Integrative Taxonomy Reveals the Chortís Block of Central America As an Underestimated Hotspot of Amphibian Diversity

Home , Amastridium, Anolis allisoni, Anolis capito, Anolis cupreus, Bolitoglossa, Bolitoglossinae

INTEGRATIVE TAXONOMY REVEALS THE CHORTÍS BLOCK OF CENTRAL AMERICA AS AN UNDERESTIMATED HOTSPOT OF AMPHIBIAN DIVERSITY

JOSIAH HAROLD TOWNSEND

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2011

Para Ileana

And in loving memory of my grandfather, Vernon Lynn Boyd

ACKNOWLEDGMENTS

I want to begin by expressing my gratitude for the support shown by my advisory committee during the course of preparing this dissertation. My Chair, James Austin, supported me both academically and personally throughout this process; his leadership and steadfast support was instrumental in my success. I want to thank my committee members, Rob Fletcher, Mike Miyamoto, Rick Stepp, and Larry David Wilson, for their guidance, advice, and support through the process of researching and writing this dissertation.

Writing a dissertation must be both a singularly selfish act and one that relies on the extended patience and support of loved ones, labmates, and colleagues. This certainly was the case for me. To the one person who is literally all three of those things, my wife Ileana Luque-Montes, I cannot adequately express the amazement with which I have watched you begin your own graduate studies over the past four months; maintaining a rigorous field schedule and supervising undergraduates, while I was of minimal help and fully focused on writing the following document. I simply would not have completed this without your remarkable dedication and support, and my gratitude will be best expressed by giving you the same degree of support as you pursue your own goals. I am blessed with the devoted support and seemingly endless patience and understanding of a wonderful and inspirational family, my parents Steve and Terri Boyd

Townsend, my sister Katielynn, and my grandparents Vernon Lynn and Norma Jean

Boyd and Willa Parker Townsend, who are largely responsible for any past and future successes I may have.

In many ways, I am who I am today thanks to the mentorship of Larry David

Wilson. As a mentor, colleague, and friend, Larry introduced me to Honduras in 1999

and guided my developing research interests there throughout my community college and undergraduate years, becoming my colleague and supporter as I progressed through my graduate studies.

I owe no small debt of gratitude to my friends and colleagues in the Austin Lab, and, in particular, for actively assisting with my labwork and providing countless impromptu instructional sessions, I want to thank Jason Butler, John Hargrove, Nathan

Johnson, Emily Saarinen, Matt Shirley, and Aria (Johnson) St. Louis. I also benefited from the dedicated work of a series of undergraduate researchers and volunteers:

Teresa Burlingame, Dania Gutierrez, Jaclyn Irwin, Rachel Shapiro, Vicki Villanova, and

Lauryn Walter.

My work for the past six years has benefited greatly from the support of the

Section of Protected Areas and Wildlife, Instituto Nacional de Conservación y Desarrollo

Forestal, Áreas Protegidas y Vida Silvestre, and particularly Iris Acosta, Carla Cárcamo,

Saíd Lainez O., Andrés Alegria, Ramón Alvarez L., Wilson Zúniga D., Sonia Martínez

Moreno, and Wendy Aronne (Instituto Nacional de Conservación y Desarrollo Forestal,

Áreas Protegidas y Vida Silvestre [ICF]). Fieldwork was carried out under a series of research permits issued by ICF, most recently Resolución DE-MP-086-2010 and

Dictamen DVS-ICF-045-2010. In Honduras, my work would not have been possible without the hard work and support variously given by Ileana Luque-Montes (UNAH),

Melissa Medina-Flores (UNAH, UNA), Luis A. Herrera (UNAH), Allan J. Fuentes and

Eduardo J. Zavala (PROLANSATE), Alcalde Adolfo Pagoada-Saybe (Municipalidad de

Arizona), Alcadesa Teresa Espinosa Aguilar (Municipalidad de Marale), Oliver Komar,

Jose Mora, Jorge Iván Restrepo, and Fredy Membreño (Centro Zamorano de

Biodiversidad), Efrain Aguilar (San José de Texíguat), Alfonso Contreras (Mezapita),

Alionso Portillo (Jilamito Nuevo), J. Dubón (La Liberación), Mario Orellana Leiva (El

Playon), Leonel Erazo Chávez (El Cedral), Rafael Ulloa (Municipalidad de Gualaco),

Eduardo Rico (ICF-Gualaco), Carlos Perdomo (Aldea Global), Paul House (UNAH,

Herbario TEFH), Robert Dale (Los Naranjos), and Alicia Ward (Santa Bárbara). I would like to acknowledge the following people for their work in the field in support of this project: Carlos Andino, Ben K. Atkinson, James D. Austin, Christopher Begley, Mark

Bonta, Jason M. Butler, Brian Campesano, César A. Cerrato, Gabriela Diaz, Anne

Donnelly, Matthew Donnelly, Yensi Flores, Sergio C. Gonzalez, Levi Gray, Vladlen

Henriquez, Luis A. Herrera, Paul House, Robert Hyman, Lorraine Ketzler, Ileana Luque,

Christina Martin, David Medina, Melissa Medina-Flores, Mayron Mckewy-Mejía, Aaron

Mendoza, Wendy Naira, Ciro Navarro, Lenin Obando, Sandy Pereira, Onán A. Reyes,

John Slapcinsky, Mario Solís, Fito Steiner, Nathaniel Stewart, Alexander Stubbs,

Katielynn Townsend, Steve Townsend, Scott L. Travers, Rony Valle, Hermes Vega,

Alicia Ward, and Christopher Wolf.

I am very grateful for the support of Amy Driscoll, Dan Mulcahy, Andrea Ormos, and the SI Barcoding Project (Smithsonian Institution Laboratory of Analytical Biology), and Roy McDiarmid (USNM) for supporting my ―Barcoding the Herpetofauna of Eastern

Nuclear Central America‖ initiative. Deposition of voucher specimens and timely acquisition of catalog numbers was facilitated by: Jim McGuire, Ted Papenfuss, Sean

Rovito, Carol Spencer, and David Wake (MVZ), Jose Rosado (MCZ), and Steve Gotte,

Jeremy Jacobs, John Poindexter, and Robert Wilson (USNM). Jason Butler, Ileana

Luque, Javier Sunyer, and Scott Travers kindly contributed photographs for use in this dissertation.

Various portions of this dissertation was funded by Critical Ecosystem Partnership

Fund (CEPF), a Summer Research Grant from the Working Forests in the Tropics

IGERT Program (National Science Foundation DGE–0221599) at the University of

Florida, a grant to Kirsten E. Nicholson (Central Michigan University; National Science

Foundation DEB-0949359), and the Explorer’s Club (via Robert Hyman; Flag #93).

During my dissertation studies, I was supported by a 2007–09 NSF GK-12 Fellowship from the UF SPICE Program, and as much as any experience during my graduate education, my participation in this program was truly formative and help to shape the conceptual framework within which I propose to carry out research. I am especially grateful to Doug Levey and Suzan Smith, my partner-teacher Nate Stewart, and fellow fellows Jackson Frechette, Rachel Naumann, and Tom Tidyman, for helping to make

SPICE a truly great experience. Following SPICE, my work with the UFTeach Program as an instructor for their core Research Methods class was key during the final two years of my dissertation, and I thank Alan Dorsey, Dimple Flesner, Griff Jones, Linda

Jones, Katrina Short, Gloria Weber, and our RM students for making these last two spring semesters so enjoyable.

Finally, I want to take this opportunity to dedicate my work in the past and future to the memory of three friends that I lost while completing this degree. I met Thad Owens as a student in my Herpetology class at UF and we quickly became friends, as he did with just about everyone who knew him. Thad was one of the most honest, unreserved, and enthusiastic people I will ever have the pleasure of knowing, and rarely a day

passes since he was lost in May 2009 that I do not think about him or find inspiration in his memory. Wade Wassenberg was a good friend since high school, and was one of the most loyal friends anyone could hope to have. After losing track of him for years, I learned that Wade had joined the US Army and served in the elite 1st Battalion of the

75th Ranger Regiment, surviving four tours of duty in Afghanistan and two more in Iraq from 2002 to 2008. After retiring to rejoin life with his beautiful young family, he was suddenly diagnosed with brain cancer and passed soon thereafter at the VA Hospital in

Gainesville, walking distance from my office. I will always regret that while my friend suffered and passed literally minutes away from my home, I was unaware and in the field in Honduras. Finally, Don Mario Guiffaro of Olancho, Honduras, was an influential and remarkable man whose legendary life story included being a feared pistolero, a frontier gold miner, a jaguar hunter, and in his later years, and outstanding advocate and conservationist in the Patuca region. In 2007, Don Mario’s forceful opposition to illegal logging and drug trafficking in the Mosquitia was ultimately answered by violence, and he was gunned down in front of his son and friends. The memories of, and examples set, by these three friends, to each of whom I would not have hesitated to entrust with my life, and in some cases did just that, will always be my compass and my motivation as I continue through life’s journey.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 14

LIST OF FIGURES ...... 15

ABSTRACT ...... 18

CHAPTER

1 INTRODUCTION ...... 20

2 BIOGEOGRAPHIC DELIMITATION OF THE CHORTÍS BLOCK: GEOLOGICAL HISTORY, PHYSIOGRAPHY, AND ECOLOGICAL ASSOCIATIONS ...... 25

Geological History ...... 27 Contemporary Ecophysiography ...... 34 Characterizing Ecological Associations: Holdridge Forest Formations ...... 46 Characterizing Ecological Associations: Updating and Operationalizing the Carr (1950) System for Classifying Honduran Ecosystems ...... 50 Lowland-associated Habitats ...... 51 Habitats Shared Between Lowlands and the Chortís Highlands ...... 57 Habitats of the Chortís Highlands ...... 61 Introduction to Biodiversity of the Chortís Block ...... 70

3 TAXONOMIC DIVERSITY, DISTRIBUTIONAL PATTERNS, AND CONSERVATION STATUS OF THE CHORTÍS BLOCK HERPETOFAUNA ...... 72

Methods and Materials ...... 74 Field Sampling ...... 74 Taxonomic Scope and Standards ...... 74 Evaluating Conservation Status ...... 77 Results ...... 78 Composition of the Herpetofauna ...... 78 Patterns of Distribution and Endemism within the Chortís Block ...... 79 Distribution of Chortís Highland Endemics ...... 79 Conservation Status ...... 80 Sampling Results by Locality ...... 80 Parque Nacional Celaque ...... 81 Parque Nacional Cerro Azul Copán ...... 82 Parque Nacional Cerro Azul Meámbar ...... 83 Parque Nacional Cusuco ...... 86 Parque Nacional La Tigra ...... 87 Parque Nacional Montaña de Botaderos ...... 87

Parque Nacional Montaña de Comayagua ...... 90 Parque Nacional Montaña de Santa Bárbara ...... 91 Parque Nacional Montaña de Yoro ...... 94 Parque Nacional Pico Bonito ...... 95 Parque Nacional Pico Pijol ...... 95 Parque Nacional Sierra de Agalta ...... 96 Refugio de Vida Silvestre Texiguat ...... 97 Reserva Biológica Cerro Uyuca ...... 98 Reserva Biológica Guajiquiro ...... 100 Reserva Biológica Güisayote ...... 101 Reserva Biológica Yerbabuena ...... 102 Reserva de la Biosfera Bosawas ...... 102 Jardín Botánico Lancetilla ...... 104 Área de Uso Multíple Isla del Tigre ...... 106 Non-Protected Areas ...... 107 Cerro El Zarciadero ...... 107 Highlands surround the Meseta de La Esperanza ...... 107 Los Naranjos ...... 108 Montaña de Jacaleapa ...... 109 Montaña Macuzal ...... 109 Saguay ...... 112 San José de Texíguat ...... 112 Selva Negra ...... 113 Yeguare Valley ...... 113 Discussion ...... 116 Baseline Herpetological Inventory of Parque Nacional Montaña de Yoro ...... 116 A New Species of Anole ...... 118 Salamanders of Uncertain Taxonomic Assignment ...... 119 Hotspot within a Hotspot: the Special Case of Refugio de Vida Silvestre Texíguat ...... 120 Discovery of Plectrohyla chrysospleura ...... 122 Underestimated Salamander Diversity? ...... 124 Highly Endemic and Highly Endangered ...... 125 Cryptozoic Snake Diversity ...... 126 A New Species of Centipede Snake (genus Tantilla) from La Liberación ...... 127 A Large New Species of Blindsnake (Typhlops tycherus) ...... 128 Geophis damiani at La Liberación ...... 130 Noteworthy Ninia ...... 131 Unidentified Salamander Populations ...... 133

4 THE CHORTIS BLOCK IS AN UNDERESTIMATED HOTSPOT OF AMPHIBIAN DIVERSITY AND ENDEMISM ...... 166

Methods and Materials ...... 172 Sampling and Sample Identification ...... 172 Extraction, Amplification, and Sequencing ...... 173 Sequence Evaluation and Alignment ...... 174

Distance-Based Barcode Metrics ...... 175 Phylogenetic Analysis ...... 176 Species Delimitation and Candidate Species ...... 176 Results ...... 177 Broad Results for Distance-based Analyses ...... 177 Identification of Potential Candidate Species of Anura ...... 179 BLASTN Results for Unassigned Salamander Sequences ...... 187 Distance-Based Analyses of Caudata ...... 187 Phylogenetic Analysis of Caudata ...... 191 Discussion ...... 196 Candidates for Further Taxonomic Study Among Anurans ...... 196 Incilius coccifer/ibarrai/porteri ...... 197 Incilius coniferus ...... 198 Rhaebo haematiticus ...... 200 Craugastor aurilegulus ...... 200 Craugastor laevissimus ...... 200 Craugastor sp. inquirenda 1 & 2 ...... 201 Diasporus diastema ...... 201 Plectrohyla cf. guatemalensis ...... 201 Ptychohyla hypomykter ...... 202 Ptychohyla spinipollex ...... 202 Smilisca baudinii ...... 203 Leptodactylus fragilis ...... 204 Lithobates brownorum X forreri ...... 204 Lithobates brownorum ...... 205 Lithobates forreri ...... 205 Lithobates maculatus ...... 208 Lithobates taylori ...... 209 Lithobates warszewitschii...... 209 Pristimantis ridens ...... 209 Candidate Species and Allopatric Populations of Salamanders ...... 210 Bolitoglossa (Magnadigita) sp. inquirenda 1 ...... 210 Bolitoglossa (Magnadigita) oresbia / sp. inquirenda 2 ...... 211 Bolitoglossa (Magnadigita) celaque ...... 214 Bolitoglossa (Magnadigita) conanti ...... 214 Bolitoglossa (Magnadigita) porrasorum ...... 215 Bolitoglossa (Nanotriton) rufescens/nympha ...... 215 Nototriton barbouri ...... 217 Nototriton sp. inquirenda 1 ...... 217 Nototriton sp. inquirenda 2 ...... 218 Nototriton sp. inquirenda 3 ...... 218 Nototriton lignicola / sp. inquirenda 4 ...... 218 Nototriton limnospectator / sp. inquirenda 5 ...... 219 Oedipina kasios / sp. inquirenda 1 ...... 219 Oedipina nica / sp. inquirenda 2 ...... 220 Oedipina koehleri / sp. inquirenda 3 ...... 220

Oedipina gephyra ...... 222 Amphibian Endemism and Conservation Priorities in the Chortís Block ...... 223 Iterative Taxonomic Approaches to Biodiversity Inventory...... 224

5 CRYPTIC DIVERSITY AND REVISIONARY SYSTEMATICS OF CHORTÍS HIGHLAND MOSS SALAMANDERS (CAUDATA: PLETHODONTIDAE) ...... 250

Salamanders as Models for Evolutionary Study ...... 250 Salamander Diversity in the Chortís Block ...... 252 Methods and Materials ...... 256 Sampling ...... 256 DNA Extraction, PCR Amplification, and Sequencing ...... 257 Sequence Alignment and Model Selection ...... 258 Sequence Analyses ...... 259 Bayesian Inference and Maximum Likelihood Phylogenetic Analyses ...... 260 Comparative Morphology ...... 260 Results ...... 261 Discussion ...... 266 DNA Barcode Identification of Chortís Highland Moss Salamanders ...... 266 Influence of Substitution Saturation in Cytochrome B Dataset ...... 267 Phylogenetic Systematics and Candidate Species ...... 268 Systematics ...... 269 A Divergent New Lineage from Refugio de Vida Silvestre Texíguat ...... 269 Nototriton tomamorum Townsend, Butler, Wilson, & Austin 2010a ...... 270 A New Species from the Sierra de Agalta ...... 275 Nototriton picucha Townsend, Medina-Flores, Murillo, and Austin 2011 . 275 Restriction of the taxon Nototriton barbouri (Schmidt, 1936) ...... 281 Nototriton barbouri (Schmidt 1936) ...... 282 Description of unassigned populations from the Cordillera Nombre de Dios .. 284 Nototriton sp. A, sp. nov...... 284 Nototriton sp. B, sp. nov...... 285 Review of the Remaining Species of Nototriton from the Chortís Highlands .. 286 Nototriton brodiei Campbell & Smith 1998 ...... 287 Nototriton lignicola McCranie & Wilson 1997 ...... 287 Nototriton limnospectator McCranie, Wilson, & Polisar 1998 ...... 289 Nototriton stuarti Wake & Campbell 2000 ...... 290

6 INTEGRATING RESEARCH, EDUCATION, AND OUTREACH IN SUPPORT OF CONSERVATION IN THE CHORTÍS HIGHLANDS ...... 294

Taxonomic Inventories in Promotion of Education and Extension ...... 295 Opportunities for Training and Education ...... 298 Pilot Project: Parque Nacional Montaña de Yoro ...... 299 Concluding Statement...... 303

APPENDIX: TAXONOMIC REVIEW OF CAUDATA FROM THE CHORTÍS BLOCK .. 304

REFERENCES ...... 326

BIOGRAPHICAL SKETCH ...... 350

LIST OF TABLES

Table page

3-1 Summary of fieldwork undertaken in the Chortís Block, 2006–2011...... 135

3-2 Conservation status and physiographic distribution of the native non-marine herpetofauna of the Chortís Highlands ...... 137

3-3 Composition of the Chortís Block herpetofauna...... 159

3-4 Broad distributional patterns of herpetofaunal diversity in the Chortís Block. ... 160

3-5 Endemic and conservation priority herpetofaunal diversity from the Chortís Block ...... 160

3-6 Distribution by mountain ranges of the endemic amphibians and reptiles of the Chortís Highlands ...... 161

4-1 Voucher information for samples used in this study ...... 228

4-2 Average nucleotide compositions of various taxonomic groups ...... 244

4-3 Potential candidate species identified through this study ...... 245

4-4 Results of BLASTN searches of the NCBI database for 16S consensus sequences representing 10 populations of taxonomically-unassigned salamanders ...... 249

5-1 Samples used in sequence divergence and phylogenetic analyses ...... 291

5-2 Models of nucleotide substitution chosen for phylogenetic analyses of Chortís Highland taxa using Akaike Information Criterion values...... 292

5-3 Within and between species sequence divergence (uncorrected p-distance) for Chortís Highland moss salamanders ...... 292

5-4 Morphological and morphometric comparison of species of Nototriton ...... 293

LIST OF FIGURES

Figure page

1-1 Map showing present-day location of the Chortís Block ...... 21

2-1 Political divisions of the Chortís Block ...... 26

2-2 Tectonic plate reconstructions showing the relative position and movement of the Chortís Block during the Cenozoic ...... 30

2-3 Physiographic regions of the Chortís Block ...... 35

2-4 Map showing mountain ranges of the Chortís Block ...... 36

2-5 Mountain ranges of the Chortís Block I ...... 37

2-6 Mountain ranges of the Chortís Block II ...... 39

2-7 Ecological associations of the Chortís Block I ...... 52

2-8 Ecological associations of the Chortís Block II ...... 58

2-9 Ecological associations of the Chortís Highlands ...... 64

3-1 Map showing sampling localities in the Chortís Block ...... 75

3-2 Sampling in the Chortís Block I ...... 84

3-3 Sampling in the Chortís Block II ...... 88

3-4 Sampling in the Chortís Block III ...... 92

3-5 Sampling in the Chortís Block IV ...... 98

3-6 Sampling in the Chortís Block V ...... 104

3-7 Sampling in the Chortís Block VI ...... 110

3-8 Sampling in the Chortís Block VII ...... 114

3-9 Exemplar paratypes and habitats from Parque Nacional Montaña de Yoro ..... 117

3-10 La Liberación de Texíguat ...... 121

3-11 Plectrohyla chrysopleura (Hylidae) from La Liberación ...... 123

3-12 New species of Tantilla (Colubridae) and Typhlops (Typhlopidae) ...... 129

3-13 Noteworthy cryptozoic snakes ...... 132

4-1 Radial phylogram showing coverage of higher-level taxonomic groups ...... 178

4-2 Maximum likelihood phylogram of COI data showing generic and subgeneric relationships of salamander samples ...... 180

4-3 COI (left) and 16S (right) neighbor-joining trees for Bufonidae ...... 182

4-4 COI (left) and 16S (right) neighbor-joining trees for Craugastoridae, Eleutherodactylidae, Leptodactylidae, and Strabomantidae ...... 183

4-5 COI (left) and 16S (right) neighbor-joining trees for Hylidae ...... 185

4-6 COI (left) and 16S (right) neighbor-joining trees for Ranidae ...... 186

4-7 COI (left) and 16S (right) neighbor-joining trees for Bolitoglossa ...... 188

4-8 COI (left) and 16S (right) neighbor-joining trees for Cryptotriton, Dendrotriton, Nototriton, and Oedipina ...... 190

4-9 Maximum likelihood phylogram for the genus Bolitoglossa ...... 192

4-10 Maximum likelihood phylogram for the genera Nototriton, Oedipina, Dendrotriton, and Cryptotriton ...... 194

4-11 Candidate Species I: Bufonidae and Craugastoridae ...... 199

4-12 Candidate Species II: Eleutherodactylidae and Hylidae ...... 203

4-13 Candidate Species III: Ranidae and Strabomantidae ...... 206

4-14 Candidate Species IV: Bolitoglossa ...... 213

4-15 Candidate Species V: Bolitoglossa cf. porrasorum ...... 216

4-16 Candidate Species VI: Oedipina ...... 221

4-17 Candidate Species VII: Oedipina cf. gephyra (= O. petiola) ...... 221

5-1 Distribution of Nototriton in the Chortís Highlands ...... 255

5-2 Comparison of ―barcoding gaps‖ for three mitochondrial genes used to delimit species boundaries in Nototriton ...... 262

5-3 Bayesian phylograms showing discordance between combined mtDNA phylogenies due to saturation at the third codon position for cytochrome b ..... 265

5-4 Nototriton tomamorum ...... 271

5-5 Nototriton picucha ...... 277

5-6 Nototriton barbouri sensu stricto ...... 283

5-7 Nototriton sp. B ...... 286

5-8 Nototriton lignicola ...... 288

5-9 Nototriton limnospectator ...... 289

6-1 Print media coverage of the discovery of new endemic species during 2008 and 2009...... 297

6-2 Examples of public outreach and dissemination of results from taxonomic inventories ...... 300

Abstract of Dissertation Presented to the Graduate School of the University of Florida, in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

INTEGRATIVE TAXONOMY REVEALS THE CHORTÍS BLOCK OF CENTRAL AMERICA AS AN UNDERESTIMATED HOTSPOT OF AMPHIBIAN DIVERSITY

Josiah Harold Townsend

December 2011

Chair: James D. Austin Major: Interdisciplinary Ecology

Central America has a remarkably complex geological history, has served as the biological dispersal route between North and South America, and is the site of extensive in situ evolution. Nuclear Central America is recognized as a region of high biodiversity, and the eastern portion of Nuclear Central America (the Chortís Block) has largely been overlooked as a biodiversity hotspot. In this dissertation, I use an integrative systematic approach to examine evolutionary patterns in a group with high diversity of endemic species that is also of global conservation priority: amphibians. I carried out 17 expeditions to over 60 localities in the Chortís Block for this dissertation, results from which are synthesized with existing data on regional herpetofaunal diversity, distribution, and conservation status. The Chortís herpetofauna is characterized by a high degree of endemism (35% of all species are endemic) and equally high extinction risk (41% threatened, including 96% of endemic species). Endemism is highest among salamanders (86% of species are endemic). A total of 456 amphibian samples representing 52 species were sequenced for two genes (16S and COI) and analyzed using distance-based and phylogenetic methods. I identify at least 36 unnamed

―candidate‖ across eight families of amphibians, while confirming that four

taxonomically-uncertain salamander populations represented new allopatric populations of endangered species previously assumed to be single-site endemics. These results reveal a regional amphibian fauna, already recognized as being of global conservation priority, with species diversity underestimated by 26% (47% in salamanders), highlighting the critical role molecular systematics plays in endangered species conservation and management. To directly address the systematics a poorly-known group with multiple candidate species, I present an integrative taxonomic revision of the genus Nototriton using data from three genes, external morphology, and osteology. I restrict the taxon N. barbouri to populations from the Sierra de Sulaco, describe three new species from the Chortís Highlands, and a review of the remaining regional species. In light of these findings, I present my vision for using systematic biology as the basis for integrating research, education, and outreach in support of biodiversity conservation in the Chortís Block.

CHAPTER 1 INTRODUCTION

The formation of the Central American land bridge, with subsequent interchange of previously-isolated organisms of Laurasian and Gondwanian origin, has been a source of inquiry for biogeographers since the founding of the discipline (Wallace 1876; Stehli and Webb 1985). Central America has had a remarkably complex geological history, owing in large part to its position as the contact zone for five of the world’s 14 major tectonic plates: the Caribbean, Cocos, Nazca, North American, and South American plates (Figure 1-1; Bird 2003). Central America has not only served as the dispersal route between North and South America, but the region’s extreme topographical and ecological heterogeneity has also fuelled significant in situ diversification, particularly associated with the disjunct highland areas of Nuclear Central America and southern

Central America (Savage 1966, 1983; Wake 1987).

Whereas Nuclear Central America has long been accepted as a region of high biodiversity, some observers have further recognized the western and eastern portions of this highland block as distinct biogeographic entities (Johnson 1989; Campbell 1999;

Townsend 2006, 2009). Eastern Nuclear Central America, corresponding to the Chortís

Block tectonic formation (Figure 1), has been shown to have a distinctive component of endemic biodiversity, particularly in amphibians and reptiles (Wilson & Johnson 2010), however molecular characterization of evolutionary patterns of diversification in this region has been limited to a few studies of a restricted taxonomic breadth and broader geographic focus (e.g. Castoe et al. 2009).

Figure 1-1. Map showing present-day location of the Chortís Block. Relative present- day positions of four major tectonic plates and selected elements of the Central American Isthmus are indicated, with the contemporary Chortís Block highlighted in green (map generated using ODSN Plate Tectonic Reconstruction Service1 and modified by author).

My principal goal is to broadly initiate molecular systematics research and studies in evolutionary biogeography in the Chortís Block (Figure 1-1), a distinctive biogeographic region with a complex and unique geomorphological history and highly variable contemporary ecophysiographic landscape. This region has a largely unrealized biogeographic importance as the first fragment of proto-Central America to contact the North American plate; furthermore, it is characterized by a high degree of in- situ endemism and evolutionary diversification in its extant biota. As a group with the highest documented endemism and highest risk for extinction, amphibians, and

1 http://www.odsn.de/odsn/services/paleomap/paleomap.html

particularly salamanders, make excellent taxa upon which to base studies of systematics and evolution.

I begin in Chapter 2 by providing a synthesis of relevant characteristics of the

Chortís Block, in order to support a working definition for use in this dissertation and beyond. I begin by summarizing the geological history of the Chortís Block, providing a comprehensive account of the contemporary physiographic and ecological landscape, and finally presenting an operationalized version of Carr’s (1950) outline of a system for characterizing Honduran animal habitats.

In Chapter 3, I provide an assessment of the diversity, conservation status, and distribution of amphibians and reptiles in the Chortís Block. This assessment is based on over 12,560 person-hours of fieldwork conducted since 2006 (over 2,577 person- hours logged myself), and existing data on distribution and conservation status of amphibians and reptiles from the region. The Chortís Block herpetofauna is characterized by a high degree of endemism that is faced with an equally high degree of extinction risk. Thirty-five percent of the Chortís Block herpetofauna is endemic, and

41% of those species are conservation priority (i.e. listed in the IUCN Red List’s three highest threat categories, Critically Endangered, Endangered, and Vulnerable), including an alarming 96% of endemic species. Endemism is by far highest among the salamanders (86% of species are endemic); while 75% of salamander species are considered conservation priority by the IUCN (2011). This figure likely is underestimated significantly.

As a result of the fieldwork detailed in Chapter 3, a comprehensive sample set of taxa and localities was assembled for Chortís Block amphibians, with particularly good

representation of salamanders (Caudata). In Chapter 4, take an integrative taxonomic approach involving genetic distance-based analysis (i.e., DNA barcoding) and model- based phylogenetic analysis to identify and characterize cryptic amphibian diversity in the Chortís Block. Analysis of 456 individual samples representing 52 named species in eight families of amphibians indicate that there are at least 36 unnamed ―candidate‖ species (20 anurans and 16 salamanders) in the Chortís Block, while confirming that four taxonomically-uncertain salamander populations actually represented new allopatric populations of endangered species previously assumed to be single-site endemics. These results reveal a regional amphibian fauna, already recognized as being of global conservation priority, to be underestimated by 26% (47% in salamanders) in terms of species diversity, highlighting the critical role molecular systematics plays in endangered species conservation and management.

In Chapter 5, I further utilize an integrative approach to delineate and estimate species-level phylogenetic relationships among and endemic radiation of highly cryptic moss salamanders (genus Nototriton) in the Chortís Highlands. Using data from three mitochondrial DNA loci, I examine the utility of each gene in a distance-based DNA barcoding approach for determining species diversity. The evolutionary relationships of diagnosed lineages are inferred using model-based phylogenetic analyses. My results indicate that species-level diversity in Nototriton is underestimated in the Chortís Block.

Three undescribed species of Nototriton were identified: one from the Sierra de Agalta in Departamento de Olancho, and one each from Pico Bonito and Texíguat in the

Cordillera Nombre de Dios. Finally, the taxonomy of this clade of Nototriton is reviewed and revised, with a restriction of the taxon N. barbouri (Schmidt) to populations from the

Sierra de Sulaco, formal description of the three new species, and a review of the remaining species.

Finally, Chapter 6 presents my long-term vision for scientific exploration, research, education, and outreach in support of herpetofaunal conservation in Honduras, the country containing the bulk of the Chortís Highlands and the majority of the Chortís

Block’s endemic species. I then provide a closing synthesis of the dissertation, reviewing the results within a context of their regional and global significance.

CHAPTER 2 BIOGEOGRAPHIC DELIMITATION OF THE CHORTÍS BLOCK: GEOLOGICAL HISTORY, PHYSIOGRAPHY, AND ECOLOGICAL ASSOCIATIONS

My enthusiasm for the topic of Chortís Block biogeography grew from the realization that patterns of endemism observed over the course of my studies in

Honduras and northern Nicaragua seemed to be largely a reflection of emerging patterns in tectonic plate dynamics that have shaped the complex geomorphological history of the region. In previous works, I have favored use of the ―Eastern Nuclear

Central America‖ biogeographic province (Townsend 2006, 2009; Townsend & Wilson

2010a) in order to set this region apart, in recognition of its distinctiveness from proximal regions. This region is geographically analogous to the Chortís Block, an allochthonous1 geological formation that today forms the only modern continental portion of the

Caribbean Tectonic Plate and the largest terrestrial segment of the contemporary

Central American land bridge (Rogers 2003; Marshall 2007). The Chortís Block has a challengingly complex history, and has recently been the subject of increased focus, and sometimes contentious debate, within the geological research community (James

2007; Mann et al. 2007; Ortega-Gutiérrez et al. 2007; Silva-Romo 2008; Morán-Zenteno et al. 2009).

Politically, the contemporary region I refer to as the Chortís Block includes all of the country of Honduras, the northern portion of El Salvador, eastern Guatemala, and northern Nicaragua (Figure 2-1). As a working delimitation of the biogeographic region, I consider the Chortís Block to be nearly homologous with Campbell’s (1999: 116) concept of an Eastern Nuclear Central American biogeographic province. I include the

1 Allochthonous – referencing an independent geological structure that has moved from its site of origin.

Figure 2-1. Political divisions of the Chortís Block. Countries are shaded differently and labeled in all capital letters, while departments are labeled in sentence-case in a smaller font size.

associated coastal plains, with the western extent of the Chortís Block at the edge of the

Río Motagua Valley (the eastern edge of the Polochic-Motagua fault complex) to a north-south line roughly running through Zacapa, Chiquimula, Concepcíon Las Minas, and the Guatemalan-El Salvador border at the Pacific Coast. It extends eastward to include all of Honduras and El Salvador, and south to a line roughly between Lago

Xolotlán (= Lago de Managua) and Lago Colcibolca (= Lago de Nicaragua) in northern

Nicaragua (Figure 2-1). Without including the associated lowlands, Campbell’s Eastern

Nuclear Central America is homologous with the highlands of the Chortís Block, referred to collectively as the Chortís Highlands (Marshall 2007) or the serranía (Carr 1950).

An integrative definition of the Chortís Block Biogeographic Province, as presented here, combines an ecological and biodiversity-based framework, as used to delineate

Eastern Nuclear Central America, with that of a physiographically and tectonically- defined Chortís Block.

Geological History

The history of the Chortís Block is in large part characterized by its eastward movement along a series of strike-slip faults on the southern margin of the North

American Plate (Figure 2-2; Dengo 1969; Donnelly et al. 1990; Gordon 1992; Rogers et al. 2007). As recently as the K-T Boundary (65 million years before present [mybp];

Figure 2-2), the Chortís Block was located somewhere south of modern south-central

México, having moved along a west-to-east trajectory around 200 km into its current position as the principal surface area of the Central American land bridge and the modern territory of Honduras, El Salvador, eastern Guatemala and northern Nicaragua

(Figure 2-1; Rogers 2003; Ortega-Gutiérrez et al. 2007).

Origin and Cenozoic development of the Chortís Block. The Chortís Block represents the only exposed Precambrian and early Paleozoic continental crust on the contemporary Caribbean Plate (DeMets et al. 2007). The oldest exposed geological formation of the Chortís Block is Precambrian in age and of Rodinian2 derivation, originating during the Grenville orogeny (1,017 ± 20 mybp or 1400 mybp, depending on dating methods) contemporaneously with the Appalachian and Adirondack mountains of eastern North America and the Llano Plateau of Texas and northeastern México

(Mandon 1996; Gordon et al. 2010). Today, this ancient formation is visible as a 60 km- long series of exposed outcrops along the Jocotán-Ceiba Fault in the Departamento de

Yoro, Honduras (Gordon et al. 2010). While it is generally accepted among geologists that the Chortís Block originated approximately 1,100 km west of its current position and became detached during the Eocene, sliding and rotating along the Motagua-Polochic

Fault Complex at the southern margin of the stationary North American Plate, there are two competing hypotheses regarding the Chortís Block’s origin (reviewed by Rogers et al. 2007). The first hypothesis places the Chortís Block along the southwestern margin of the North American Plate and physically contiguous with México, and is supported by the existence of similarly aged Precambrian and Paleozoic rock formations, potential aligned fault systems in present day Honduras and southwestern México, and geological evidence that the Chortís Block rotated 30–40° counterclockwise while sliding eastward (Figure 2-2; Gose 1985, Silva-Romo 2008). The second hypothesis places the Chortís Block’s original position some 700–800 km south of México, with it moving northeastwardly while rotating 40° clockwise (Keppie & Moran-Zenteno 2005). A

2 Rodinia was a paleo-supercontinent that contained most of Earth’s landmass and existed between approximately 2,000 mybp and 750 mybp

third, less accepted hypothesis holds that the Chortís Block has remained in virtually the same position relative to the North American Plate, and that structures and evidence to the contrary have essentially been misinterpreted (James 2007).

The relatively dramatic Cenozoic tectonic history of the Chortís Block was dominated by what can only be described as catastrophic, prolonged, and repeated volcanism as the block slid and rotated its way eastward. The beginning of this extended period was the mid-Eocene (approximately 55 mybp), following initial detachment of the Chortís Block from its parent structure (Jordan et al. 2008). A second flare-up took place during the mid-Oligocene (around 40 mybp), and the third, and largest, flare-up took place during the early to middle Miocene (Jordan et al. 2008).

Miocene Ignimbrite Flare-up and subsequent uplift. The Mesozoic history of the Chortís Block features an approximately 10 million year period of intense explosive volcanism along the margins of the Chortís Block and Central American Volcanic Front, considered the second largest ignimbrite1 event in the known geological history of Earth

(Jordan et al. 2008). During the mid-Miocene over 5,000 km3 of ignimbrites up to 2,000 m thick were deposited on top of the low-relief surface of the southern and western

Chortís Block and tens of thousands of square kilometers were repeatedly covered in thick layers of ash (Williams & McBirney 1969; Rogers et al. 2002; Jordan et al. 2008).

The most intense period of the ignimbrite flare-up lasted from around 20 mybp to 15 mybp, with activity ceasing approximately 10.5 mybp (Gordon & Muehlberger 1994).

This period is well documented by a series of deep-sea sediment cores from sites in the western Caribbean Sea (Jordan et al. 2008). The site of a fissure-like volcano that was

3 Ignimbrite – hard rock formed from the deposition of super-heated pyroclastic flow

Figure 2-2. Tectonic plate reconstructions showing the relative position and movement of the Chortís Block (shaded green) during the Cenozoic. Follows the generally accepted tectonic model of Hay et al. (1999; alternative models reviewed in Rogers et al. 2007), with black lines represent tectonic boundaries and gray shading delineates modern-day shorelines and landmasses (maps generated using ODSN Plate Tectonic Reconstruction Service2 and modified by author).

4 http://www.odsn.de/odsn/services/paleomap/paleomap.html

―ground zero‖ for the Miocene Ignimbrite Flare-up is represented today by the Padre

Miguel Group geological formation in southwestern Honduras and peripherally in El

Salvador and Guatemala (Rogers 2003). Under these circumstances, it would seem unlikely that extant terrestrial organisms and ecosystems on the Chortís Block are survivors of this extreme volcanism, and more likely that the extant biota represents the product of post-volcanic colonization and diversification. Following the Miocene flare-up and up and until approximately 3.8 mybp, the Chortís Block went through a period of rapid uplift driven by the detachment and subsequent subduction of a portion of the

Cocos Plate, which induced upwelling in the mantle that raised the Chortís Block up to

1,100 m (Rogers et al. 2002).

The contemporary Chortís Block. The Chortís Block presently continues its eastward movement along the strike-slip faults of the Motagua-Polochic Fault Zone and

Swan Island Fault Zone, interacting in the continental context with the Maya Block of the

North American Plate to the north and the Chorotega Block to the south, albeit interrupted by the Nicaraguan Depression (Figure 1-1, 2-2; Rogers 2003, Marshall

2007). Marshall (2007) defined 15 physiographic provinces in Central America, four of which (the Chortís Highlands, Chortís Volcanic Front, Chortís Fore Arc, and Mosquitia

Coast Lowlands) are geomorphological associates of the Chortís Block.

The Chortís Highlands Province consists of a large dissected plateau that forms the greater part of the Chortís Block and includes the majority of the territory of the countries of Honduras and El Salvador, as well as western Guatemala and northern

Nicaragua (Marshall 2007). The Chortís Highlands Province is subdivided into four regions: the Western Rifted Highlands, the Central Chortís Plateau, the Eastern

Dissected Highlands, and the Honduran Borderlands. The Chortís Volcanic Front

Province is an active volcanic front that borders the southern margins of the Chortís

Highlands Province, and includes two regions: the Guatemalan Cordillera, which borders the western margin of the Chortís Highlands Province; and the Salvadoran

Cordillera, which borders the Median Trough, an elongate graben that extends along the boundary faults at the margin of the active Nicaraguan Volcanic Front (Marshall

2007). The Chortís Volcanic Front Province represents part of the proverbial ―Ring of

Fire.‖ a loose chain of active volcanoes and tectonic plate subduction that rings the

Pacific Ocean. The Chortís Fore Arc Province encompasses the Pacific coastal plain of the Chortís Volcanic Front Province, and is similarly subdivided into the Guatemalan

Coastal Plain and Salvadoran Coastal Plain (Marshall 2007). The Mosquito Coast

Lowlands Province is the wide alluvial plain along the eastern Caribbean slope of

Honduras and Nicaragua, a region also referred to as La Mosquitia. The Mosquito

Coast Lowlands Province is dominated by a massive paleo-Coco/Patuca river delta built up during the glacial cycles of the Pliocene-Pleistocene (Marshall 2007).

The aforementioned Chortís Highlands Province and its four constituent subregions are of principal interest for this dissertation, and these subregions are described in detail below.

