodel M-O. 2010. Dried or fried: amphibians in local and regional food markets in West Africa. TRAFFIC Bulletin 22: 117-128. Neguiffo DN, Josue WP, Mbida M. 2015. Gastro-intestinal helminths of Goliath Frogs (Conraua goliath) from Amphib. Reptile Conserv. the localities of Loum, Yabassi, and Nkondjock in the Littoral Region of Cameroon. Global Ecology and Conservation 4: 146-149. Neguiffo ND, Mpoame M, Wondji SC. 2019. Genetic diversity and population structure of Goliath Frogs (Conraua goliath) from Cameroon. Mitochondrial DNA Part A 30(4): 1-7. Ofori-Boateng C, Oduro W, Hillers A, Norris A, Oppong SK, Adum GB, Rédel M-O. 2013. Differences in the effects of selective logging on amphibian assemblages in three West African forest types. Biotropica 45: 94— 101. Perret J-L. 1957. Observations sur Rana_ goliath Boulenger. Bulletin de la Société Neuchdteloise des Sciences Naturelles 80: 195-202. Perret J-L. 1960. Etudes herpétologiques africaines. Bulletin de la Société Neuchdteloise des Sciences Naturelles 83: 93-100. QGIS Development Team. 2021. QGIS Geographic Information System. Open Source Geospatial Foundation Project, Chicago, Illinois, USA. Available: http://qgis.osgeo.org [Accessed: 14 April 2022]. R Development Core Team. 2014. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available: http://www.R-project.org [Accessed: 4 April 2018]. Rodel M-O. 2003. The amphibians of Mont Sangbé National Park, Ivory Coast. Salamandra 39: 91-110. Rodel M-O, Bangoura MA. 2004. A conservation assessment of amphibians in the Forét Classée du Pic de Fon, Simandou Range, southeastern Republic of Guinea, with the description of a new Amnirana species (Amphibia Anura Ranidae). Tropical Zoology 17: 201-232. Rodel M-O, Ernst R. 2004. Measuring and monitoring amphibian diversity in tropical forests. I. An evaluation of methods with recommendations for standardization. Ecotropica 10: 1-14. Sabater-Pi J. 1962. La rana goliat, aportacion a su estudio. Revista del Parque Zoologico de Barcelona |: 23—25. Sabater-Pi J. 1967. Notas sobres la ecologia de la rana gigante de Rio Muni. Revista del Parque Zoologico de Barcelona 7: 24—25. Sabater-Pi J. 1985. Contribution to the biology of the Giant Frog (Conraua goliath, Boulenger 1906). Amphibia-Reptilia 6: 143-153. Schafer M, Tsekané SJ, Tchassem FAM, Drakulic S, Kameni M, Gonwouo NL, Rodel M-O. 2019. Goliath Frogs build nests for spawning — the reason for their gigantism? Journal of Natural History 53: 1,263-— 1,276. Spieler M. 1997. Radio-telemetrische Untersuchungen zur Laichplatzwahl eines westafrikanischen Raniden. Mertensiella 7: 203-220. Stuart SN, Chanson JS, Cox NA, Young BA, Rodriguez ASL, Fischman DL, Waller RW. 2004. Status and trends of amphibians: declines and _ extinctions November 2022 | Volume 16 | Number 2 | e319 Goliath Frog populations in Cameroon worldwide. Science 306: 1,783-1,786. Pp. 102-110 In: The Plants of Kupe, Manenguba, and Stuart SN, Hoffmann M, Chanson JS, Cox NA, Berridge the Bakossi Mountains, Cameroon, a Conservation RJ, Ramani P, Yong BE. 2008. Threatened Amphibians Checklist. Editors, Cheek M, Pollard BJ, Darbyshire of the World. Lynx Edicions, Barcelona, Spain. 758 p. I, Onana J-M, Wild C. Royal Botanic Gardens, Kew Wild C, Morgan M, Fotso R. 2004. The vertebrate fauna. Cromwell Press, Richmond, United Kingdom. 508 p. Amphib. Reptile Conserv. Nono LeGrand Gonwouo is a Conservation Biologist and Senior Lecturer at the University of Yaoundé I Cameroon. He holds a Ph.D. on the biogeography and ecology of reptiles from the same University and has long term experience studying the herpetofauna of central and West Africa, particularly in the fields of biodiversity, taxonomy, and ecology. He has combined research and community engagement to leverage conservation outcomes for highly threatened herpetofauna in Cameroon. Nono is a member of the IUCN SSC for Amphibians, Reptiles and Chameleon Specialist Groups. His recent research interest focuses on the responses of biodiversity to environmental change, with particular interest in the distributions and dynamics of endemic herpetofauna near their geographic range margins. Marvin Schafer is a Ph.D candidate at Humboldt University Berlin, Germany working within the group of M.