The Western Rifted Highlands region of southeastern Guatemala, southwestern

Honduras, and northern El Salvador is a west-to-east oriented plateau, generally exceeding 1,000 m elevation, which is interrupted by a series of independent, north-to- south oriented rift valleys featuring flat, xeric valley floors (Marshall 2007). The rift valleys, or grabens, of the Western Rifted Highlands include the contemporary

Comayagua Valley and Otoro Valley. This region corresponds to the Padre Miguel

Group, a 1,000–2,000 m thick layer of mid-Miocene ignimbrites laid down during the super-volcanic eruptions along the margin of the Chortís Highlands and Chortís

Volcanic Front. Those super-eruptions essentially reset the landscape, allowing the development of new meandering river drainages as the rift valleys began spreading following the end of the Miocene ignimbrite flare-up around 10.5 mybp (Gordon &

Muehlberger 1994; Rogers et al. 2002).

The Central Chortís Plateau region of the Honduran interior represents the most tectonically stable portion of the Chortís Highlands, forming an essentially level plateau with little dissection or embedding by rivers (Marshall 2007). The Central Chortís

Plateau lies atop of Paleozoic bedrock overlain with layered Cretaceous-aged sedimentary deposits (Rogers et al. 2002; Marshall 2007).

The Eastern Dissected Highlands region of eastern Honduras and Nicaragua includes the lower elevation, higher relief mountains bordering the Mosquito Coast

Lowlands, and features three large, deeply embedded river drainages that drain a large portion of the Chortís Highlands Province (Marshall 2007).

The Honduran Borderlands region lies along the northern margin of the Chortís

Highlands, and is characterized by five major west-to-east trending faults that border major mountain ranges, including the Cordillera Nombre de Dios in northern Honduras

(Rogers 2003, Marshall 2007). One large graben valley, the Sula Graben, extends north-to-south from the Caribbean coast to the north end of Lago de Yojoa, and today contains the lower courses of two of the largest watersheds in the Chortís Block: the Río

Chamelecón and the Río Ulua.

Contemporary Ecophysiography

Carr (1950), in his pioneering classification of Honduran ecological associations, recognized three principal ecophysiographic components that make up the Chortís

Block: the Caribbean versant lowlands, the Pacific versant lowlands, and the mountainous interior region known as the serranía (Figure 2-3). Carr’s ecophysiographic regions correspond well with the geologically-based physiographic provinces of Marshall (2007; described in the previous section), with the serranía being congruent with the Marshall’s Chortís Highlands Physiographic Province. Subsequently, from this point forward I use the name ―Chortís Highlands‖ analogously with serranía.

Carr (1950) further subdivided the Chortís Highlands into the Northern Cordillera, the

Southern Cordillera, and the Pacific Colinas. This arrangement was also used by Wilson

& Meyer (1985), McCranie & Wilson (2002), and others. Mejía-Ordóñez & House (2002) introduced a modified arrangement, based on their comprehensive evaluation of the ecosystems of Honduras using the UNESCO system of Physiognomic-Ecological

Classification of Plant Formations of the Earth (Mueller-Dombois & Ellenberg 1974), which recognized a Cordillera del Norte, Cordillera Central, and Cordillera del Sur. This arrangement is preferable and is used here as the basis for describing the ecophysiography of the Chortís Block, which I divide into three principal regions: the

Caribbean Lowlands, the Pacific Lowlands, and the Chortís Highlands, which itself is subsequently subdivided into the Northern, Central, and Southern Cordilleras.

Northern Cordillera of the Serranía. As first defined by Mejía-Ordóñez & House

(2002) and expanded here, the Northern Cordillera consists of the following mountain ranges and groups of ranges:

Figure 2-3. Physiographic regions of the Chortís Block. After Carr (1950).

 The Cordillera (or Sierra) Nombre de Dios (Figure 2-4) stretches west-to-east across the departments of Atlántida, Colón, and Yoro, Honduras, and includes the cloud forest protected areas Refugio de Vida Silvestre (RVS) Texíguat (maximum elevation 2,208 m) at the western end and Parque Nacional (PN) Pico Bonito (2,435 m) and PN Nombre de Dios (1,725 m) in the central portion, with a few scattered low peaks extending to the east, terminating with PN Capiro y Calentura (1,235 m) near Trujillo. Based on my preliminary observations, I consider the Sierra de Mico Quemado, a north-to-south oriented range in western Yoro, to be the western terminus of the Cordillera Nombre de Dios. This range, which includes Zona de Reserva Ecológica (ZRE) Montaña de Mico Quemado y Las Guanchias, was considered part of the Central Cordillera by Mejía-Ordóñez & House (2002).

 The Sierra de Omoa (Figure 2-4) in the departments of Cortés and Santa Bárbara, Honduras, with the cloud forest protected area PN Cusuco (2,242 m), as well as the Área de Producción de Agua Merendón (1,749 m).

 The Sierra de Espíritu Santo (Figure 2-4) in the departments of Copán and Santa Bárbara, in Honduras and Izabal and Zacapa in Guatemala, which includes the cloud forest reserve PN Cerro Azul Copán (2,285 m), as well as unprotected highland forests at Río Amarillo (1,479 m) in Copán, Honduras, and Cerro del Mono (1,653 m) in Zacapa, Guatemala.

While these mountain ranges appear disjunct and isolated (Figure 2-4), particularly with respect to the Sula Graben, their herpetofaunal composition and patterns of

Figure 2-4. Map showing mountain ranges of the Chortís Block. Green outlines correspond to the Northern Cordillera, blue to the Central Cordillera, red to the Southern Cordillera, and black to ranges extralimital to this study. 1 = Sierra Nombre de Dios, 2 = Sierra de Omoa, 3 = Sierra de Espíritu Santo, 4 = Sierra de Joconal, 5 = Montaña de Santa Bárbara, 6 = Montañas de Meámbar, 7 = Sierra de Montecillos, 8 = Sierra de Comayagua, 9 = Sierra de Sulaco, 10 = Cordillera de La Flor-La Muralla, 11 = Sierra de Agalta, 12 = Sierra de Botaderos, 13 = Sierra Punta Piedra, 14 = Montañas de Patuca, 15 = Sierra de Montecristo, 16 = Sierra del Merendón, 17 = Sierra de Celaque, 18 = Sierra de Erandique, 19 = Sierra de Puca-Opalaca, 20 = Montaña de la Sierra, 21 = Sierra de Lepaterique, 22 = Sierra de Dipilto, 23 = Montaña de Colón, 24 = Cordillera Dariense, 25 = Salvadoran Cordillera, 26 = Cordillera de Las Marabios.

Figure 2-5. Mountain ranges of the Chortís Block I. A) Cordillera Nombre de Dios, seen from offshore looking south; Cerro Búfalo is the tallest mountain on the left (east) side, and Pico Bonito is the sharp peak on the right (west) side. B) Sierra de Omoa, seen from near Buenos Aires de Bañaderos looking east- northeast. C) Southern slopes of Cerro Azul de Copán in the Sierra de Espíritu Santo, seen from Quebrada Grande. D) Montaña de Santa Bárbara, seen from the road west of Peñas Blancas looking south-southwest. E) Montañas de Meámbar, seen from the road south of Santa Elena looking south. F) Sierra de Comayagua, seen from road south of San Jerónimo looking south. (Photos © J.H. Townsend). distribution reflect a greater shared evolutionary history among these mountains than any of them share with other ranges (as evidenced in Chapter 4).

Central Cordillera of the Serranía. I follow Mejía-Ordóñez & House (2002) in recognizing a Central Cordillera made up of the remaining Caribbean versant serranía, otherwise included in Carr’s (1950) Northern Cordillera.

 The Sierra de Joconal (1,688 m; Figure 2-4) extends to the roughly west-to-east from eastern Departamento de Copán (municipalities Nueva Arcadia and San Nicolás), across Departamento de Santa Bárbara and into western Departamento de Cortés (municipality of Villanueva).

 Montaña de Santa Bárbara (2,744 m; Figure 2-4) is an isolated karstic massif rising from the southern terminus of the Ulúa-Chamelecón Plain to the west of Lago de Yojoa, its unique ecological communities wholly contained within the boundaries of PN Montaña de Santa Bárbara.

 The Montañas de Meámbar (2,080 m; Figure 2-4), also called the Montañas de Yule, are a rugged set of peaks on the eastern side of Lago de Yojoa on the border between the departments of Cortés and Comayagua, primarily contained within PN Cerro Azul Meámbar. Mejía-Ordóñez & House (2002) apparently included this group of mountains in the Sierra de Montecillos; however I consider it to be a separate formation.

 The Sierra de Montecillos (Figure 2-4) is one of two roughly parallel mountain ranges that are oriented northwest-southeast and form the margins of the ―Honduran Depression‖ in central Honduras. Some of the highest portions of this mountain range, which straddles the border between the departments of Comayagua and Intibucá, make up RB Montecillos (2,459 m).

 The Sierra de Comayagua (Figure 2-4) runs roughly parallel to the Sierra de Montecillos, separated by the dry intermontane Comayagua Valley, and extends over 130 km north-to-south along the border between the departments of Comayagua and Francisco Morazán. The highest portions of this range are found within PN Montaña de Comayagua (2,407 m) and RVS Corralitos (2,117 m).

 The Sierra de Sulaco (Figure 2-4) runs roughly west-to-east in the southwestern part of the Departamento de Yoro, and includes the highlands of PN Pico Pijol (2,282 m) at the western end of the range and Montaña Macuzal (1,945 m) at the eastern end.

 The Cordillera de La Flor-La Muralla (Figure 2-4) stretches across northern Departamento de Francisco Morazán, southern Departamento de Yoro, and into western Departamento de Olancho, with highland forest protected areas PN Montaña de Yoro (2,245 m), RVS La Muralla (2,064 m), Reserva Forestal Anthropológica (RFA) Montaña de la Flor (1,637 m), RB El Cipresal (1,930 m), and RB Misoco (2,153 m).

Figure 2-6. Mountain ranges of the Chortís Block II. A) Looking west along the spine of the Sierra de Sulaco, taken from the top of Montaña Macuzal, with Pico Pijol being the largest peak in the distance. B) Southeastern reaches of the highest peak in the Sierra de Botaderos. C) The Sierra de Celaque viewed from the east, with the tallest peak in the Chortís Block, Cerro de la Minas, visible as the peak in the middle of the photograph. (Photos © J.H. Townsend).

 The Sierra de Agalta (Figure 2-4) in central Olancho is a long, relatively narrow and steep range that includes, from west to east, the protected areas Monumento Natural Boquerón (1,261 m), PN Sierra de Agalta (2,335 m), Reserva Anthropológica El Carbón (1,817 m), and PN Sierra de Río Tinto (1,925 m).

 The Sierra de Botaderos (Figure 2-4) is located in northern Olancho along the border with the department of Colón, and includes Cerro Ulloa (1,735 m) and Cerro Azul (1,433 m) within the highland reserve PN Montaña de Botaderos, as well as the lower mountains of the Sierra de La Esperanza.

 The Sierra Punta Piedra (Figure 2-4) is a relatively low elevation range in the departments of Colón and Gracias a Dios, and includes Montaña Punta Piedra (1,500 m), Cerro Antílope (1,075 m), Cerro Mirador (1,200 m), and Cerro Baltimór (1,082 m). These mountains are found in Reserva de Hombre y la Biosfera Río Plátano.

 The Montañas de Patuca (1,155 m; Figure 2-4) in Olancho are located between the Río Guayape, which flows directly southwest to meet the Río Guayambre and form the Río Patuca, and a lower course of the Río Patuca that flows northeast to the Caribbean Sea. The southeastern portion of this range is found within PN Patuca.

Southern Cordillera of the Serranía. The Southern Cordillera is a geomorphologically linked series of mountain ranges extending from the vicinity of the

El Salvador-Guatemala-Honduras border region east-southeast into northern

Nicaragua.

 The Sierra de Montecristo (Figure 2-4) has its highest elevations at the point where El Salvador, Guatemala, and Honduras meet, in a tri-nationally managed protected area called Montecristo Trifinio (2,419 m). Most of this range is found in Guatemala, where it extends northward into the department of Chiquimula.

 The Sierra del Merendón (Figure 2-4) is a north-south oriented range that extends from Guatemala (Chiquimula, Zacapa) across Honduras (Copán, Ocotepeque, Lempira) and into El Salvador (Chalatenango), and includes the following cloud forest areas: RVS Erapuca (2,380 m), Cerro El Pital (2,730 m), Cerro Sumpul (2,167 m), and Reserva Biológica (RB) Güisayote (2,310 m),

 The Sierra de Celaque (Figure 2-4) is a north-south oriented range located in Lempira and easternmost Ocotepeque, and contains the highest elevations in the Chortís Highlands in PN Celaque (including peaks of 2,849 m, 2,825 m, and 2,804 m elevation) and RB Volcán Pacayita (2,516 m).

 The Sierra de Erandique (2,134 m; Figure 2-4) is a north-south oriented range in southeastern Lempira, extending from the municipality of La Campa at the northern end to the municipality of Piraera in the south.

 The Sierra de Puca-Opalaca (Figure 2-4) is located in Intibucá, northeastern Lempira, and extreme southern Santa Bárbara, and includes cloud forest areas found in RC Cordillera de Opalaca (2,390 m), RVS Puca (2,234 m), RVS Mixcure (2,312 m), and RVS Montana Verde (2,127 m).

 The Montaña de la Sierra (Figure 2-4) is found in the department of La Paz and extreme southern Intibucá, and includes a number of peaks and high plateaus, including a number within RB Guajiquiro (2,265 m), RB El Chiflador (1,811 m), RB El Pacayal (1,955 m), RB Mogola (1,648 m), RB Sabanetas (2,047 m), RB San Pablo (1,741 m), and RB San Pedro (1,719 m).

 The Sierra de Lepaterique (Figure 2-4) is the roughly U-shaped range that circles the southern side of the upper Choluteca valley, which is also the valley containing the Honduran capital, Tegucigalpa. This range includes PN La Tigra (2,290 m), RB Yerba Buena (2,243 m), RB Cerro Uyuca (2,006 m), RB El Chile (2,190 m), and RB Monserrat-Yuscarán (1,825 m).

 The Sierra de Dipilto (Figure 2-4) extends over 300 km west-to-east from PN La Botija (1,710 m) in Choluteca, to the Cordillera Entre Ríos in PN Patuca, straddling

the Honduras-Nicaragua border and including Reserva Natural (RN) Cerro Mogotón (2,106 m), the highest point in Nicaragua.

 The Montaña de Colón (Figure 2-4) is a low (maximum elevation 941 m), isolated karstic range located in southeastern Olancho and adjacent Gracias a Dios, and is found primarily within the Reserva de Biosfera Tawahka-Asangni.

 The Cordillera Dariense (Figure 2-4) is a collection of cloud forested peaks and highland areas in northern Nicaragua, in the departments of Jinotega, Matagalpa, and Región Autónoma Atlántico Norte, including Reserva Natural (RN) Apante (1,442 m), RN Cerro Musún (1,438 m), RN Dantalí-El Diablo (1,680 m), RN Kilambé (1,755 m), RN Peñas Blancas (1,744 m), RN Saslaya (1,658 m), and RN Volcán Yali (1,709 m).

Intermontane Valleys and Plains. The Chortís Highlands can be characterized as well for its valleys as it can be for its mountains; the isolated mountains form

―islands‖ of cool mesic habitat and the subhumid intermontane valleys represent isolated areas of hot, dry habitat. Aspects of the physiography, ecological associations, and biogeography of these subhumid valleys were the subject of study by Stuart (1954),

Johannessen (1963), Wilson & McCranie (1998), Sasa & Bolaños (2004), and

Townsend & Wilson (2010b).

 The Middle Motagua Valley is among the driest areas in Central America, along with the Middle Aguán Valley in Honduras. This valley lies between the Sierra de las Minas (extralimital to the Chortís Highlands) and the Sierra Espíritu Santo.

 The Sula Valley in northwestern Honduras is formed from combined drainages of two large watersheds, the Río Chamelecón and Río Ulúa, which have courses that flow closely together in their lower reaches into the Caribbean Sea. The Sula Valley is a north-to-south oriented graben valley that has been spreading since the late Miocene.

 The Otoro Valley is a moderately high elevation subhumid graben valley (lowest elevations 500–600 m) lying in a narrow upper portion of the Sula Valley, possessing an ecological character distinctive from that of the broader middle Sula Valley.

 The Comayagua Valley is a relatively high subhumid graben valley (lowest elevations 580–680 m) that forms a principal portion of the Honduran Depression, lying between the Sierra de Montecillos to the west and the Sierra de Comayagua to the east. This valley, like the Otoro Valley to the west, is actually a narrow upper portion of the Sula Valley, distinctive enough in character to warrant recognition.

 The Middle Aguán Valley is a west-to-east oriented fault valley that lies in the rain- shadow of the Cordillera Nombre de Dios in Yoro and is one of the driest areas in the Chortís Block.

 The Siria-Talanga Valley is a high plain (lowest elevations 620–720 m) in central Francisco Morazán that is the headwaters of two of the largest watersheds in the Chortís Highlands, the Río Guayambre/Río Patuca and the Río Ulúa.

 The Olancho Valley (or Guayape–Guayambre Valley) is a large valley in central Olancho surrounded by several mountain ranges, including the Sierra de Agalta, Cordillera de La Flor-La Muralla, and Montañas de Patuca, to form the headwaters of the Río Patuca.

 The Agalta Valley, also referred to as the ―San Esteban Valley‖ by various authors (Wilson & McCranie 1998, McCranie & Wilson 2002, Townsend & Wilson 2010b), is found between the Sierra de Botaderos and Sierra de Agalta in central Olancho. The lowest elevations of the relatively high subhumid intermontane valley, formed by the Río Grande (a river whose name changes to Río Sico, Tinto, and Negro downstream), are 550–650 m.

 The Middle Lempa Valley lies on a west-to-east orientation in central El Salvador between the Southern Cordillera of the Serranía and the Salvadoran Cordillera.

 The Upper Segovia Valley in Nicaragua is a subhumid region in the headwaters of the Río Coco (the upper reaches are also called the Río Segovia, and the lower river is called the Wangki by the indigenous Miskitu).

 The Choluteca Valley is the only major subhumid intermontane valley on the Pacific versant, and is initially oriented south-to-north from the headwaters of the Río Choluteca in the Sierra de Lepaterique, before curving around the north side of the mountains protected within PN La Tigra and turning south and then southwest on its path to the Pacific Ocean.

 The Meseta de La Esperanza is the highest plain in the Chortís Highlands, extending 12 km in length across central Intibucá at elevations ranging between 1800 and 2000 m.

 The Meseta de Siguatepeque is located in Comayagua between the Sierra de Comayagua, Sierra de Montecillos, and Montañas de Meámbar at around 1,100 m elevation.

 The Meseta de Santa Rosa is a wide plain on a plateau at about 1,100 m elevation in western Copán.

Caribbean Lowlands. Corresponding to the Mosquito Coast Lowlands

Physiographic Province of Marshall (2007), the major ecophysiographic regions of the

Caribbean lowlands include (McCranie & Wilson 2002, Wilson & Townsend 2006): the

Motagua Plain (lower alluvial plain of the Río Motagua, east of the river and northwest and west of the Sierra de Omoa and Sierra de Espíritu Santo), the Ulúa-Chamelecón

Plain (large alluvial plain formed by Chamelecón and Ulúa rivers, which drain close to half of the physical territory of Honduras), the Nombre de Dios Piedmont (the narrow strip of coastal plain backed by the Cordillera Nombre de Dios), the Aguán-Negro Plain, and the wide expanse of the Mosquitia (broad alluvial plain essentially lying between the

Sico-Paulaya watershed in Honduras and the Río Grande de Matagalpa watershed in

Nicaragua). Two climatic regimes are present in the Caribbean Lowlands (McCranie and Wilson 2002, Wilson & Townsend 2006): the Lowland Wet climate is found on the

Caribbean coastal plain from sea level to about 600 m elevation, with mean annual

precipitation exceeding 2000 mm and mean annual temperature exceeding 24°C.

Important protected areas for Caribbean Lowlands ecosystems include: Parque

Nacional (PN) Cuyamel-Omoa, PN Jeannette Kawas, Refugio de Vida Silvestre (RVS)

Cuero y Salado, PN Punto Izopo, Jardín Botánico Lancetilla, RVS Laguna de

Guaymoreto, Reserva de Hombre y la Biosfera Río Plátano, PN Patuca, PN Warunta,

Reserva Biológica (RB) Rus Rus, RB Laguna de Karataska, PN Río Kruta, Reserva de

Biosfera Tawahka-Asangni, Reserva de Biosfera Bosawas, Reserva Natural (RN) Cabo

Viejo-Tela Sulumas, RN Laguna Bismuna-Raya, and RN Laguna Pahara.

Pacific Lowlands. Corresponding to the Salvadoran Coastal Plain of the Chortís

Fore Arc Physiographic Province of Marshall (2007), the Pacific versant lowlands consist of a relatively broad coastal plain extending from the western to the southern limits of the Chortís Highlands province, becoming narrowest around the Gulf of

Fonseca. These lowlands constitute a single ecophysiographic region with a relatively homogenized biota (Wilson & McCranie 1998; Sasa & Bolaños 2004; Townsend &

Wilson 2010b). The Pacific Lowlands are subject to the Lowland Dry climate regime

(Wilson & Meyer 1985), found from sea level to about 600 m elevation, with mean annual precipitation below 2000 mm and mean annual temperature exceeding 24°C.

Important protected areas for Pacific Lowlands ecosystems include: Área Protegida con

Recursos Manejados Barra de Santiago, Parque Privada Walter T. Deininger, Área de

Protección y Restauración (APR) Nancuchiname, Área de Manejo Laguna El Jocotal,

APR Conchagua, Área de Manejo de Habitát de Especie (AMHE) Bahía de Chismuyo,

AMHE Bahía de San Lorenzo, AMHE Las Iguanas-Punta Condega, AMHE Los

Delgaditos, AMHE El Jicarito, AMHE La Berbería, AMHE San Bernardo, and RN Delta de Estero Real.

Salvadoran Cordillera. The Salvadoran Cordillera is not geomorphologically part of the the Chortís Highlands, instead constituting the Chortís Volcanic Front and being dominated by more recent Pliocene and Quaternary volcanic deposits (Marshall 2007). I include the Salvadoran Cordillera for the sake of completeness in this discussion, given its position across the Pacific Lowlands of El Salvador and putative inclusion in the

Eastern Nuclear Central America biogeographic province of Campbell (1999; as expanded upon by Townsend 2006). This west-to-east oriented range is a continuation of the Guatemalan Cordillera, and is made up of several dozen volcanic cones and peaks, including Santa Ana (2,365 m), San Vincente (2,182 m), and San Miguel (2,130 m).

Cordillera Los Marabios. Like the Salvadoran Cordillera, this range not geomorphologically part of the Chortís Highlands, and is represented by a string of northwest-to-southeast oriented Quaternary (and in some cases active) volcanic cones arising from the Pacific Lowlands of northwestern Nicaragua. Volcanoes in this cordillera include Cosigüina (858 m), San Cristóbal (1,745 m), Casita (1,405 m), Telica

(1,060 m), Cerro Negro (726 m), El Hoyo (1,079 m), and Momotombo (1,279 m).

Watersheds. Major river systems on the Caribbean versant of the Chortís Block include: Motagua (485 km), Chamelecón (200 km), Ulúa (300 km), Leán (60 km), Aguán

(225 km), Sico or Tinto or Negro or Grande (215 km), Plátano (85 km), Sikre (70 km),

Patuca (500 km), Warunta (85 km), Mocorón (92 km), Nacunta (65 km), Kruta (125 km),

Coco or Segovia or Wangki (550 km), Wawa (160 km), Kukalaya (140 km), Prinzapolka

(330 km), and Grande de Matagalpa or Awaltara (430 km). Major Pacific versant watersheds include Estero Real (137 km), Negro (85 km), Choluteca (250 km),

Nacaome (90 km), Goascorán (115 km), Lempa (422 km) and Paz (134 km).

Lakes and coastal lagoons. There are very few large inland water bodies in the

Chortís Block, the most notable of which is Lago de Yojoa (700 m elevation) in central

Honduras. The two other large bodies of freshwater, Embalse El Cajón (285 m) in

Honduras and Lago de Apanás (970 m) in Nicaragua, are both reservoirs created by hydroelectric dams. There are a number of large coastal lagoons and lagoon complexes on the Caribbean coast, including Los Micos, Guaymoreto, Ibans, Brus, Tilbalakan,

Laguntara, Warunta, Tansín, Karataska, Kohunta, Bismuna, Pahara, Karatá, Huouhnta, and Laguna de Las Perlas. I consider Lago de Izabal in Guatemala and Lago Xolotlán

(= Lago de Managua) in Nicaragua extralimital to the Chortís Block and do not include them.

Islands. The principle islands associated with the Chortís Highlands include the

Honduran Islas de la Bahía (Utila, Roatán, Guanaja, and Cayos Cochinos), the Cayos

Miskitos of Nicaragua, and Isla El Tigre and other small islands of the Gulf of Fonseca.

Characterizing Ecological Associations: Holdridge Forest Formations

Forest formations described below follow the system developed by Holdridge

(1967), as applied to Honduras in previous works (Meyer & Wilson 1971, 1973; Wilson

& Meyer 1985; Wilson & McCranie 1998; Wilson et al. 2001; McCranie & Wilson 2002).

The widely used Holdridge (1967) system uses climatic, edaphic, and atmospheric conditions to define and determine the distribution of terrestrial ecosystems. The Chortís

Block is typified by a wide range of climatic and elevational regimes, resulting in nine

Holdridge forest formations being recognized within the region. I am using this system,

described below, to partially define Chortís Block ecosystems, supplemented with other published reports, gray literature, and my own observations.

Lowland Moist Forest. Commonly referred to as lowland rainforest, the Lowland

Moist Forest (LMF) formation is defined by a high mean annual temperature (>24°C), high mean annual precipitation (>2000 mm; with no month of the year having precipitation <50 mm), and by being found from sea level to about 600 m elevation. In the Chortís Block, the LMF formation is restricted to the Caribbean versant, the majority of remaining intact forest being found in the vast region of eastern Honduras and

Nicaragua known as La Mosquitia. In addition to lowland rainforest, the pine savannas in La Mosquitia, open woodlands intersected by veins of gallery forest, are found within the Lowland Moist Forest formation. Intact LMF is characterized by having a heterogeneous canopy dominated by evergreen broadleaf trees regularly reaching 30–

40 m in height (Agüdelo C. 1987).

Lowland Dry Forest. The Lowland Dry Forest (LDF) formation includes habitat commonly referred to as scrub forest, and is defined by high mean annual temperature

(>24°C), moderate but seasonally variable annual precipitation (1000–2000 mm; with at least 3–4 months having precipitation <50 mm), and an elevational range of sea level to about 600 m elevation. In the Chortís Block, the LDF formation is found on the Pacific versant and in several interior valleys. Intact LDF is characterized by having a heterogeneous canopy dominated by deciduous trees typically reaching around 25 m in height (Agüdelo C. 1987).

Lowland Arid Forest. The Lowland Arid Forest (LAF) formation, commonly called thorn forest, is defined by high mean annual temperature (>24°C), low annual

precipitation (500–1000 mm; with at least 3–4 months having precipitation <50 mm), and an elevation range of sea level to approximately 600 m elevation. This formation has one of the most limited extents in the Chortís Block, known only from the Middle

Aguán Valley and the Upper Motagua Valley. Intact LAF is characterized by having a low, heterogeneous canopy dominated by deciduous trees typically reaching around 10 m in height, with vegetation dominated by xeric-adapted plants such as cacti (Agüdelo

C. 1987).

Premontane Wet Forest. The Premontane Wet Forest (PWF) formation, sometimes called highland rainforest (McCranie & Wilson 2002), is defined by a moderate mean annual temperature (18°–24°C), high annual precipitation (>2000 mm), and an elevational range of approximately 600 to 1500 m elevation. The PWF formation bridges the LMF with higher elevation montane forests, and thus contains characteristics of both. Intact PWF is characterized by having a closed canopy dominated by evergreen broadleaf trees reaching 25–30 m in height, sometimes reaching 40 m (Agüdelo C. 1987).

Premontane Moist Forest. The Premontane Moist Forest (PMF) formation, commonly referred to as upland pine-oak forest, is defined by a moderate mean annual temperature (18°–24°C), moderate annual precipitation (1000–2000 mm), and an elevational range of about 600 to 1850 m elevation. The PMF formation is relatively widespread in the Chortís Highlands, particularly on interior slopes. Various habitat types are found within the PMF, and are defined below following the classification system of Carr (1950).

Premontane Dry Forest. The Premontane Dry Forest (PDF) formation, which can be termed ―upland scrub forest,‖ is defined by moderate mean annual temperature

(18°–24°C), low annual precipitation (500–1000 mm), and an elevation range of approximately 600 to 1250 m elevation. This habitat is generally limited to the upper periphery of some xeric interior valleys that otherwise support LDF or LAF, two formations with which PDF shares its typical characteristics.

Lower Montane Wet Forest. The Lower Montane Wet Forest (LMWF) formation, is defined by having a low mean annual temperature (12°–18°C), high annual precipitation (>2000 mm), and an elevation range of approximately 1500 to 2700 m elevation (note: habitat typical of this formation can also occur at lower elevations, particularly in the Cordillera Nombre de Dios). In the Chortís Highlands, LMWF is primarily distributed on the Caribbean versant, being replaced by Lower Montane Moist

Forest (below) in the somewhat drier Pacific versant highlands. Intact LMWF is characterized by having a closed canopy dominated by evergreen broadleaf trees reaching 50 m in height (Agüdelo C. 1987).

Lower Montane Moist Forest. The Lower Montane Moist Forest (LMMF) formation, also referred to as cloud forest or montane forest, is defined by having a low mean annual temperature (12°–18°C), moderate annual precipitation (1000–2000 mm), and an elevation range of approximately 1500 to 2700 m elevation. In the Chortís

Highlands, LMMF is distributed in highland areas, typically on the Pacific versant and on the leeward slopes of some of the interior-most Caribbean versant peaks. The LMMF formation contains both pine and broadleaf dominated habitats (better characterized using the classification system of Carr 1950).

Montane Rainforest. The Montane Rainforest (MRF) formation is defined by having a very low mean annual temperature (6°–12°C), high annual precipitation (>2000 mm), and an elevation range of above approximately 2700 m elevation. This formation is the most geographically limited formation in the Chortís Highlands, being restricted to the highest slopes of Cerro Celaque (Honduras; maximum elevation 2,849 m), Cerro

Santa Bárbara (Honduras; maximum elevation 2,744 m), and Cerro El Pital (El

Salvador and Honduras; maximum elevation 2,730 m).

Characterizing Ecological Associations: Updating and Operationalizing the Carr (1950) System for Classifying Honduran Ecosystems

Based on four years of first-hand observation made during his time as a professor at Escuela Agrícola Panamericana (Zamorano), Carr (1950) presented a preliminary characterization of the ecosystems of Honduras. Carr’s initial interest was in

―determining and attempting to define herpetological habitats,‖ which expanded to generalize the habitats so as to ―help the visiting naturalist in his preliminary reconnaissance and shorten his orientation period‖ (Carr, 1950: 569). This work was the first to classify ecological associations in Honduras, and the accuracy of Carr’s observations is such as to allow for his ―outline‖ to be developed into an operational system for classifying animal habitats in Honduras and the greater Chortís Block. As opposed to the Holdridge (1967) system, the Carr (1950) system is descriptive and is designed with the intent of being modified and contextualized by specialists to fit their particular study system or taxonomic group. I developed such an operational system, which I will refer to as the Carr Classification System for Honduran Ecological

Associations or simply the Carr System, based on a synthesis of the available literature

and my own observations from 1999–2011. This system is discussed below, and the habitat names defined here are used throughout this dissertation.

Lowland-associated Habitats

Selva or Lowland Broadleaf Rainforest. Found primarily within the Lowland

Moist Forest (LMF) formation in the Caribbean Lowlands, selva is characterized by a tall, multi-layered, closed-canopy that is dominated by evergreen broadleaf trees, with upper canopy trees normally reaching 30–40 m (Agüdelo C. 1987, Mejía-Ordóñez &

House 2002) and reaching 60 m in some areas (Carr 1950, Wilson & Meyer 1985).

Large expanses of selva are found within Reserva de Hombre y la Biosfera Río Plátano,

PN Patuca, PN Warunta, Reserva Biológica (RB) Rus Rus, Reserva de Biosfera

Tawahka-Asangni, and Reserva de Biosfera Bosawas. Mejía-Ordóñez & House (2002) listed the following tree species as typical of Honduran selva: Brosimun alicastrum,

Bursera simarouba, Calophyllum brasiliense, Cedrela odorata, Coccoloba anisophylla,

Cordia alliodora, Ficus colubrinae, Ficus insipida, Ficus tonduzii, Guarea grandifolia,

Hernandia stenura, Licania platypus, Luehea candida, Nectandra sp., Ocroma pyranidale, Pithecoellobium donnel-smithii, Pouteria campechiana, Pouteria sapota,

Rinorea guatemalensis, Symphonia globulifera, Swietenia macrophilla, Tabebuia chrysantha, Terminalia amazonia, Virola koshnyi and Vochysia hondurensis; the relatively open understory is made up of palms (Acoelorrhaphe wrightii, Chamaedorea spp., Bactris spp., and Geonoma spp.), woody plants (Cespedesia macrophylla, Isertia haenkeana, Piper spp., Cephaelis spp., Psychotria spp.), and herbaceous plants

(Adiantum spp., Polypodium spp., Begonia spp., Selaginella spp., Philodendron spp., and Syngonium spp.).

Figure 2-7. Ecological associations of the Chortís Block I. A) Selva or Lowland Broadleaf Forest; Río Tapalwás, Reserva Biológica Rus Rus, Depto. Gracias a Dios, 180 m; Lowland Moist Forest formation. B) Mosquitia Pine Savanna; between Rus Rus and Awasbila, with the Montañas de Colón in the background, Depto. Gracias a Dios, 200 m; Lowland Moist Forest formation. C) Coastal Scrub; Caribbean coast near Kaukira, Depto. Gracias a Dios; Lowland Moist Forest formation. D) Mangrove Swamp, near mouth of the Río Kruta, Depto. Gracias a Dios; Lowland Moist Forest formation. E) Seasonal Deciduous Forest; near Teocintecito, Depto. Olancho, 690 m; Premontane Dry Forest formation. F) Freshwater Swamp in Seasonal Deciduous Forest; seepage bog in the upper Valle de Agalta, northeast of Saguay, Depto. Olancho, 570 m; Premontane Dry Forest formation. Photos © J.H. Townsend.

Mosquitia Pine Savannas. Although wholly classified as a rainforest area by

Wilson & Townsend (2006) due to being found within the Lowland Moist Forest formation, the Mosquitia of eastern Honduras and Nicaragua supports large areas of

Pinus caribaea savanna that bear a stronger resemblance to subhumid ecosystems than to rainforests (Parsons 1955, Zamora Villalobos 2000, Townsend & Wilson 2010b).

The Mosquitia pine savannas resemble an open woodland dominated by P. caribaea, with a mix of broadleaf trees and shrubs including Agarista mexicana var pinetorum,

Amaioua corymbosa, Arthrostemma ciliatum, Arundinella deppeana, Byrsonima crassifolia, B. verbasifolia, Calea integrifolia, Cecropia peltata, Cephaelis tomentosa,

Chamaecrista nictitans, Clethra calocephala, Clidemia sericea, Cococypsellum sp.,

Cuphea pinetorum, Davilla kunthii, Guazuma ulmifolia, Gnaphalium semiamplexicaule,

Lasianthaeas fruticosa, Lobelia laxiflora, Miconia albicans, M. glaberrima, Myrica cerifera, Psychotria suerrensis, Quercus oleoides, Salvia sp., Vernonia agyropappa,

Vigna vexillata, and Xylopia frutescens, with an open understory of fire-tolerant grasses

(Poaceae) and sedges (Cyperaceae), particularly Paspalum pectinatum, Blechnum serrulatum, Rhynchospora rugosa, Rhynchospora bulbosa, Scleria cyperina, and

Setaria geniculata (Parsons 1955; Zamora Villalobos 2000; Mejía-Ordóñez & House

2002). Herpetofaunal diversity present in the Mosquitia pine savannas is almost completely congruent with that of the subhumid forests of the intermontane valleys

(Townsend & Wilson 2010b), and phylogenetic analyses of subhumid-specialized taxa support conspecific relationships among Pacific and pine savanna populations (Incilius coccifer, Mendelson et al. 2005; Porthidium ophryomegas, Castoe et al. 2005).

Townsend & Wilson (2010b: 702) presented two principal, and as-yet untested,

hypotheses for explaining the apparent continuing connectivity between the Pacific lowlands, subhumid intermontane valleys, and Mosquitia pina savannas:

1) The existence subhumid ―corridor‖ located in Nicaragua between the southern end of the Nuclear Middle American highlands and Lago de Nicaragua, allowing for dispersal of subhumid species from the Pacific Lowlands of Honduras, El Salvador and

Nicaragua to the pine savannas of the Nicaraguan Mosquitia, which is contiguous with pine savannas extending to the northern coast of Honduras.

2) Utilization of open areas along large rivers (Patuca, Coco, and Grande de

Matagalpa) that have their upper reaches in subhumid areas as routes for dispersal.

These rivers originate in subhumid intermontane valleys in the Chortís Highlands, flowing through extensive areas of broadleaf rainforest and on through pine savannas of

La Mosquitia. Secondary connectivity may also occur through coastal strand habitat, which creates a network among individual river drainages at or near their mouths.

Broadleaf Swamp Forest. These swamp forests are found along poorly-drained margins and backwater areas of large rivers, with notable expanses of Broadleaf

Swamp Forest found along the Río Patuca in Reserva de Hombre y la Biosfera Río

Plátano, as well as along Río Kruta and in PN Jeannette Kawas. While riverine swamp forests are restricted to the LMF in the Caribbean Lowlands, there is also broadleaf swamp forest at the northern and southern ends of Lago de Yojoa (700 m elevation), at the lower edge of the Premontane Wet Forest (PWF) formation. Mejía-Ordóñez &

House (2002) listed the following species as being found in permanently inundated broadleaf swamp forest: the trees Crias cauliflora, Pachira aquatica, Pterocarpus hayesii, and Pterocarpus officinalis, and the palms Roystonea dunlapiana, R. regia var

hondurensis, Acoelorrhaphe wrightii, and Desmoncus orthacantus. The following are found in seasonally inundated broadleaf swamp forest: the trees Castilla elastica,

Coccoloba sp., Combretum cacoucia, Symphonia globulifera, and Vochysia ferruginea; and in the flooded forests at the northern end of Lago de Yojoa are found the tree

Erythrina fusca and an understory including Calathea spp., Costus spp., Heliconia spp.,

Hymenocallis litoralis, Maranta spp., Thalia geniculata, Smilax spp., Philodendron spp., and Syngonium spp.

Palm Swamp. A variety of palm-dominated swamp forests occur in coastal areas.

Tique palm swamps are dominated by the tique (Acoelorrhaphe wrightii), or paurotis palm, in association with Annona glabra, Chrysobalanus icaco, Coccoloba uvifera,

Conocarpus erectus, Dalbergia ecastaphylla, Dalbergia monetaria, Davilla kunthii,

Morinda citrifolia, Doliocarpus guianensis, Eugenia aeruginea, Henriettea succosa,

Miconia glaberrima, Miconia albicans, Montrichardia arborescens, Myrmecophila wendlandii, Palicourea tripilla, Symphonia globulifera, Terminalia bucidoides, Thrinax parviflora, Tococa guianensis, Clidemia sericea, Acrocomia mexicana, Bursera simaruba, Casearia sylvestris, Chrysophyllum mexicanum, Cordia alliodora, C. curassavica, Hibiscus tiliaceus, and Ochroma pyramidala (Mejía-Ordóñez & House

2002). Seasonally inundated mixed broadleaf-palm swamps near the Río Kruta and

Cabo Gracias a Dios can have a 40–50 m high canopy dominated by the palm

Roystonea dunlapiana and the understory palm Acoelorrhaphe wrightii, as well as

Mimosa schomburki, Psychotria spp., Alibertia edulis, Spondias mombim, Pachira aquatica, Desmoncus orthacantus, Bactris sp., Ficus sp., Calophyllum brasiliense,

Coccoloba schiedeana., Hirtella racemosa, Xylopia frutescens, Dialium guianensise,

Virola koschnyi, Annona glabra, Grias integrifolia, Dalbergia ecastaphyllum, and Trophis racemosa. In defining the Huiscoyol Swamp as a habitat, Carr (1950: 587) described thick stands of slender Bactris palms (called huiscoyol in Costa Rica) with ―ghastly, glass-hard stem spines‖ and recounts one of his most ―harrowing misadventures‖ being lost in the ―dreary and forbidding environment‖ of a Bactris swamp.

Coastal Scrub. A heterogeneous habitat association found above the beach-line along the Caribbean coast, Islas de la Bahía, Cayos Cochinos, and Cayos Miskitos,

Coastal Scrub includes low coastal strand forest (typical plants include: Cannavalia maritima, C. rosea, Euphorbia buxifolia, Ipomoea pescaprae, Sesuvium portulacastrum,

Sporobolus virginicus, Chrysobalanus icaco, Coccoloba uvifera, Citharexylum caudatum, Hybiscus tiliaceus and Phyllanthus acidus) and its associated grass-covered dune system (typical ground cover includes Andropogon brevifolius, Aristida sp.,

Eleocharis sp., Eragrostis sp., Fimbristylis spadicea, and Paspalum sp.). This association is found in relatively long, undisturbed extensions along essentially the entire Caribbean coast of the Chortís Highlands (Mejía-Ordóñez & House 2002). Carr

(1950) described the somewhat peculiar and specialized coastal scrub habitat found on

Isla El Tigre in the Gulf of Fonseca and a few exposed hillsides facing the gulf as a distinctive habitat association: Sea-Breeze Scrub Forest. While this forest lies within the

Lowland Dry Forest (LDF) formation, the forest receives most of its moisture in the form of occult precipitation brought in on Pacific winds. Common plant species include

Bursera simarouba, Cresentia alata, Enterolobium saman, Spondias purpurea, Prosopis juliflora, Acacia spp., Heamatoxilon brasiletti, and Zizyphum sp.

Mangrove Swamp. These estuarine swamp forests are found along both the

Caribbean and Pacific coasts. Mangrove swamp communities on both coasts typically include the mangroves Avicennia germinans and Rhizophora mangle, with Caribbean swamps including salt-tolerant species such as Laguncularia racemosa, Acrostichum aureum, Cecropia spp., and Coccoloba uvifera, and Pacific mangroves similarly accompanied by Sesuvium portulacastrum, Sporobolus virginicus, Acrostichum aureum,

Cecropia spp., Coccoloba uvifera, Conocarpus erectus, and Laguncularia racemosa

(Mejía-Ordóñez & House 2002).

Habitats Shared Between Lowlands and the Chortís Highlands

Vegas and Gallery Forest. This association is found on rich alluvial soils along stream and river courses, and in areas of low relief around the confluence of two streams (a vega). I tentatively include Carr’s (1950) habitat classes Dry Gullies and

Fence Rows and Hondonadas in this category, recognizing that all of the constituent associations are essentially arteries of mesic habitat, often through comparatively xeric areas. Frequent plants of Caribbean Lowland vegas and gallery forests include Carapa guianensis, Hirtella racemosa, Xylopia frutescens, Dentropanax arboreus, Dialium guianense, Ficus sp., Licania platipus, Ochroma lagopus, Pterocarpus rohrii,

Symphonia globulifera, Vochysia hondurensis, Schizolobium parahybum, Cecropia obtusifolia, Hyeronima alcornoides, Lacmellea panamensis, Prioria copaifera,

Enterolobium schomburki, Apeiba membranaceae, Casearia sylvestris, Cedrela macrophilla, Dendropanax arboreus, Vismia macrophylla, Xylopia frutescens, and

Zuelania guidonia (Mejía-Ordóñez & House 2002).

Carr (1950: 588) observed that these types of alluvial forest appeared to serve as

―mesic highways for the rainforest biota,‖ that might ―afford often contiguous connection

Figure 2-8. Ecological associations of the Chortís Block II. A) Thorn Scrub Forest, upper Valle de Aguán, northwest of Coyoles Central, Depto. Yoro, 250 m; Lowland Arid Forest formation. B) Pantano or Freshwater Marsh, northern end of Lago de Yojoa, with the Montañas de Meámbar in the background, Depto. Cortés, 640 m. C) Infrequently burned Ocotal, near Guaymas, Depto. Francisco Morazán, 1,450 m; Premontane Moist Forest formation. D) Frequently burned Ocotal, Cerro de las Cruces, Depto. Olancho, 1,260 m; Premontane Moist Forest formation. E) Broadleaf Transitional Forest, La Liberación, Refugio de Vida Silvestre Texíguat, Depto. Atlántida, 1,030 m; Premontane Wet Forest formation. F) Mixed Transitional Forest, Montaña de Jacaleapa, Depto. Olancho, 1,120 m; Premontane Wet Forest formation. Photos © J.H. Townsend.

between lower tropical rainforest and upper tropical cloud forest,‖ and therefore ―must be of prime importance in the ecology of the region.‖ In light of over 60 years of advancement in our knowledge and understanding, Carr’s insights are still grounds generating biogeographic hypotheses testable with modern molecular methods.

Seasonal Deciduous Forest. Called Monsoon forest by Carr (1950) and commonly referred to as tropical dry forest, this habitat has a distribution limited to the

Lowland Dry Forest and Premontane Dry Forest formations in the intermontane valleys and on the Pacific Lowlands. Mejía-Ordóñez & House (2002) reported the following species as frequent in Seasonal Deciduous Forest: Enterolobium cyclocarpun, Bursera simarouba, Ceiba pentandra, Cordia alliodora, Lysiloma auritum, Lysiloma seemanii,

Samanea saman, Swetenia macrophylla, Cochlospermum vitifolium, Gyrocarpus americana, Apeiba membranacea, Alvaradoa amorphoides, Calycophylum candidissimum, Tabebuia neochrysanta, Samanea saman, Spondias mombin,

Lonchocarus minimiflorus, and Guazuma ulmifolia.

Thorn Scrub Forest. This low (<4 m in canopy height) habitat is dominated by spine-bearing shrub-like trees such as Pachycereus sp., Hylocereun spp., Mammillaria spp., and Opuntia spp., and dry tolerant shrubs and herbaceous plants like Ananas sp.,

Argyreia especiosa, Cnidoscolus tubulosus, Digitaia insularis, Epidendrum xipheses,

Evolvulus sp. Gonolobus sp., Acacia farnesiana, Albizzia neopoides, Combretum fruticosum, Diphysa ribinoides, Jacquinia macricarpa, Karwinskia calderonii,

Lepidagastris alopecuroidea, Loeselia sp., Melanthera nivea, Thouviinidium decandrum, and Watheria americana (Mejía-Ordóñez & House 2002). Thorn Scrub Forest is

restricted to the Lowland Arid Forest and Lowland Dry Forest formations in the intermontane valleys of the Chortís Highlands.

Pantano or Freshwater Marsh. Freshwater marshes dominated by Typha domingensis, Phragmites australis, and/or Thalia geniculata form an expansive habitat association in some areas, especially the wide alluvial plains of La Mosquitia (Carr

1950; Zamora Villalobos 2000; Mejía-Ordóñez & House 2002). Other grasses present include Andropogon brevifolius, Aristida sp., Eleocharis sp., Eragrostis sp., Fimbristylis spadicea y Paspalum sp. (Mejía-Ordóñez & House 2002). In La Mosquitia, the vast pantanos have a similar character to that of the Florida Everglades, including being dotted with islands of pine (Pinus caribaea) or paurotis palms (Acoelorrhaphe wrightii), with extensive areas of habitat found around the Laguna de Karataska, the Río Kruta, and the lower Río Coco.

Habitats of the Chortís Highlands

Ocotal. The ubiquitous Mesoamerican pine-oak forests are found throughout moderate elevation areas of the serranía and are dominated by ocote pine (Pinus oocarpa), with representation from P. pseudostrobus and other pines at higher elevations. Ocotales are extensively distributed in the serranía in the Premontane Moist

Forest formation, and peripherally in the Premontane Dry Forest and Lower Montane

Moist Forest formations, roughly between 800 and 1,600 m elevation. Ocotal is subject to regular burning by humans, in some areas annually, and, therefore, the biotic composition is limited to species able to tolerate or escape frequent fires. Besides pines, the various species of oaks (Quercus spp.) and the hardwood trees and herbaceous plants Acacia farnesiana, Brahea salvadorensis, Byrsonima crassifolia,

Clethra occidentalis, Myrica cerifera, Enterolobium cyclocarpun, Eritrina berteroana,

Ficus spp. Lysiloma auritum, Mimosa tenuiflora, Psidum guianense, and Tabebuia chrysantha are typically found in ocotales (Mejía-Ordóñez & House 2002). Carr (1950) identified a number of subdivisions of the Ocotal habitat, including shaded ocotal, park ocotal, ocotal steppe, and ocotal-pedregal, based on edaphic and climatic variation as well as burn frequency.

Mixed Transitional Cloud Forest. This transitional forest between ocotales and cloud forest, which Carr (1950) called High Ocotal association, is essentially a humid ocotal with higher concentrations of bromeliads and other epiphytes, as well as a denser and more diverse understory. High ocotales are typically found within the

Premontane Moist Forest and Lower Montane Moist Forest formations, between around

1,000–1,500 m elevation on the Caribbean versant and 1,200–1,800 m on the Pacific versant. Trees of mixed transitional forest include the pines Pinus oocarpa, P. pseudostrobus, and P. tecunumanii, the hardwoods Arbutus xalapensis, Clethra macrophylla, Ficus aurea, Heliocarpus apendiculatus, Oreopanax lachnocephalus,

Oreopanax xalapensis, and Quercus cortesii, and an understory that includes Buddleia americana, Conostegia sp., Miconia sp., Psychotria macrophylla, Vernonia arborescens, Calyptranthes hondurensis, Lobelia laxiflora, Piper launosum, and

Verbesina sp. (Mejía-Ordóñez & House 2002). In addition to High Ocotales, Carr (1950) identified two other types of transitional forest, Pinabetales and ―Diquidambales‖ (=

Liquidambales), which I recognize as distinctive localized associations within Pinus and

Liquidambar Transitional Cloud Forest. Pinabetales are ridge-line groves dominated by the pinabete pine (Pinus pseudostrobus) having epiphyte and understory communities similar to those of the High Ocotal association, but also incorporating representative

from higher elevation hardwood forests. Pinabetales are typically found between 1200–

1600 m within the Premontane Moist Forest and Lower Montane Moist Forest formations. Besides P. pseudostrobus, the pines P. maximinoi and P. tecunumanii can also be present, as might plants otherwise characteristic of both ocotales and mixed cloud forest. Liquidambales have a similar composition to Pinabetales, but are dominated by sweet-gums (Liquidambar styraciflua) and are more typical of leeward slopes than exposed ridges.

Broadleaf Transitional Cloud Forest. Also referred to as Premontane rainforest, this high diversity forest blends diversity from both selva below and the cloud forest above. It is found primarily in the Premontane Wet Forest formation from between 500 m and 1,500 m in elevation, primarily on the windward slopes of mountains along the

Caribbean versant.

Broadleaf Cloud Forest. This is be considered the ―typical‖ cloud forest of the

Lower Montane Wet Forest (LMWF) and Lower Montane Moist Forest (LMMF) formations, characteristically found between around 1,500–2,300 m elevation on the

Caribbean versant and 1,800–2,600 m elevation on the Pacific versant. This forest characteristically has a high diversity of large canopy trees, with canopy heights regularly reaching 40–50 m. Typical vegetation of Broadleaf Cloud Forest in LMWF includes the trees Alnus arguta, A. jorullensis, Cornus sp. Prunus sp., Olmediella betschieriana, Abies guatemalensis, Taxus globosa, Podocarpus oleifolius, Acalypha firmula, Bocona glaucifolia, Cleyera theaeoides, Weinmannia pinnata, W. tuerckheimii,

Daphnopsis strigillosa, Fuchsia paniculata, F. splendens, Hedyosmun mexicanum,

Hoffmannia lineolata, Miconia glaberina, Quercus cortesii, Q. lancifolia, Q. laurina,

Figure 2-9. Ecological associations of the Chortís Highlands. A) Broadleaf Cloud Forest, above Quebrada Varsovia, Montañas de Meámbar, Depto. Comayagua, 1,620 m; Lower Montane Wet Forest formation. B) Broadleaf Cloud Forest, canyon across top of Montañas de Santa Bárbara, Depto. Santa Bárbara, 2,190 m; Lower Montane Wet Forest formation. C) Broadleaf Cloud Forest, Montaña de Botaderos, Depto. Olancho, 1,715 m; Lower Montane Wet Forest formation. D) Mixed Cloud Forest, Quebrada Cataguana, Montañas de Yoro, Depto. Francisco Morazán, 1,860 m; Lower Montane Wet Forest formation. E) Mixed Cloud Forest, Sierra de Omoa, Depto. Cortés, 1,660 m; Lower Montane Wet Forest formation. F) Mixed Cloud Forest, Sierra de Celaque, Depto. Lempira, 2,130 m; Lower Montane Moist Forest formation. G) Río Arcagual in Mixed Cloud Forest, Sierra de Celaque, Depto. Lempira, 2,580 m; Lower Montane Wet Forest formation. H) Hepatic Forest, Sierra de Celaque, Depto. Lempira, 2,780 m; Montane Rainforest formation. I) Heather Wind Scrub, Montaña Macuzal, Depto. Yoro, 1,730 m; Lower Montane Wet Forest formation. Photos © J.H. Townsend.

Rondeletia buddleioides, R. laniflora, Rubus eriocarpus, and Saurauia kegeliana, the herbaceous plants Senecio jurgensenii, Smilax spinosa, Ternstroemia megaloptycha,

Begonia convallariodora, B. fusea, B. oaxacana, Cibotium regale, Deppea grandiflora,

Lobelia nubicola, L. tatea, Parathesis hondurensis, and Peperomia spp., and the ferns

Adiantum piretii, Asplenium harpeodes, A. olivaceum, A. pterocarpus, Blechnum lehmannii, and Elaphoglossum eximium (Mejía-Ordóñez & House 2002). From

Broadleaf Cloud Forest in LMMF on the Pacific versant, Mejía-Ordóñez & House (2002) listed the following species as common in PN La Tigra in the Sierra de Lepaterique:

Mauria sessiflora, Ilex chiapensis, Ilex williamsii, Oreopanax xalapensis, Carpinus caroliniana var tropicalis, Weinmannia balbisina, Hieronyma guatemalensis, Hieronyma poasana, Quercus cortesii, Q. lanciflia, Q. laurrina, Q. bumelioides, Homalium racemosum, Olmediella betschieriana, Calatola laevigata, Nectandra heydeana, Ocotea veraguensis, Phoebe helicterifolia, Magnolia hondurensis, Miconia argentea, Guarea pittieri, Trophis chorizantha, Ardisia paschalis, Chamaedorea pinnatifrons, Clusia rosea,

Lophosoria quadripionnata, and Cyathea mexicana.

Mixed Cloud Forest. Called Bosque mixto by Mejía-Valdivieso (2001), Mixed

Cloud Forest is typically found within the Lower Montane Wet Forest and Lower

Montane Moist Forest formations, between above about 1,500 m elevation on the

Caribbean versant and 1,800 m on the Pacific versant up to around 2,500 m on both versants. Trees of Mixed Cloud Forest include the pines Pinus pseudostrobus, P. tecunumanii, and P. ayacahuite, with a high diversity of oaks (Quercus brumeliodes, Q. cortesii, Q. rugosa, Q. sapotifolia, and Q. acutifolia) and other hardwoods including

Arbutus xalapensis, Bernoulia flamea, Brunellia mexicana, Clusia spp., Cornus discifolia, Cyrilla racemiflora, Dendropanax arboreus, Dendropanax hondurensis,

Hedyosmun mexicanum, Magnolia sp., Liquidambar styraciflua, Myrica cerifera, Ocotea sp., Oreopanax caspitatus, O. xalapensis, O. lachnocephalus, Picramnia teapensis,

Symplocos vernicosa, Toxicodendron striatum, Viasmia baccifera, and Weinmannia pinnata (Mejía-Ordóñez & House 2002). At elevations exceeding 2,300 m, particularly on Cerro Celaque and Montaña de Santa Bárbara, Laurasian trees at the southernmost extent of their range, such as firs (Abies guatemalensis) and yews (Taxus globosa), are found syntopically with Gondwanan trees at their northern distributional limit (e.g.

Podocarpus oleifolius).

Coniferous Cloud Forest. This is a rarely encountered association that is characterized by essentially pure stands of pines (Pinus hartwegii, P. maximinoi, and P. ayacahuite) as well as other conifers (Cupressus lusitanica and Taxus globosa) and is recorded from only a few drier, open ridges above 2,400 m on Cerro Celaque and

Montaña de Santa Bárbara (Mejía-Valdivieso 2001). It is called Bosque de coníferas by

Mejía-Valdivieso (2001), who described these stands as appearing to be subject to natural fires every 3–5 years. I encountered Coniferous Cloud Forest meeting this description between 1900–2100 m elevation on the southeastern side of Montaña de

Yoro, and within this recently burned high pine forest collected herpetofaunal species otherwise considered endemic to nearby Broadleaf Cloud Forest.

Montane Mixed Forest. The habitat of the uppermost reaches of LMWF (>2,600 m) and the Montane Rainforest formation, Bosque mixto montano alto Mejía-Valdivieso

(2001) is dominated by the primitive conifers Abies guatemalensis, Taxus globosa, and

Podocarpus oleifolius, the pines Pinus ayacahuite, P. hartwegii, P. maximinoi, and P. tecunumanii, and the broadleaf trees Alnus arguta, Cornus sp., Prunus sp., Olmediella betschieriana, Oreopanax lempirana, Acalypha firmula, Alnus jorullensis, Bocona glaucifolia, Cleyera theaeoides, Weinmannia pinnata, W. tuerckheimii, Daphnopsis strigillosa, Fuchsia paniculata, F. splendens, Hedyosmun mexicanum, Hoffmannia lineolata, Miconia glaberina, Quercus cortesii, Q. lancifolia, Q. laurina, Rondeletia buddleioides, R. laniflora, Rubus eriocarpus, and Saurauia kegeliana (Mejía-Ordóñez &

House 2002). This habitat is limited to high elevation on Cerro Celaque, Cerro El Pital, and Montaña de Santa Bárbara.

Hepatic Forest. A type of mountain-top dwarf forest, Hepatic, or Mossy, Forest

(Bosque hepático o musgoso of Mejía-Valdivieso 2001), that appears to be restricted to the wet upper slopes of the tallest peaks, including at least Cerro La Picucha (2,100–

2,200 m) in the Sierra de Agalta, Cerro Celaque (2,700 m), and Cerro Jilinco (2,200 m) and Cerro Cusuco (1,990 m) in the Sierra de Omoa (Mejía-Valdivieso 2001, Townsend

& Wilson 2008). The canopy does not exceed 10 m in height and is typically shorter, with trees taking on a twisted appearance. Nearly all available surface area is covered, even layered, in epiphytic plants and fungi, up to 50% of which can be liverworts

(Marchantiophyta; Mejía-Valdivieso 2001). In some cases, this luxuriant epiphytic community creates a living exoskeleton that remains long after the death and decay of the tree within. Hepatic Forest is often found in association with Heather Wind Scrub, and appears to be a transitional habitat between the exposed scrub and the cloud forest below.

Heather Wind Scrub. Found on exposed portions of the highest peaks, this is a wind-swept association that Carr (1950) termed Peña Wind Scrub and is variously referred to as elfin forest or dwarf forest (Townsend & Wilson 2006, 2008), names that reflect the somehow mystic character of these mountain-top ecosystems. Carr’s (1950:

582) own description of this habitat artfully captures this character:

It is a seemingly incongruous combination of dwarfed and twisted microphyllous and sclerophyllous trees and shrubs, Ericaceae, Myrtiaceae, Myrsinaceae, and the like, implying xeric conditions, but with an astounding array of mosses, filmy ferns, selaginellas, and similar delicate hygrophyllous epiphytes. Although at first glance this is an altogether ill-assorted looking flora, the incongruity is only apparent, since each of the two floristic elements is in its own way adapted to withstand drastic reversals in its water economy. On these peñas the wind blows almost constantly, often violently, and while it usually brings in abundance of moisture, it imposes a heavy penalty when the supply fails for even a short period. The wind- pruned trees meet the situation by conservation of their moisture, while their cryptogamic guests yield freely to desiccation, lapsing into dormancy almost on a moment's notice, and without permanent injury.

As indicated by the name, this habitat is dominated by plants in the family Ericaceae with a 2–4 m tall ―canopy,‖ with the thick ―understory‖ being comprised of a bewildering array of bromeliads and other epiphytes. This habitat association is known from Cerro

La Picucha (2,200+ m) in the Sierra de Agalta, Cerro Azul Meámbar (1,950 m+), Cerro

Celaque (various exposed ridges above 2,700 m), and Cerro Jilinco (2,240 m) and

Cerro Cusuco (2,010 m) in the Sierra de Omoa (Hazlett 1980, Mejía-Valdivieso 2001,

Townsend & Wilson 2006). Heather Wind Scrub likely is found in at least small patches on the tops of most exposed peaks above around 1,900 m elevation.

Elfin Forest. This term is herein reserved for use to describe the unique forest association found on Cerro La Picucha in the highest portions of the Sierra de Agalta.

Delineated as Bosque enano by Mejía-Valdivieso (2001), this habitat superficially resembles the Heather Wind Scrub in having a ―canopy‖ not exceeding 3 m in height;

however, in place of Ericaceae, this true dwarf forest is made up of twisted, epiphyte covered, bonsai-like versions of the trees Billia hippocastanum, Podocarpus oleifolius, and Pinus hartwegii, with exposed ground covered in clubmosses (Huperzia and

Lycopodium) and dense patches of ground-dwelling tank bromeliads.

Introduction to Biodiversity of the Chortís Block

Given the wide variation in physiographic and ecological attributes that are evident in the Chortís Block, it comes as little surprise that the region also supports a rich and diverse biota. While the Caribbean and Pacific lowlands have biotas that are typically characterized as being composed of relatively widespread species from both the west and south, the Chortís Highlands and associated piedmont is an area of considerable endemic biodiversity.

While investigation of the endemic fauna of the Chortís Highlands has been limited compared with other areas of the Neotropics, particularly southern Central America, the dedicated work of a small group of specialists led to the documentation of endemism across a variety of taxonomic groups. Botanical data may support this better than other groups, with over 263 described endemic species found in Honduras alone (Nelson-S.

2001, 2008). Areas of elevated plant endemism include mesic highland forests and xeric intermontane valleys of the Chortís Highlands, and, in particular, the piedmont of the Cordillera Nombre de Dios (Nelson-S. 2008). Despite the high degree of localized plant endemism seen across the Chortís Highlands, to date there has been no published analysis of biogeographic patterns among these taxa. Over 166 native freshwater fishes have been documented from Honduras, including three described and at least six undescribed endemic species (Martin 1972, Matamoros et al. 2009,

Matamoros & Schaefer 2010). Among terrestrial vertebrates, birds (one species;

Monroe 1968) and mammals (three species; Goodwin 1942, Reid 2009) stand out in having very few described endemic species; however, at least for mammals, this low level of endemism is almost certainly an artifact of a lack of focused sampling in the molecular age, particularly for small mammals, by systematic mammalogists in the

Chortís Highlands.

It is the herpetofauna, the amphibians and reptiles, which provides the best opportunity for elucidating patterns of evolutionary diversification in the Chortís Block.

Systematic herpetologists have been active in the Chortís Block since at least the early

1900’s, and, in Townsend & Wilson (2010a), I detailed the history of herpetological exploration and research in Honduras. My own work in the Chortís Block began as a student of Dr. Wilson’s in 1999, and had progressed to the point where, when I began this dissertation in 2006, I had identified that the Chortís Block, and particularly the isolated cloud forests of the Chortís Highlands, as being in serious need of intensive study using molecular-based approaches. However, taking an approach that was both broad and comprehensive in terms of taxonomic coverage required that virtually all the areas of the Chortís Block, principally those that were known to support endemic species, would be sampled or resampled. This sampling was carried out from 2006–

2011, and the results are presented herein.

CHAPTER 3 TAXONOMIC DIVERSITY, DISTRIBUTIONAL PATTERNS, AND CONSERVATION STATUS OF THE CHORTÍS BLOCK HERPETOFAUNA

The constituent countries of the Chortís Block (El Salvador, Guatemala, Honduras, and Nicaragua) each have their own rich histories of herpetological investigation

(Mertens 1952, Stuart 1963, Meyer 1969, Villa 1972, Wilson et al. 2010). Most research has expectedly taken a geopolitically-delimited approach that is often arbitrary and is seldom congruent with biogeographic boundaries. While this approach is practical and largely necessary due to a variety of considerations (e.g., visas, research and export permits, logistics), it can also come at the expense of a biogeographically-meaningful approach to research, or of a unified strategy for conservation in transboundary areas.

The Chortís Block is an exemplar of this issue, where national boundaries effectively divide efforts to document and conserve isolated cloud forest areas in five mountain ranges that serve to physically delineate national borders.

In some cases, border regions are avoided due to security-related issues.

Honduran frontier zones have become increasingly favored by transnational narcotics traffickers, particularly over the past decade as the influence of Mexican drug cartels has expanded into Central America (e.g., Archibold & Cave 2011). In an even more extreme case, the highest mountain range in Nicaragua (Sierra de Dipilto, maximum elevation 2,107 m), which also forms the border with Honduras, was considered a strategic vantage point that was heavily contested, and subsequently landmined, during the Contra-Sandinista War of the 1980’s (United Nations Mine Action Service 1998).

Despite the apparent biogeographic importance of this mountain range and its potential for supporting endemic species, little to no biological inventory work has been carried out in the Sierra de Dipilto. Beyond this mountain range, a large area of the Chortís

Highlands in northern Nicaragua has also been heavily undersampled as a result of being the principal zone of conflict in the Contra-Sandinista War, and has only recently begun to be sampled in a concerted fashion (Sunyer et al. 2009, Travers et al. 2011).

Political boundaries have placed the majority of the endemic-rich Chortís

Highlands within the borders of one country, Honduras, with significant extensions into three neighboring countries, most notably Nicaragua. Subsequently, no concerted attempt has been made to date to present a unified understanding of the biogeographic province’s patterns of biological diversification.

A consequence of research constrained by political boundaries is that systematic cataloguing of regional diversity has been handicapped. As a result, a number of endemic species are known from localities in one country but unconfirmed as occurring across the border in ecophysiographically contiguous areas, as in the cases, for example, of Cryptotriton monzoni (endemic to the Sierra de Espíritu Santo in

Guatemala) and Anolis johnmeyeri (endemic to the Sierra de Espíritu Santo and Sierra de Omoa in Honduras). In at least one case, it has been suggested that two species of

Cryptotriton described from opposite sides of the Sierra de Omoa (called the Sierra de

Caral in Guatemala), C. nasalis and C. wakei, actually represent the same species

(McCranie & Wilson 2002).

Minimizing the constraints of these largely political and logistical issues is critical to accurately analyzing and interpreting regional patterns of biogeography, endemism, and conservation priority. My goal in this chapter is to synthesize the available data for the

Chortís Block herpetofauna, drawing first from a considerable regional literature base.

This available data are then augmented by the results of sampling efforts from 2006–

2011 in Honduras and Nicaragua, and presented to provide the basis from which to further investigate herpetofaunal diversity and assess research and conservation priorities moving forward.

Methods and Materials

This section serves not only to define the methodologies used in Chapter 2, but also to provide some of the broader methods and definitions used throughout this dissertation.

Field Sampling

Between June 2006 and April 2011, I made 17 fieldtrips totaling over 2,577 person-hours of effort (over 12,560 person-hours were logged by expedition participants) over the course of 275 field-days in Honduras and Nicaragua, sampling over 60 localities in the Chortís Block (Table 3-1; Figure 3-1). Tissue samples were taken from freshly euthanized vouchers and stored in SED buffer (250 mM EDTA/20%

DMSO/saturated NaCl; Seutin et al. 1991; Williams 1997). Voucher specimens were preserved in 10% formalin solution and later transferred to 70% ethanol for permanent storage. Vouchers were deposited in the Florida Museum of Natural History, University of Florida (UF), the Museum of Vertebrate Zoology, University of California, Berkeley

(MVZ), and the National Museum of Natural History, Smithsonian Institution (USNM).

Taxonomic Scope and Standards

I recognize a Chortís Block herpetofauna inclusive of species that occur south or east of the Río Motagua and north of a latitudinal line across the northernmost edge of

Lago Xolotlán (= Lago de Managua). I have included taxa from the Islas de la Bahía and

Cayos Miskitos, as they are continental islands of Chortís Block origin, but not the

Figure 3-1. Map showing sampling localities in the Chortís Block. Red crosses = localities sampled 2006–2011, blue crosses = localities sampled 1999–2005, green diamonds = localities sampled by collaborators for inclusion here.

offshore islands of Belize, the smaller cays far offshore from eastern Honduras and

Nicaragua, or the Islas del Cisne, as they neither geological nor biogeographically related directly to the Chortís Block. I have excluded marine taxa (sea turtle families

Cheloniidae and Dermochelyidae and the sea snake Pelamis platura) and introduced species, as well as taxa known only from the Salvadoran Cordillera or southwestern

Salvadoran coastal plain, as they are not considered herein to be part of the Chortís

Block.

The taxonomic nomenclature used in this dissertation generally follows that used by Wilson & Johnson (2010). A number of additions and changes have occurred since that publication that pertain directly to composition of the Chortís Block herpetofauna, including description or resurrection of the following taxa: Bolitoglossa nympha

(Campbell et al. 2010), Cerrophidion sp. (Jadin et al., submitted), Ctenosaura praeocularis (Hasbún & Köhler 2009), Anolis beckeri (Köhler 2010), A. unilobatus

(Köhler & Vesely 2010), Epictia magnamaculata (Adalsteinnsson et al. 2009),

Dendrophidion clarkii (McCranie 2011a), Leptodeira rhombifera (Daza et al. 2009),

Mastigodryas alternatus (McCranie 2011a), Omoadiphas cannula (McCranie & Cruz-

Díaz 2010), Tantilla sp. (Townsend et al., submitted), and Tantilla psittaca (McCranie

2011b). In addition, four new species of salamanders have recently been described, in part based on results of this dissertation reported in later chapters, and are included in this chapter for the sake of completeness: Nototriton picucha (Townsend et al. 2011a),

Oedipina koehleri (Sunyer et al. 2011), O. nica (Sunyer et al. 2010), and O. petiola

(McCranie & Townsend 2011). I follow Myers (2011) the species of the Rhadinaea godmani group as representing a separate genus, Rhadinella.

Evaluating Conservation Status

I used three different measures to assess the conservation status of the herpetofauna of the Chortís Block: IUCN Red List categorization, Environmental

Vulnerability Scores (Wilson & McCranie 2003), and Conservation Status Scores

(Wilson & Townsend 2010).

IUCN Red List categorizations were taken from one of three sources: the IUCN

Red List of Threatened Species (v. 2011.1; www.iucnredlist.org) for amphibians, marine turtles, crocodilians, and Ctenosaura; Townsend & Wilson (2010) for Honduran reptiles, and Sunyer & Köhler (2010) for Nicaraguan reptiles. Species not previously evaluated by the IUCN or other authors were assessed using the standard criteria of the IUCN

(2001).

Environmental Vulnerability Scores (EVS) are primarily from Townsend & Wilson

(2010a). Species not assessed in that study were calculated using the methodology developed and refined by Wilson & McCranie (1992, 2003, 2004a), which is calculated by taking the total of three rankings: 1) extent of geographic range, 2a) degree of specialization of reproductive mode for amphibians or 2b) the degree of persecution by humans for reptiles, and 3) extent of ecological distribution in Honduras; EVS scores from 10–13 indicate medium vulnerability, and scores from 14–19 are high vulnerability

(Wilson & McCranie 2003).

Conservation Status Scores (CSS) were developed by Wilson & Townsend (2010) to provide a simple measure for assessing conservation status of amphibians and reptiles across Mesoamerica. The CSS represents sum of individual scores for 1) numbers of countries, 2) physiographic regions, and 3) vegetation zones occupied by a given species of amphibian or reptile. The country score ranges from 1 to 8 (the number

of countries of Central American plus México), the physiographic region score ranges from 1 to 21, and the vegetation zone score from 1 to 15 (Wilson & Townsend 2010).

Given this, the CSS can range from 3 (the most restricted endemic species, inhabiting a single vegetative zone in a single physiographic region in a single country) to the theoretical maximum of 44 (for a species found literally everywhere in México and

Central America).

Results

Composition of the Herpetofauna

The native, non-marine herpetofauna of the Chortís Block is comprised of 382 species in 134 genera and 41 families, including 145 species of amphibians and 237 reptiles (Tables 3-2, 3-3). The Chortís Block herpetofauna comprises about approximately 2.3% of global herpetofaunal species diversity (16,301 species as of 6

October 2011; Table 3-3; AmphibiaWeb 2011, Uetz et al. 2011). This includes 2.1% of global amphibians, with 1% of caecilians, 7% of salamanders, and 1.6% of anurans, along with 2.5% of global reptiles, with 3.1% of turtles, 8.3% of crocodilians, and 2.5% of squamates (2.3% of lizards, and 4.2% of snakes). There are 41 families of amphibians and reptiles in the Chortís Block (Table 3-3), comprising 30.4% of the families of the world and 65.1% of the families of Mesoamerica (Table 3-3).

Within Mesoamerica (considered in this discussion to include México and Central

America), the Chortís Block herpetofauna contains approximately 20.3% of the regional herpetofauna (Wilson & Johnson 2010). This includes 19.8% of Mesoamerican amphibian species, with 12.5% of caecilians, 17.8% of salamanders, and 21.1% of amphibians, and 20.6% of reptiles, including 18.2% of turtles, two of three crocodilian species, and 20.6% of squamates (Table 3-3).

Patterns of Distribution and Endemism within the Chortís Block

Within the Chortís Block, species were considered to occur within one of eight physiographic regions: the Caribbean Lowlands, Caribbean Versant Intermontane

Valleys, Northern Cordillera of the Chortís Highlands, Central Cordillera of the Chortís

Highlands, Southern Cordillera of the Chortís Highlands, Pacific Lowlands, Pacific

Versant Intermontane Valleys, and Islas de la Bahía (Tables 3-2, 3-4). The Caribbean

Lowlands holds the highest diversity, with 188 species, with the Northern Cordillera being the most diverse portion of the Chortís Highlands with 163 species (Table 3-4).

The majority of amphibians (51%) that occur in the Chortís Block are endemic, i.e., restricted in distribution to within the Chortís Block, with a lesser share of reptiles (24%) being endemic (Table 3-5). The salamanders have a particularly high degree of endemism, with 37 of 43 species (86%) being Chortís Block endemics, and 36 of those

37 being endemic to the Chortís Highlands province (Tables 3-2, 3-4). A large share of named anuran species (37%) is also endemic (Table 3-5).

Distribution of Chortís Highland Endemics

The Northern Cordillera is the most endemic-rich portion of the Chortís Highlands with 72 species, versus 42 and 35 in the Central and Southern Cordilleras, respectively

(Table 3-6). The Sierra de Omoa is clearly the mountain range supporting the highest number of Chortís Block endemic species (35), and each of the four mountain ranges of the Northern Cordillera support more endemic species than any other mountain range in the Chortís Highlands (Table 3-6). In the Central Cordillera, the Sierra de Sulaco and

Cordillera de La Flor-La Muralla tie for the lead with 13 endemic species, with the Sierra de Agalta having 11 (Table 3-6). In the Southern Cordillera, the Sierra del Merendón

supports the most endemic species with 12, followed by the Cordillera Dariense with 9

(Table 3-6).

Conservation Status

A startling portion of the Chortís Block herpetofauna is threatened at the local, regional, and global levels (Tables 3-2, 3-5). At least 41% of the entire herpetofauna is at endangered at a globally significant level (listed in one of the top three threat categories on the IUCN Red List: Critically Endangered, Endangered, or Vulnerable), including an alarming 74% of salamanders (Table 3-5). Although not surprising given their typically narrow geographic distributions, 96% of species endemic to the Chortís

Block are also listed in the top three IUCN Red List categories, with 48% of endemic species listed in the highest risk category: Critically Endangered (Table 3-5). Regionally, the Conservation Status Score (CSS) was used to gauge the relative degree of threat facing members of the Chortís Block herpetofauna. Two hundred and ten species (55% of Chortís Block herpetofauna) have CSS between 3 and 11, placing them in the category of ―Very High conservation significance,‖ the highest risk category employed at the Mesoamerican scale by Wilson & Townsend (2010). Within the Chortís Block,

Environmental Vulnerability Score (EVS) was used to provide a local-scale measure of the relative degree of threat facing herpetofaunal species. Results of the EVS indicated an elevated degree of susceptibility to degradation for the Chortís Block herpetofauna, with 146 species (38%) having EVS from 14 and 19, placing them in the ―high vulnerability‖ category (Townsend & Wilson 2010a).

Sampling Results by Locality

Below is a summary of results, presented by locality, from herpetofaunal sampling in the Chortís Block from June 2006 to April 2011. For each general locality, typically a

protected area, the location and geographical extent are provided, along with details of the legal protection (if any), a list of amphibian and reptile species recorded, the sites and dates visited, and a summary of findings.

Parque Nacional Celaque

Location and Extent: West-central Departamento de Lempira and eastern

Departamento de Ocotepeque, Honduras; 26,267 ha in extent, highest elevation 2,849 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Bolitoglossa celaque, Plectrohyla psiloderma.

Site visit summary: Campamento Don Tomás (2,050 m), vicinity of Campamento

Naranjo (2,560-2,850 m), Departamento de Lempira, core zone of PN Celaque. 21–28

June 2008.

Comments: PN Celaque has a large core zone, and the physiographic structure of the mountain benefits conservation efforts. Steep sides to this mesa-like mountain protect a large, relatively flat area above 2,500 m elevation that is almost completely intact. Upon this mesa originates the primary water source for the nearby city of

Gracias, and protection of this and other important watersheds provides added benefit to the people inhabiting the park’s surroundings. The endemic salamander Bolitoglossa celaque was found to be locally abundant around the Río Arcagual. Only one of four species of Plectrohyla reported from Celaque (McCranie & Wilson 2002) was found during my visit (P. psiloderma), and additional targeted searches for the other three species (P. guatemalensis, P. hartwegi, and P. matudai) should be carried out

immediately in an attempt to determine their status. Additional fieldwork in Celaque would almost certainly produce additional new herpetofaunal species.

Parque Nacional Cerro Azul Copán

Location and Extent: Western Departamento de Copán, Honduras, along the border with Guatemala; 11,766 ha, highest elevation 2,285 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa nympha, B. dofleini; Anura:

Dendropsophus microcephalus, Incilius valliceps, Lithobates brownorum, Lithobates maculatus, Ptychohyla hypomykter, Smilisca baudinii; Squamata: Anolis ocelloscapularis, Lepidophyma flavimaculatum, Sceloporus schmidti, Adelphicos quadrivirgatum, Mastigodryas dorsalis, Ninia diademata, N. sebae, Typhlops stadelmani.

Site visit summary: Quebrada Grande, Departamento de Copán, 1,280–1,400 m elevation, in the buffer zone of PN Cerro Azul, on the south-southeastern side of Cerro

Azul. 26–30 July 2008.

Summary findings: PN Cerro Azul is both a reserve with high importance with regard to its amphibian diversity and one at high risk due to a lack of management and increasing human population and forest clearing. The highest portion of the southeastern flank of Cerro Azul appears steep enough to deter campesinos from clearing most of the remaining forest. Unfortunately, there appear to be no streams flowing through this portion of forest due to the extreme topographical grade. Riparian areas in the vicinity of Quebrada Grande have been completely cleared and converted to livestock pastures. Below this area, a larger creek (Quebrada Cañon Oscuro) forms

from smaller streams running through pastures and rapidly descends into a deep canyon, inside which remains some riparian forest. The water source for Quebrada

Grande is a spring just above town, with a very small (<1 ha) patch of forest remaining in the spring’s vicinity. The hillsides above and surrounding the spring have been converted to agriculture. Additional work is urgently needed to determine the status of remaining forest both at higher elevations and of the side of Cerro Azul opposite to

Quebrada Grande.

Parque Nacional Cerro Azul Meámbar

Location and Extent: East of Lago de Yojoa in extreme southern Departamento de Cortés and northern Departamento de Comayagua; 17,872 ha, highest elevation

2,090 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa mexicana, B. oresbia, Nototriton limnospectator; Anura: Craugastor laevissimus, C. laticeps, Incilius ibarrai, I. porteri, I. valliceps, Lithobates maculatus, Ptychohyla hypomykter, Smilisca baudinii, Squamata:

Anolis lemurinus, A. cf. limifrons, Sceloporus variabilis, Sphaerodactylus millepunctatus,

Thecadactylus rapicauda, Atropoides mexicanus, Bothriechis schlegelii, Imantodes cenchoa,

Site visit summary: Los Pinos Visitor’s Center, Departamento de Cortés, 700–

1,400 m elevation, 3–5 December 2006, 14 July 2007, 4–5 April 2008, 14–17 April

2008, 4–12 June 2008, 25–27 November 2009; Cerro Azul, 800–1,740 m elevation, 18

April 2008, 5–12 July 2008; San José de los Planes, 930-1,290 m elevation, 1–3 June

2008.