-O. Rodel at the Natural History Museum in Berlin. Marvin has studied biology and toxicology at the Master’s level, with a broad interest in ecology and physiology, particularly of amphibians. For his Ph.D., he is researching chemical ecology in a group of West African frogs. He has a keen interest in the natural history of Goliath Frogs, and was the principal investigator on the recent paper on their nest building behavior. Sedrick Junior Tsekané is currently enrolled as a Ph.D. student in the Department of Animal Biology and Physiology, University of Douala, Cameroon, with research work focused on the ecology of Goliath Frogs. Sedrick obtained a Master’s degree from the Laboratory of Zoology, University of Yaoundé I, during which he studied certain aspects of the ecology of Goliath frog tadpoles, and is co-author of a recent paper on the nest building behavior of Goliath Frogs. Mareike Hirschfeld is working as a scientist in the Collection Future department at the Museum ftir Naturkunde Berlin, Germany. She studied Biology at the Universities of Marburg and Wurzburg, Germany, and obtained her Ph.D. from the Humboldt University Berlin. Mareike is interested in the ecology and taxonomy of West and Central African amphibians. For her Doctoral thesis, she studied the amphibians of highland forests in Cameroon and investigated their ability to cope with changes to their environment. Mareike is now leading the science data management team at the Museum and is involved in institutional, national, and international initiatives for opening up natural science collections to scientists and society worldwide. As a Freelancer, Mareike supports the association Frogs and Friends and their projects in Cameroon. 118 November 2022 | Volume 16 | Number 2 | e319 Amphib. Reptile Conserv. Gonwouo et al. Arnaud Marius Tchassem Fokoua recently obtained his Ph.D. on the ecology of amphibians from the Laboratory of Zoology of the University of Yaoundé I, Cameroon. Arnaud received his Bachelor’s degree at the same university, completing multiple research projects in the domain of herpetology. His research focuses on the community ecology and conservation of African amphibians, especially aiming to understand how altitude and anthropogenic activities influence amphibian community composition. Mark-Oliver Rédel works with his team at the Museum fiir Naturkunde, Berlin, Germany, where he is the curator of herpetology. He has studied the systematics, taxonomy, biogeography, and ecology of African amphibians for almost 30 years. With his group, comprising students from around the world, he currently runs projects in Germany, Ecuador, Guinea, Ivory Coast, Cameroon, and Mozambique. His special interest is on how species and ecological communities react to environmental changes. He has authored or co-authored about 400 scientific and popular publications, including several books. 119 November 2022 | Volume 16 | Number 2 | e319 Official journal website: amphibian-reptile-conservation.org Amphibian & Reptile Conservation 16(2) [Taxonomic Section]: 120-132 (e320). urn:Isid:zoobank.org:pub:7F752CB7-4667-46F5-A44E-E4D165D28CF4 Out of the Past: A new species of Tantilla of the calamarina group (Squamata: Colubridae) from southeastern coastal Guerrero, Mexico, with comments on relationships among members of the group ‘Ricardo Palacios-Aguilar, *7Lydia Allison Fucsko, *Victor H. Jiménez-Arcos, *’Larry David Wilson, and *Vicente Mata-Silva 'Museo de Zoologia “Alfonso L. Herrera” de la Facultad de Ciencias, Universidad Nacional Aut6noma de México, 3° Circuito Exterior s/n, Coyoacan, CP 04510, MEXICO; Totlok A.C., C. P. 04350, Delegacién Coyoacdn, Ciudad de México, MEXICO *Department of Humanities and Social Sciences, Swinburne University of Technology, Melbourne, Victoria, AUSTRALIA *Laboratorio de Herpetologia Vivario de la Facultad de Estudios Superiores Iztacala, Universidad Nacional Auténoma de Mexico, Ay. de los Barrios 1, Tlanepantla, CP 54090, MEXICO “Centro Zamorano de Biodiversidad, Escuela Agricola Panamericana Zamorano, Departamento de Francisco Morazan, HONDURAS; 1350 Pelican Court, Homestead, Florida 33035-1031, USA *Department of Biological Sciences, The University of Texas at El Paso, El Paso, Texas 79968-0500, USA Abstract.