Figure 3-2. Sampling in the Chortís Block I. A) Campamento Don Tomás, 2,080 m, Parque Nacional (PN) Celaque (A–D, June 2008). B) The author in Montane Hepatic Forest, 2,770 m, PN Celaque. C) Campamento Los Naranjos, 2,570 m, PN Celaque. D) I. Luque searches for Bolitoglossa celaque in ground- dwelling bromeliads; Montane Mixed Forest, 2,630 m, PN Celaque. E) Community of Quebrada Grande, 1,320 m, with PN Cerro Azul (July 2008). F) I. Luque preparing to descend over 350 m to Quebrada Grande, 1,690 m (July 2008). G) Hiking along Quebrada Varsovia, 1,670 m, PN Cerro Azul Meámbar (August 2008). Figure 3-2B © I. Luque, all other photos © J.H. Townsend.

Summary findings: Highland forest areas of this park previously had not been sampled, and my first two expeditions to cloud forest elevations in July 2008 yielded significant results, the discovery of Bolitoglossa oresbia, considered the most Critically

Endangered salamander in Honduras whose previous known distribution consisted of a

1 ha patch of forest on top of the otherwise-deforested Cerro El Zarciadero, and

Nototriton limnospectator, previously known only to occur across Lago de Yojoa in

Parque Nacional Montaña de Santa Barbara (Townsend et al. 2011b).

Parque Nacional Cusuco

Location and Extent: Northwestern Depto. Cortés and adjacent northern Depto.

Santa Bárbara, Honduras, near the Guatemalan border; 17,704 ha, highest elevation

2,242 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa conanti, B. diaphora; Anura:

Duellmanohyla soralia, Plectrohyla dasypus, P. exquisita, Ptychohyla hypomykter;

Squamata: Anolis amplisquamosus, A. cusuco, A. johnmeyeri, Mesaspis moreletii,

Sceloporus schmidti, Cerrophidion sp., Ninia diademata, N. espinali, N. sebae.

Site visit summary: Vicinity of Centro de Visitantes (1,520–1,600 m), buffer zone of PN Cusuco; Quebrada Cantiles (1,820 m), core zone of PN Cusuco, 11–17 March

2006, 30 August–3 September 2008, 28–30 November 2009.

Summary findings: Details of herpetofaunal work through 2006 are found in

Townsend & Wilson (2008); work in 2008 and 2009 did not appreciably add to this knowledge base.

Parque Nacional La Tigra

Location and Extent: South-central Departamento de Francisco Morazán; 24,340 ha, highest elevation 2,290 m.

Status: National Park (legally declared in 1980; Decreto 976-80); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Anura: Incilius porteri, Lithobates maculatus; Squamata:

Anolis laeviventris, A. tropidonotus.

Site visit summary: El Rosario Visitor’s Center (1,500–1,700 m), core zone of PN

La Tigra, 6–8 December 2006.

Comments: The streamside frog Craugastor emleni, has been reported to have undergone extreme declines, and was feared extinct by McCranie & Wilson (2002). This species was recently rediscovered from a small stream at 1,600 m elevation in Parque

Nacional La Tigra (McCranie et al. 2010), and additional fieldwork should be undertaken at this easily accessible cloud forest to determine the extent of the distribution of C. emleni.

Parque Nacional Montaña de Botaderos

Location and Extent: Northern Departamento de Olancho, and peripherally in adjacent Departamento de Colón; 64,227 ha, maximum elevation 1,724 m.

Status: National Park (legally declared in 2011; Acuedro 002–2011).

Herpetofaunal results: Caudata: Nototriton sp.; Anura: Craugastor lauraster, C. noblei, Craugastor pechorum, Lithobates maculatus, Pristimantis ridens, Ptychohyla hypomykter; Squamata: Anolis capito, A. tropidonotus, Imantodes cenchoa, Ninia maculata, Stenorrhina degenhardtii.

Figure 3-3. Sampling in the Chortís Block II. A) Visitor’s Center at Parque Nacional (PN) Cusuco, 1,550 m elevation; the day before the post-coup presidential election, with red-and-white Partido Liberal party members meeting in seclusion (November 2009). B) Looking down from PN La Tigra, with El Rosario in the foreground and the Choluteca Valley beyond (B–C, December 2006). C) S. Townsend at Quebrada de la Cascada, 1,850 m, PN La Tigra. D) R. Ulloa on trail leading to cloud forest, 1,640 m, PN Montaña de Botaderos (D–G, April 2011, Explorer’s Club Flag #93). E) Broadleaf Cloud Forest, 1,720 m, PN Montaña de Botaderos. F) Our supplies and team near Alao, 1,100 m, on expedition into PN Montaña de Botaderos. G) M. Medina (left) recovering from nighttime sampling efforts while O. Reyes (right) prepares botanical samples at our base camp, 1,300 m, PN Montaña de Botaderos. Photos © J.H. Townsend.

Site visit summary: Cerro de la Cruces and vicinity of Cerro ―Ulloa‖, the highest ridge in the range, 1,160–1,730 m elevation, 16–19 April 2011.

Summary findings: In April 2011, I participated in an Explorer’s Club expedition

(Flag #93) to the Municipalidad de Gualaco in Departamento de Olancho, which included the first ever biological expedition to the highest portion of this newly declared park (approved by the Honduran Congress the same week we were there).

Parque Nacional Montaña de Comayagua

Location and Extent: Southeastern Departamento de Comayagua and adjacent

Departamento de Francisco Morazán; 29,767 ha, maximum elevation 2,410 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa mexicana; Anura: Craugastor laevissimus, Ptychohyla hypomykter, Squamata: Anolis sminthus, A. tropidonotus,

Sceloporus malachiticus.

Site visit summary: La Oki, Departamento de Comayagua, 1,680–2,040 m elevation, edge of core zone for PN Montaña de Comayagua, 21–24 January 2008; Río

Negro, Departamento de Comayagua, 1,100–1,560 m elevation, buffer zone of PN

Montaña de Comayagua, 19–21 April 2008, 15–20 May 2008, 15–22 July 2008.

Comments: Accessing areas of cloud forest above 1,500 m elevation has proven very challenging; intact cloud forest occurs as low as 1,200 m elevation along large streams and rivers. The community of Río Negro is a model of community based conservation and sustainable development, and is a prime candidate to receive funding to support these areas. Areas of cloud forest above 2,000 m elevation need to be

sampled for the presence of salamanders, as no salamanders have been found above

1,300 m elevation despite the presence of endemic highland species in surrounding mountain ranges.

Parque Nacional Montaña de Santa Bárbara

Location and Extent: Southeastern Departamento de Santa Bárbara; 13,202 ha, maximum elevation 2,744 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa mexicana, Dendrotriton sanctibarbarus; Anura: Craugastor laevissimus, C. laticeps, Lithobates maculatus,

Ptychohyla hypomykter, Squamata: Anolis rubribarbaris, Sceloporus malachiticus,

Typhlops tycherus.

Site visit summary: El Cedral, in the buffer zone of Parque Nacional Montaña de

Santa Barbara on the western slope of the mountain, 1,600–1,720 m elevation, 28–29

January 2008; Las Quebradas, in the buffer zone of Parque Nacional Montaña de Santa

Bárbara on the western slope of the mountain, 1,400–1,450 m elevation, 6–7 June

2010; core zone of Parque Nacional Montaña de Santa Barbara, west-to-east canyon across the crest of the mountain, 1,450–2,180 m elevation, 6–9 November 2010.

Summary findings: The physiography of PN Montaña de Santa Bárbara appears to be helping ensure the long-term persistence of intact forest inside the park’s core zone. The steep side and jagged karst composition deter deforestation above 2,000 m, and both endemic species known from within the park appear to have stable populations. My work in PN Montaña de Santa Bárbara resulted in two notable discoveries: a remarkable new large species blindsnake (Typhlops tycherus; Townsend

Figure 3-4. Sampling in the Chortís Block III. A) I. Luque and M. Medina in Rio Negro, hiking towards Quebrada El Gavilán, 1,200 m, Parque Nacional (PN) Montaña de Comayagua (April 2008). B) Quebrada El Gavilán, 1,260 m, PN Montaña de Comayagua (May 2008). C) J. Austin (foreground) and field team ascending near-vertical trail above El Playón, 1,480 m, en route to PN Montaña de Santa Bárbara (C–E, November 2010). D) J. Austin at basecamp in canyon across top of PN Montaña de Santa Bárbara, 2,200 m. E) L. Herrera setting infrared camera trap for mammal survey in PN Montaña de Santa Bárbara, 2,190 m. F) G) Mesic ravine through Coniferous Cloud Forest, 1,930 m, PN Montaña de Yoro (September 2008). Photos © J.H. Townsend.

et al. 2008a), and the documentation of variation and distributional records of Anolis rubribarbaris, a lizard species previously known from only a single poorly preserved specimen (Townsend et al. 2008b). An expedition into a canyon across the top of Cerro

Santa Bárbara in November 2010 encountered the bromeliad salamander Dendrotriton sanctibarbarus to be locally abundant, particularly around large, bromeliad-laden branches that had come to rest on the forest floor.

Parque Nacional Montaña de Yoro

Location and Extent: Northern Departamento de Francisco Morazán and adjacent southern Departamento de Yoro; 11,766 ha, highest elevation 2,285 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa cataguana, Nototriton lignicola,

Oedipina kasios, Anura: Incilius porteri, I. valliceps, Lithobates maculatus, Ninia sebae,

Plectrohyla guatemalensis, Squamata: Anolis morazani, A. tropidonotus, A. yoroensis,

Mesaspis moreletii, Sceloporus malachiticus, Cerrophidion sp., Rhadinella godmani.

Site visit summary: Río Maralito, outside Marale (610–630 m), seat of municipality; Los Planes (1,450–1,500 m), buffer zone community; Cataguana (1,780–

2,020 m), northwestern core zone of PN Montaña de Yoro, 8–14 June 2006, 8–15

March 2007; Montaña de la Sierra (1,920 m), southeastern core zone of PN Montaña de Yoro, 21–25 September 2008.

Comments: Results from this protected area are discussed in detail below.

Parque Nacional Pico Bonito

Location and Extent: Eastern Departamento de Atlántida and adjacent

Departamento de Yoro; 107,107 ha, maximum elevation 2,480 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Anura: Craugastor aurilegulus, C. noblei, Duellmanohyla salvavida, Pristimantis ridens, Ptychohyla spinipollex; Squamata: Ameiva festiva, Anolis lemurinus, A. tropidonotus, A. zeus, Basiliscus vittatus, Corytophanes cristatus, Incilius leucomyos, Mabuya unimarginata, Plestiodon sumichrasti, Sphenomorphus cherriei,

Thecadactylus rapicauda, Leptophis ahaetulla, Oxybelis aeneus.

Site visit summary: Cangrejal, vicinity of the visitor’s center, 200 m elevation, 2

December 2009; Quebrada de Oro, 930–1,380 m elevation, 21–28 May 2010; CURLA

Station, 150 m elevation, 10–12 August 2010; Pico Bonito Lodge, 120 m, 7 April 2011.

Comments: My fieldwork in Parque Nacional Pico Bonito has thus far been limited to the lowlands, supported by two fieldtrips undertaken by César Cerrato (UNAH) in

2010.

Parque Nacional Pico Pijol

Location and Extent: Southwestern Departamento de Yoro; 11,669 ha, maximum elevation 2,282 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetofaunal results: Anura: Craugastor laevissimus, C. rostralis, Lithobates maculatus, Ptychohyla hypomykter, Smilisca baudinii; Squamata: Anolis yoroensis,

Oxybelis fulgidus.

Site visit summary: Road above El Porvenir de Morazán, northeastern buffer zone of PN Pico Pijol in the vicinity of Cerro Las Pajarillos, 1,380–1,520 m elevation,

23–25 July 2008; 17–20 September 2008.

Summary findings: Extensive shade-coffee production dominates the northern side of the buffer zone of PN Pico Pijol. A single frog, Craugastor laevissimus, was collected along a path through converted cloud forest; however, a search of a medium sized upper tributary of Quebrada Las Payas surrounded by moderately disturbed forest failed to find any of the normally stream-side Craugastor. No individuals of Bolitoglossa porrasorum were collected, despite targeted searching for this species. A single

Ptychohyla hypomykter and two groups of tadpoles were the only amphibians observed along the stream.

Parque Nacional Sierra de Agalta

Location and Extent: East-central Departamento de Olancho; 51,793 ha, maximum elevation 2,354 m.

Status: National Park (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetological results: Caudata: Bolitoglossa mexicana, B. longissima,

Nototriton sp.; Squamata: Anolis cf. sminthus, Atropoides indomitus, Cerrophidion sp.

Site visit summary: Trail to, and vicinity of, Cerro La Picucha, 615–2,230 m elevation.

Summary findings: Fieldwork led by Melissa Medina-Flores (Universidad

Nacional Autónoma de Honduras) resulting in collection of the first record of Atropoides indomitus from the Sierra de Agalta and two specimens of an unknown species of

Nototriton (chapters 4 and 5; Townsend et al. 2011a).

Refugio de Vida Silvestre Texiguat

Location and Extent: Southwestern Departamento de Atlántida and adjacent northwestern Departamento de Yoro; 15,736 ha, maximum elevation 2,208 m.

Status: Wildlife Refuge (legally declared in 1987; Decreto 87-87); Conservation

International Key Biodiversity Area.

Herpetological results: Caudata: Bolitoglossa nympha, B. cf. porrasorum,

Nototriton cf. barbouri, N. tomamorum, Oedipina gephyra; Anura: Craugastor aurilegulus, Craugastor cf. rostralis, Duellmanohyla salvavida, Hyalinobatrachium fleischmanni, Incilius leucomyos, I. valliceps, Leptodactylus fragilis, Plectrohyla chrysopleura, Ptychohyla spinipollex, Smilisca baudinii, Teratohyla pulverata;

Squamata: Anolis beckeri, A. kreutzi, A. loveridgei, A. yoroensis, A. zeus, Corytophanes cristatus, Laemanctus longipes, Lepidophyma flavimaculatum, Sceloporus malachiticus,

Sphenomorphus cherriei, Adelphicos quadrivirgatum, Atropoides mexicanus,

Bothriechis marchi, B. schlegelii, Bothrops asper. Dendrophidion percarinatum,

Drymobius chloroticus, Geophis damiani, Hydromorphus concolor, Imantodes cenchoa,

Leptodeira septentrionalis, Micrurus nigrocinctus, Ninia pavimentata, Ninia sebae,

Pliocercus elapoides, Scaphiodontophis annulatus, Sibon dimidiatus, Sibon nebulatus,

Stenorrhina degenhardtii, Tantilla sp., Tropidodipsas sartorii.

Site visit summary: 2.5 km NNE of La Fortuna, buffer zone of RVS Texiguat,

1,500–1,890 m elevation, 8–11 April 2008; 10–21 June 2010, 25 July–1 August 2010.

Comments: RVS Texiguat might be simultaneously the single protected area of the most biodiversity significance and facing the strongest threat to its long-term survival in the Honduran protected areas system. The reserve is discussed in detail later in this chapter.

Figure 3-5. Sampling in the Chortís Block IV. A) Cerro de Pajarillos, 1,460 m, Parque Nacional (PN) Pico Pijol (July 2008). B) J. Slapcinsky (left) and J. Butler (right) at basecamp, La Fortuna, 1,650 m, Refugio de Vida Silvestre Texíguat (B–C, April 2008). C) Type locality of Isthmohyla insolita and Nototriton tomamorum, La Fortuna, 1,550 m, Refugio de Vida Silvestre Texíguat. D) Cattle pond with explosive breeding aggregation of Exerodonta catracha and Hypopachus barberi, dark masses in water are H. barberi eggs; 2,210 m, Reserva Biológica (RB) Guajiquiro (D–E, May 2008). E) Male Exerodonta catracha guarding egg clutches in pond from Figure 3-5D. F) Degraded Mixed Cloud Forest, 2,100 m, RB Güisayote; Anolis heteropholidotus and Mesaspis moreletii were abundant in the pasture in the foreground (F–G, June 2008). G) Camping under the communications tower at the top RB Güisayote, 2,270 m. Photos © J.H. Townsend.

Reserva Biológica Cerro Uyuca

Location and Extent: Southeastern Departamento de Francisco Morazán, 772 ha, maximum elevation 2,006 m.

Status: Biological Reserve (legally declared in 1987; Decreto 87-87);

Conservation International Key Biodiversity Area.

Herpetological results: Anura: Incilius porteri, Lithobates brownorum X forreri

(this population is considered to be interspecific hybrids by McCranie & Wilson 2002), L. maculatus, Ptychohyla salvadorensis, Rhinella marina, Tlalocohyla loquax; Squamata:

Anolis laeviventris, A. tropidonotus, Sceloporus malachiticus.

Site visit summary: Vicinity of Cabot Biological Station, 1,620 –1,700 m elevation, 17–18 March 2007, 17–19 July 2007.

Summary findings: Sampling in Reserva Biológica Cerro Uyuca took place in the vicinity of the biological station managed by the Centro Zamorano de Biodiversidad, which includes access to a small reservoir supporting a dense population of Lithobates brownorum/forreri.

Reserva Biológica Guajiquiro

Location and Extent: Departamento de La Paz; 17,165 ha, maximum elevation

2,330 m.

Status: Biological Reserve (legally declared in 1987; Decreto 87-87);

Conservation International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa celaque; Anura: Exerodonta catracha, Hypopachus barberi, Incilius ibarrai, Lithobates maculatus; Squamata: Anolis crassulus, Drymobius chloroticus, Ninia espinali, Thamnophis fulvus.

100

Site visit summary: Guajiquiro and surrounding countryside, 1,900–2,240 m elevation, 23–25 May 2008, 14–15 August 2008.

Summary findings: Highland areas above about 2000 m elevation in the vicinity of the town of Guajiquiro are a matrix of both traditional and modern agriculture and maintained patches of old growth cloud forest. In some places, proximity of existing patches and make-up of the intervening agricultural areas is such as to allow species such as Bolitoglossa cf. celaque and Exerodonta catracha to persist and, in some cases, apparently thrive. Cattle ponds in small pastures were readily being used for breeding by the anurans E. catracha and Hypopachus barberi (Ketzler et al. 2011,

Luque-Montes et al. 2011), and B. cf. celaque was found in a forest patch surrounding a communications tower. Additional highland areas with forest patches in the area need to be visited to fully and accurately assess the conservation needs of the relatively large but biogeographically connected highland area.

Reserva Biológica Güisayote

Location and Extent: Departamento de Ocotepeque; 12,677 ha, maximum elevation 2,330 m.

Status: Biological Reserve (legally declared in 1987; Decreto 87-87);

Conservation International Key Biodiversity Area.

Herpetofaunal results: Caudata: Bolitoglossa conanti; Anura: Hypopachus barberi; Squamata: Anolis heteropholidotus, Mesaspis moreletii, Sceloporus malachiticus.

Site visit summary: Communication towers southeast of El Portillo de

Ocotepeque, core zone of RB Güisayote, 2,080-2,230 m elevation, 18-21 June 2008.

101

Summary findings: The core zone of RB Güisayote has an access road running along the top of the reserve through high pastures and remnant cloud forest to a set of communications towers. This reserve appears to support a surprising amount of forest given that it lies in a region that is otherwise heavily impacted by human activity. The status of Leptodactylus silvanimbus is unclear, it has been documented previously inhabiting open fields and pastures with flooded areas in the vicinity.

Reserva Biológica Yerbabuena

Location and Extent: Southwestern Departamento de Francisco Morazán 3,522 ha, maximum elevation 2,243 m.

Status: Biological Reserve (legally declared in 1987; Decreto 87-87);

Conservation International Key Biodiversity Area

Herpetological results: Caudata: Bolitoglossa carri; Anura: Incilius ibarrai;

Squamata: Anolis sminthus, A. tropidonotus, Sceloporus malachiticus.

Site visit summary: Finca La Alondra, cafetal in region of Cerro Cantagallo, and other fincas in the area, 1,750–2,020 m elevation, 16–17 July 2007.

Summary findings: Despite the fact that most of the forest around Cerro

Cantagallo is disturbed and much of that is fully converted to agriculture, Bolitoglossa carri can still be found relatively easily inside bromeliads in remaining forest patches.

Additional work should be done to determine the extent of remaining forest on the highest portions of Cerro Cantagallo.

Reserva de la Biosfera Bosawas

Location and Extent: Departamento de Jinotega and Región Autónoma Atlántico

Norte; 1,992,000 ha.

102

Status: UNESCO Man and the Biosphere Reserve (legally declared in 1991 and revised and expanded in 1997 and 2001; Decreto 44–91, Decreto 32–96, and Ley 407).

Herpetological results: Gymnophiona: Gymnopis multiplicata; Caudata:

Bolitoglossa striatula; Anura: Agalychnis callidryas, Cochranella granulosa, Craugastor fitzingeri, C. lauraster, C. megacephalus, C. mimus, C. noblei, Dendropsophus microcephalus, Diasporus diastema, Incilius valliceps, Leptodactylus melanonotus, savagei, Lithobates vallanti, Pristimantis cerasinus, P. ridens, Rhaebo haematiticus,

Rhinella marina, Smilisca baudinii, S. phaeota, S. sordida, Teratohyla pulverata;

Squamata: Ameiva festiva, Anolis capito, A. limifrons, A. lionotus, A. quaggulus,

Basiliscus plumifrons, B. vittatus, Corytophanes cristatus, Iguana iguana,

Sphaerodactylus millepunctatus, Sphenomorphus cherriei, Thecadactylus rapicauda,

Adelphicos quadrivirgatum, Boa constrictor, Bothrops asper, Drymarchon melanurus,

Imantodes cenchoa, Leptodeira septentrionalis, Mastigodryas melanolomus, Micrurus nigrocinctus, Ninia sebae, Oxybelis aeneus, O. brevirostris, Porthidium nasutum,

Pseustes poecilonotus, Sibon longifrenis, S. nebulatus, Tretanorhinus nigroluteus;

Testudines: Kinosternon leucostomum, Rhinoclemmys annulata, R. funerea, Trachemys venusta.

Site visit summary: Muru Ta, 12–13 June 2007, 180 m elevation; Muru Lak, 13–

18 June 2007, 190 m elevation; Kalum Kitang, 18–21 June 2007, 180 m elevation; Aran

Dak, 11 and 22 June 2007, 150 m elevation; Raiti, 23 June 2007, 140 m elevation.

Summary findings: Results from this expedition, the first in a series, to the core zone of Bosawas were reported in Sunyer et al. (2009) and Travers et al. (2011).

103

Figure 3-6. Sampling in the Chortís Block V. A) Cerro Yaluk, Departamento de Olancho, Honduras, seen from the Río Coco (A–D, June 2007). B) Aran Dak, most remote Mayangna community on Río Lakus, 150 m, Reserva de la Biósfera Bosawas, Nicaragua. C) L. Wilson (seated) being poled down the upper Río Lakus by Miskitu parataxonomist S. Charley (standing); core zone of Reserva de la Biósfera Bosawas, 170 m. D) Preparing samples; (left to right) L. Wilson, S.Travers (seated), Miskitu parataxonomists R. Picado and S. Charley, J. Sunyer (seated), and Mayangna parataxonomist J. Lopez; Muru Lak, 180 m, core zone of Reserva de la Biósfera Bosawas. E) I. Luque ascending the volcanic peak of Isla El Tigre, with a view of the Golfo de Fonseca and the volcanoes of eastern El Salvador; 550 m (August 2010). F) Campsite at the summit of Cerro Zarciadero, 1,890 m, Honduras (July 2007). Photos © J.H. Townsend.

104

105

Jardín Botánico Lancetilla

Location and extent: Western Departamento de Atlántida; 2,255 ha, maximum elevation 710 m.

Status: National Park (legally declared in 1990; Decreto 48-90); Conservation

International Key Biodiversity Area.

Herpetological results: Anura: Craugastor aurilegulus, Leptodactylus melanonotus, Lithobates brownorum, L. vallanti, Smilisca baudinii; Squamata: Ameiva festiva, Anolis lemurinus, A. cf. zeus, Lepidophyma flavimaculatum, Sphaerodactylus millepunctatus, Sphenomorphus cherriei, Clelia clelia, Coniophanes imperialis.

Site visit summary: 7–9 June 2010, 22–24 July 2010.

Summary findings: Besides the extensive grounds of the botanical gardens,

Lancetilla also protects the entire watershed of the Río Lancetilla. This includes 1,281 ha of virgin lowland rainforest, one of the best preserved fragments of lowland rainforest along the northern coast of Honduras.

Área de Uso Multíple Isla del Tigre

Location and Extent: Island in the Golfo de Fonseca on the Pacific Coast,

Departamento de Valle; 601 ha, maximum elevation 783 m.

Status: Multiple-Use Area (legally declared in 1999; Decreto 5–99–E).

Herpetological results: Anura: Incilius coccifer; I. porteri; Squamata: Sceloporus squamosus.

Site visit summary: Amapala and trail to summit, 5–783 m elevation, 16–17

August 2010.

106

Non-Protected Areas

Cerro El Zarciadero

Location: Northern Departamento de Comayagua.

Herpetological results: Caudata: Bolitoglossa oresbia; Anura: Exerodonta catracha, Incilius porteri; Squamata: Drymobius margaritiferus.

Site visit summary: Cerro El Zarciadero, 1,845–1,890 m elevation, 14-15 July

2007.

Comments: The small patch of forest on top of Cerro El Zarciadero remains intact, and is essentially protected as private property around a set of communications towers. A family lives at the entrance to the tower complex and the husband is paid to guard and maintain the grounds. The residents were already familiar with Bolitoglossa oresbia and the idea that they are protecting the only known habitat of this animal, which was found active on vegetation on the only night we spent on top of the mountain.

Exerodonta catracha was also found actively calling and breeding in and around a very small water basin and stream just below the road and towers. The situation with the remaining forest patch, tiny as it is, appears stable for the time, since it provides a buffer area around the communications towers at the top of the mountain. The discovery of

Bolitoglossa oresbia in PN Cerro Azul Meámbar also lessens the urgency with which earlier conservation efforts surrounding El Zarciadero were being pursued.

Highlands surround the Meseta de La Esperanza

Location: Central Departamento de Intibucá, mountainous areas surrounding the

Meseta de La Esperanza; 1,700–2,100 m elevation.

Herpetological results: Caudata: Bolitoglossa celaque; Anura: Exerodonta catracha, Incilius ibarrai, Lithobates brownorum X forreri (this population is considered

107

to be made up of interspecific hybrids [McCranie & Wilson 2002]); Squamata: Anolis crassulus, A. sminthus, Sceloporus malachiticus, S. variabilis.

Site visit summary: San Pedro La Loma, 1,960–2,020 m elevation, 21–24

January 2008, 16–17 August 2008; Cerro El Pelón, 2,065 m elevation, 30 June 2008;

Zacate Blanco, 1,950–2,100 m elevation, 29–30 June 2008, 18–19 August 2008.

Summary findings: As is the case in RB Guajiquiro, highland areas above around 2,000 m elevation in the vicinity of La Esperanza are a matrix of both traditional and modern agriculture and patches of remnant cloud forest. In some places, proximity of existing patches and make-up of the intervening agricultural areas are such as to allow species such as Bolitoglossa cf. celaque and Exerodonta catracha to persist and, in some cases, apparently thrive. Additional highland areas with forest patches in the area need to be visited to fully and accurately assess the conservation needs of the relatively large but biogeographically connected highland area.

Los Naranjos

Location: Near the border of Departamento de Cortés and Departamento de

Santa Bárbara, between Lago de Yojoa to the east and Parque Nacional Montaña de

Santa Bárbara to the west; 700–730 m elevation.

Herpetological results: Caudata: Bolitoglossa mexicana; Anura: Craugastor laevissimus, Dendropsophus microcephalus, Engystomops pustulosus,

Hyalinobatrachium fleischmanni, Hypopachus variolosus, Incilius valliceps, Lithobates brownorum, L. maculata, Rhinella marina, Smilisca baudinii; Squamata: Ameiva undulata, Anolis laeviventris, A. lemurinus, A. tropidonotus, A. unilobatus, A. cf. zeus,

Sphaerodactylus millepunctatus, Sphenomorphus cherriei, Ninia diademata, N. sebae,

Tantilla taeniata.

108

Site visit summary: 13–15 August 2007, 30–31 January 2008, 5 April 2008, 11–

12 April 2008, 20–22 May 2008, 7 June 2010.

Summary findings: Collections in this area were primarily made on the Plowden family farm, now called Compañia Agrícola El Paraíso. This large property includes numerous springs, shade coffee plots, and shaded horticultural plots.

Montaña de Jacaleapa

Location: Central Departamento de Olancho, vicinity of Nahoan community of El

Norte, 980–1,180 m elevation.

Herpetological results: Anura: Craugastor laevissimus, C. lauraster, C. noblei,

Lithobates maculatus, Pristimantis ridens, Ptychohyla hypomykter; Squamata: Anolis sp., A. tropidonotus, Sceloporus malachiticus, Sphenomorphus cherriei.

Site visit summary: 11–12 April 2011.

Summary findings: This isolated patch of mesic highland forest had not previously been surveyed by biologists, and supports over 500 ha of intact premontane rainforest.

Montaña Macuzal

Location: Southern Departamento de Yoro, west of Yorito.

Herpetological results: Caudata: Bolitoglossa porrasorum, Nototriton barbouri;

Anura: Craugastor rostralis; Squamata: Anolis laeviventris, A. pijolensis, A. yoroensis,

Sceloporus malachiticus.

Site visit summary: El Panal, southwest side of Montaña de Macuzal, 1500–1800 m elevation, Departamento de Yoro, 31 January–3 February 2008, 6–7 April 2008.

Comments: While the flanks of Montaña de Macuzal are converted to shade coffee cultivation or completely cleared of all natural vegetation, the area above 1780 m

109

Figure 3-7. Sampling in the Chortís Block VI. A) ―Island‖ fragment of Broadleaf Cloud Forest surrounded by converted corn fields, 2,230 m, Sierra de Opalaca west of La Esperanza (July 2008). B) Small reservoir supporting population of Lithobates brownorum X forreri, 2,010 m, Cerro San Pedro, east of La Esperanza (January 2008). C) Converted cloud forest, Zacate Blanco, highest point 2,250 m (July 2008). D) L. Wilson and I. Luque searching for Bolitoglossa cf. celaque in bromeliads that had been cut and left for cattle feed, 2,010 m, Cerro San Pedro. E) Intact bromeliad cover, 2,020 m, Cerro San Pedro. F) L. Wilson at Pozo Azul, a karstic spring near Los Naranjos, 680 m. G) Montaña Macuzal, viewed from the northeast, showing strip of remnant Broadleaf Cloud Forest at highest elevations (maximum 1,905 m). H) Montaña Macuzal, viewed from the southwest near El Panal, 1,480 m, with yellow wildflowers covered the completely deforested slopes; access to the cloud forest visible in the saddle (January 2008). I) N. Stewart (front) and J. Butler (back) crossing upper deforested slope of Montaña Macuzal (April 2008). Photos © J.H. Townsend.

110

111

elevation still supports an approximately 5–10 ha patch of moderately disturbed cloud forest on karstic soils, and Bolitoglossa porrasorum and Nototriton barbouri are persisting, and in the case of the former species, abundantly so. A stream flowing out of the northern corner of Montaña de Macuzal near the community of El Portillo should be checked for signs of Duellmanohyla salvavida and/or Plectrohyla guatemalensis

(McCranie & Wilson 2002), as well as other priority species.

Saguay

Location: Upper Valle de Agalta, Departamento de Olancho.

Herpetological results: Anura: Hypopachus variolosus, Incilius valliceps,

Leptodactylus fragilis, Lithobates maculatus, Ptychohyla hypomykter, Smilisca baudinii,

Tlalocohyla loquax; Squamata: Anolis capito, A. quaggulus, A. wermuthi, Sceloporus malachiticus; Testudines: Kinosternon scorpioides.

Site visit summary: Teocintecito ravine, 720 m elevation, and a seepage bog outside Saguay, 580 m elevation, 10 April 2011.

Summary findings: Survey sites in remnant dry forest patches in the foothills around the upper Río Grande (also called the Sico, Tinto, and Negro at different portions downstream) watershed. One locality was a relatively large seepage bog, covered with a think mat of floating vegetation that arises from the middle of the dry valley floor.

San José de Texíguat

Location: Southern side of the upper Río Leán valley in the foothills of Texíguat,

150–200 m elevation.

Herpetological results: Caudata: Bolitoglossa nympha; Anura: Craugastor aurilegulus, Duellmanohyla salvavida, Hyalinobatrachium fleischmanni, Leptodactylus

112

fragilis; Squamata: Anolis cf. zeus, Iguana iguana, Lepidophyma flavimaculatum,

Atropoides mexicanus, Hydromorphus concolor.

Site visit summary: 10–11 November 2010.

Summary findings: Sampling was carried out in a deep, protected ravine just south of the town. Craugastor aurilegulus and Duellmanohyla salvavida were found in abundance around the small, spring-fed creek. A juvenile Hydromorphus concolor was found in a rocky fissure at the spring’s source.

Selva Negra

Location: Between cities of Jinotega and Matagalpa, Departamento de

Matagalpa, 1,200–1,300 m elevation.

Herpetological results: Anura: Agalychnis callidryas, Craugastor lauraster,

Incilius valliceps, Lithobates cf. taylori, Ptychohyla hypomykter, Smilisca baudinii,

Tlalocohyla loquax; Squamata: Anolis capito, A. quaggulus, A. wermuthi, Sceloporus malachiticus.

Site visit summary: 13–15 August 2007.

Summary findings: Collections were made in the premontane rainforest above the Selva Negra coffee farm and lodge.

Yeguare Valley

Location: Valley associated with an upper tributary of the Río Choluteca, 810 m elevation.

Herpetological results: Anura: Engystomops pustulosus; Squamata: Basiliscus vittatus, Gonatodes albogularis, Porthidium ophryomegas; Testudines: Kinosternon scorpioides, Trachemys venusta.

Site visit summary: 16–17 March 2007; 3–4 June 2010.

113

Figure 3-8. Sampling in the Chortís Block VII. A) O. Reyes (left) and M. Bonta (right) hiking through the Nahoan community of El Norte en route to Montaña de Jacaleapa, 1,100 m (A–E, April 2011). B) Premontane rainforest stream, 1,120 m, Montaña de Jacaleapa. C) The author with a giant (10+ m) teocinte cycad (Dioon mejiae) near Saguay, 630 m, Valle de Agalta. D) La Puzunca, upper reach of the Río Grande, 560 m, Valle de Agalta; distinctive green plants on far hillsides are teocinte cycads. E) Giant paddle cactus (Nopalea cf. hondurensis), 580 m, Valle de Agalta. F) Artificial lagoon, 1,220 m, Selva Negra, Nicaragua (F–H, August 2007). K. Townsend (front) and S. Travers (back) climb a trail through premontane rainforest, 1,460 m, Selva Negra. H) Premontane rainforest creek, 1,420 m, Selva Negra. Photos © J.H. Townsend.

114

115

Summary findings: Collections made in agricultural areas and remnant premontane dry forest in and around the campus of Escuela Agrícola Panamericana

Zamorano.

Discussion

Baseline Herpetological Inventory of Parque Nacional Montaña de Yoro

In June 2006, March 2007, and September 2008, I led the initial herpetological survey work carried out in Parque Nacional Montaña de Yoro, a relatively large yet herpetologically-unknown protected area in central Honduras. Parque Nacional

Montaña de Yoro was established in 1987 and sits along the boundary between the

Honduran departments of Francisco Morazán and Yoro, with about two-thirds of its area in the municipality of Marale in northernmost Francisco Morazán (COHECO 2003). With a total area of over 154 km2, more than 47 km2 of cloud forest supporting area above

1,800 m, Parque Nacional Montaña de Yoro potentially contains one of the largest areas of cloud forest remaining in Honduras (COHECO 2003). Next to nothing has been documented with regard to the park’s biodiversity, and only 14 herpetofaunal species are mentioned in the park management plan, with most of those being relatively widespread species that commonly are found around human habitation or in habitats below cloud forest (Townsend & Wilson 2009). Additionally, Parque Nacional Montaña de Yoro protects some of the highest portions of the Montañas de la Flor, which is home to the last traditional communities of indigenous Tolupanes that maintain the Tol language and culture (Chapman 1992). Within the park’s borders are over 4,700 people spread across 55 communities, with the population consisting of Tolupan, mestizos, and mixed Tolupan-mestizo farmers (COHECO 2003). Most of these people inhabit the

116

Figure 3-9. Exemplar paratypes and habitats from Parque Nacional Montaña de Yoro. A) Adult male paratype (UF 150000) of Anolis morazani sp. nov. B) Adult female paratype (MCZ 185612) of A. morazani sp. nov. C) Dewlap of UF 150000. D) Adult female paratype (MVZ 258030) of Bolitoglossa cataguana (―B. sp. inquirenda 1‖ in Chapter 4). E) Arboreal tank bromeliads in disturbed mixed cloud forest, Cataguana, 1,880 m. F) Quebrada Cataguana, 1,820 m. G) Remote homestead at Cataguana, 1,910 m. Photos © J.H. Townsend. nearly completely deforested buffer zone of Parque Nacional Montaña de Yoro, and as the buffer zone population grows, increasing pressure is being placed on the remaining cloud forest in the nuclear zone. Parque Nacional Montaña de Yoro has become so heavily impacted by the encroaching agricultural frontier that at one point it was

117

recommended that the park be re-designated as a biological reserve and that its borders be reduced to the remaining extent of intact cloud forest (Vreugdenhil et al.

2002).

My work in Parque Nacional Montaña de Yoro was limited to the vicinity of two sites, Cataguana on the northwestern side of the core zone, and the highlands above

Guaymas (sometimes referred to as Montaña de la Sierra; however, this is not to be confused with the Montañas de la Sierra of Departamento de La Paz). Despite the relatively limited amount of fieldwork carried out, the expeditions were nonetheless fruitful and resulted in the discovery of a new species of anole (Anolis morazani;

Townsend & Wilson 2009), and salamanders from three genera that could not be unambiguously assigned to a named species.

A New Species of Anole

Fieldwork in Parque Nacional Montaña de Yoro uncovered a relatively common anole lizard assignable to the Anolis crassulus group, characterized by having moderately to strongly enlarged medial dorsal scales, no more than two scales separating the supraorbital semicircles, four to seven rows of loreals, suboculars, and supralabials in contact under the central portion of the orbit, enlarged postanals in males, and heterogeneous flank squamation (McCranie et al. 1992; Köhler et al. 1999).

The majority of the species in the A. crassulus group are endemic to the Chortís

Highlands, including A. amplisquamosus, A. heteropholidotus, A. muralla, A. rubribarbaris, A. sminthus, and A. wermuthi (Köhler 2008). In addition, one widespread species, A. crassulus, is found across the highlands of Nuclear Central America from

Chiapas, México, to El Salvador and southwestern Honduras. The population from

Parque Nacional Montaña de Yoro was demonstrated to be morphologically distinctive

118

from all other species in the A. crassulus group based on hemipenial structure and scalation, and I subsequently described it as the new species Anolis morazani

(Townsend & Wilson 2009).

Salamanders of Uncertain Taxonomic Assignment

Populations of at least three different salamanders were discovered in Parque

Nacional Montaña de Yoro. A series of Bolitoglossa, assignable to the B. dunni species group based on having well-developed subdigital pads and bluntly rounded, free toe tips but with coloration differing from other described species, was collected at Cataguana in

2006 and 2007 and above Guaymas in 2008. Three specimens of Nototriton were also collected at Cataguana from within ground cover, and a small Oedipina was collected in a deep mesic ravine through coniferous cloud forest. Taxonomic assignment of these three populations, referred to as Bolitoglossa sp. inquirenda 1, Nototriton sp. inquirenda

5, and Oedipina sp. inquirenda 1, are addressed in Chapter 4.

Parque Nacional Montaña de Yoro appears to be both overlooked and undervalued by national and international conservation planners, due to perceived lack of documented ―conservation-priority‖ species; however, this apparent lack of diversity is a result of undersampling and is not reflective of the true significance of the park’s biodiversity. Prior to initiation of survey work in Parque Nacional Montaña de Yoro,

Parque Nacional Montaña de Yoro was a virtual ―black hole‖ of biodiversity data, surrounded by literally dozens of cloud forest areas, many of which have well- documented and diverse endemic biotas (Wilson & McCranie 2004b). Given the biodiversity present in other cloud forest areas in Honduras, it is highly probable that additional undescribed species await discovery in Parque Nacional Montaña de Yoro.

Future work in Parque Nacional Montaña de Yoro should continue to shed light on

119

these relationships and provide a clearer picture of highland biogeography in eastern

Nuclear Central America.

Hotspot within a Hotspot: the Special Case of Refugio de Vida Silvestre Texíguat

The Refugio de Vida Silvestre (RVS) Texíguat was established in 1987 and consists of approximately 33,267 ha of premontane and lower montane rainforest straddling the border of the Honduran departments of Atlántida and Yoro (CIPF 2009).

RVS Texíguat is administered by the Instituto Nacional de Conservación y Desarrollo

Forestal, Áreas Protegidas y Vida Silvestre (ICF) as part of the Sistema Nacional de

Áreas Protegidas de Honduras (SINAPH), with management authority for the reserve in the hands of the non-governmental organization Fundación para la Protección de

Lancetilla, Punta Sal y Texíguat (PROLANSATE). Most herpetological surveys in RVS

Texíguat have been conducted since 1991 on the leeward side of the park in

Departamento de Yoro, at elevations above 1,500 m in the vicinity of a coffee farm known locally as La Fortuna (Holm & Cruz 1994; McCranie et al. 1993; McCranie &

Castañeda 2004a, b; Townsend et al. 2010a; Wilson et al. 1994, 1998).

Situated at the western end of the Cordillera Nombre de Dios, the leeward side of

RVS Texíguat and its counterpart at the eastern end of the Cordillera, Parque Nacional

(PN) Pico Bonito, are the most significant areas of herpetofaunal diversity in a country whose national level of endemism has already been demonstrated to be the highest in

Central America (Wilson & Johnson 2010). Our preliminary results from the windward side of RVS Texíguat indicate that this area represents a unique opportunity for

120

Figure 3-10. La Liberación de Texíguat. A) Valley of the Río Jilamito, with the remote encampment known as La Liberación de Texíguat partially visible as the lower clearing in the center of the picture beyond the emergent palm; the entire expanse from La Liberación to the farthest peak (Cerro Texíguat, 2,208 m) is undisturbed premontane and lower montane rainforest. B) The initial leg of the expedition to La Liberación requires a 2 hour mule ride through the lowlands (L.D. Wilson on mount). C) The mule ride is followed by a 6–8 hour hike into the mountain, with around two-thirds the trail within this deep, high- grade trench (M. Medina-Flores following our guides and pack animals). Photos © J.H. Townsend.

121

conserving the remarkable diversity that has already been documented on the leeward side of RVS Texíguat, and for expanding the known herpetofaunal diversity of the reserve.

Discovery of Plectrohyla chrysospleura

Plectrohyla chrysopleura (Anura: Hylidae) is a large, critically endangered spikethumb frog only known from the vicinity of its type locality in Parque Nacional Pico

Bonito (Cruz et al. 2004). The type locality, Quebrada de Oro, is a premontane rainforest locality that has been under relatively intensive study since 1980 and is among the most remarkable sites of amphibian endemism in Central America

(McCranie & Wilson 2002). Unfortunately, this locality also has the distinction of being one of the best-documented cases of catastrophic amphibian decline in Central America

(McCranie & Wilson 2002; McCranie & Castañeda 2005, 2007; Townsend & Wilson

2010). It is the type locality for six species of amphibians (Craugastor aurilegulus, C. chrysozetetes, C. fecundus, Duellmanohyla salvavida, Plectrohyla chrysopleura, and

Rhinella chrysophora), and is just below the type locality (Cerro Búfalo) of two other species (Craugastor cruzi and C. saltuarius). Almost all of these species are considered to be in decline, with even some considered to be extinct or close to extinction

(McCranie & Castañeda 2007).

Like other species from Quebrada de Oro that have declined or disappeared, P. chrysopleura is apparently extirpated from that locality, and given that Quebrada de Oro is the only locality where this species has been found despite consistent focused work in the area (McCranie and Castañeda 2005, 2007), it raises the concern that the species might be near extinction, if not already extinct. The last time P. chrysopleura was documented as extant was in May 1996, when two metamorphs and two tadpoles

122

Figure 3-11. Plectrohyla chrysopleura (Hylidae) from La Liberación. A) Adult female Plectrohyla chrysopleura (USNM 573993) from La Liberación, 1,030 m elevation, Refugio de Vida Silvestre Texiguat, Honduras. B) Adult male P. chrysopleura (USNM 573995) from Cerro El Chino, 1,420 m elevation, Refugio de Vida Silvestre Texiguat, Honduras. C) Juvenile P. chrysopleura (USNM 573994) from La Liberación. D) Recently metamorphosed P. chrysopleura (not collected) from La Liberación. E) Río Jilamito at La Liberación, 1,020 m. E) Tributary of Río Jilamito, La Liberación, 1,030 m. Photos © J.H. Townsend.

123

were collected along Quebrada de Oro (McCranie and Wilson 2002, McCranie and

Castañeda 2005). At that time, one of two tadpoles collected had deformed mouthparts

(McCranie and Wilson, 2002). Based on these data and additional considerations, Cruz et al. (2004), IUCN (2011), and Townsend and Wilson (2010) all judged P. chrysopleura to be Critically Endangered, based on IUCN criteria (IUCN, 2001). For P. chrysopleura, the particular red list status was Critically Endangered (A2ace, B1ab[iii,v]+2ab[iii,v]), meaning that a reduction in population size of ≥ 80% was observed, estimated, inferred, or suspected over the last 10 years based on direct observation, a decline in area of occupancy, extent of occurrence and quality of habitat, and the suspected impact and susceptibility to decline from chytridiomycosis.

In June and July 2010, a series of Plectrohyla chrysopleura was collected around

La Liberación and Cerro El Chino on the windward side of RVS Texíguat, providing the first evidence of this species’ survival in 14 years and the second known locality for this species, approximately 55 km west-southwest of the type locality (Townsend et al.

2011c). Of the six endemic species of premontane forest amphibians described from the Quebrada de Oro area, we now know that two of them (Duellmanohyla salvavida and Plectrohyla chrysopleura) occur at La Liberación. Given the robust nature of the populations of these two treefrogs and the intactness of the premontane rainforest in this area, I remain hopeful that perhaps others of the endemic anurans that have undergone decline at Quebrada de Oro will be discovered still resident at La Liberación.

Underestimated Salamander Diversity?

Three species of salamanders were previously known from RVS Texíguat,

Bolitoglossa porrasorum sensu lato, Nototriton barbouri s.l., and Oedipina gephyra s.l., each of which is considered endemic to two or more isolated localities in northern

124

Honduras (McCranie & Wilson 2002). Bolitoglossa porrasorum is known from the type locality at Pico Pijol, Montaña Macuzal, Texíguat, and Pico Bonito; however, only the

Pico Bonito population has been included previously in phylogenetic studies (Parra-

Olea et al. 2004). Nototriton barbouri is considered to have the same distribution as B. porrasorum, whereas O. gephyra is known from its type locality at La Fortuna and from a single specimen from Pico Bonito (McCranie & Wilson 2002). Both N. barbouri and O. gephrya have been shown to demonstrate species-level divergence between the

Texíguat and Pico Bonito populations (García-París & Wake 2000), and I supplement this existing data with newly collected samples representing both putative taxa in

Chapter 4. In April 2008, a single specimen of Nototriton possessing the distinctive morphological traits of enlarged nostrils, syndactylous feet, and a contrasting pale venter was collected within leaf litter packed into a crevice in the side of a small mesic canyon near La Fortuna (1,550 m elevation). This species, described as N. tomamorum by Townsend et al. (2010a), is analyzed genetically in Chapter 4 and reviewed morphologically in Chapter 5.

Highly Endemic and Highly Endangered

The leeward side of RVS Texíguat (herein referred to as Yoro Texíguat) is highly imperiled due to continued illegal logging of valuable hardwoods and forest clearing for subsistence agriculture, which I witnessed firsthand in April 2008 (Townsend et al.

2010a). Despite the rapid advancement of deforestation in Yoro, little exploration of the virtually unknown Atlántida side of the refuge has been conducted to date. I led three expeditions to the windward side of RVS Texíguat (Atlántida Texíguat) during 2010; these visits were limited to 12 days (10–21 June 2010) and seven days (26 July–2

August 2010) in the vicinity of La Liberación (15.53°N/87.29°W; camp was established

125

at 1,030 m elevation), and three days (10–12 November 2010) in the vicinity of San

José de Texíguat (15.52°N/87.45°W), yet we recorded 46 species (Townsend et al., submitted A), compared to 40 species amassed during the approximately two-decade span of limited study of Yoro Texíguat (Wilson & McCranie 2004b; Townsend et al.

2010a). This pattern suggests that herpetofaunal diversity and endemism demonstrated on the Yoro side will likely be exceeded on the windward side.

The 2010 work in this area, along with past work on the leeward slope of RVS

Texíguat (summarized in Townsend et al. 2010a) and in Parque Nacional Pico Bonito

(summarized in McCranie & Castañeda 2005), demonstrates that 29 of the 93 endemic species (close to one-third) of amphibians and reptiles reported from Honduras by

Townsend & Wilson (2010) occur within these three areas of the same cordillera

(Townsend et al., In press A). At least 10 species of amphibians and reptiles are

Texíguat-restricted endemics, and RVS Texíguat is home to 14 more species endemic to the Chortís Block (Townsend et al., In press A).

Given the preliminary successes of efforts towards documenting and promoting the conservation of herpetofaunal diversity in RVS Texíguat, I believe that the serious problems facing the reserve can be addressed through continued efforts in the coming years to provide enduring protection for this vitally important component of the

Honduran patrimony.

Cryptozoic Snake Diversity

Cryptozoic snake taxa include some of the most species-rich groups of snakes, and their remarkable diversity is exceeded by their ability to remain concealed and evade detection, even in relatively well studied field sites (Myers 2003; Stafford 2004;

Townsend 2009). Genera such as Geophis, Ninia, and Tantilla are all cryptozoic in

126

nature, typically small in size and often inhabit the interface between leaf litter and soil.

Many of these secretive species are known only from one or a handful of specimens, yet are distinctive enough in terms of external morphology to be promptly recognized as distinct species (e.g., Campbell 1998; Stafford 2004; Townsend 2009).

A New Species of Centipede Snake (genus Tantilla) from La Liberación

The genus Tantilla currently consists of 63 described species (Townsend et al. In press B), broadly distributed in all three of the major geographic sections of the Western

Hemisphere (North America, Mesoamerica, and South America; Wilson 1999), including at least seven species in the Chortís Block (McCranie 2011b). During survey work in

July 2010 around La Liberación on the windward side of RVS Texíguat, we collected a single specimen of Tantilla possessing a set of distinctive morphological characteristics that stand out as unique within the genus. Based on having pale middorsal and/or lateral stripes and pale markings on the nape, we posit that this snake is assignable to the T. taeniata species group (sensu Wilson & Meyer 1971, Campbell 1998, Wilson &

McCranie 1999), but represents a previously unknown taxon, which is described in a manuscript submitted for publication in August 2011. The new species differs significantly from all congeners on the basis of its dorsal body pattern, consisting of a pale middorsal stripe composed of narrow spots and confined to the middorsal row and a lateral coloration of pale spots on each of rows 1, 2, and 4. In addition, it possesses a middorsally interrupted pale nuchal band and dark brown pigment on the lateral edges of the ventrals. The new species appears to have no close affinities within the T. taeniata group, and elucidation of its phylogenetic relationships will have to await a group or genus-wide assessment, which, given the difficulties in securing fresh material of these taxa, might not occur in the near future.

127

A Large New Species of Blindsnake (Typhlops tycherus)

Blindsnakes of the genus Typhlops (Squamata: Typhlopidae) have a cosmopolitan and primarily tropical distribution (McDiarmid et al. 1999), with the majority of diversity in the Western Hemisphere confined to the islands of the West Indies (Dixon & Hendricks

1979; Thomas & Hedges 2007). Four native species of Typhlops previously were known to occur in Mesoamerica (Dixon & Hendricks 1979; Köhler 2008): T. costaricensis

(Honduras to Costa Rica), T. microstomus (Yucatan Peninsula), T. stadelmani (western and northern Honduras), and T. tenuis (Veracruz, México, to Guatemala). A fifth typhlopid, the parthenogenetic Ramphotyphlops braminus, has been introduced to scattered localities throughout Mesoamerica, primarily around urban areas (Köhler,

2008).

Two species of blindsnakes (T. costaricensis and T. stadelmani) are known to occur in the Chortís Highlands. Both of these species occur in Honduras; Typhlops costaricensis is known from localities in the mesic lowlands of Departamento de Gracias a Dios and from mid-elevation pine-oak forest localities between Tegucigalpa and

Parque Nacional La Tigra in Departamento de Francisco Morazán (McCranie et al.

2006; Wilson et al. 1988), and T. stadelmani, which is reported from mid-elevation localities in the departments of Atlantída, Copán, and Yoro (McCranie & Castañeda

2005; McCranie & Wilson 2001). Typhlops stadelmani was, until recently, considered a junior synonym of T. tenuis (Dixon & Hendricks 1979), and was resurrected by

McCranie & Wilson (2001) after collection of a large series that verified the distinctiveness of T. stadelmani from T. tenuis, based primarily on differences in the number of scales between rostral and tail tip and in color.

128

Figure 3-12. New species of Tantilla (Colubridae) and Typhlops (Typhlopidae). A) Dorsal, lateral, and central views of the head of adult male holotype (USNM 574000) of Tantilla sp. nov., 1,150 m elevation, La Liberación de Texiguat. B) sulcate and asulcate views of the hemipenis of USNM 574000. C) dorsal and lateral coloration of USNM 574000. D) subadult Typhlops tycherus sp. nov. from El Cedral, Montaña de Santa Bárbara, 1,550 m. Photos © J.H. Townsend.

129

In January 2008, a single specimen of Typhlops was found freshly killed on an unpaved road between two small communities near the lower edge of intact cloud forest in Parque Nacional Montaña de Santa Bárbara, Honduras. This specimen differed from all other Middle American blindsnakes in its large size, in having 22–22–22 scales around the body, and by having a dark brownish gray dorsum with a well-defined pale yellowish gray to immaculate white ventral coloration. This specimen formed the basis for describing the species Typhlops tycherus, which was confirmed in April 2011 by collection of a second specimen from the vicinity of the type locality by M. Medina-

Flores (UNAH).

Geophis damiani at La Liberación

The genus Geophis (Squamata: Colubridae: Dipsadinae) is currently comprised of

47 species of secretive semifossorial snakes distributed across nearly every terrestrial habitat from México to northern South America (Downs 1967, Wilson & Townsend

2007, Townsend 2009). Six members of the genus are known to occur in the Chortís

Block: Geophis damiani, G. dunni, G. fulvoguttatus, G. hoffmanni, G. nephodrymus, and

G. rhodogaster, four of which (G. damiani, G. dunni, G. fulvoguttatus and G. nephodrymus) are endemic (Townsend 2009). Geophis damiani is one of the least known snakes in the Chortís Block, and was only known from two adult specimens

(Wilson et al. 1998, McCranie & Castañeda 2004). On 29 July 2010, an adult male

Geophis damiani (USNM 573999) was collected during survey work around La

Liberación, representing the third adult specimen and second adult male specimen known for this taxon. The snake was found at 22h30 while active in the bottom of a deep, trench-like trail at 1,075 m elevation in moderately disturbed Premontane Wet

Forest. The snake had apparently fallen into and become trapped inside the trench,

130

which was about 2 m deep at the collection site and over 3.5 m deep in some places.

This locality lies around 10 km NNW of the previously reported localities, and the elevation of the new record is 475 m below the previously-known lowest elevational distribution attributed to this species, 1,550 m (based on UF 142543, an egg and embryo), and 605 m below the lowest verified elevation for the species (1,680 m; USNM

573999). All previously reported localities for G. damiani are in the Lower Montane Wet

Forest formation. In Townsend et al. (2010d), we provided morphological data and color notes for USNM 573999, and compared it to the two previously known specimens of G. damiani.

Noteworthy Ninia

Ninia pavimentata. Ninia pavimentata is a small semifossorial snake reported from pine-oak and cloud forest areas in central Guatemala, as well as from a single locality in northwestern Honduras (Smith & Campbell 1996, Townsend et al. 2005). This taxon was previously considered a synonym of N. maculata (Peters 1861), until Smith &

Campbell (1996) resurrected N. pavimentata (Bocourt 1883) to species level on the basis of non-overlapping segmental count ranges and other morphological characteristics. The distributions of N. maculata and N. pavimentata are presently known to be separated by a roughly 315 airline km gap in Honduras, with N. pavimentata reaching its easternmost known distribution in the Sierra de Omoa, outside of Parque Nacional Cusuco, Departamento de Cortés (15.51ºN, 88.18ºW), 1,250 m elevation, and N. maculata its northernmost known locality at Quebrada Machín,

Reserva de la Biósfera Río Plátano, Departamento de Colón (15.32ºN, 85.28ºW), 540 m elevation (McCranie et al. 2001; Townsend et al. 2005). On 10 April 2008, J. Butler and I collected a female Ninia pavimentata (UF 152810) from under a rock at the edge

131

Figure 3-13. Noteworthy cryptozoic snakes. A) Third adult specimen, and second male, Geophis damiani (USNM 573999), 1,075 m elevation, La Liberación de Texiguat. B) Ninia espinali, a Chortís Highlands endemic recorded for the first time in Cerro Azul Meámbar. C) Ninia pavimentata, recorded for the first time in Refugio de Vida Silvestre Texíguat. Photos © J.H. Townsend. of a fragment of cleared cloud forest in Refugio de Vida Silvestre Texiguat (15.44°N,

87.30°W), 1,715 m elevation, Departamento de Yoro, Honduras (Townsend et al.

2009b). This habitat is similar to that of most N. pavimentata specimens, which are known from 1,120–1,825 m elevation in pine-oak or cloud forest (Smith & Campbell

1996), as well as to the disturbed habitat (a shade cafetal) where the other Honduran specimen originated. Examination of the specimen showed it to fall with the known

132

range of variation in N. pavimentata (Smith & Campbell 1996; Townsend et al. 2005;

Townsend et al. 2009b), providing further evidence for the specific status of this taxon.

This record extends the known range of Ninia pavimentata approximately 65 airline km east from its known distribution. The new locality also narrows the geographic gap between N. pavimentata and its most similar congener and presumed sister taxon, N. maculata, to approximately 250 airline km.

Ninia espinali. Ninia espinali is known to occur in western Honduras and northern

El Salvador, where it inhabits highland rainforests from 1,580–2,270 m elevation (Köhler

2008). On 9 July 2008, I. Luque and I collected an adult female N. espinali on a steep slope approximately 200 m above the Río de Varsovia (14°47.95’N, 87°53.47’W; 1,660 m elevation) in Parque Nacional Cerro Azul Meámbar, extending the known range of this species approximately 85 km SSE of localities in Parque Nacional Cusuco,

Deptartamento de Cortés, and approximately 75 km N of localities in the vicinity of

Guajiquiro, Departamento de La Paz (Luque-Montes & Townsend 2009). The snake was active at night in an exposed root mass on a steep slope near the top of a ridge.

Unidentified Salamander Populations

As indicated above, fieldwork in Parque Nacional Montaña de Yoro and Refugio de Vida Silvestre Texíguat resulted in collection of samples representing five populations of salamanders that could not be assigned unambiguously to a known species. Besides these five entities, at least five other populations of salamanders were sampled that cannot be unequivocally assigned to a known taxon. From Parque

Nacional Cerro Azul Meámbar, a series of Nototriton and a single distinctively colored

Bolitoglossa require additional attention, as do samples representing at least two forms of Oedipina from northern Nicaragua. In 2010 and 2011, fieldwork in Departamento de

133

Olancho resulted in discovery of two previously unknown allopatric populations of

Nototriton, one from the Sierra de Agalta and a second from the Sierra de Botaderos.

These 10 populations of salamanders, as well as other allopatric populations of questionable taxonomic assignment (e.g., Bolitoglossa porrasorum and Nototriton barbouri from Refugio de Vida Silvestre Texíguat), are the subject of molecular investigation and taxonomic evaluation in Chapter 4.

134

Table 3-1. Summary of fieldwork undertaken in the Chortís Block, 2006–2011. DATES LOCALITIES PARTICIPANTS 3 Jun – 18 Jun 2006 PN Montaña de Yoro: Cataguana JHT, Larry David Wilson 2 – 10 Dec 2006 PN Cerro Azul Meámbar: Los Pinos Brian Campesano, Lorraine Ketzler, Scott Travers, JHT, Steve Townsend PN La Tigra " " 8 – 20 Mar 2007 PN Montaña de Yoro: Cataguana Jason Butler, Lorraine Ketzler, Scott Travers, JHT, Larry David Wilson, et al. RB Cerro Uyuca Jason Butler, Lorraine Ketzler, Scott Travers, JHT 7 – 29 Jun 2007 Biosfera Bosawas (7 localities) Lenin Obando, Javier Sunyer, Scott Travers, JHT, Larry David Wilson, et al. 13 – 21 Jul 2007 PN Cerro Azul Meámbar: Los Pinos Lorraine Ketzler, JHT, Larry David Wilson RB Cerro Uyuca " " RB Yerbabuena " " Cerro Zarciadero " " 11 – 23 Aug 2007 Leon Scott Travers, JHT, Katielynn Townsend Selva Negra Scott Travers, JHT, Katielynn Townsend 20 Jan–5 Feb 2008 PN Montaña de Comayagua: La Okí Lorraine Ketzler, Ileana Luque-Montes, JHT, Larry David Wilson PN Montaña de Santa Bárbara: El " " Cedral Marcala " " San Pedro La Loma " " Los Naranjos Ileana Luque-Montes, JHT, Larry David Wilson Montaña Macuzal Ileana Luque-Montes, JHT, Larry David Wilson Yeguare Valley Ileana Luque-Montes, JHT, Larry David Wilson 4 – 20 Apr 2008 PN Cerro Azul Meámbar: Los Pinos Carlos Andino, César Cerrato, Gabriela Diaz, Lorraine Ketzler, Ileana Luque-Montes, Melissa Medina-Flores, Aaron Mendoza, Wendy Naira, JHT, Larry David Wilson PN Cerro Azul Meámbar: Aldea Cerro César Cerrato, Ileana Luque-Montes, Melissa Azul Medina-Flores, JHT, Larry David Wilson PN Montaña de Comayagua: Río Negro Ileana Luque-Montes, Melissa Medina-Flores, JHT, Larry David Wilson RVS Texíguat: La Fortuna Jason Butler, Lorraine Ketzler, Nathaniel Stewart, JHT, Larry David Wilson Montaña Macuzal " " 14 – 28 May 2008 PN Montaña de Comayagua: Río Negro James Austin, Lorraine Ketzler, JHT, Larry David Wilson RB Guajiquiro César Cerrato, Lorraine Ketzler, Ileana Luque- Montes, JHT, Larry David Wilson Los Naranjos James Austin, Lorraine Ketzler, JHT, Larry David Wilson 12 Jun–29 Jul 2008 PN Celaque Lorraine Ketzler, Ileana Luque-Montes, JHT, Larry David Wilson PN Cerro Azul Copán: Quebrada Grande Ileana Luque-Montes, JHT, Larry David Wilson PN Pico Pijol: Quebrada Las Payas " " PN Montaña de Comayagua: Río Negro " " PN Cerro Azul Meámbar: Aldea Cerro Ileana Luque-Montes, JHT Azul/Varsovia RB Güisayote Lorraine Ketzler, Ileana Luque-Montes, Melissa Medina-Flores, JHT, Larry David Wilson Copán Ruinas Lorraine Ketzler, Ileana Luque-Montes, Melissa Medina-Flores, JHT, Larry David Wilson La Esperanza area (3 localities) Lorraine Ketzler, Ileana Luque-Montes, JHT, Larry David Wilson 14 Aug–1 Oct 2008 PN Cerro Azul Meámbar: Aldea Cerro Ileana Luque-Montes, JHT Azul/Varsovia PN Cusuco César Cerrato, Ileana Luque-Montes, Melissa Medina-Flores, JHT, Larry David Wilson

135

Table 3-1. Continued. TRIP DATES LOCALITIES PARTICIPANTS PN Montaña de Yoro: above Ileana Luque-Montes, JHT, Larry David Wilson Guaymas PN Pico Pijol: Pino Alto César Cerrato, Ileana Luque-Montes, JHT, Larry David Wilson RB El Pital César Cerrato, Melissa Medina-Flores, Larry David Wilson RB Güisayote " " RB Mixicuri " " Erandique " " La Esperanza area " " 10 – 20 April 2009 PN Montaña de Comayagua: Río Sergio Gonzalez, Christina Martin, Mario Solis, JHT, Negro Rony Valle, Christopher Wolf 25 Nov – 6 Dec 2009 PN Cerro Azul Meámbar: Los Pinos César Cerrato, Vladlen Henriquez, JHT PN Cusuco " " PN Pico Bonito " " 1 – 26 Jun 2010 RVS Texíguat: La Liberación Benjamin Atkinson, César Cerrato, Luis Herrera, Mayron McKewy-Mejía, JHT, Larry David Wilson, et al. JB Lancetilla Benjamin Atkinson, César Cerrato, Luis Herrera, Mayron McKewy-Mejía, Ciro Navarro, JHT 20 Jul – 21 Aug 2010 PN Cerro Azul Meámbar: Los Pinos Anne Donnelly, Matthew Donnelly, Ileana Luque- Montes, JHT RVS Texíguat: La Liberación Levi Gray, Luis Herrera, Melissa Medina-Flores, Alexander Stubbs, JHT, etc. JB Lancetilla " " Roatán Anne Donnelly, Matthew Donnelly, Yensi Flores, Ileana Luque-Montes, Melissa Medina-Flores, Sandy Pereira, JHT Utila Anne Donnelly, Matthew Donnelly, Ileana Luque- Montes, Melissa Medina-Flores, Sandy Pereira, JHT 5 – 16 Nov 2010 PN Cerro Azul Meámbar: Los Pinos James Austin, Luis Herrera, Melissa Medina-Flores, JHT PN Montaña de Santa Bárbara James Austin, Luis Herrera, JHT, Alicia Ward, et al. San José de Texíguat James Austin, Luis Herrera, JHT 7 – 22 April 2011 PN Montaña de Botaderos Christopher Begley, Mark Bonta, Robert Hyman, David Medina, Melissa Medina-Flores, Onán Reyes, Fito Steiner, JHT PN Pico Bonito Robert Hyman, David Medina, Melissa Medina- Flores, Fito Steiner, JHT RB Colibrí Esmeralda Robert Hyman, Fito Steiner, JHT Montaña deJacaleapa Mark Bonta, Onán Reyes, JHT Rio Grande, Valle de Agalta Christopher Begley, Mark Bonta, Robert Hyman, David Medina, Melissa Medina-Flores, Onán Reyes, Fito Steiner, JHT

136

Table 3-2. Conservation status and physiographic distribution of the native non-marine herpetofauna of the Chortís Highlands. Data primarily sourced from Wilson & Johnson (2010) and supplemented by Acevedo et al. (2010), McCranie (2011a), Sunyer & Köhler (2010), Townsend & Wilson (2010a), and my own observations. Taxa followed by an asterisk (*) indicated new taxa described or being described through the course of this dissertation; details of taxonomic assignment of these species are provided in Chapter 3. IUCN Red List status follows is from the IUCN (2011; www.iucnredlist.org) when available, with other sources indicated by footnotes. Elevational Distributions are range–wide and not confined to the Chortís Highlands. Conservation Status Scores (CSS) are from Wilson & Townsend (2010a), and Environmental Vulnerability Scores (EVS) are sources cited above. Species are allocated to one of two General Distribution categories: CB = endemic to the Chortís Block, WS = widespread outside Chortís Block. Physiographic distribution is related to the Chortís Highlands (Chapter 1 provides details of each province), and includes: CL = Caribbean Lowlands physiographic province, CV = Caribbean versant intermontane valleys, NC = Northern Cordillera of the Serranía, CC = Central Cordillera of the Serranía, SC = Southern Cordillera of the Serranía, PL = Pacific Lowlands physiographic province, PV = Pacific versant intermontane valleys, IB = Islas de la Bahía; a plus (+) indicates the taxon is found within the province. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB GYMNOPHIONA Caeciliaidae (2) Dermophis mexicanus Vulnerable A2ac1 20 12 0–1500 WS + + + Gymnopis multiplicata Least Concern1 14 12 0–1400 WS + + CAUDATA Plethodontidae (43) Bolitoglossa carri Critically 3 17 1840–2070 CB + Endangered B1ab(iii)+2ab(iii) 1 Bolitoglossa cataguana* Critically 3 16 1800–2080 CB + Endangered B1ab(iii)+2ab(iii)2 Bolitoglossa celaque Endangered 3 16 1900–2820 CB + B1ab(iii) 1 Bolitoglossa conanti Endangered 5 14 950–2010 CB + + B1ab(iii) 1

137

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Bolitoglossa decora Critically 3 17 1430–1550 CB + Endangered B1ab(iii)+2ab(iii) 1 Bolitoglossa diaphora Critically 3 16 1470–2200 CB + Endangered B2ab(iii) 1 Bolitoglossa dofleini Near Threatened1 10 14 100–1550 WS + + + Bolitoglossa dunni Endangered 5 14 1020–1600 CB + B1ab(iii) 1 Bolitoglossa heiroreias Endangered 5 15 1840–2300 CB + B1ab(iii) 1 Bolitoglossa longissima Critically 3 17 1840–2240 CB + Endangered B1ab(iii) 1 Bolitoglossa mexicana Least Concern1 14 9 0–1900 WS + + + + Bolitoglossa nympha Near Threatened3 63 123 30–1400 CB + + + Bolitoglossa oresbia Critically 3 17 1560–1880 CB + Endangered B1ab(iii)+2ab(iii) 1 Bolitoglossa porrasorum Endangered 4 15 980–1920 CB + + B1ab(iii) 1 Bolitoglossa striatula Least Concern1 10 14 2–1055 WS + Bolitoglossa synoria Critically 4 15 2150–2715 CB + Endangered B1ab(iii) 1 Cryptotriton monzoni Critically 3 174 1570 CB + Endangered B1ab(iii) 1 Cryptotriton nasalis Endangered 5 15 1220–2200 CB + B1ab(iii) 1

138

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Cryptotriton wakei Critically 3 174 1150 CB + Endangered B1ab(iii) 1 Dendrotriton sanctibarbarus Endangered 3 16 1830–2744 CB + B1ab(iii) 2 [Vulnerable D21] Nototriton barbouri Endangered 3 15 1530–1920 CB + B1ab(iii)1 Nototriton brodiei Critically 3 17 875–1140 CB + Endangered B1ab(iii) 1 Nototriton lignicola Critically 3 17 1760–2020 CB + Endangered B1ab(iii) 1 Nototriton limnospectator Endangered 3 16 1640–1980 CB + B1ab(iii) 1 Nototriton picucha* Critically 33 173 1890–1920 CB + Endangered B1ab(iii)3 Nototriton saslaya Vulnerable D21 3 175 1280–1500 CB + Nototriton stuarti Data Deficient1 3 174 744 CB + Nototriton tomamorum* Critically 33 173 1550 CB + Endangered B1ab(iii)+2ab(iii)2 Nototriton sp A (Pico Bonito)* Critically 33 173 1210–1540 CB + Endangered B1ab(iii)3 Nototriton sp B (Texiguat)* Critically 33 173 1420–1800 CB + Endangered B1ab(iii)+2ab(iii)3

139

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Nototriton sp. C (Botaderos)* Critically 33 173 1700–1735 CB + Endangered B1ab(iii)3 Oedipina elongata Least Concern1 9 15 10–770 WS + + Oedipina gephyra Critically 3 16 1580–1810 CB + Endangered B1ab(iii)+2ab(iii)3 [Endangered B1ab(iii) 1] Oedipina ignea Data Deficient1 7 14 1000–2000 WS + Oedipina kasios Endangered 4 15 950–1920 CB + B1ab(iii) 2 Oedipina koehleri* Endangered 33 153 628–945 CB + B1ab(iii)3 Oedipina leptopoda Endangered 3 15 700–1300 CB + B1ab(iii) 2 Oedipina nica* Endangered 33 153 1360–1660 CB + B1ab(iii)3 Oedipina quadra Vulnerable B1ab(iii) 2 3 15 70–540 CB + + + Oedipina petiola* Critically 33 173 1580 CB + Endangered B1ab(iii)3 Oedipina stuarti Data Deficient1 6 15 0–1000 CB + + Oedipina taylori Least Concern1 8 15 140–1140 WS + Oedipina tomasi Critically 3 16 1800 CB + Endangered B2ab(iii) 1 ANURA Bufonidae (10) Incilius campbelli Near Threatened1 11 10 70–1200 WS + +

140

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Incilius coccifer Least Concern1 15 6 0–1350 WS + + + + + Incilius ibarrai Endangered 7 11 1500–1730 WS + + + B1ab(iii) 1 Incilius leucomyos Endangered 6 11 0–1600 CB + + + B1ab(iii) 1 Incilius luetkenii Least Concern1 16 7 0–1300 WS + + + + + Incilius porteri Endangered 3 13 1524–1890 CB + + B1ab(iii)3 [Data Deficient1] Incilius valliceps Least Concern1 26 5 0–2000 WS + + + + + + Rhaebo haematiticus Least Concern1 12 11 0–1300 WS + Rhinella chrysophora Endangered A2ac & 4 12 750–1760 CB + B1ab(iii,v) 1 Rhinella marina Least Concern1 35 5 0–2000 WS + + + + + + + + Centrolenidae (8) Cochranella granulosa Least Concern1 11 12 0–1500 WS + Espadarana prosoblepon Least Concern1 14 12 0–1900 WS + Hyalinobatrachium chirripoi Near Threatened1 9 12 0–700 WS + Hyalinobatrachium Least Concern1 10 12 0–1710 WS + colymbiphyllum Hyalinobatrachium fleischmanni Least Concern1 27 9 0–1730 WS + + + + + Sachatamia albomaculata Least Concern1 12 12 0–1500 WS + + + Teratohyla pulverata Least Concern1 11 12 0–950 WS + + + Teratohyla spinosa Least Concern1 9 13 0–560 WS + Craugastoridae (29) Craugastor anciano Critically 4 15 1400–1840 CB + Endangered B1ab(iii,v)+2ab(iii,v)1 Craugastor aurilegulus Endangered 6 14 50–1550 CB + + B1ab(iii,v)+2ab(iii,v)1 Craugastor bransfordii Least Concern1 10 115 20–1535 WS +

141

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Craugastor chac Near Threatened1 8 14 0–1000 WS + Craugastor charadra Endangered 7 13 30–1370 CB + + B1ab(iii, v) 1 Craugastor chrysozetetes EXTINCT1 3 17 880–1130 CB + Craugastor coffeus Critically 3 17 1000 CB + Endangered B1ab(iii) + 2ab(iii) 1 Craugastor cruzi Critically 3 17 1520 CB + Endangered A2ace, B1ab(iii,v) +2ab(iii,v)1 Craugastor cyanochthebius Critically 3 17 900–1200 CB + Endangered B1ab(iii)+2ab(iii)2 [Near Threatened1] Craugastor emleni Critically 3 14 800–2000 CB + Endangered A2ace, B2ab(v) 1 Craugastor epochthidius Critically 5 15 150–1450 CB + Endangered A3ce1 Craugastor fecundus Critically 5 15 200–1260 CB + Endangered A2ace1 Craugastor fitzingeri Least Concern1 14 13 1–1520 WS + + Craugastor laevissimus Endangered A2ace1 11 8 0–2000 CB + + + + + Craugastor laticeps Near Threatened1 13 14 10–1600 WS + + Craugastor lauraster Endangered 6 14 40–1200 CB + + + B1ab(iii) 1 Craugastor megacephalus Least Concern1 10 14 1–1200 WS + Craugastor merendonensis Critically 3 17 150–200 CB + Endangered A2ace, B1ab(v)+ 2ab(v) 1

142

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Craugastor milesi Critically 4 15 1050–1720 CB + Endangered A2ae1 Craugastor mimus Least Concern1 9 13 15–700 WS + Craugastor nefrens Data Deficient1 3 174 800–1000 CB + Craugastor noblei Least Concern1 11 13 4–1200 WS + + + Craugastor olanchano Critically 3 16 1180–1350 CB Endangered A2ace1 Craugastor omoaensis Critically 3 16 760–1150 CB + Endangered A2ace, B1ab(iii) 1 Craugastor pechorum Endangered 5 15 150–680 CB + B1ab(iii) 1 Craugastor rhodopis Least Concern1 16 14 0–1370 WS + Craugastor rostralis Near Threatened1 6 14 850–1800 WS + + Craugastor saltuarius Critically 3 16 1550–1800 CB + Endangered A2ace1 Craugastor stadelmani Critically 4 15 1125–1900 CB + + Endangered A2ace1 Eleutherodactylidae (1) Diasporus diastema Least Concern1 11 14 0–1620 WS + Hylidae (35) Agalychnis callidryas Least Concern1 17 10 0–1200 WS + + + + + Agalychnis moreletii Critically 19 13 200–2130 WS + + + Endangered A3e1 Agalychnis saltator Least Concern1 9 13 0–819 WS + Anotheca spinosa Least Concern1 11 15 95–2068 WS + Bromeliohyla bromeliacia Endangered A2ace1 8 15 900–1790 WS + Cruziohyla calcarifer Least Concern1 9 12 30–820 WS + Dendropsophus ebraccatus Least Concern1 16 11 0–1320 WS + Dendropsophus microcephalus Least Concern1 19 5 0–1200 WS + + + + + + +

143

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Duellmanohyla salvavida Critically 5 12 90–1400 CB + Endangered B2ab(iii,v) 1 Duellmanohyla soralia Critically 7 10 40–1570 CB + Endangered B2ab(iii,v) 1 Ecnomiohyla miliaria Vulnerable; 9 15 0–1330 WS + B1ab(iii)1 Ecnomiohyla salvaje Critically 5 16 1370–1520 CB + Endangered B1ab(iii) 1 Exerodonta catracha Endangered 3 13 1800–2160 CB + + B1ab(iii)+2ab(iii) 1 Isthmohyla insolita Critically 3 16 1550 CB + Endangered B1ab(iii)+2ab(iii) 1 Isthmohyla melacaena Critically 3 16 1550 CB + Endangered B2ab(iv) 2 [Near Threatened1] Plectrohyla chrysopleura Critically 4 13 930–1550 CB + Endangered A2ace, B1ab(iii,v)+ 2ab(iii,v)1 Plectrohyla dasypus Critically 3 13 1410–1990 CB + Endangered A2ace, B1ab(iii,v) +2ab(iii,v)1 Plectrohyla exquisita Critically 3 13 1490–1680 CB + Endangered A3e1

144

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Plectrohyla guatemalensis Critically 11 9 900–2800 WS + + + Endangered A3e1 Plectrohyla hartwegi Critically 9 12 925–2700 WS + Endangered A3e1 Plectrohyla matudai Vulnerable B1ab(iii) 1 8 10 700–2300 WS + + Plectrohyla psiloderma Endangered 4 12 2400–2530 CB + B1ab(iii)1 Ptychohyla euthysanota Near Threatened1 10 114 200–2200 WS + Ptychohyla hypomykter Critically 11 9 340–2070 CB + + + Endangered A3e1 Ptychohyla salvadorensis Endangered 8 11 700–2050 CB + B1ab(iii) 1 Ptychohyla spinipollex Endangered 6 11 160–1580 CB + B1ab(iii)+2ab(iii) 1 Scinax boulengeri Least Concern1 11 11 0–700 WS + Scinax staufferi Least Concern1 25 5 0–1530 WS + + + + + + + + Smilisca baudinii Least Concern1 30 4 0–1925 WS + + + + + + + + Smilisca phaeota Least Concern1 11 10 0–1116 WS + + Smilisca sordida Least Concern1 13 11 0–1525 WS + Tlalocohyla loquax Least Concern1 18 6 0–1585 WS + + + + + Tlalocohyla picta Least Concern1 15 9 0–1300 WS + + Trachycephalus venulosus Least Concern1 25 5 0–1610 WS + + + Triprion petasatus Least Concern1 12 12 0–740 WS + Leiuperidae (1) Least Concern1 Engytomops pustulosus Least Concern1 24 6 0–1540 WS + + + + + Leptodactylidae (4) Least Concern1 Leptodactylus fragilis Least Concern1 25 6 0–1700 WS + + + + + + + Leptodactylus melanonotus Least Concern1 25 6 0–1440 WS + + + + + + Leptodactylus savagei Least Concern1 15 11 0–1200 WS +