—A new species of Tantilla in the calamarina group from southeastern coastal Guerrero, Mexico is described. This new species is part of a clade that contains T. calamarina, T. cascadae, T. ceboruca, T. coronadoi, T. deppei, T. sertula, and T. vermiformis. All of these species are endemic to Mexico, except for T. vermiformis, which is found along the Pacific coastal plain from El Salvador to northwestern Costa Rica. Members of this group of Tantilla show varying adaptations to fossoriality, which might reflect their phylogenetic relationships. Keywords. calamarina group, centipede snake, fossoriality, phylogenetics, Reptilia, taxonomy Resumen.— Describimos una nueva especie de Tantilla del grupo calamarina de la costa sureste de Guerrero, Mexico. Esta nueva especie forma parte de un clado que contiene a T. calamarina, T. cascadae, T. ceboruca, T. coronadoi, T. deppei, T. sertula y T. vermiformis. Todas estas especies son endemicas de Mexico, excepto T. vermiformis, que se encuentra a lo largo de la planicie costera del Pacifico desde El Salvador hasta el noroeste de Costa Rica. Los miembros de este grupo de TJantilla muestran diversas adaptaciones a la fosorialidad, lo que podria reflejar sus relaciones filogeneticas. Palabras claves. Culebra ciempiés, fosorialidad, grupo calamarina, relaciones filogenéticas, Reptilia, taxonomia Citation: Palacios-Aguilar R, Fucsko LA, Jiménez-Arcos VH, Wilson LD, Mata-Silva V. 2022. Out of the Past: A new species of Tantilla of the calamarina group (Squamata: Colubridae) from southeastern coastal Guerrero, Mexico, with comments on relationships within members of the group. Amphibian & Reptile Conservation 16(2) [Taxonomic Section]: 120-132 (e320). Copyright: © 2022 Palacios-Aguilar et al. This is an open access article distributed under the terms of the Creative Commons Attribution License [Attribution 4.0 International (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/], which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are 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: amphibian-reptile-conservation.org. Accepted: 11 October 2022; Published: 12 November 2022 Introduction A recent paper by Palacios-Aguilar et al. (2021) noted a specimen of Zantilla from Guerrero that was collected over acentury ago (BMNH 1906.6.1.241). This specimen, a female clearly referable to the Tantilla calamarina group (Wilson and Meyer 1981), was discussed by Wilson and Mata-Silva (2014) and determined to be either an atypical individual of Tantilla coronadoi or a representative of an unnamed taxon. Palacios-Aguilar et al. (2021) suspected the latter alternative because the three hitherto known specimens of 7’ coronadoi are consistent in key aspects of their scutellation, but they also declined to name a new species based on specimen BMNH 1906.6.1.241, which was collected in 1904 by Hans Gadow and reported as Homalocranium miniatum in his 1905 paper. According to Dixon et al. (2000), the name H. miniatum is a synonym of 7: rubra. Correspondence. ricardopalaciosaguilar@gmail.com (RPA), lydiafucsko@gmail.com (LAF), victorhja@iztacala.unam.mx (VHJA), * bufodoc@aol.com (LDW), vmata@utep.edu (VMS) Amphib. Reptile Conserv. November 2022 | Volume 16 | Number 2 | e320 Palacios-Aguilar et al. When the Palacios-Aguilar et al. (2021) paper appeared, one author of this paper (LDW) contacted the senior author and proposed reexamining the BMNH specimen to determine whether its status could be established more definitively, based on a modern understanding of the content and relationships within the Tantilla calamarina group. Consequently, the authors of this paper decided to collaborate in an attempt to determine on the status of this curious specimen, which has been lying in anonymity since it was collected 118 years ago. The first four words of the title of this paper (Out of the Past) are used in reference to the period of time this specimen, which now becomes the holotype of a new species of Zantilla, has resided in the BMNH collection. These words also refer to the period of time that has elapsed since the person after whom this snake is being named began her journey in 1956 to find freedom from the tyranny in her homeland in her adopted land of Australia. For aficionados of cinema, Out of the Past will be recognized as the title of a well-regarded film