145

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Leptodactylus silvanimbus Critically 4 13 1470–2000 CB + Endangered B2ab(iii,v) 1 Microhylidae (3) Gastrophryne elegans Least Concern1 13 11 0–1500 WS + Hypopachus barberi Vulnerable B1ab(iii) 1 10 11 1300–2500 WS + Hypopachus variolosus Least Concern1 31 6 0–2200 WS + + + + + Ranidae (6) Lithobates brownorum Least Concern1 18 3 0–1200 WS + + + + Lithobates forreri Least Concern1 20 8 0–1960 WS + + Lithobates maculatus Least Concern1 16 6 40–3000 WS + + + + Lithobates vaillanti Least Concern1 21 7 0–990 WS + + + + Lithobates warszewitschii Near Threatened1 13 11 0–2500 WS + Lithobates sp. nov. Endangered CB + B1ab(iii)+2ab(iii)3 Rhinophrynidae (1) Rhinophrynus dorsalis Least Concern1 18 9 0–700 WS + + Strabomantidae (2) Least Concern1 Pristimantis cerasinus Least Concern1 9 14 19–680 WS + + Pristimantis ridens Least Concern1 12 12 0–1600 WS + + + REPTILIA TESTUDINES Chelydridae (2) Chelydra acutirostris Data Deficient3 12 13 0–1164 WS + + Chelydra rossignonii Vulnerable A2d1 11 14 0–660 WS + + Emydidae (1) Trachemys venusta Near Threatened2 26 12 0–650 WS + + Geoemydidae (4) Rhinoclemmys annulata Near Threatened2 11 13 2–920 WS [Lower risk/least concern1]

146

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Rhinoclemmys areolata Near Threatened1 12 12 0–600 WS + Rhinoclemmys funerea Near Threatened2 7 16 2–600 WS [Lower risk/least concern1] Rhinoclemmys pulcherrima Near Threatened2 19 9 0–1480 WS + + + Kinosternidae (3) Kinosternon leucostomum Least Concern2 21 9 0–1500 WS + + Kinosternon scorpioides Least Concern2 24 9 0–1500 WS + + Staurotypus triporcatus Near Threatened2 10 15 0–300 WS + [Lower risk/least concern1] CROCODILIA Alligatoridae (1) Caiman crocodilus Least Concern2 15 16 0–300 WS + + [Lower risk/least concern1] Crocodylidae (1) Crocodylus acutus Vulnerable A1ac1 22 13 0–650 WS + + + + SQUAMATA: LIZARDS Anguidae (6) Abronia montecristoi Endangered 4 15 1370 CB + + B2ab(iii) 2 [Endangered B1+2c1] Abronia salvadorensis Endangered 3 16 2020–2125 CB + B2ab(iii) 2 Celestus bivittatus Near Threatened2 7 13 1510–1980 CB + Celestus montanus Endangered 5 14 915–1372 CB + B1ab(iii)2

147

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Celestus scansorius Endangered 3 14 1550–1590 CB + + B2ab(iii)2 [Near Threatened1] Mesaspis moreletii Least Concern2 13 13 1450–3060 WS + + + Corytophanidae (7) Basiliscus plumifrons Least Concern2 11 13 0–780 WS + Basiliscus vittatus Least Concern2 25 7 0–1500 WS + + + + + + + + Corytophanes cristatus Least Concern2 20 11 0–1640 WS + + Corytophanes hernandesii Least Concern2 14 12 0–1400 WS + Corytophanes percarinatus Vulnerable B1ab(iii)2 11 14 200–2200 CB + Laemanctus longipes Least Concern2 17 9 0–1200 WS + + Laemanctus serratus Least Concern1 17 12 0–1600 WS + Eublepharidae (1) Coleonyx mitratus Least Concern2 16 10 0–1435 WS + + + + + Gymnophthalmidae (1) Gymnophthalmus speciosus Least Concern2 24 8 0–1320 WS + + + + Helodermatidae (1) Heloderma horridum Near Threatened4 11 144 100–1530 WS + [Least Concern1] Iguanidae (8) Ctenosaura bakeri Critically 3 19 0–5 CB + Endangered B1ab(i,ii,iii,v)+ 2ab(i,ii,iii,v) 1 Ctenosaura flavidorsalis Endangered 8 13 370–750 CB + + B1ab(iii,v)+2ab(iii,v)1 Ctenosaura melanosterna Critically 5 17 0–300 CB + + Endangered B1ab(iii,v) 1

148

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Ctenosaura oedirhina Critically 3 18 0–20 CB + Endangered B1ab(iii) 1 Ctenosaura palearis Endangered 3 174 150–700 CB + B1ab(i,ii,iii,iv,v)+ 2ab(i,ii,iii,iv,v) 1 Ctenosaura praeocularis Endangered 33 173 800–1000 CB + B1ab(iii,v)3 [Data Deficient1] Ctenosaura quinquecarinata Endangered 5 165 0–250 WS + + B1ab(iii,v)+2ab(iii,v)1 Ctenosaura similis Least Concern1 22 11 0–1320 WS + + + + + + + + Iguana iguana Least Concern2 26 12 0–1000 WS + + + + + + + + Phrynosomatidae (3) Sceloporus malachiticus Least Concern2 16 8 540–3800 WS + + + Sceloporus variabilis Least Concern2 25 7 0–1500 WS + + + + Sceloporus squamosus Least Concern2 14 10 0–2500 WS + + + Phyllodactylidae (3) Phyllodactylus palmeus Endangered 3 15 0–30 CB + B2ab(iii)2 Phyllodactylus tuberculosus Least Concern2 23 10 0–1230 WS + + + Thecadactylus rapicauda Least Concern2 21 10 0–1052 WS + + + + Polychrotidae: (38) Anolis allisoni Least Concern2 7 13 0–30 WS + Anolis amplisquamosus Endangered 3 16 1530–2200 CB + B2ab(iii, v)1 Anolis beckeri Least Concern2 22 11 0–1780 WS + + Anolis bicaorum Endangered 3 16 0–20 CB + B2ab(iii)2 Anolis biporcatus Least Concern2 23 10 0–2000 WS + + + + Anolis capito Least Concern2 19 11 0–1250 WS + + + +

149

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Anolis carpenteri Least Concern1 6 135 4–682 WS + Anolis crassulus Least Concern2 11 13 1200–3200 WS + Anolis cupreus Least Concern2 13 9 0–1435 WS + Anolis cusuco Endangered 3 16 1550–1935 CB + B1ab(iii) 1 Anolis heteropholidotus Endangered 4 14 1860–2200 CB + B2ab(iii)2 Anolis johnmeyeri Endangered 3 15 1340–1825 CB + B1ab(iii)2 Anolis kreutzi Critically 3 16 1670–1690 CB + Endangered B1ab(iii)+2ab(iii)2 Anolis laeviventris Least Concern2 17 9 500–2000 WS + + + Anolis lemurinus Least Concern2 26 9 0–2000 WS + + + + Anolis limifrons Least Concern2 13 12 0–1340 WS + Anolis lionotus Least Concern1 5 13 20–1200 WS + Anolis loveridgei Endangered 6 14 550–1600 CB + B1ab(iii)1 Anolis morazani* Critically 3 16 1780–2150 CB + Endangered B1ab(iii)2 Anolis muralla Vulnerable D21; 3 15 1440–1740 CB + Critically Endangered B1ab(iii)+2ab(iii)2 Anolis ocelloscapularis Endangered 4 15 1150–1450 CB + B1ab(iii)2 Anolis petersii Vulnerable B1ab(iii)2 14 13 200–2130 WS + Anolis pijolensis Critically 4 14 1180–2050 CB + Endangered B1ab(iii)2

150

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Anolis purpurgularis Endangered 3 15 1550–2040 CB + B1ab(iii)+2ab(iii)2 Anolis quaggulus Least Concern2 8 12 0–1350 WS + Anolis roatanensis Endangered 3 15 0–30 CB + B1ab(iii)+2ab(iii)2 Anolis rodriguezii Least Concern2 16 10 0–2000 WS + + Anolis rubribarbaris Critically 3 16 1700 CB + Endangered B1ab(iii)2 Anolis sminthus Endangered 4 15 1450–2200 CB + B1ab(iii)+2ab(iii)2 [Data Deficient1] Anolis tropidonotus Least Concern2 20 5 0–1900 WS + + + + + + Anolis uniformis Least Concern2 13 11 0–1370 WS + Anolis unilobatus Least Concern2 26 7 0–1200 WS + + + + Anolis utilensis Critically 3 16 0–5 CB + Endangered B1ab(iii) 2 Anolis wampuensis Endangered 3 16 95–110 CB + B2ab(iii) 2 Anolis wermuthi Vulnerable B1ab(iii)5 3 155 1230–1660 CB + Anolis yoroensis Endangered 4 14 1180–1600 CB + B1ab(iii)+2ab(iii) 2 Anolis zeus Endangered 5 14 90–900 CB + + B1ab(iii) 2 Polychrus gutturosus Least Concern2 12 12 6–700 WS + Scincidae (6) Mabuya unimarginata Least Concern2 27 7 0–1800 WS + + + + + + + Mesoscincus managuae Least Concern2 10 12 0–920 WS + Plestiodon sumichrasti Least Concern2 16 11 0–1000 WS + + Sphenomorphus assatus Least Concern2 16 13 0–2500 WS +

151

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Sphenomorphus cherriei Least Concern2 23 7 0–1860 WS + + + + + + Sphenomorphus incertus Vulnerable B1ab(iii)2 6 12 1350–1670 WS + Sphaerodactylidae (5) Gonatodes albogularis Least Concern2 20 10 0–1000 WS + + + + Sphaerodactylus dunni Vulnerable B1ab(iii)2 4 14 60–230 CB + + [Least Concern1] Sphaerodactylus glaucus Least Concern2 14 13 0–1000 WS + Sphaerodactylus millepunctatus Least Concern2 19 7 0–1000 WS + + + + + + Sphaerodactylus rosaurae Endangered 3 15 0–20 CB + B2ab(iii) 2 Teiidae (5) Ameiva festiva Least Concern2 20 10 0–1400 WS + + + + Ameiva undulata Least Concern2 27 7 0–1800 WS + + + Aspidoscelis deppii Least Concern1 22 8 0–1200 WS + + + + Aspidoscelis motaguae Least Concern1 12 9 175–1200 WS + + Cnemidophorus lemniscatus Least Concern2 9 12 0–1000 WS + Xantusiidae (2) Lepidophyma flavimaculatum Least Concern2 19 11 0–1400 WS + + + Lepidophyma mayae Vulnerable 7 13 100–800 + B1ab(iii)2 SQUAMATA: SNAKES Anomalepididae (1) Anomalepis mexicanus Data Deficient2 8 11 5–500 WS + Boidae (2) Boa constrictor Least Concern2 32 8 0–1500 WS + + + + + + + + Corallus annulatus Least Concern2 9 11 0–400 WS + Colubridae (110) Adelphicos quadrivirgatum Least Concern2 19 8 0–1740 WS + + + + + + [Data Deficient1] Amastridium sapperi Least Concern2 18 12 100–1600 WS + Chironius grandisquamis Least Concern2 12 12 0–1600 WS + + +

152

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Clelia clelia Least Concern2 18 11 0–1000 WS + + + + Coniophanes bipunctatus Least Concern2 18 11 0–1000 WS + + Coniophanes fissidens Least Concern2 28 9 0–2200 WS + + + + + Coniophanes imperialis Least Concern1 19 11 0–2000 WS + + + + + Coniophanes piceivittis Least Concern1 23 11 0–1305 WS + + + Conophis lineatus Least Concern1 22 9 0–1500 WS + + + + Crisantophis nevermanni Least Concern2 11 14 0–1385 WS + Dendrophidion clarkii Least Concern2 17 12 30–1500 WS + + + Dendrophidion percarinatum Least Concern2 13 12 4–1200 WS + + + Dendrophidion vinitor Least Concern1 18 13 15–1500 WS + + Dipsas bicolor Least Concern2 8 11 4–1100 WS + + Drymarchon melanurus Least Concern1 35 9 0–2500 WS + + + + + + + + Drymobius chloroticus Vulnerable; 18 11 500–2500 WS + + + B1ab(iii)2 [Least Concern1] Drymobius margaritiferus Least Concern2 33 7 0–2000 WS + + + + + + + Drymobius melanotropis Least Concern1 8 14 0–1400 WS + + Enuliophis sclateri Least Concern2 10 11 0–1235 WS + Enulius bifoveatus Critically 3 15 0–10 CB + Endangered B1ab(iii) 2 Enulius flavitorques Least Concern2 26 6 0–3000 WS + + + + + Enulius roatanensis Endangered 3 15 0–10 CB + B1ab(iii)+2ab(iii) 2 Erythrolamprus mimus Least Concern2 12 12 70–1400 WS + + Ficimia publia Least Concern2 18 11 0–1000 WS + Geophis damiani Critically 3 15 1075–1750 CB + Endangered B1ab(iii)+2ab(iii) 2 Geophis dunni Data Deficient1 3 165 900 CB +

153

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Geophis fulvoguttatus Endangered 6 12 1680–2200 CB + + B2ab(iii) 2 Geophis hoffmanni Least Concern2 14 12 18–670 WS + + + + Geophis nephodrymus Endangered 3 14 1560–1580 C + B2ab(iii) 2 Geophis rhodogaster Endangered 8 12 1480–2600 WS + B1ab(iii) 2 [Least Concern1] Hydromorphus concolor Least Concern2 16 9 1–1500 WS + + + Imantodes cenchoa Least Concern2 30 6 0–2063 WS + + + + Imantodes gemmistratus Least Concern2 26 10 2–1435 WS + + + Imantodes inornatus Least Concern1 12 10 5–1450 WS + + + Lampropeltis triangulum Least Concern2 36 9 0–2500 WS + + + + + + + Leptodeira nigrofasciata Least Concern2 18 10 0–1300 WS + + + Leptodeira rhombifera Least Concern3 28 8 0–2000 WS + + + + + + Leptodeira septentrionalis Least Concern2 33 9 0–2000 WS + + + + + Leptodrymus pulcherrimus Least Concern2 14 10 10–1300 WS + + + + + Leptophis ahaetulla Least Concern2 24 8 0–1680 WS + + + + Leptophis depressirostris Least Concern5 9 135 4–1120 WS + Leptophis mexicanus Least Concern2 28 8 0–1700 WS + + + + + + + Leptophis modestus Endangered 8 15 1500–2500 WS + + + B1ab(iii)2 Leptophis nebulosus Least Concern2 12 14 0–1600 WS + Masticophis mentovarius Least Concern2 32 11 0–2500 WS + + + + + + Mastigodryas alternatus Least Concern3 N/E 146 20–3006 WS + Mastigodryas dorsalis Vulnerable B1ab(iii)2 10 12 635–1900 WS + + + Mastigodryas melanolomus Least Concern1 N/E 96 0–10406 WS + + + + Ninia diademata Least Concern1 19 8 0–2200 WS + + + + Ninia espinali Endangered 5 12 1590–2242 CB + + + B1ab(iii) 2 [Near Threatened1]

154

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Ninia maculata Least Concern2 13 12 36–1800 WS + Ninia pavimentata Endangered 5 12 1300–1500 WS + B1ab(iii)2 Ninia sebae Least Concern2 28 4 0–2200 WS + + + + + + + Nothopsis rugosus Least Concern1 11 12 2–830 WS + Omoadiphas aurula Endangered 3 15 1250–1900 CB + B2ab(iii) 2 Omoadiphas cannula Critically 33 143 1250 CB + Endangered B1ab(iii)+2ab(iii)3 Omoadiphas texiguatensis Critically 3 14 1690 CB + Endangered B1ab(iii)+2ab(iii) 2 [Data Deficient1] Oxybelis aeneus Least Concern2 35 9 0–2500 WS + + + + + + + Oxybelis brevirostris Least Concern2 12 13 4–800 WS + Oxybelis fulgidus Least Concern2 27 10 0–1600 WS + + + + + Oxybelis wilsoni Endangered 3 15 0–95 CB + B1ab(iii)+2ab(iii)2 Oxyrhopus petola Least Concern2 18 13 0–800 WS + + + Pliocercus elapoides Least Concern1 25 10 0–2000 WS + + + + + Pliocercus euryzonus Least Concern1 13 14 0–1250 WS + Pseudelaphe flavirufa Least Concern1 21 12 0–1200 WS + + + Pseustes poecilonotus Least Concern1 22 12 0–1330 WS + + + + Rhadinella anachoreta Endangered 9 12 500–1180 CB + + B1ab(iii) 2 Rhadinella decorata Least Concern2 20 11 0–1400 WS + + Rhadinella godmani Vulnerable B1ab(iii) 2 13 9 1200–2200 WS + + + Rhadinella kinkelini Least Concern1; 10 12 1370–2085 CB + + + Vulnerable B1ab(iii) 2

155

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Rhadinella lachrymans Least Concern1; 12 13 500–3000 WS + Vulnerable B1ab(iii) 2 Rhadinella montecristi Endangered 7 12 1370–2620 CB + + B1ab(iii)2 Rhadinella pegosalyta Critically 3 14 CB + Endangered B1ab(iii)+2ab(iii) 2 Rhadinella rogerromani Vulnerable B1ab(iii)5 3 165 1450 CB + Rhadinella tolpanorum Endangered 3 15 1690–1900 CB + B1ab(iii)+2ab(iii) 2 Rhinobothryum bovallii Least Concern1 9 15 4–550 WS + Scaphiodontophis annulatus Least Concern2 19 12 0–1400 WS + + + + Scaphiodontophis venustissimus Least Concern2 11 11 2–830 WS + Scolecophis atrocinctus Least Concern2 14 14 100–1530 WS + + Senticolis triaspis Least Concern2 30 10 10–2500 WS + + + Sibon annulatus Least Concern2 10 12 2–1300 WS + Sibon anthracops Least Concern2 12 14 4–915 WS + + Sibon carri Endangered 8 12 30–800 CB + + B1ab(iii)2 Sibon dimidiatus Least Concern1 18 11 0–1600 WS + + + + Sibon longifrenis Least Concern2 9 11 60–750 WS + Sibon manzanaresi Critically 3 15 250–300 CB + Endangered B1ab(iii)+2ab(iii) 2 Sibon miskitus Critically 3 15 150 CB + Endangered B1ab(iii)+2ab(iii) 2 Sibon nebulatus Least Concern2 27 8 0–1690 WS + + + + Spilotes pullatus Least Concern2 30 9 0–1500 WS + + + + + + Stenorrhina degenhardtii Least Concern2 23 10 0–1900 WS + + + + + + Stenorrhina freminvillei Least Concern1 22 11 0–2000 WS + + + +

156

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Storeria dekayi Least Concern1 13 9 0–1900 WS + + + Tantilla armillata Least Concern2 14 9 0–1435 WS + + + + Tantilla impensa Least Concern1 10 12 300–1600 CB + Tantilla lempira Endangered 4 13 1450–1730 CB + B1ab(iii)+2ab(iii) 2 Tantilla sp. nov. * Critically 33 163 1150 CB + Endangered B1ab(iii)+2ab(iii)3 Tantilla psittaca Vulnerable B1ab(iii) 53 123 5–420 CB + Tantilla schistosa Least Concern2 24 10 40–1680 WS + + + Tantilla taeniata Least Concern2 13 10 0–1280 WS + + + + Tantilla tritaeniata Critically 3 15 0 CB + Endangered B1ab(iii)+2ab(iii)2 Tantillita lintoni Least Concern1 15 13 0–550 WS + Thamnophis fulvus Least Concern1 10 14 1680–3500 WS + Thamnophis marcianus Least Concern2 22 13 0–1400 WS + Thamnophis proximus Least Concern1 27 9 0–2500 WS + + + + + Tretanorhinus nigroluteus Least Concern2 18 8 0–1200 WS + + + Trimorphodon quadruplex Least Concern2 14 10 0–2000 WS + + Tropidodipsas fischeri Least Concern1 12 12 1000–3000 WS + Tropidodipsas sartorii Least Concern2 26 12 0–2000 WS + + + + Urotheca decipiens Least Concern2 10 11 15–1500 WS + Urotheca guentheri Least Concern1 13 12 25–1600 WS + Xenodon rabdocephalus Least Concern2 24 12 0–1300 WS + + + + + Elapidae (6) Micrurus alleni Least Concern2 11 15 1–1620 WS + Micrurus browni Least Concern3 15 13 0–2200 WS + Micrurus diastema Least Concern2 19 12 50–600 WS + + + Micrurus mipartitus Least Concern5 6 155 2–1160 WS + Micrurus nigrocinctus Least Concern2 21 9 0–1600 WS + + + +

157

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Micrurus ruatanus Critically 3 17 0–20 CB + Endangered B1ab(iii)1 Leptotyphlopidae (2) Epictia ater Least Concern6 33 6 0–1600 WS + + + + Epictia magnamaculata Endangered 5 9 0–25 CB + B1ab(iii)+2ab(iii) Loxocemidae (1) Loxocemus bicolor Least Concern2 16 11 0–750 WS + + + Typhlopidae (3) Typhlops costaricensis Least Concern2 11 11 540–1500 WS + + Typhlops stadelmani Endangered 6 12 320–1370 CB + + + B1ab(iii) 2 Typhlops tycherus* Data Deficient2 3 14 1550 CB + Ungaliophiidae (1) Ungaliophis continentalis Vulnerable B1ab(iii)2 11 12 990–2300 WS + + Viperidae (12) Agkistrodon bilineatus Near Threatened1 17 15 0–1500 WS + Atropoides indomitus Endangered 33 17 670–1200 CB + B1ab(iii)2 Atropoides mexicanus Least Concern2 17 12 0–1600 WS + + + + Atropoides occiduus Vulnerable4 7 154 100–1600 WS + Bothriechis marchi Endangered 5 16 500–1840 CB + + B1ab(iii, iv)2 Bothriechis schlegelii Least Concern2 21 12 0–1530 WS + + + + Bothriechis thalassinus Vulnerable 5 15 1370–1750 CB + + B1ab(iii)+2ab(iii)2 Bothrops asper Least Concern2 25 12 0–1300 WS + + + + + Cerrophidion sp. Vulnerable B1ab(iii)2 15 12 1300–2875 WS + + +

158

Table 3-2. Continued. Elevational IUCN Red List Distribution General Taxon Status CSS EVS (m) Distribution CL CV NC CC SC PL PV IB Crotalus simus Least Concern2 21 12 500–2600 WS + + + + + Porthidium nasutum Least Concern1 18 12 0–1100 WS + + + + Porthidium ophryomegas Least Concern2 15 9 0–1400 WS + + + + Sources for Red List Status: 1IUCN Red List (2011) 2Townsend & Wilson (2010) 3Evaluated for this study 4Acevedo et al. (2010) 5Sunyer & Köhler (2010) 6McCranie (2011a)

Table 3-3. Composition of the Chortís Block herpetofauna, compared with regional (Mesoamerica = MA; Wilson & Johnson 2010) and global (worldwide = WW; AmphibiaWeb 2011, Uetz et al. 2011). FAMILIES GENERA SPECIES CB MA WW CB MA CB MA WW Gymnophiona 1 1 9 2 4 2 16 188 Caudata 1 4 10 5 17 43 241 614 Anura 11 16 38 33 68 100 474 6,086 AMPHIBIA 13 21 57 40 89 145 731 6,888 Crocodylia 2 2 3 2 2 2 3 24 Testudines 4 9 14 5 19 10 55 323 Squamata 22 31 61 87 179 225 1,090 8,883 REPTILIA 28 42 78 94 200 237 1,148 9,413 TOTALS 41 63 135 134 289 382 1,879 16,301

159

Table 3-4. Broad distributional patterns of herpetofaunal diversity in the Chortís Block. CB = Chortís Block restricted, WS = widespread, CL = Caribbean Lowlands, CV = Caribbean Intermontane Valleys, NC = Northern Cordillera, CC = Central Cordillera, SC = Southern Cordillera, PL = Pacific Lowlands, PV = Pacific Intermontane Valleys, IB = Islas de la Bahia. Distribution Physiographic Regions CB WS CL CV NC CC SC PL PV IB Gymnophiona 0 2 2 2 0 0 0 1 0 0 Caudata 37 6 6 1 19 17 9 2 2 0 Anura 37 63 51 18 50 37 30 12 16 6 AMPHIBIA 74 71 59 21 69 54 39 15 18 6 Crocodylia 0 2 2 2 0 0 0 2 0 1 Testudines 0 10 6 6 0 0 0 2 1 0 Squamata: Sauria 26 61 36 31 32 21 20 18 20 13 Squamata: Serpentes 32 107 85 64 62 60 41 27 36 14 REPTILIA 57 180 129 103 94 81 61 49 57 28 TOTALS 131 251 188 124 163 135 100 64 75 34

Table 3-5. Endemic and conservation priority herpetofaunal diversity from the Chortís Block. Total % of Total % of Endemic Endemic Species Critically Species that Species that (% of endemic Endangered Endangered Vulnerable are are Families Genera Species species) Species Species Species CR/EN/VU CR/EN/VU Gymnophiona 1 2 2 0 0 0 1 50% – Caudata 1 5 43 36 (84%) 19 12 2 74% 87% Anura 11 33 100 37 (37%) 26 15 3 44% 100% AMPHIBIA 13 40 145 73 (51%) 45 27 6 54% 94% Crocodylia 2 2 2 0 0 0 1 50% – Testudines 4 5 10 0 0 0 1 10% – Squamata: 13 26 87 32 (37%) 9 23 6 44% 97% Sauria Squamata: 9 61 139 32 (23%) 10 18 9 27% 90% Serpentes REPTILIA 28 94 237 64 (27%) 19 41 17 33% 96% TOTALS 41 134 382 134 (35%) 64 68 23 41% 96%

160

Table 3-6. Distribution by mountain ranges of the endemic amphibians and reptiles of the Chortís Highlands (exclusive of Caribbean and Pacific Lowlands and islands); mountain ranges are defined in Chapter 2.

Northern Cordillera Central Cordillera Southern Cordillera

: : Texíguat Bonito : Pico

LaMuralla

Opalaca -

Taxon

Nombre Dios de Nombre Di de Nombre Omoa Santo Espíritu Joconal Bárbara Santa Meámbar Montecillos Comayagua Sulaco Flor La Agalta Botaderos Piedra Punta Patuca Montecristo Merendón Celaque Erandique Puca Sierra la De Lepaterique Dipilto Colón Dariense CAUDATA Plethodontidae (36) Bolitoglossa carri + Bolitoglossa cataguana* + Bolitoglossa celaque + + + + Bolitoglossa conanti + + + Bolitoglossa decora + Bolitoglossa diaphora + Bolitoglossa dunni + + Bolitoglossa heiroreias + Bolitoglossa longissima + Bolitoglossa nympha + + + + + Bolitoglossa oresbia + Bolitoglossa porrasorum + + + Bolitoglossa synoria + Cryptotriton monzoni + Cryptotriton nasalis + Cryptotriton wakei + Dendrotriton sanctibarbarus + Nototriton barbouri + Nototriton brodiei + Nototriton lignicola + Nototriton limnospectator + + Nototriton picucha* + Nototriton saslaya + Nototriton stuarti + Nototriton tomamorum* +

161

Table 3-6. Continued.

Northern Cordillera Central Cordillera Southern Cordillera

: : Texíguat Bonito : Pico

LaMuralla

Opalaca -

Taxon

Nombre Dios de Nombre Dios de Nombre Omoa Santo Espíritu Joconal Bárbara Santa Meámbar Montecillos Comayagua Sulaco Flor La Agalta Botaderos Piedra Punta Patuca Montecristo Merendón Celaque Erandique Puca Sierra la De Lepaterique Dipilto Colón Dariense Nototriton sp A + (Pico Bonito)* Nototriton sp B (Texiguat)* + Nototriton sp. C + (Botaderos)* Oedipina gephyra + Oedipina kasios + Oedipina koehleri* + Oedipina leptopoda + + Oedipina nica* + Oedipina quadra + + Oedipina petiola* + Oedipina tomasi + ANURA (37) Bufonidae (3) Incilius leucomyos + + + Incilius porteri + + Rhinella chrysophora + + Craugastoridae (20) Craugastor anciano + Craugastor aurilegulus + + Craugastor charadra + + Craugastor chrysozetetes + Craugastor coffeus + + Craugastor cruzi + Craugastor cyanochthebius + Craugastor emleni + Craugastor epochthidius + + Craugastor fecundus + Craugastor laevissimus + + + + + + +

162

Table 3-6. Continued.

Northern Cordillera Central Cordillera Southern Cordillera

: : Texíguat Bonito : Pico

LaMuralla

Opalaca -

Taxon

Nombre Dios de Nombre Dios de Nombre Omoa Santo Espíritu Joconal Bárbara Santa Meámbar Montecillos Comayagua Sulaco Flor La Agalta Botaderos Piedra Punta Patuca Montecristo Merendón Celaque Erandique Puca Sierra la De Lepaterique Dipilto Colón Dariense Craugastor lauraster + + + + + + + + Craugastor merendonensis + Craugastor milesi + + Craugastor nefrens + Craugastor olanchano + + Craugastor omoaensis + Craugastor pechorum + + Craugastor saltuarius + + Craugastor stadelmani + + + + HYLIDAE (13) Duellmanohyla salvavida + + Duellmanohyla soralia + + Ecnomiohyla salvaje + + Exerodonta catracha + + + + Isthmohyla insolita + Isthmohyla melacaena + Plectrohyla chrysopleura + + Plectrohyla dasypus + Plectrohyla exquisita + Plectrohyla psiloderma + + Ptychohyla hypomykter + + + + + + + + + + + + Ptychohyla salvadorensis + + + + + Ptychohyla spinipollex + + Leptodactylidae (1) Leptodactylus silvanimbus + + Ranidae (1) Lithobates sp. nov. + + + + SQUAMATA: LIZARDS (32) Anguidae (5)

163

Table 3-6. Continued.

Northern Cordillera Central Cordillera Southern Cordillera

: : Texíguat Bonito : Pico

LaMuralla

Opalaca

- a Sierra a

Taxon

Nombre Dios de Nombre Dios de Nombre Omoa Santo Espíritu Joconal Bárbara Santa Meámbar Montecillos Comayagua Sulaco Flor La Agalta Botaderos Piedra Punta Patuca Montecristo Merendón Celaque Erandique Puca l De Lepaterique Dipilto Colón Dariense Abronia montecristoi + + + Abronia salvadorensis + Celestus bivittatus + + + Celestus montanus + + Celestus scansorius + + Corytophanidae (1) Corytophanes percarinatus + + Polychrotidae: (17) Anolis amplisquamosus + Anolis cusuco + Anolis heteropholidotus + + Anolis johnmeyeri + + Anolis kreutzi + Anolis loveridgei + + + Anolis morazani* + Anolis muralla + Anolis ocelloscapularis + + Anolis pijolensis + Anolis purpurgularis + + Anolis rubribarbaris + Anolis sminthus + + + + Anolis wampuensis + Anolis wermuthi + + Anolis yoroensis + + + Anolis zeus + + + Sphaerodactylidae (1) Sphaerodactylus dunni + SQUAMATA: SNAKES (27) Colubridae (21) Geophis damiani +

164

Table 3-6. Continued.

Northern Cordillera Central Cordillera Southern Cordillera

: : Texíguat Bonito : Pico

iedra

LaMuralla

Opalaca -

Taxon

Nombre Dios de Nombre Dios de Nombre Omoa Santo Espíritu Joconal Bárbara Santa Meámbar Montecillos Comayagua Sulaco Flor La Agalta Botaderos P Punta Patuca Montecristo Merendón Celaque Erandique Puca Sierra la De Lepaterique Dipilto Colón Dariense Geophis dunni + Geophis fulvoguttatus + + Geophis nephodrymus * + Ninia espinali + + + + Omoadiphas aurula + Omoadiphas cannula + Omoadiphas texiguatensis + Rhadinella anachoreta + + Rhadinella montecristi + + + + + Rhadinella pegosalyta + Rhadinella rogerromani + Rhadinella tolpanorum + Sibon manzanaresi + Sibon miskitus + Tantilla impensa + + Tantilla lempira + Tantilla psittaca + Tantilla sp. nov. * + Typhlopidae (2) Typhlops stadelmani Typhlops tycherus* Viperidae (4) Atropoides indomitus + Bothriechis marchi + + + + + + + + Bothriechis thalassinus + + + + Cerrophidion sp. + + + + + + + + + + RANGE TOTALS 26 22 35 22 1 8 9 2 3 13 13 11 5 4 3 7 12 7 3 5 6 9 3 3 10 CORDILLERA TOTALS 72 42 37

165

CHAPTER 4 THE CHORTIS BLOCK IS AN UNDERESTIMATED HOTSPOT OF AMPHIBIAN DIVERSITY AND ENDEMISM

Over the past 20 years, amphibians have become emblematic of the myriad of environmental issues that have led to the recognized global loss of biodiversity (Wake &

Vredenburg 2008). The global phenomenon of amphibian declines and disappearances, termed the Global Amphibian Crisis, began coming to light at the First World

Herpetology Congress in 1989 as amphibian researchers met and related similar stories of unexplained declines from opposite sides of the globe, often from seemingly pristine habitats (Stuart et al. 2004; Mendelson 2011). Since that time, amphibian declines have been traced to a variety of factors, including habitat loss and fragmentation (Cushman

2006), emerging diseases such as chytridiomycosis (Berger et al. 1998; Skerratt et al.

2007) and Ranavirus (Gray et al. 2009), and environmental contamination (Reylea &

Diecks 2008). Current rates of extinction in amphibians may approach 45,000 times the background rate (Wake & Vredenburg 2008; Honeycutt et al. 2010), indicating that amphibians as a group may be facing the worst threat to their survival as a group in their 365 million-year existence. Furthermore, extinction risk is not evenly distributed globally, with species found in tropical areas, and particular tropical streams, facing the highest degree of threat (Dudgeon et al. 2006).

Paradoxically, the past 20 years has also seen the discovery and taxonomic description of amphibian species increase at a remarkable pace (Köhler et al. 2005), with number of recognized species rising 52% from 1992 (4,533; Duellman 1993) to

October 2011 (6,885; AmphibiaWeb 2011), meaning about one-quarter of all known amphibians have been described in the past two decades. A principal reason for this increase is the advent and widespread implementation of molecular approaches in

166

systematic biology, allowing for rapid evaluation of large numbers of samples and identification of candidate species (e.g. Vieites et al. 2009; Crawford et al. 2010). This rapid increase in amphibian discovery and description was documented in Köhler et al.

(2005), whose meta-analysis demonstrated that most recently described species were not simply subdivisions of described species and were as or more distinctive (as measured by DNA sequence divergence) as species described in previous eras of taxonomic activity, therefore refuting the notion that the recent rise in new amphibian discoveries is an artifact of taxonomic inflation. Rather, the increase is being fuelled by wider implementation of molecular methods in biodiversity inventory, intensified efforts to sample poorly-known areas, particularly in tropical regions, and a rise in the number of working taxonomists from developing countries, i.e. those countries that contain the majority of undescribed diversity (Joppa et al. 2011).

The juxtaposition of increasing loss of diversity and habitats with the accelerated documentation of new diversity has a number of important implications. First, this suggests that amphibians in some of the world’s most diverse regions may be going extinct without ever becoming known to the scientific community (Köhler et al. 2005;

Crawford et al. 2010). As pointed out in a number of recent works (e.g., Janzen et al.

2009; Monaghan et al. 2009; Honeycutt et al. 2010), this potential loss underscores the need for rapid taxonomic assessments across a wide taxonomic breadth. The use of a single or multiple DNA fragments (i.e. DNA barcoding) has been proposed, and widely implemented, as a standardized method for rapidly assessing taxonomic diversity

(Hebert et al. 2003).

167

DNA barcoding initially received a mixed reception among specialists and potential practitioners (Moritz & Cicero 2004), largely fuelled by a backlash to the effort of a limited number of advocates (Tautz et al. 2002, 2003) for the development of a single- marker based ―DNA taxonomy‖ for broad-scale use in the assignment of unknown samples to named taxa and the enhancement of taxonomic discovery (Lipscomb et al.

2003; Seberg et al. 2003). Even promotion of the term ―DNA barcode‖ is a point of contention and misunderstanding, given its inherent implication that each species possesses a fixed identification marker analogous to the Universal Product Code (UPC) on the side of a cereal box rather than a perpetually evolving sequence of biological molecules (Moritz & Cicero 2004).

A single fragment of the mitochondrial gene cytochrome oxidase subunit I (COI),

~650 base-pair (bp) in length, has been intended for use as a universal metazoan DNA marker, with the goal of generating a comprehensive reference dataset for use in DNA barcode-based biodiversity inventory projects (Hebert et al. 2003; Smith et al. 2008).

While COI has been demonstrated as effective for species-delimitation in mammals

(Clare et al. 2007), birds (Kerr et al. 2007), fishes (Ward et al. 2009), and a variety of invertebrates (Virgilio et al. 2010), its use as a ―universal‖ marker for amphibians has been somewhat controversial. In some groups of amphibians, COI shows a misleadingly high degree of intraspecific variation that can overlap with interspecific divergence and mask species boundaries (Vences et al. 2005a, 2005b), while in other groups is has been used effectively (Smith et al. 2008; Xia et al. 2011), demonstrating a clear gap between intraspecific and interspecific sequence divergence, i.e. the

―barcoding gap‖. A principal problem limiting the universal implementation of COI in

168

amphibians is in fact its high degree of variation, with variability in the priming regions across multiple groups leading to high rates of failure in PCR amplification (Vences et al. 2005b), which can be addressed by the use of degenerate primers (Meyer 2003) or by designing group-specific primer sets (Smith et al. 2008), however taking these steps also diminishes the ―universality‖ that is promoted as an advantage of DNA barcoding.

The proposed alternative to COI is use of a ~530 bp fragment of the large subunit rRNA gene (16S) as a ―universal amphibian barcoding gene,‖ which has a number of benefits compared with COI, including 1) conserved priming regions that amplify successfully across amphibian groups using a single set of primers (Mueller 2006), 2) reduced frequency of overlap between intraspecific and interspecific divergence

(Vences et al. 2005a, 2005b), and 3) widespread use of 16S in amphibian studies prior to widespread implementation of DNA barcoding methods, providing a relatively large and well-explored reference dataset for comparison of newly acquired sequence data

(e.g. García-París & Wake 2000).

An alternative to selecting a single marker for use in amphibian barcoding studies is to take a two-gene barcoding approach, for which I see the following advantages:

1. Distance-based analyses of COI and 16S data can be directly compared for congruence in species delimitation, reducing concerns over erroneous species identification based on use of a single marker. 2. Expansion of available reference data by allowing comparisons to sequences only available for one marker or the other. 3. Taken together, COI and 16S represent fragments of two classes (protein-coding and rRNA) of mitochondrial genes with differential rates of evolution, a preferable combination of characteristics for use in phylogenetic analysis.

As the benefits and limitations of DNA barcoding have become more clearly defined (Hebert & Gregory 2005; Hajibabaei et al. 2007), the approach has gained acceptance among systematic biologists as a useful, even powerful, addition to the

169

methods used for the exploration and description of biological diversity (e.g. Vieites et al. 2009). Rather than consider DNA barcoding as an alternative to traditional taxonomic methods, I view it as a means of providing verifiable taxonomic assignment of samples and for identifying samples in need of more rigorous evaluation; the molecular analog to the process of providing preliminary taxonomic assignments to samples morphologically using a dichotomous key. In this capacity, sequence data is evaluated using distance- based methods with the sole intention of providing a means of DNA-based taxonomic assignment and identification of potential candidate species, rather than generating hypotheses of evolutionary relationships, which is the goal of phylogenetics.

Consequently, distance-based neighbor-joining (NJ) trees used in DNA barcoding are simply graphical representations of genetic distances used to quickly assess the presence of reciporically-monophyletic sequence clusters. In cases where further analysis appears warranted, the sequence data used in the barcoding study should be supplemented with data from additional loci and subject to more rigorous phylogenetic analysis.

Scientists engaged in rapid biodiversity inventories using DNA barcoding should also be committed to following up that work, or at least tangibly supporting the follow-up to the work, with focused work in descriptive systematics. The taxonomic description of new species is considered a top priority as systematic biology moves through the era of elevation biodiversity loss, and is given the same level of importance as the biodiversity inventory work itself by Systematics Agenda 2000 (1994) and the IUCN Global

Amphibian Assessment (Parra-Olea et al. 2007). An accurate and functioning taxonomy provides the critical basis for not only conservation and management planning, but also

170

communicating about biodiversity conservation with the general public (Parra-Olea et al.

2007; Honeycutt et al. 2010). While the relative ease with which new species can be identified using molecular methods has led to the dramatic rise in known amphibian diversity, the proverbially painstaking and specialized work required to address alpha taxonomy leads to a lag between the molecular identification of potential candidate species and the formal description of newly identified taxa (Wheeler 2004).

The amphibian fauna of the Chortís Block has largely mirrored the global patterns of both decline and discovery (Townsend & Wilson 2010a). More than half (54%) of

Chortís Block amphibians are in one of the three highest IUCN threat categories

(Critically Endangered, Endangered, and Vulnerable), and include an alarming 74% of salamanders (Chapter 3, Table 3-5). Honduras, the country making up the majority of the geographic territory of the Chortís Block, has 49 endemic amphibians, 46 of which were discovered and described since 1976 (Townsend & Wilson 2010a). While the

Chortís Block has been demonstrated to have a regionally distinctive component of endemic biodiversity, particularly in amphibians (Campbell 1999; Wilson & Johnson

2010), molecular characterization of Chortís Block amphibian diversity has been restricted to a few studies of wider geographic focus and limited within-region sampling

(García-París & Wake 2000; Parra-Olea et al. 2004; Hillis & Wilcox 2005; Frost et al.

2006).

Given these considerations, I followed an iterative process to inventory and characterize phylogenetic diversity in Chortís Block amphibians. I present p-distance- based results from COI and 16S sequencing of 456 samples representing 52 named species in nine amphibian families, and use these results to identify samples using the

171

existing taxonomy and to delimit potential candidate species for future study. Among

Chortís Block amphibians, salamanders exhibit both the highest degree of endemism and extinction risk, and, based on the current taxonomy, are the most comprehensively sampled group among my data. I provide tentative taxonomic assignments for ten unidentified populations of salamanders (discussed in Chapter 3), identifying both new candidate species and new populations and morphotypes of known species. I then employ maximum likelihood-based phylogenetic analysis to evaluate evolutionary relationships among Chortís Block salamanders. Implications for taxonomy, biogeography, and conservation are discussed.

Methods and Materials

Sampling and Sample Identification

Specimens were collected in the field between 2006 and 2011 (Chapter 3 for details), with fresh tissue samples typically being preserved in SED buffer (20% DMSO,

0.25 M EDTA, pH 7.5, NaCl2 saturated; Seutin et al. 1991; Williams 2007), or 95%

ETOH. Vouchers were deposited in the Florida Museum of Natural History (FLMNH) at the University of Florida (UF), Museum of Vertebrate Zoology, University of California,

Berkeley (MVZ), Senckenberg Museum, Frankfurt am Main, Germany (SMF), and

National Museum of Natural History, Smithsonian Institution (USNM). In addition, tissue samples were deposited at the FLMNH Genetics Resources Repository in May 2009.

Species assignments for samples are based on morphological identification, and, in the case of endemic species with limited distributions, collecting locality was also used to help inform taxonomic assignments. The label ―sp. inquirenda‖ or ―sp. inq.‖ is applied to samples of uncertain taxonomic assignment, determined a priori.

172

Extraction, Amplification, and Sequencing

For salamanders, Template DNA was extracted from muscle tissue using the

Gentra PureGene Tissue Kit (QIAGEN, Valencia, CA) following manufacturer’s instructions. Concentration and purity of DNA extract was estimated using a NanoDrop

1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA).

Regions of two mitochondrial genes were selected for amplification and sequencing: 16S large subunit RNA (16S) and cytochrome oxidase subunit I (COI). The locus 16S is widely used in studies of amphibian phylogeny, and is considered the

―amphibian barcoding gene‖ (Vences et al. 2005a, Vieites et al. 2009); COI is widely promoted as the ―universal metazoan barcoding gene‖ (Hebert et al. 2003), but has not been widely used in amphibians (Smith et al. 2008, Xia et al. 2011). Fragments of the mitochondrial genes 16S and COI were amplified using the primers 16Sar-L and 16Sbr-

H for 16S (Palumbi et al. 1991) and dgLCO-1490 and dgHCO-2198 for COI (Meyer

2003). PCR reactions were typically 20L in total volume, containing ~25ng of DNA template, 4 L 5X PCR buffer (final concentration 1X), 1.2 L of 25 mM MgCl2 (final concentration 1.5 mM), 0.09 L of 10 mM dNTPs (0.045 mM), 0.8 L of each 10 M primer (0.04 M), 0.01U of GoTaq Flexi polymerase (1U/L; Promega, Madison, WI,

USA), and 11.91 L H2O. Amplification profile for 16S consisted of an initial denaturation for 3 minutes at 94°C, 35 cycles of denaturation at 94°C for 45 seconds, annealing at 50°C for 45 seconds, and extension at 72°C for 45 seconds, with a final elongation at 72°C for 5 minutes. COI used an initial denaturation for 1.5 minutes, 37 cycles of 94°C for 40 seconds, 45°C for 40 seconds and 72°C for 40 seconds, and a final elongation at 72°C for 5 minutes. PCR products were verified using electrophoresis

173

on a 1.5% agarose gel stained with ethidium bromide. Unincorporated nucleotides were removed from PCR product using 1 uL of ExoSAP-IT (USB, Santa Clara, CA, USA) per

10L of PCR product. Product was cycle sequenced forward and reverse stands using the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Inc., Carlsbad,

CA [ABI]); sequencing reactions typically included 0.8–1.2 L of PCR product, 1.1 L

5X ABI sequencing buffer, 0.22 L primer, and the balance in ddH2O for a total volume of 5 L. The sequenced product was cleaned using spin column filtration through

Sephadex (G-50; GE Healthcare, London, UK). Cleaned product was electrophoresed on an ABI 3130xl Genetic Analyzer.

Anuran samples were amplified and sequenced at the Smithsonian Institution

Laboratory of Analytical Biology (Suitland, Maryland) following standardized DNA

Barcode of Life (BOLD) protocols (Hebert et al. 2003, Borisenko et al. 2009). For these samples, Template DNA was obtained using phenol-chloroform extraction implemented by an AutoGen Geneprep 965 (AutoGen, Holiston, MA) automated DNA isolation robot.

Template DNA was then amplified for COI using the primers LCO-1490 and HCO-2198

(Folmer et al. 1994) and for 16S using the primers 16Sar-L and 16Sbr-H (Palumbi et al.

1991). Unincorporated nucleotides were removed from PCR product using 2 uL of

ExoSAP-IT per sample. Product was cycle sequenced using BigDye Terminator v3.1

Cycle Sequencing kit (ABI), cleaned using spin column filtration through Sephadex, and electrophoresed on an ABI 3730xl DNA Analyzer.

Sequence Evaluation and Alignment

Sequences were manually assembled and trimmed using CLC Combined

Workbench 3 (CLCbio, Denmark). DNA barcode identification was carried out by

174

comparing generated sequences to those in the National Center for Biological

Information (NCBI) database (http://www.ncbi.nlm.nih.gov/). Representatives from all sampled species and unassigned samples were searched against the NCBI database using BLASTN (Zhang et al. 2000) in order to identify reference sequences for inclusion in this analysis; a complete list of samples used in this analysis is presented in Table 4-

1. A large number of reference sequences were included from a recent study of

Panamanian amphibians (Crawford et al. 2010), and referenced sequences for Ranidae were from Hillis & Wilcox (2005). Samples representing ten populations of salamanders of uncertain taxonomic assignment were searched using BLASTN against the NCBI database. For use in BLASTN searching, consensus sequences were generated in CLC

Combined Workbench 3 for each set of non-divergent 16S samples representing a single species. Finally, my newly generated sequences and reference sequences identified in BLASTN searches were aligned under the default parameters using

ClustalW (Thompson et al. 1994) as implemented by the program MEGA5 (Tamura et al. 2011). Alignments were then visually inspected and adjusted. No stop codons were present in the COI sequences, indicating that these all were amplified from mitochondrial COI and not nuclear pseudogenes (Song et al. 2008, Buhay 2009).

Distance-Based Barcode Metrics

Data analysis was undertaken at various scales. First, to provide a graphical view of the family and generic level taxonomic discrimination of Chortís Block amphibians, all newly generated sequences (Caudata + Anura) were used to construct neighbor-joining

(NJ) trees in MEGA5 under the uncorrected (p-distance) model using default settings and with nodal support calculated with 1,000 nonparametric bootstrap replicates. Then, newly generated data were combined with BLASTN reference sequences and

175

separated into six datasets representing distantly-related monophyletic clades (Frost et al. 2006, Pyron & Wiens 2011): Bolitoglossa, other Plethodontidae, Bufonidae, Hylidae,

Ranidae, and a ―Terrarana+‖ dataset containing Craugastoridae, Eleutherodactylidae,

Leptodactylidae, and Strabomantidae (these four groups do not form a monophyletic group in combination). For each dataset, NJ trees were generated for both 16S and COI under the uncorrected (p-distance) method described above, with 5,000 nonparametric bootstrap replicates. Sequence divergences were estimated using uncorrected p- distances, using the default parameters in MEGA5.

Phylogenetic Analysis

Evolutionary model-based hypotheses of phylogenetic relationships were estimated for the salamander data, to provide a more rigorous evaluation of candidate species identified in p-distance-based results. The 16S and COI alignments were concatenated into a single dataset, and redundant haplotypes were removed to streamline analysis. The combined dataset then was subjected to maximum likelihood

(ML) phylogenetic analysis using the program RAxML v7.2.8 (Stamatakis 2006), performed with 1,000 bootstrap replicates under the GTR-GAMMA substitution model

(RAxML has a limited number of substitution models that can be implemented), with data partitioned by 16S, COI codon position 1, COI codon position 2, and COI codon position 3.

Species Delimitation and Candidate Species

Degrees of divergence between two sequences or groups of sequences (based on uncorrected p-distance) are herein characterized as ―shallow‖ (16S: <0.5%; COI:

<1.0%), ―moderate‖ (16S: 0.5–2.0%; COI: 1.0–6.0%), or ―deep‖ (16S: >2.0%; COI:

>6.0%). To label samples of uncertain taxonomic assignment identified a posteriori, I

176

use the terminology Confirmed Candidate Species, Unconfirmed Candidate Species, and Deep Conspecific Lineage, as proposed by Vieites et al. (2009). Unconfirmed

Candidate Species (UCS) are lineages identified as being deeply divergent genealogical lineages that have not yet been studied for differences in morphology or other characters, but may represent distinct, unnamed species. Confirmed Candidate

Species (CCS) are unnamed but moderately to deeply divergent genealogical lineages that are supported by morphological characters or other evidence of distinctiveness.

Deep Conspecific Lineages (DCL) are shallow to deeply divergent genealogical lineages that lack clear morphological or other differences with described sister species, yet represent putative evolutionary species that warrant further investigation. In general discussion, I refer to all three categories as Potential Candidate Species.

Results

Broad Results for Distance-based Analyses

Newly-generated sequence fragments of the mitochondrial genes COI and 16S were analyzed for a total of 437 individual samples representing 52 nominal species, 17 genera, and nine families of amphibians, as well as ten populations of salamanders with undetermined taxonomic assignments (Figure 4-1, Table 4-1; previously-discussed in

Chapter 3). For my 437 samples, 391 COI sequences (10.5% failure rate) and 434 16S sequences (0.7% failure rate) were successfully obtained. Average sequence lengths and nucleotide compositions are summarized in Table 4-2. In each of the six taxonomically-delimited datasets, 16S datasets included a greater number of reference sequences than COI datasets, due to widespread use of 16S as the ―universal amphibian barcode gene‖ (e.g., Zimkus & Schick 2010, Vieites et al. 2009).

177

Figure 4-1. Radial phylogram showing coverage of higher-level taxonomic groups. Family-level for anurans, genus/subgenus-level for salamanders. Phylogram generated using neighbor-joining method with uncorrected p-distances and1,000 bootstrap pseudoreplicates. Stars indicate potential candidate species identified in this study.

Analysis of both 16S and COI datasets revealed at least 36 new Potential

Candidate Species in eight of nine sampled amphibian families (Table 4-3), including 13

Confirmed Candidate Species (CCS), 17 Unconfirmed Candidate Species (UCS), and six Deep Conspecific Lineages (DCL). If both CCS and UCS are assumed to represent distinct undescribed species, then a 24.4% increase in diversity among the nine targeted families is projected. However, this study focused on just 52 named species of the 123 represented in those families, and these results represent a 57.7% increase among ingroup taxa.

178

Mean uncorrected p-distance between Anura and Caudata was 0.244 for 16S and

0.238 for COI, and mean genetic distance within Anura was 0.206 for COI and 0.146 for

16S, and within Caudata was 0.155 for COI and 0.084 for 16S. For Anuran families with more than one species-level lineage sampled, intra-family mean genetic distance for

COI was 0.122 for Bufonidae, 0.113 for Craugastoridae, 0.154 for Hylidae, and 0.130 for Ranidae, and for 16S was 0.051 for Bufonidae, 0.104 for Craugastoridae, 0.068 for

Hylidae, and 0.064 for Ranidae. Between anuran families, mean genetic distance for

COI ranged from 0.199 (Bufonidae /Leptodactylidae) to 0.269

(Eleutherodactylidae/Ranidae), and for 16S from 0.128 (Hylidae/Leptodactylidae) to

0.278 (Craugastoridae/Ranidae). All samples of Caudata represent a single family,

Plethodontidae, which accounted for 36% of all COI and 38% of all 16S sequences analyzed (Table 4-1), far more representation than any other family. Analysis of the

Caudata dataset recovered the same deeply-diverged higher-order clades (family, genus, subgenus) as published phylogenetic hypotheses (García-París & Wake 2000,

Parra-Olea et al. 2004, Wiens et al. 2007, McCranie et al. 2008), however relationships within these clades are poorly resolved (Figure 4-2).

Identification of Potential Candidate Species of Anura

Twenty new Potential Candidate Species of anurans were identified through uncorrected p-distance analysis of the COI and 16S data (Table 4-3). A comprehensive comparative dataset for 16S is available for Central American Bufonidae, based largely on the work of Mendelson et al. (2005), Mendelson & Mulcahy (2010), Mulcahy &

Mendelson (2000), and Mulcahy et al. (2006). Reference sequences for COI were limited to Panamanian samples assigned to Incilius coniferus and Rhaebo haematiticus

(Crawford et al. 2010), and two samples of I. nebulifer, the sister species to I. valliceps.

179

Figure 4-2. Maximum likelihood phylogram of COI data showing generic and subgeneric relationships of salamander samples.

180

Samples from this study were shown to be conspecific with the nominal species I. coccifer (16S=0.2%), I. ibarrai (16S=0.3%; COI=0.5%), I. leucomyos (16S=0.2%;

COI=0.7%), I. luetkenii (16S=0.4%), I. porteri (16S=0.5%; COI=0.8%), and I. valliceps

(16S=0.8%; COI=0.7%). Two Unconfirmed Candidate Species were identified among the Bufonidae data: I. cf. coniferus and R. cf. haematiticus (Figure 4-3).

The majority of Craugastoridae samples represented taxa that lacked comparative reference sequences; only one sample of Craugastor fitzingeri (N060) was confirmed by

COI and 16S data (Figure 4-4). Most clades correspond with a priori taxonomic assignments. Three Unconfirmed Candidate Species were identified: a divergent population of C. cf. aurilegulus; and two from northern Nicaragua, C. sp. inquirenda 1 and C. sp. inquirenda 2 (Figure 4-4).

Reference sequences for the genus Diasporus (Eleutherodactylidae) were available for four putative species (Crawford et al. 2010). Four samples of Diasporus cf. diastema from northern Nicaragua form a well-supported clade based on 16S and the single COI sequence is also deeply divergent (16S=7.1%; COI=15.7%) from the closest available reference sequences (Fig . 4-4), representing an Unconfirmed Candidate

Species awaiting comparative morphological evaluation.

While a large number of congeneric reference sequences are available for 16S, and to a lesser extend COI, all newly generated samples of Hylidae represented either the first sequence data available for a putative species for both COI and 16S (the case for at least 10 taxa), or a Potential Candidate Species assigned to another taxon.

Purported congeners tended to cluster together, except in the cases of Ptychohyla salvadorensis, two species of Duellmanohyla, and two species of Exerodonta in the 16S

181

Figure 4-3. COI (left) and 16S (right) neighbor-joining trees for Bufonidae. ** = boostrap support ≥ 99, * = boostrap support ≥ 90; candidate species indicated with red stars.

182

Figure 4-4. COI (left) and 16S (right) neighbor-joining trees for Craugastoridae, Eleutherodactylidae, Leptodactylidae, and Strabomantidae. ** = boostrap support ≥ 99, * = boostrap support ≥ 90; candidate species indicated with red stars.

183

NJ tree (Figure 4-5). At least five Unconfirmed Candidate Species were identified in the

16S NJ tree and supported by the COI NJ tree (Figure 4-5, Table 4-3): a Plectrohyla, two Ptychohyla, and two Smilisca.

Samples representing a single putative species, Leptodactylus fragilis

(Leptodactylidae), were the only members of this family included in this study. Four samples from Honduras and Nicaragua are deeply divergent (16S=3.6%; COI=15.7%) from that of Panamanian samples assigned to L. fragilis (Figure 4-4).

Only two 16S reference sequences provided taxonomic matches: Lithobates macroglossa and L. taylori (Figure 4-6). Two GenBank samples are referred to L. macroglossa; one sample clusters with samples from Caribbean versant localities in

Honduras that were identified as L. brownorum; the second L. macroglossa sample forms a weakly supported sister lineage to the L. taylori clade (Figure 4-6). I tentatively consider the latter sequence to represent L. macroglossa, while the widespread clade represents L. brownorum. Two series of samples, one from Cerro Uyuca and one from

San Pedro La Loma, that were considered hybrids between L. brownorum and L. forreri by McCranie & Wilson (2002) form well-supported clades for both COI and 16S (Figure

4-6). The 16S data places these as sister to a sample representing an undescribed highland leopard frog from Costa Rica (Lithobates sp. 5 sensu Hillis & Wilcox 2005).

Among samples thought to represent the highland and piedmont stream-inhabiting L. maculatus, considerable genetic diversity is seen in both COI and 16S data, and this taxon apparently masks at least four Unconfirmed Candidate Species in the Chortís

Highlands (Figure 4-6). Samples assigned to L. warszewitschii from northern Nicaragua

184

Figure 4-5. COI (left) and 16S (right) neighbor-joining trees for Hylidae. ** = bootstrap support ≥ 99, * = boostrap support ≥ 90; candidate species indicated with red stars.

185

Figure 4-6. COI (left) and 16S (right) neighbor-joining trees for Ranidae. ** = boostrap support ≥ 99, * = boostrap support ≥ 90; candidate species indicated with red stars.

186

also a deeply divergent clade from L. warszewitschii samples from Panamá (Figure 4-

6).

Within the family Strabomantidae, samples from northern Nicaragua assigned to

Pristimantis ridens are shown to represent an Unconfirmed Candidate Species, deeply divergent (16S=4.8%; COI=9.1%) from Panamanian samples referred to P. ridens, but in themselves represent three candidate species (Figure 4-4).

BLASTN Results for Unassigned Salamander Sequences

Two of ten unassigned salamander populations matched available 16S references sequences from NCBI: Nototriton sp. inquirenda 4 with N. lignicola, and Oedipina sp. inquirenda 1 with O. kasios (Table 4-4). Both unidentified samples were from mixed cloud forest in Parque Nacional (PN) Montaña de Yoro and matched with species previously considered endemic to nearby PN La Muralla. A third population from the same localities in PN Montaña de Yoro, Bolitoglossa sp. inquirenda 1, also matches its counterpart from PN La Muralla, B. decora (Table 4-4), however these species exhibit markedly different external morphology, discussed in further detail below. The remaining seven salamander populations do not closely match with any sequences in

NCBI database (≤97% max ident by BLASTN).

Distance-Based Analyses of Caudata

A relatively comprehensive 16S reference dataset is available for Mexican and

Central American salamanders (García-París & Wake 2000, Parra-Olea et al. 2004,

Wiens et al. 2007, Adams et al. 2009); however, only five COI sequences (two

Bolitoglossa, two Oedipina, and one Nototriton) were available through NCBI. Intra- generic divergence was lowest for Nototriton (16S=3.5%, n=25; COI=9.7%, n=16), compared with Bolitoglossa (16S=5.5%, n=192; COI=12.7%, n=114), Cryptotriton

187

Figure 4-7. COI (left) and 16S (right) neighbor-joining trees for Bolitoglossa. ** = boostrap support ≥ 99, * = boostrap support ≥ 90; candidate species indicated with red stars.

188

(16S=5.1%, n=5) and Oedipina (16S=7.5%, n=50; COI=13.5%, n=10). Between salamander genera, mean genetic divergence for COI ranged from 20.5%

(Nototriton/Oedipina and Cryptotriton/Nototriton) to 21.9% (Cryptotriton/Oedipina), and for 16S ranged from 11.3% (Nototriton /Oedipina) to 20.2% (Bolitoglossa/Cryptotriton).

New samples were recovered as conspecific with the following nominal species

(intraspecific divergence indicated parenthetically): Cryptotriton veraepacis (16S =0.0%;

COI=0.0%), Nototriton limnospectator (16S=0.3%; COI=0.2%), N. lignicola (16S=0.2%),

Oedipina gephyra (16S=0.1%; COI=0.0%), O. kasios (16S=0.1%), Bolitoglossa celaque

(16S=1.5%; COI=1.9%), B. conanti (16S=1.6%; COI=1.9%), B. diaphora (16S=0.0%;

COI=0.1%), B. dofleini (16S=0.3%; COI=0.1%), B. dunni (16S=0.3%), B. heiroreias

(16S=0.0%), B. longissima (16S=0.3%; COI=1.2%), B. mexicana (16S=2.1%;

COI=1.2%), B. nympha (16S=3.0%; COI=1.4%), and B. synoria (16S=0.1%;

COI=0.0%).

Sixteen Potential Candidate Species of salamanders were identified in the 16S and COI data (Table 4-2). Populations assigned to B. porrasorum and N. barbouri from

Texíguat and Pico Bonito are divergent from each other at an interspecific level, and are deeply divergent from samples representing the terra tipica of each nominal taxa

(Figures 4-7, 4-8). Samples from the Cordillera de Agalta (Nototriton sp. inq. 2) and

Sierra de Botaderos (Nototriton sp. inq. 3), representing the first known samples of

Nototriton from either mountain range, are sister clades (intraclade divergence 0.0% for

16S and 0.2–0.6% for COI; interclade divergence 1.7% for 16S and 0.29% for COI) and are both Confirmed Candidate Species (CCS). Other CCS include a divergent sample

189

Figure 4-8. COI (left) and 16S (right) neighbor-joining trees for Cryptotriton, Dendrotriton, Nototriton, and Oedipina. ** = boostrap support ≥ 99, * = boostrap support ≥ 90; candidate species indicated with red stars. from Texíguat (Nototriton sp. inq. 1), Oedipina cf. gephyra from Pico Bonito, Oedipina sp. inq. 2 from the Cordillera de Dariense in Nicaragua, Oedipina sp. inq. 3 from the southeastern piedmont of the Chortís Highlands in Nicaragua, and Bolitoglossa cf. rufescens from the northern foothills of the Sierra de Omoa. Clusters of geographically-

190

discrete, moderately divergent subclades (within subclade divergence 16S=0.0–0.8%;

COI=0.0–0.3%; between subclade divergence 16S=0.6–1.6%; COI=1.7–3.6%) were also present in samples assigned to B. celaque and B. conanti (Figure 4-7), which have not previously been differentiated morphologically (McCranie & Wilson 2002) and therefore are considered Deeply Conspecific Lineages pending further evaluation of morphological variation.

Phylogenetic Analysis of Caudata

After removing identical redundant haplotypes, data from 16S and COI were concatenated into a single dataset consisting of 276 individual samples, with a total aligned sequence length of 1,178 bp. Maximum likelihood (ML) analysis recovered a topology largely congruent with previous analyses of available reference data (Figure 4-

9; García-París and Wake 2000, Parra-Olea et al. 2004, Wiens et al. 2007, McCranie et al. 2008).

For Bolitoglossa, clades representing the subgenera Bolitoglossa, Eladinea,

Magnadigita, Nanotriton, and Pachymandra were recovered as monophyletic (Figure 4-

9). A clade corresponding to the Bolitoglossa dunni species group was also recovered, containing the nominal taxa B. carri, B. cataguana, B. celaque, B. conanti, B. decora, B. diaphora, B. dunni, B. flavimembris, B. heiroreias, B. longissima, B. morio, B. oresbia,

B. porrasorum, B. cf. porrasorum (Pico Bonito), B. cf. porrasorum (Texíguat), and B. synoria (Figure 4.9). Within the B. dunni group, four phylogenetic divisions are apparent, each corresponding to a discretely distributed group of species: Clade A (B. celaque, B. heiroreias, B. synoria; across the ignimbrite highlands of the Southern

Cordillera), Clade B (B. carri, B. conanti, B. diaphora, B. dunni, B. oresbia; from the

191

Figure 4-9. Maximum likelihood phylogram for the genus Bolitoglossa. Combined 16S/COI dataset, partitioned by gene with 1,000 bootstrap replicates.

192

Sierra de Omoa southward to the Sierra del Merendón, then eastward to the Sierra de

Lepaterique and north again to the Montañas de Meámbar), Clade C (B. cataguana, B. decora, B. longissima, B. porrasorum, B. cf. porrasorum [Pico Bonito], B. cf. porrasorum

[Texíguat]; the highlands of northern and north-central Honduras east to the Sierra de

Agalta), and Clade D (B. flavimembris, B. morio; Guatemala).

Taxonomic discrepencies in distance-based analyses (Figure 4-7) were confirmed by ML analysis (Figure 4-9). Samples assigned to B. celaque from three mountain ranges in the Southern Cordillera are recovered as an eastern clade (La Paz + Intibuca) and a western clade (Celaque), with the La Paz samples forming a shallow sub-clade within the eastern clade (Figure 4-9). Samples assigned to B. conanti also form a clade of geographically-discrete, shallowly-divergent, monophyletic groups corresponding to the Sierra de Espíritu Santo, Sierra del Merendón, Sierra de Omoa, and Cerro del

Mono, with the B. conanti complex clade recovered as sister to B. carri, itself endemic to the Sierra de Lepaterique (Figure 4-9). Samples from the B. rufescens complex (sensu

McCranie & Wilson 2002) are assignable to either B. nympha or to an unnamed lineage

(Figure 4-9).

Clades corresponding to the putative genera Nototriton, Oedipina, and Cryptotriton were supported in the ML analysis (Figure 4-10). For Nototriton, clades representing the

N. barbouri group, N. picadoi group, and N. richardi group were recovered as monophyletic (Figure 4-10). As with Bolitoglossa, phylogenetic results were congruent with distance-based results in terms of delimiting species-level lineages (Figures 4-8, 4-

10). Nototriton sp. inq. 1 was shown to be a phylogenetically distinctive and deeply divergent lineage, sister to the entire N. barbouri group. Together, the Nototriton sp. inq.

193

Figure 4-10. Maximum likelihood phylogram for the genera Nototriton, Oedipina, Dendrotriton, and Cryptotriton. Combined 16S/COI dataset, partitioned by gene with 1,000 bootstrap replicates.

194

1 and N. barbouri group clades represent an radiation endemic to the Chortís

Highlands. Inclusion of material from the type locality of N. barbouri in phylogenetic analysis revealed the taxon to be widely paraphyletic with respect to populations from

Pico Bonito and Texíguat (Figure 4-10). Nototriton barbouri sensu stricto from the vicinity of the type locality is the sister species to N. limnospectator, while the Pico

Bonito and Texiguat populations (themselves forming two distinct monophyletic lineages) form a clade with N. brodiei from the highlands along the

Guatemala/Honduras border. Nototriton sp. inquirenda 4 and N. sp. inquirenda 5 from central Honduras are shown to be conspecific with N. limnospectator and N. lignicola, respectively. In reference to the N. picadoi group, N. saslaya was recovered as the sister lineage to the remaining species of the group, and represents the only representative of the N. picadoi group found in the Chortís Block.

For Oedipina, clades corresponding to the subgenera Oedipina, Oedipinola, and

Oeditriton were recovered as monophyletic (Figure 4-10). Samples of Oedipina from five localities in northern Nicaragua were included, and represent two unnamed lineages, O. sp. inquirenda 2 and O. sp. inquirenda 3; one in the Oeditriton clade and one in the Oedipina clade. Oedipina sp. inquirenda 2, found at three isolated mountain localities in northern Nicaragua, is a well-supported sister lineage to the other highland dwelling species of Oeditriton, O. kasios, with the lowland form O. quadra remaining as sister to the two highland lineages. The second Nicaraguan candidate species, O. sp. inquirenda 3, is sister to a clade containing O. cyclocauda and O. pseudouniformis

(Figure 4-10). The nominal species O. gephyra, from Parque Nacional Pico Bonito and

195

Reserva de la Vida Silvestre Texiguat, contains two monophyletic lineages (one at each locality), and is itself a monophyletic group. Oedipina tomasi is the sister taxon to the

Oedipina gephyra complex, forming a monophyletic clade that is sister to the remaining

Oedipinola. A sample from a previously unknown allopatric population of Oedipina from central Honduras, O. sp. inquirenda 1, is shown to be conspecific with O. kasios (Figure

4-10).

Discussion

Candidates for Further Taxonomic Study Among Anurans

Undescribed diversity in the Chortís Highlands is both geographically and phylogenetically widespread. While a disproportionately large share of potential candidate species are salamanders (n=16; 44%), unexpected diversity is present in all families included in this study, with the exception of Microhylidae (containing three widespread species that occur in the Chortís Block). Not surprisingly, the majority of

Potential Candidate Species have distributions limited to one or a few localities in the

Chortís Highlands, however even lowland anuran taxa heretofore considered to be widespread throughout Central America (Köhler 2011), such as Diasporus diastema,

Leptodactylus fragilis, Pristimantis ridens, Lithobates warszewitschii, Rhaebo haematiticus, and Smilisca baudinii, demonstrate deep divergence indicative of species- level diversification (Table 4-3).

Below I present a brief account of each Potential Candidate Species, as well as unassigned species identified through this study, providing details of distribution and taxonomy relative to nominal taxa.

196

Incilius coccifer/ibarrai/porteri

The toads of the Incilius coccifer species complex include three apparently parapatric species in Honduras, although the distribution and taxonomic status of these species are disputed by authorities (Mendelson et al. 2005; McCranie 2009). The taxonomy of these populations has long been recognized as inadequate in terms of characterizing the existing diversity (McCranie & Wilson 2002), and Mendelson et al.

(2005) revised the Incilius coccifer complex, including resurrection of the taxon I. ibarrai

(Figure 4-11A) and description of a new taxon, I. porteri (Figure 4-11B), to accommodate premontane and highland populations of I. coccifer complex toads in the

Chortís Highlands. A leading authority on Honduran amphibians, James McCranie, steadfastly has refused to recognize these taxa as valid (McCranie & Castañeda 2007;

McCranie 2006, 2009), stating that ―despite the recent review of the I. coccifer complex by Mendelson et al. (2005), a thorough review of this complex is needed that includes all known populations,‖ citing ―the reasons discussed by McCranie & Castañeda

(2007)‖, a Spanish-language field guide the Honduran amphibian fauna. McCranie provided the following principal reasons for not recognizing the validity of I. ibarrai and I. porteri:

1. ―Numerous problems‖ with the dissertation work of Mendelson (in which Mendelson presented morphological analyses of the I. coccifer complex) detailed by ―McCranie (in McCranie & Wilson 2002)‖(as stated in McCranie & Castañeda 2007:125), and the lack of address given to those concerns in Mendelson et al.’s (2005) monograph.

2. The single sample of Honduran I. coccifer sensu stricto included by Mendelson et al. (2005) is from the mesic portion of eastern Honduras, which McCranie (by inference) considers distinctive from populations of the Pacific lowlands and therefore not representative of typical I. coccifer.

3. The lack of inclusion of ―large specimens‖ from southern Honduras in the work of Mendelson et al. (McCranie 2009:4).

197

Results from my 16S and COI data support recognition of both I. ibarrai and I. porteri as valid species (Figure 4-3), and also demonstrate that populations of I. coccifer from eastern Honduras are indeed conspecific with those from the Pacific lowlands and apparently have a continuous distribution through subhumid areas around the southern terminus of the Nuclear Central American highlands in northern Nicaragua.

The distribution of these taxa is more complicated and intertwined than was initially recognized by Mendelson et al. (2005), and includes at least one area of sympatry or near-sympatry between I. ibarrai and I. porteri on the southern side of

Parque Nacional Cerro Azul Meámbar in Departamento de Comayagua (CAC044, I. porteri from Río Bonito, 1,570 m elevation; and IRL002 and IRL 004-05, I. ibarrai from

San Jose de Los Planes, 1,290 m elevation, approximately 6 airline km east-northeast of Río Bonito, and JHT 2205 (Figure 4-11A) from Cerro Zarciadero, 1,835 m, approximately 6 km south of Río Bonito). Even more curious, samples representing both I. coccifer and I. porteri were collected on Isla del Tigre in the Golfo de Fonseca, with I. coccifer (JHT3301) know from 190 m elevation and I. porteri (JHT 3302) known from 520 m elevation. Clearly, further sampling and investigation is needed to determine the extent of each species’ distribution, identify areas of sympatry, and explore ecological or behavioral mechanisms isolating these closely related species.

Incilius coniferus

The toads referred to the taxon Incilius coniferus are distributed from northern

Nicaragua (Travers et al. 2011) south to the Pacific versant of Colombia and northern

Ecuador (Savage 2002). A single sample from Bosawas (Figure 4-11C) is included in a clade with the holotype of I. karenlipsae (Mendelson & Mulcahy 2010) and eight samples referred to as I. coniferus (Figure 4-3). In comparisons with the available COI

198

Figure 4-11. Candidate Species I: Bufonidae and Craugastoridae. A) Incilius ibarrai, Cerro Zarciadero, 1,800 m (Photo © J.H. Townsend [JHT]). B) I. porteri, Cerro Uyuca 1,640 m. C) I. cf. coniferus, Biosfera Bosawas, 180 m (Photo © Scott Travers). D) Rhaebo cf. haematiticus, Biosfera Bosawas, 180 m (Photo © Javier Sunyer). E) Craugastor cf. aurilegulus, adult female, San José de Texíguat, 200 m (Photo © JHT). F) C. cf. aurilegulus, adult male, same data as E (© JHT); G) C. laevissimus, Río Negro de Comayagua, 1,220 m (Photo © JHT). H) C. laevissimus, Cerro Azul Meámbar, 800 m (Photo © JHT).

199

data, the Bosawas sample (N416) is a strongly supported, distant sister to eight other samples from Panamá (Figure 4-3).

Rhaebo haematiticus

This species is considered a relatively common and widespread inhabitant of lowland and premontane rainforest from eastern Honduras to northern South America

(McCranie et al. 2006, Savage 2002). The Nicaraguan samples of Rhaebo haematiticus

(Figure 4-11D) forms a well-supported (bootstrap ≥99) reciprocally monophyletic sister clade to Panamanian reference samples for both COI and 16S (Figure 4-3).

Craugastor aurilegulus

This species is currently recognized as endemic to the Cordillera Nombre de Dios.

Samples of C. aurilegulus from the vicinity of the type locality (Pico Bonito) and the vicinity of Texíguat (Figures 4-11E, 4-11F) form well supported (bootstrap ≥99) clades for both COI and 16S (Figure 4-4), and accordingly the Texíguat population is assigned

Deep Conspecific Lineage status and warrants further investigation to clarify its taxonomic status.

Craugastor laevissimus

These data confirm the species C. laevissimus as a widespread member of the C. rugulosus group with a distribution congruent with that defined by Campbell & Savage

(2000). Samples from the vicinity of the type locality in the Sierra de Omoa (JHT1824) and from throughout the serranía in Honduras (Figures 4-11G, 4-11H; JHT series samples) and into northern Nicaragua (N series samples) demonstrate relatively low divergence between samples and no outwardly clean pattern of haplotype distribution

(Figure 4-4).

200

Craugastor sp. inquirenda 1 & 2

Two Unconfirmed Candidate Species (Figure 4-12A) from northern Nicaragua form a well-supported (bootstrap ≥99) but deeply-divergent clade with C. bransfordii samples from Panamá (Figure 4-4). It may be that one of these populations will be assigned to C. lauraster, a small species known from eastern Honduras and northern

Nicaragua (Köhler 2011), following investigation of morphological affinities of these samples.

Diasporus diastema

Only a single putative species of Eleuthodactylidae, D. diastema, is considered to naturally occur in northern Central America (Köhler 2011, McCranie et al. 2006). The type locality of D. diastema is at the extreme eastern edge of its distribution in the

Panama Canal Zone, and samples from northern Nicaragua (Figure 4-12B) clearly represent a monophyletic lineage divergent from Panamanian samples (Figure 4-4).

Further investigation is needed, preferably with inclusion of samples in the dataset from eastern Honduras, southern Nicaragua, and Costa Rica.

Plectrohyla cf. guatemalensis

The taxon P. guatemalensis has long been recognized as a potential composite taxon masking undescribed diversity (Duellman & Campbell 1992: 8, McCranie &

Wilson 2002: 301), and the ability to study the phylogenetic relationships of populations assigned to this taxon has suffered from declines and disappearances of this cloud forest treefrog from many localities where it previously was abundant. I have only encountered one population of treefrogs assigned to P. guatemalensis (type locality:

Patzicia, Departamento de Chimaltenango, Guatemala; Stuart 1963): cloud forest above 1,750 m in Parque Nacional Montaña de Yoro (Figures 4-12C, 4-12D). In both

201

the 16S and COI NJ trees, this population is recovered as a divergent clade, and below the intra-generic level does not demostrate any particular relationship to the single reference available sequence (Figure 4-4). Populations assigned to this species are known from across the Chortís Highlands (Duellman 2001), and a lack of representative sampling will likely continue to hinder phylogeny-based taxonomic evaluation.

Ptychohyla hypomykter

Samples from across the serranía in Honduras and Nicaragua, including the type locality at Quebrada Grande, appear to represent a single widespread species.

However single reference sequence assigned to this taxon (ENS8486 from Izabal,

Guatemala; Faivovich et al. 2005) is not recovered as part of the P. hypomykter clade, nor does it cluster closely with any other species of Ptychohyla, representing an

Unconfirmed Candidate Species (Figure 4-5).

Ptychohyla spinipollex

This species is presently considered endemic to the vicinities of Texíguat and Pico

Bonito in the Cordillera Nombre de Dios. A single 16S reference sequence from

Quebrada de Oro in Parque Nacional Pico Bonito was surprising divergent from a series of samples from La Liberación in Refugio de Vida Silvestre Texíguat, well into the range normally applied to intraspecific relationships (Figure 4-5). Given the large genetic distance between the Pico Bonito sample and the 18 samples from Texíguat, it is somewhat surprising to see that the extreme variability in color pattern variation exhibited by P. cf. spinipollex from Texíguat (Figure 4-12E–H) was not reflected by genetic variation in 16S and COI.

202

Figure 4-12. Candidate Species II: Eleutherodactylidae and Hylidae. A) Craugastor sp. inquirenda 1, Cerro Saslaya, 1,000–1,600 m (Photo © Javier Sunyer). B) Diasporus cf. diastema, Biosfera Bosawas, 180 m (Photo © Javier Sunyer). C) Plectrohyla cf. guatemalensis, adult female, Montaña de Yoro, 1,820 m (Photo © J.H. Townsend [JHT]). D) P. cf. guatemalensis, adult male, Montaña de Yoro, 1820 m (Photo © JHT). E–G) Ptychohyla cf. spinipollex, La Liberación de Texíguat, 1,080–1,120 m. (Photo © JHT).

203

Smilisca baudinii

This large treefrog is considered something of a ―trash‖ species that occurs from

México to Panamá, and is often abundant around areas of human disturbance. It is likely that the disregard shown this taxon in the past, despite indications that there is somewhat considerable geographically-linked morphological variation across its distribution (Duellman 2001), has led to an underestimation of species-level diversity.

Both 16S and COI data support this notion, with two potential candidate species (one in

Honduras, one in northern Nicaragua) that are relatively deeply divergent from available reference sequences assignable to S. baudinii (Figure 4-5).

Leptodactylus fragilis

This species is currently considered to have a range extending from Texas to

Venezuela (Heyer 2002), with a type locality of ―Tehuantepec, Mexique‖ (Brocchi 1877).

Samples assigned to L. fragilis from Honduras and Nicaragua form a clade that is deeply divergent with respect to Panamanian samples referred to L. fragilis (Figure 4-4).

Given that the type locality is in México but remains unsampled, I will continue to refer to Honduran and Nicaraguan samples as L. fragilis, and consider Panamanian samples to be an Unconfirmed Candidate Species pending further investigation.

Lithobates brownorum X forreri

An unresolved taxonomic problem in the Chortís Highlands relates to the status of populations of Lithobates that occur at high elevation ponds and wetlands in the

Southern Cordillera of the Chortís Highlands. McCranie and Wilson (2002: 478–479) considered these frogs to represent morphological intermediates that possess a blend of diagnostic features from both L. brownorum on the Caribbean versant and L. forreri on the Pacific versant, and judged these intermediate populations to represent hybrids

204

between the lowland forms L. brownorum and L. forreri (Figures 4-13A, 4-13B). These high-elevation pond-breeding frogs apparently have an extended larval period, with tadpoles regularly reaching or exceeding the size of adult males (Figure 4-13B). A total of 15 samples were analyzed , eight from Cerro Uyuca in Departamento de Francisco

Morazán and seven from San Pedro La Loma in Departamento de Intibucá (Table 4-1).