noir, the 1947 work of French director Jacques Tourneur starring Robert Mitchum, Jane Greer, and Kirk Douglas. Materials and Methods During the last five decades, author LDW has been examining specimens of the genus Zantilla, resulting in the publication of several taxonomic treatments and species descriptions within this genus (e.g., Wilson and Meyer 1981; Wilson 1999; Wilson and Campbell 2000; Wilson and Mata-Silva 2014, 2015). While compiling information for the first taxonomic study of the 7 calamarina species group (Wilson and Meyer 1981), measurements and scale counts were obtained for the specimen that now is the focus of this paper (BMNH 1906.6.1.241). Soon after that work started, the Senior Curator of Reptiles at the Natural History Museum, London, was contacted to obtain morphological data on the specimen, as well as high quality photographs. The photographs aided in making comparisons with the representatives of the 7’ calamarina species group. In addition to the aforementioned works on the genus Tantilla, the comparisons in this paper benefitted from examinations of more recent literature specifically involving the calamarina species group (Canseco- Marquez et al. 2007; Cisneros-Bernal et al. 2020; Cruz-Saenz et al. 2015; Davalos-Martinez et al. 2021; Palacios-Aguilar et al. 2021; Ramirez-Bautista et al. 2014; Rocha et al. 2016), as well from the examination of comparative material housed at the Coleccién Nacional de Anfibios y Reptiles (CNAR) and the Museo de Zoologia “Alfonso L. Herrera,” Facultad de Ciencias (MZFC), both at the Universidad Nacional Autonoma de México. Scale counts and digital photographs of specimens deposited in the herpetological collection of the University of Texas at Arlington (UTA-R) were also obtained. The measurements are indicated in millimeters, and the ventral scales were counted following the method proposed by Dowling (1951), with the segmental counts referring to the sum of the ventrals + subcaudals, excluding the cloacal scute. Amphib. Reptile Conserv. Results After detailed comparisons of the BMNH specimen with the pertinent published sources and specimens held in scientific collections, we determined that BMNH 1906.6.1.241 does indeed represent a new species of the genus Jantilla. Tantilla carolina, new species Figs. 1-2. Suggested common name. Carolina’s Little Snake. um:|sid:zoobank.org:act: B8520FC0-94E14A 6F-8312-F6D3EA3434EF Holotype. BMNH 1906.6.1.241, an apparent subadult or juvenile female from Tezonapan (= Tecoanapa), north of Ayutla, Guerrero, Mexico, collected by Hans Gadow in 1904. Diagnosis. Zantilla carolina is a member of the 7. calamarina group (Table 1). This species differs from Tantilla calamarina by the presence of more ventrals in females (156 versus [hereinafter = vs.] 118-140), more total segmental scales (194 vs. 146-179), a normally- sized preocular scale in contact with the postnasal scale (vs. a preocular with a tendency toward a decrease in size to complete the loss of the scale), two postocular scales (vs. one), seven supralabials (vs. usually six), a uniform dorsal head color followed by two pale postparietal spots (vs. a head pattern consisting of a spatulate dark anterior extension of the middorsal dark stripe flanked by Fig. 1. Dorsal (A), lateral (B), and ventral (C) views of the head of the holotype of Zantilla carolina sp. nov. (BMNH 1906.6.1.241). November 2022 | Volume 16 | Number 2 | e320 Anew Tantilla species in the calamarina group from Guerrero, Mexico = — =t ) Se eee Fig. 2. Dorsal view of the holotype of Zantilla carolina sp. nov. (BMNH 1906.6.1.241). prominent pale narrow longitudinal markings confluent with the pale postparietal spots), and a body pattern involving a dark lateral stripe that does not extend to the end of the body (vs. a dark brown lateral stripe on rows 3 and 4 extending the length of the body). The new species differs from Tantilla cascadae by the presence of more ventrals in females (156 vs. 139-144), more total segmental scales (194 vs. 176-192), seven supralabial scales (vs. six), and the dorsum of the head without a pattern but containing a pair of small pale postparietal spots (vs. a spatulate dark anterior extension of the middorsal dark stripe flanked by pale narrow longitudinal markings confluen