All samples are supported as a single clade representing a Potential Candidate Species

(Figure 4-6). Analysis of 16S places this candidate species as sister to an undescribed species from the central highlands of Costa Rica (Rana sp. 5 sensu Hillis & Wilcox

2005). The addition of data from the mitochondrial gene 12S would allow for these

Lithobates sp. inquirenda samples to be directly compared to a geographically comprehensive dataset of Mexican Lithobates (Zaldívar-Riverón et al. 2004) in addition to that of Hillis & Wilcox (2005), and should further elucidate the relationships of these

Honduran highland populations.

Lithobates brownorum

Examination of the Honduran ―hybrid‖ problem calls into focus a series of unresolved taxonomic problems related to Central American frogs of the L. berlandieri group. The identity of lowland forms from the Caribbean versant of Honduras and northern Nicaragua, variously referred to as L. berlandieri (McCranie & Wilson 2002) and L. brownorum (after Zaldívar-Riverón et al. 2004), is unconfirmed due to a lack of comparative sequence data overlapping this dataset and both those of Zaldívar-Riverón et al. (2004) and Hillis & Wilcox (2005).

Lithobates forreri

The status of the lowland frogs from the Pacific versant is even less clearly defined, and they have been assigned to L. forreri from México to Costa Rica (McCranie

205

Figure 4-13. Candidate Species III: Ranidae and Strabomantidae. A) Lithobates sp. 1 (―brownorum X forreri‖) Cerro Uyuca, 1,640 m (Photo © Jason Butler). B) Lithobates sp. 1 (―brownorum X forreri‖), comparing tadpole and adult male, Cerro Uyuca, 1,640 m (Photo © Jason Butler). C) Lithobates cf. maculatus, adult male, Cerro Saslaya, 1,000–1,600 m (Photo © Javier Sunyer). D) L. cf. maculatus, adult male, Cerro Saslaya, 1,000–1,600 m (Photo © Javier Sunyer). E) L. cf. maculatus, adult female, Montaña de Yoro, 1,820 m (Photo © J.H. Townsend). F) Lithobates cf. warszewitschii, Biosfera Bosawas, 180 m (Photo © Scott Travers). G) Pristimantis cf. ridens, Biosfera Bosawas, 180 m (Photo © Javier Sunyer). H) P.cf. ridens, Cerro Saslaya, 180 m (Photo © Javier Sunyer).

206

207

& Wilson 2002; Savage 2002). Based on phylogenetic analysis of 12S mtDNA data,

Zaldívar-Riverón et al. (2004:47) demonstrated that the taxon L. forreri is referable only to frogs from southern Sonora, México, south along the Pacific coastal plain not farther than northern Jalisco, and in México alone there are no fewer than four well-supported species masked under the name L. forreri. The relationship between Pacific lowland populations assigned to L. forreri from Guatemala, El Salvador, Honduras, Nicaragua, and Costa Rica, as well as the taxonomic assignment of various populations, remains unresolved.

Lithobates maculatus

The putative species Lithobates maculatus currently is considered to be found at premontane and lower montane elevations (and peripherally in the lowlands) throughout

Nuclear Central America (IUCN 2011; Köhler 2011). Both 16S and COI data indicate the existence of at least four discretely-distributed Potential Candidate Species among the Chortís Highland samples (Figure 4-6): one from northeastern Nicaragua, including

Cerro Saslaya and adjacent lowland rainforest areas (Figure 4-13C, 4-13D); one from highland areas in north-central Nicaragua, including Cerro Kilambé and Peñas Blancas; one from the southern side of the Texíguat cloud forest and the northern side of Pico

Pijol in north-central Honduras; and one from the rest of the Honduran serranía (Figure

4-13E). The ―serranía‖ clade itself is composed of four geographically discrete, reciprocally monophyletic subclades (Figure 4-6). Two 16S sequence fragments assigned to L. maculatus were available for comparison: KU 195258, from 19 km NW

Rizo de Oro, Oaxaca, México, and USNM 559483, from Quebrada de Oro in Parque

Nacional Pico Bonito, Honduras (Table 4-1). The Quebrada de Oro sample appears to represent a fifth Potential Candidate Species (Figure 4-6). The Oaxaca sample may or

208

may not be typical L. maculatus, whose type locality is Totonicapán, Guatemala, some

350 km southwest of the Oaxacan locality.

Lithobates taylori

The boundaries between the aforementioned L. brownorum and another

Caribbean lowland leopard frog, L. taylori, lack clear definition (McCranie & Wilson

2002, Savage 2002, Hillis & Wilcox 2005). While L. taylori is generally considered to occur from somewhere in the central portion of Nicaragua and extend southward into

Costa Rica and presumably Panamá (Köhler 2011), the boundary between it and L. brownorum to the north (if a boundary exists) is undefined, and the status of populations of Lithobates from Panamá also is unclear (Hillis & Wilcox 2005). Interestingly, all

Lithobates berlandieri group samples from highland areas in northern Nicaragua cluster with a single sample from a lowland site in eastern Nicaragua (Figure 4-6) that is assigned to L. taylori by Hillis & Wilcox (2005), and I tentatively refer those samples to

L. taylori pending additional sampling throughout eastern Nicaragua.

Lithobates warszewitschii

This species is considered a relatively common and widespread inhabitant of lowland and premontane rainforest from eastern Honduras south to central Panamá

(McCranie et al. 2006, Savage 2002). Previously published barcoding data (Crawford et al. 2010) indicates that there are potentially two cryptic species concealed under L. warszewitschii in Panamá, and samples from northeastern Nicaragua (Figure 4-13F) form a third clade for both 16S and COI (Figure 4-6).

Pristimantis ridens

This species is considered a relatively common and widespread inhabitant of lowland and premontane rainforests from northern Honduras to northern South America

209

(Wang et al. 2008). Samples from low and moderate elevation in northern Nicaragua

(Figures 4-13G, 4-13H) form a clade in both 16S and COI analyses, deeply divergent from samples from Panama (Figure 4-2). The Panamanian samples themselves appear to represent three candidate species (Crawford et al. 2010).

Candidate Species and Allopatric Populations of Salamanders

I began this study with samples representing ten populations of salamanders that could not be assigned unequivocally to a known species, due either to morphological distinctiveness or from being found in a new locality that was geographically and ecologically isolated from congeners. Four of these populations (Bolitoglossa sp. inquirenda 1, Nototriton sp. inquirenda 1, Oedipina sp. inquirenda 1, and Oedipina sp. inquirenda 2), do not correspond to any known species-level taxa, and have subsequently been described as distinct taxa (Sunyer et al. 2010, 2011; Townsend et al. 2009a, 2010a). One population, Nototriton sp. inquirenda 2, represents an undescribed species from the Sierra de Agalta in eastern Honduras, and is further evaluated in Chapter 5. Another population, Nototriton sp. inquirenda 3 from Montaña de Botaderos, also represents an undescribed species. In the cases of Bolitoglossa sp. inquirenda 2, Nototriton sp. inquirenda 4, N. sp. inquirenda 5, and Oedipina sp. inquirenda 1, these samples were found to represent new allopatric populations of species previously thought to be endemic to a single highland area (Townsend et al.

2011b). These cases, as well as a summary of information related to potential candidate species and other resulting taxonomic considerations are detailed below.

Bolitoglossa (Magnadigita) sp. inquirenda 1

In June 2006, during the first herpetological expedition into the core zone of PN

Montaña de Yoro, two specimens of an unknown salamander diagnosable to the

210

Bolitoglossa (Magnadigita) dunni species group were collected in the vicinity of

Cataguana (1,780–2,020 m elevation) in northern Departamento de Francisco Morazán.

On a return trip in March 2007, five additional samples were collected at the same locality. BLASTN searches of 16S returned a high degree of similarity to B. decora from

PN La Muralla (Table 4-4). No 16S sequence divergence was observed between samples series from two localities in Montaña de Yoro and the single available sequence for B. decora, typically an indication that samples are conspecific. However, the Montaña de Yoro series differs from B. decora in coloration, head width, and digit length, as well as being allopatric and found within a different elevational band. This evidence, supported by subsequent sequencing and comparative analysis of a widely used fragment of the mitochondrial gene cytochrome b (cyt b), led Townsend et al.

(2009a) to recognize the PN Montaña de Yoro population as a young but distinct species, Bolitoglossa cataguana (Figures 4-14A, 4-14B).

Bolitoglossa (Magnadigita) oresbia / sp. inquirenda 2

On 17–18 July 2007, I visited Cerro El Zarciadero in central Honduras, the type locality of Bolitoglossa oresbia, finding a single subadult B. oresbia (UF 156333) active on vegetation at night. This species was described recently on the basis of three specimens collected from an irregular patch of remnant cloud forest (Lower Montane

Wet Forest formation) less than 1 hectare in total extent on the peak of Cerro El

Zarciadero (McCranie et al. 2005). This patch is the only remaining forest near the peak of this mountain, and is adjacent to a set of communications towers and surrounded by agricultural clearings given over primarily to corn and other staple crops. In the original description, the authors indicated that B. oresbia was ―the most critically endangered salamander in Honduras‖ (McCranie et al. 2005:111), and, given its extremely limited

211

and vulnerable distribution, this species is classified as Critically Endangered on the

IUCN Red List and could be considered one of the most threatened amphibians on earth. From 6–13 July 2008, we surveyed a ridge above Quebrada Varsovia

(14.79913°N, 87.89128°W), on the southwestern side of Parque Nacional Cerro Azul

Meámbar in Departamento de Comayagua. Seven specimens of Bolitoglossa were collected, One adult salamander (Bolitoglossa sp. inquirenda 2; UF 156532) with a distinctive golden yellow coloration covering most of its body (Figure 4-14C) was collected in moist leaf litter during the morning of 12 July 2008 along the trail leading to the aforementioned campsite (14.79618°N, 87.89527°W), 1,560 m elevation. This golden individual was so distinctive in terms of color pattern that it was assumed to represent a distinct taxon from B. oresbia; however both distance-based (sequence divergence between Zarciadero and Meámbar samples 0.0–0.2% for 16S and 0.1% for

COI) and phylogenetic analyses confirm that this individual is conspecific with typical B. oresbia (Figures 4-7). One adult male (UF 156529) and four small juveniles (UF

156526–27, 156530–31) were found while active on vegetation at night along a trail that follows a steep hillside through undisturbed Lower Montane Wet Forest, 1,640–1,680 m elevation. An adult female (UF 156528; Figure 4-14D) was collected along the same trail during the daytime, when it was dislodged from a standing, rotten tree trunk approximately 1 m above the ground. Two of the juveniles were red-orange in color and resembled the golden morph of B. oresbia (Figure 4-14E), while the adults (Figure 4-

14D) and other two juveniles (Figure 4-14F) demonstrate coloration more typical of B. oresbia. These new localities are only around 8 airline km north of Cerro Zarciadero,

212

Figure 4-14. Candidate Species IV: Bolitoglossa. A) Bolitoglossa sp. inquirenda 1 (= B. cataguana), adult female holotype, Montaña de Yoro, 1,820 m; B) B. sp. inquirenda 2 (= B. oresbia), subadult male, Cerro Azul Meámbar, 1,560 m; C) B. sp. inquirenda 2 (= B. oresbia), subadult male, Cerro Azul Meámbar, 1,640 m; D) B. sp. inquirenda 2 (= B. oresbia), adult female, Cerro Azul Meámbar, 1,640 m; E) B. sp. inquirenda 2 (= B. oresbia), juvenile, Cerro Azul Meámbar, 1,640 m; F) B. sp. inquirenda 2 (= B. oresbia), juvenile, Cerro Azul Meámbar, 1,640 m; G) B. celaque, adult female, Cerro Celaque, 2,560 m. Photos © J.H. Townsend.

213

and the intervening area, while wholly converted to agriculture, is no lower than 1,300 m elevation at any given point.

Bolitoglossa (Magnadigita) celaque

As currently recognized, Bolitoglossa celaque is a highland salamander found across three mountainous areas of the Southern Cordillera of the Chortís Highlands, including the Sierra de Celaque (Figure 4-14G; the type locality), the Sierra de Puca-

Opalaca, and the Montaña de la Sierra. Both 16S and COI data indicate that samples from these three ranges each form clades (Figure 4-7). Although divergence among these three clades is moderate (1.7–3.6%), the clear phylogeographic structure (within- mountain range divergence ≤0.8% for 16S and ≤0.3% for COI) associated with these samples indicates they may represent a complex of relatively young species, and should be the focus of further study using finer-scale population-level approaches, environmental niche modeling, and comparative morphology to clarify the systematic status of populations assigned to B. celaque.

Bolitoglossa (Magnadigita) conanti

This highland salamander is presently considered to occur in the Sierra de Omoa,

Sierra de Espíritu Santo, and Sierra del Merendón, and is characterized by a high degree of variability in coloration within populations (McCranie & Wilson 1993, 2002).

Both 16S and COI data revealed clades corresponding to geographically isolated mountains ranges (Figure 4-7). As with B. celaque, divergence among these clades is moderate (2.5–3.0% for COI) and within-mountain range is very low (0.0–0.2% for 16S and 0.0% for COI), indicating relatively recent diversification, and should be the focus of further study using multiple avenues of finer-scale investigation.

214

Bolitoglossa (Magnadigita) porrasorum

As with the two other species of Chortís Highland Magnadigita currently thought to occur at more than one isolated cloud forest locality, B. porrasorum populations are not recovered as conspecific in distance-based analyses of 16S or COI data (Figure 4-7) or

ML analysis of the combined dataset (Figure 4-9). Populations of B. porrasorum from the Sierra de Sulaco (where the type locality is found) and Texíguat and Pico Bonito in the Cordillera Nombre de Dios each form geographically-discrete clades, although the lineages assigned to B. porrasorum demonstrate greater genetic distances than those of the B. celaque and B. conanti complexes and are potentially paraphyletic or polyphyletic with respect to B. cataguana, B. decora, and B. longissima (Figure 4-7).

Samples from Texíguat demonstrate a remarkable degree of color polymorphism

(Figure 4-15), however all samples from this locality appear to represent a single taxon

(Figure 4-7).

Bolitoglossa (Nanotriton) rufescens/nympha

Salamanders of the Bolitoglossa rufescens complex in the Chortís Highlands have a confusing taxonomic history (McCranie & Wilson 2002). Recently, Campbell et al. (2010) described a new species from this complex, B. nympha, from a locality on the western side of the Sierra de Caral in eastern Guatemala. Unfortunately, this description did little to resolve the systematic relationships of the B. rufescens complex in the

Chortís Highlands, as it was based on 12S and cyt b sequence data from a single sample. The use of only a single sample precludes characterization of diversity and distribution of B. nympha, and the choice of 12S in favor of the previously ubiquitous (at least for neotropical salamanders) 16S minimizes the ability of other researchers to directly compare Campbell et al.’s (2010) results with the comprehensive existing

215

Figure 4-15. Candidate Species V: Bolitoglossa cf. porrasorum. A–F) variation in color pattern in conspecific samples of Bolitoglossa cf. porrasorum, La Liberación de Texíguat, 1,080–1,420 m. Photos © J.H. Townsend. dataset available at NCBI. Comparative morphological analyses of other species and populations in the B. rufescens complex, particularly the multiple known populations from across the border (albeit in some cases in the same mountains) in Honduras are also not available. Based on the type locality and description of B. nympha, I am

216

inferring that four samples from two localities, one in the Sierra de Omoa and one in the

Sierra de Espíritu Santo, are conspecific with B. nympha (Figure 4-7). Surprisingly, of five samples collected syntopically in El Paraiso Valley on the northern slope of the

Sierra de Omoa, three appear to be conspecific with B. nympha while two others represent a deeply divergent (divergence between syntopic samples 5.5% for 16S and

17.6% for COI) cryptic species that is recovered in the 16S NJ tree as the sister to a sample assigned to B. rufescens (MVZ 194254) from Chiapas, México.

Nototriton barbouri

All analyses support that two populations currently assigned to N. barbouri from the Cordillera Nombre de Dios are paraphyletic with respect to a sample from the vicinity of the type locality of N. barbouri (Figures 4-8, 4-10). Furthermore, those two populations, one from Texíguat and one from Pico Bonito (McCranie 1996a), represent two Potential Candidate Species, forming a clade with N. brodiei from the Sierra de

Omoa and Sierra de Caral (Figure 4-8). A systematic revision of this taxon is presented in Chapter 5.

Nototriton sp. inquirenda 1

During a trip in April 2008 to the leeward side of RVS Texiguat in the vicinity of La

Fortuna, a single specimen of Nototriton was collected that possesses a series of morphological characteristics unique among Honduran congeners. This single salamander had greatly large nares, syndactylous hands and feet with pointed toe tips, and a pale ventral surface with light mottling, characteristics typical of species in the N. richardi group (Costa Rica) or the genus Cryptotriton. Data from 16S and COI indicate that this species forms a sister clade to the remaining species of Chortís Highlands

217

Nototriton (Figure 4-8), and this species was described as N. tomamorum by Townsend et al. (2010a).

Nototriton sp. inquirenda 2

Two specimens of Nototriton were collected in July 2010 from cloud forest in PN

Sierra de Agalta, the first record of this genus from eastern Honduras. Both distance and model-based analyses of COI and 16S data supports these samples as conspecific and representative an unnamed species (Figure 4-8), which is further analyzed and formally described in Chapter 5 (Townsend et al. 2011a).

Nototriton sp. inquirenda 3

Four specimens of Nototriton were collected in April 2011 along the highest ridge

(1,700+ m elevation) in the Sierra de Botaderos in northeastern Olancho. Both distance and model-based analyses of COI and 16S data indicate that these samples are conspecific and represent a new candidate species that is sister to Nototriton sp. inq. 2

(Figure 4-8). This candidate species is being described outside of this dissertation.

Nototriton lignicola / sp. inquirenda 4

On 10 June 2006, we collected a juvenile Nototriton (UF 156544) from inside a fractured rock alongside Quebrada Cataguana (15.01°N, 87.10°W), 1,820 m elevation,

Parque Nacional Montaña de Yoro, in northern Departamento de Francisco Morazán.

On 14 March 2007, we collected two adult Nototriton (UF 156542–43) from inside rotten logs on Cerro El Filón above Quebrada Cataguana, 2,020 m elevation. These samples are shown to be conspecific with N. lignicola (Figure 4-8), a Critically Endangered species previously known only from 13 specimens collected inside two rotten logs at

Cerro de Enmedio, Parque Nacional La Muralla, 1,760–1,780 m elevation (McCranie &

Wilson 2002).

218

Nototriton limnospectator / sp. inquirenda 5

Nototriton limnospectator is an Endangered moss salamander previously known only from forest above 1,600 m elevation in Parque Nacional Montaña de Santa

Bárbara, an isolated karstic mountain to the west of Lago de Yojoa. During 2008, four specimens of Nototriton were collected from three sites in Parque Nacional Cerro Azul

Meámbar, including two individuals (UF 156539–40) found on 7 and 9 July 2008 alongside juvenile Bolitoglossa oresbia (UF 156526–27, 156530–31) while active on low vegetation at night. Another individual (UF 156541) was collected on 27 August 2008 from near a campsite farther up the Quebrada Varsovia (14.80°N, 87.89°W), 1,710 m elevation, while it was active at night on a root buttress. All three of these localities are in the Lower Montane Wet Forest formation. In addition, a juvenile N. limnospectator

(UF collection) was collected from underneath a moss mat at night on 7 June 2008 along Sendero Bosque Nublado, 1,105 m elevation, above Centro de Visitantes de

Parque Nacional Cerro Azul Meámbar ―Los Pinos‖ (14.87°N, 87.91°W), Departamento de Cortés, a locality that lies in the Premontane Wet Forest formation. These samples were shown to be conspecific with reference samples of N. limnospectator from Parque

Nacional Montaña de Santa Bárbara (Figure 4-8). The new localities are approximately

20–25 km east of the nearest locality in Parque Nacional Montaña de Santa Bárbara

(McCranie & Wilson 2002), from which they are isolated by Lago de Yojoa, the largest freshwater lake in Honduras.

Oedipina kasios / sp. inquirenda 1

On 25 September 2008, a single specimen of Oedipina (Figure 4-16A; UF 156500) was collected in a small mesic ravine on Montaña de la Sierra (14.95°N, 87.061°W),

1,920 m elevation, in coniferous cloud forest on the southeastern side of Parque

219

Nacional Montaña de Yoro. This specimen was shown to be conspecific with Oedipina kasios Figure 4-8), a species recently described from Parque Nacional La Muralla as a member of a divergent, Chortís Highlands endemic clade (subgenus Oeditriton;

McCranie et al. 2008). This species occurs from 950–1,780 m elevation, and has been found in the same logs as Nototriton lignicola (McCranie et al. 2008). The Critically

Endangered amphibians Bolitoglossa cataguana and Plectrohyla guatemalensis were also collected within and along the outer margins of this and similar ravines. The forest in this part of PN Montaña de Yoro is typified by regularly burned pine-oak forest with an open grassy understory interrupted by deep, narrow seepage ravines that support dense mesic vegetation, providing refuge for species like O. kasios during fire events.

Oedipina nica / sp. inquirenda 2

Nicaraguan worm salamanders (Oedipina) have long been a source of taxonomic uncertainty, the byproduct of conserved morphology and sparse sampling. During

2008–2009, a series of 18 Oedipina were collected from three highland localities in northern Nicaragua (Figure 4-16B). These samples could not be assigned to a taxon based on comparisons from BLASTN searches, NJ tree clustering, or sequence divergence, and appeared to represent an undescribed species of the subgenus

Oeditriton (Figure 4-8). Sequencing and subsequent analysis of cyt b data, as well as comparative morphology, confirmed that these samples represent a new species, which we recently described as Oedipina nica (Sunyer et al. 2010).

Oedipina koehleri / sp. inquirenda 3

In his pioneering monograph of Oedipina, Brame (1968) described the taxon O. pseudouniformis from central Costa Rica, and included eight paratypes collected in July

1957 from ―Hacienda La Cumplida, 1.5 km north of Matagalpa, 731 m elevation,‖

220

Figure 4-16. Candidate Species VI: Oedipina. A) Oedipina sp. inquirenda 1 (= O. kasios), Montaña de Yoro, 1,920 m (Photo © J.H. Townsend). B) O. sp. inquirenda 2 (= O. nica), Macizos de Peñas Blancas, 1,515 m (Photo © Scott Travers). C) O. sp. inquirenda 3 (= O. koehleri), Cerro Musún, 724 m (Photo © Javier Sunyer).

Figure 4-17. Candidate Species VII: Oedipina cf. gephyra (= O. petiola). A) dorsal view of right hind foot of O. cf. gephyra (= O. petiola), Cerro Búfalo, 1,580 m; B) dorsal view of right hind foot of O. gephyra, Texíguat, 1,820 m; C) ventral view of right hind foot of O.cf. gephyra (= O. petiola); D) ventral view of right hind foot of O. gephyra, Texíguat, 1,820 m. Photos © J.H. Townsend.

221

Departamento de Matagalpa, Nicaragua. Premontane forests in this area have been severely degraded since 1957, and remaining patches of mesic forest are restricted to the most marginal upper portions of the surrounding peaks. Köhler et al. (2004) later assigned two specimens from pristine forest between approximately 600 and 945 m elevation in Parque Nacional Cerro Saslaya, Región Autónoma Atlántico Norte,

Nicaragua, to O. pseudouniformis, recognizing that these samples appeared conspecific with those from Hacienda La Cumplida. Analysis of recently collected samples from PN

Cerro Saslaya and an isolated mountain to the south, Cerro Musún (Figure 4-16C), revealed them to represent a single undescribed species (Figure 4-8). We described this species as O. koehleri based on evidence from phylogenetic analyses, macroecological models, and morphology (Sunyer et al. 2011).

Oedipina gephyra

The endemic species Oedipina gephyra was described from the leeward side of

Reserva de Vida Silvestre Texíguat in the western portion of the Cordillera Nombre de

Dios, Honduras (McCranie et al. 1993), and soon after was reported from Parque

Nacional Pico Bonito in the central portion of the Cordillera Nombre de Dios (McCranie

1996b). A single representative from each of the two populations assigned to O. gephyra was included in the phylogenetic analyses of García-París & Wake (2000), who reported interspecific-level sequence divergence between the two samples (García-

París & Wake 2000: 70), suggesting that more than one species may be concealed under O. gephyra. Multiple attempts to secure additional material from PN Pico Bonito were unsuccessful, however the addition of two new samples of O. gephyra from the vicinity of the type locality demonstrates that the two populations are reciprocally monophyletic (Figure 4-8), supporting recognition of two species. Following the

222

phylogenetic results, McCranie & Townsend (2011) also found diagnostic differences in foot morphology (Figure 4-17), and subsequently described the Pico Bonito population as O. petiola.

Amphibian Endemism and Conservation Priorities in the Chortís Block

The Chortís Block is already known to possess a diverse endemic amphibian fauna, with 74 regionally endemic species with distributions restricted to within the region (Chapter 3, Table 3-5). The identification of as many as 36 new species from a subset of about 36% of named Chortís Block amphibians indicates that the ―true‖ diversity is largely underestimated. One-third of the Potential Candidate Species (12/36) appear to be restricted to localities in the Northern Cordillera of the Chortís Highlands, a region already known to support the highest level of amphibian endemism in the Chortís

Block (Chapter 3, Table 3-6). It comes as little surprise to me that six Potential

Candidate Species are from the vicinity of Texíguat at the western end of the Cordillera

Nombre de Dios (Table 4-3), a mountain range identified as one of most significant endemism hotspots in Mesoamerica (Chapter 3; Townsend et al. 2010a, 2011c,

Submitted A). Four more species are from the vicinity of Pico Bonito, and another two from the Sierra de Omoa (Table 4-3). Given that established protected areas encompass most of the remaining highland forest in each of these three mountains ranges, tangible opportunities for conserving extant populations of both named and unnamed endemic species already exist and could be strengthened by the description of additional endemic diversity from within each park’s boundaries.

If it is assumed that at least the CCS and UCS represent distinct species, then, including the six candidate species that have already been formally described, a total of

30 new endemic amphibian species are identified in the Chortís Block. Of these 30, at

223

least 12 would immediately qualify as Critically Endangered based on IUCN Red List criteria (IUCN 2001), and at least an additional nine species would be classified as

Endangered (Table 4-3).

A somewhat surprising finding was the identification of Potential Candidate

Species from sampled populations assumed a priori to represent widespread lowland taxa such as Diasporus diastema, Lithobates warszewitschii, Pristimantis ridens,

Rhaebo haematiticus, and Smilisca baudinii. Should these populations be confirmed as distinct species, the implication that there exists a previously unknown endemic amphibian fauna inhabiting the Caribbean lowlands of northeastern Nicaragua (and presumably eastern Honduras) would rather dramatically alter the understanding of patterns influencing Chortís Block biogeography and diversification. Given the large number of species that are presently considered to meet the northern limit of their distribution in this area (as was the case with the aforementioned candidate species), this revelation would warrant a comprehensive re-evaluation of conservation and management objectives in the transboundary Mosquitia region, which is currently aimed at the maintenance of ecological corridors principally to support populations of large mammals such as jaguar, tapir, and white-lipped peccaries (CIPF 2009).

Accomplishment of a rapid amphibian inventory, emphasizing the collection of molecular data, is urgently needed to assess the extent of endemic cryptic diversity currently overlooked in the Caribbean lowlands of the Chortís Block.

Iterative Taxonomic Approaches to Biodiversity Inventory

In this chapter, I have provided the first comprehensive assessment of amphibian evolutionary diversity for the Chortís Block of Central America, a region already recognized for its elevated endemism. I have deliberately employed an iterative,

224

heuristic approach to the systematic evaluation of regional diversity, with two principal aims in mind. First, to initiate development of a comprehensive molecular reference dataset upon which future work in the region can be directed and compared. Doing so contributes to the goals set out by both the Systematics Agenda 2000 (1994) and the

IUCN Global Amphibian Assessment (2007) chief of which are the execution of rapid inventories to catalog global biodiversity and the generation of a ―Tree of Life‖ phylogeny of all species to serve as the basis for classification and as a framework for use by researchers in the life sciences. Second, the use of this approach is advantageous in that it allows for, what I consider to be, the appropriate application of

DNA barcoding data and methods in taxonomy.

After establishing the ecophysiographic setting in Chapter 2 and providing a comprehensive faunal treatment of the region’s amphibians and reptiles, I presented and analyzed a comprehensive two-gene DNA barcoding dataset for Chortís Block amphibians. As part of an iterative process, the distance-based analyses used in association with DNA barcoding provide a vital step that can be used to guide how and where to focus analytical resources moving forward. In this study, the Class Caudata

(the salamanders) represents both the most taxonomically comprehensive dataset among groups of amphibians analyzed, and presents both the most Potential Candidate

Species (Table 4-3) and the highest degree of extinction risk (Chapter 3, Table 3-5). As the next step, I used the two-gene barcoding dataset as the basis for model-based maximum likelihood (ML) phylogenetic analysis in order to generate a phylogenetic hypothesis that can be used to more rigorously establish species limits than can be done using distance-based methods alone, and, to address a principal deficiency of

225

distance-based analysis, estimate evolutionary relationships at a deeper level than that of species and species-delimitation. Utilization of the two-gene approach to DNA barcoding in this case allows for the initial dataset to be more informative phylogenetically and therefore of more utility at later stages in the process of taxonomic evaluation.

In the case of Chortís Block salamanders, DNA barcoding analyses indicated that there are at least 12 CCS and USC and another four DCL that are in need of further systematic study (Table 4-3). Phylogenetic analysis supports each of these Potential

Candidate Species as a monophyletic group (Figures 4-9, 4-10), and revealed further details of evolutionary relationships within and among amphibian clades. One group in particular, the moss salamanders (Nototriton), exhibited a relatively high degree of cryptic diversification, and will serve as the model group for use in my subsequent systematic evaluation. Phylogenetic analyses recovered three geographically- circumscribed clades (Figure 4-10): a Northern Cordillera clade consisting of N. brodiei from the Sierra de Omoa and Sierra de Caral, N. sp. 2 CCS from Pico Bonito, and N. sp. 3 CCS from Texíguat; a central clade, consisting of N. barbouri sensu stricto from the Sierra de Sulaco and N. limnospectator from Montaña de Santa Bárbara and the

Montañas de Meámbar; and an eastern clade, consisting of N. sp. inquirenda 2 (=N. picucha; Townsend et al. 2011a) and N. sp. inquirenda 3 from the Sierra de Botaderos.

Nototriton sp. inquirenda 1 (=N. tomamorum; Townsend et al. 2010a), which possesses distinctive morphological traits, was recovered as the sister lineage to the rest of the

Chortís Block endemic clade corresponding to the N. barbouri species group (sensu

García-París & Wake 2000). The Cordillera de La Flor-La Muralla endemic N. lignicola

226

is recovered as sister to the rest of the in-group. Chapter 5 deals with the molecular and morphological systematics and taxonomic revision associated with this group, providing what is, in my estimation, the most critically overlooked yet definitively fundamental activity for contemporary systematic biologists to engage in: the formal taxonomic treatment of candidate taxon identified in molecular studies.

Conservation planners, ecologists, legislators, and the general public have little use for metrics based on un-named, un-described diversity, and I do not foresee an entity such as ―Nototriton sp. inq. 1 CCS‖ being promoted as part of a regional conservation strategy or public outreach campaign. Despite the best intentions and most diligent labor of biologists, in the end it is the greater share of society, made up of non-scientists, which will determine the success or failure of this generation’s challenge to preserve our shared biodiversity resources. Taxonomy provides the most basic means for which the specialist can identify and communicate information about biological diversity, and to carry out molecular inventories of biodiversity without promoting the direct taxonomic evaluation of the results is, in my view, a disservice to the study organisms themselves, our field of study, and to society as a whole.

227

Table 4-1. Voucher information for samples used in this study. Voucher abbreviations are as follows: JHT = author’s field series; IRL = Ileana R. Luque-Montes field series; AJC = Andrew J. Crawford field series; AMNH = American Museum of Natural History; JSF = John Frost tissue collection; KRL = Karen R. Lips field series; KU = University of Kansas Natural History Museum; MVUP = Museo de Vertebratos de la Universidad de Panamá, SIUC = Herpetology Collection, Southern Illinois University at Carbondale; USNM = National Museum of Natural History, Smithsonian Institution. Taxon Voucher Locality 16S COI Data source CAUDATA Plethodontidae (29) Bolitoglossa sp. inq. 1 JHT2087 Honduras: Cataguana + — This study (= B. cataguana) JHT2113 Honduras: Cataguana + — This study JHT2114 Honduras: Cataguana + — This study JHT2115 Honduras: Cataguana + + This study JHT2125 Honduras: Cataguana + — This study JHT2964 Honduras: above Guaymas + — This study JHT2965 Honduras: above Guaymas + + This study Bolitoglossa celaque JHT2743 Honduras: Cerro Celaque + + This study JHT2744 Honduras: Cerro Celaque + + This study JHT2745 Honduras: Cerro Celaque + + This study JHT2747 Honduras: Cerro Celaque + + This study JHT2748 Honduras: Cerro Celaque + + This study JHT2749 Honduras: Cerro Celaque + — This study JHT2750 Honduras: Cerro Celaque + + This study JHT2751 Honduras: Cerro Celaque + + This study JHT2752 Honduras: Cerro Celaque + + This study JHT2753 Honduras: Cerro Celaque + + This study JHT2754 Honduras: Cerro Celaque + + This study JHT2755 Honduras: Cerro Celaque + + This study JHT2756 Honduras: Cerro Celaque + + This study JHT2757 Honduras: Cerro Celaque + + This study JHT2903 Honduras: Intibucá + + This study JHT2904 Honduras: Intibucá + + This study JHT2910 Honduras: Intibucá + + This study JHT2911 Honduras: Intibucá + + This study JHT2912 Honduras: Intibucá + + This study JHT2618 Honduras: Guajiquiro + + This study JHT2867 Honduras: Guajiquiro + + This study JHT2868 Honduras: Guajiquiro + + This study JHT2869 Honduras: Guajiquiro + + This study JHT2306 Honduras: Intibucá + — This study JHT2307 Honduras: Intibucá + + This study JHT2877 Honduras: Intibucá + + This study

228

Table 4-1. Continued. JHT2878 Honduras: Intibucá + + This study JHT2879 Honduras: Intibucá + + This study JHT2880 Honduras: Intibucá + + This study JHT2887 Honduras: Intibucá + + This study JHT2888 Honduras: Intibucá + + This study JHT2889 Honduras: Intibucá + + This study JHT2890 Honduras: Intibucá + + This study Bolitoglossa colonnea CH 6526 Panamá: El Copé FJ784318 FJ766578 Crawford et al. (2010) Bolitoglossa conanti JHT2922 Honduras: Cusuco + + This study JHT2923 Honduras: Cusuco + + This study JHT2924 Honduras: Cusuco + + This study JHT2925 Honduras: Cusuco + + This study JHT2725 Honduras: Ocotepeque + + This study JHT2741 Honduras: Ocotepeque + + This study JHT2742 Honduras: Ocotepeque + + This study UTA A58141 Guatemala: Cerro El Mono + + This study Bolitoglossa diaphora JHT2917 Honduras: Cusuco + + This study JHT2918 Honduras: Cusuco + — This study JHT2919 Honduras: Cusuco + + This study JHT2920 Honduras: Cusuco + + This study JHT2921 Honduras: Cusuco + + This study JHT2989 Honduras: Cusuco + + This study JHT2990 Honduras: Cusuco + + This study JHT2991 Honduras: Cusuco + + This study Bolitoglossa dofleini JHT2430 Honduras: Texíguat (Yoro) + + This study JHT2854 Honduras: Quebrada Grande + + This study UTA A56787 Guatemala + + This study Bolitoglossa dunni UTA A51488 Guatemala: Sierra de Caral + + This study Bolitoglossa heiroreias UTA A54712 Guatemala: Chiquimula + + This study Bolitoglossa longissima JHT3194 Honduras: Cerro La Picucha + — This study JHT3195 Honduras: Cerro La Picucha + + This study JHT3196 Honduras: Cerro La Picucha + + This study JHT3197 Honduras: Cerro La Picucha + + This study JHT3198 Honduras: Cerro La Picucha + — This study JHT3199 Honduras: Cerro La Picucha + — This study JHT3200 Honduras: Cerro La Picucha — + This study JHT3201 Honduras: Cerro La Picucha + — This study Bolitoglossa mexicana JHT2390 Honduras: Los Naranjos + — This study JHT2572 Honduras: Comayagua + + This study JHT2573 Honduras: Comayagua + — This study JHT2960 Honduras: above Guaymas + + This study JHT2512 Honduras: Azul Meámbar + — This study JHT3182 Sierra de Agalta + — This study

229

Table 4-1. Continued. Bolitoglossa nympha1 JHT1766 Honduras: El Paraiso Valley + — This study JHT1793 Honduras: El Paraiso Valley + + This study JHT1823 Honduras: El Paraiso Valley + + This study JHT2803 Honduras: Quebrada Grande + + This study Bolitoglossa oresbia JHT2225 Honduras: Cerro Zarciadero + + This study (includes Bolitoglossa sp. inq. 2) IRL071 Honduras: Azul Meámbar + + This study IRL072 Honduras: Azul Meámbar + + This study IRL074 Honduras: Azul Meámbar + + This study IRL075 Honduras: Azul Meámbar + + This study IRL079 Honduras: Azul Meámbar + + This study IRL080 Honduras: Azul Meámbar + — This study IRL081 Honduras: Azul Meámbar + + This study Bolitoglossa porrasorum JHT2359 Honduras: Macuzal + + This study JHT2360 Honduras: Macuzal + + This study JHT2371 Honduras: Macuzal + + This study JHT2372 Honduras: Macuzal + + This study JHT2373 Honduras: Macuzal + + This study JHT2374 Honduras: Macuzal + — This study JHT2419 Honduras: Macuzal + — This study Bolitoglossa cf. porrasorum JHT2431 Honduras: Texíguat (Yoro) + + This study JHT2432 Honduras: Texíguat (Yoro) + — This study JHT2449 Honduras: Texíguat (Yoro) + + This study JHT2450 Honduras: Texíguat (Yoro) + + This study JHT2453 Honduras: Texíguat (Yoro) + + This study JHT2454 Honduras: Texíguat (Yoro) + — This study JHT2455 Honduras: Texíguat (Yoro) + + This study JHT3104 Honduras: Texíguat (Atlántida) + + This study JHT3105 Honduras: Texíguat (Atlántida) + + This study JHT3106 Honduras: Texíguat (Atlántida) + + This study JHT3143 Honduras: Texíguat (Atlántida) + + This study JHT3144 Honduras: Texíguat (Atlántida) + + This study JHT3145 Honduras: Texíguat (Atlántida) + + This study JHT3146 Honduras: Texíguat (Atlántida) + — This study JHT3147 Honduras: Texíguat (Atlántida) + + This study JHT3148 Honduras: Texíguat (Atlántida) + + This study JHT3149 Honduras: Texíguat (Atlántida) + + This study JHT3150 Honduras: Texíguat (Atlántida) + — This study JHT3151 Honduras: Texíguat (Atlántida) + + This study JHT3152 Honduras: Texíguat (Atlántida) + + This study JHT3153 Honduras: Texíguat (Atlántida) + + This study JHT3171 Honduras: Texíguat (Atlántida) + + This study JHT3172 Honduras: Texíguat (Atlántida) + + This study JHT3229 Honduras: Texíguat (Atlántida) + + This study

230

Table 4-1. Continued. JHT3230 Honduras: Texíguat (Atlántida) + — This study JHT3231 Honduras: Texíguat (Atlántida) + + This study JHT3250 Honduras: Texíguat (Atlántida) + + This study JHT3251 Honduras: Texíguat (Atlántida) + + This study JHT3252 Honduras: Texíguat (Atlántida) + + This study JHT3253 Honduras: Texíguat (Atlántida) + + This study JHT3254 Honduras: Texíguat (Atlántida) + + This study JHT3255 Honduras: Texíguat (Atlántida) + + This study JHT3256 Honduras: Texíguat (Atlántida) + + This study JHT3257 Honduras: Texíguat (Atlántida) + + This study JHT3264 Honduras: Texíguat (Atlántida) + + This study JHT3265 Honduras: Texíguat (Atlántida) + + This study JHT3266 Honduras: Texíguat (Atlántida) + + This study Bolitoglossa cf. rufescens JHT1767 Honduras: El Paraiso Valley + + This study JHT1797 Honduras: El Paraiso Valley + + This study Bolitoglossa schizodactyla USNM 572791 Panamá: El Copé FJ784482 FJ766579 Crawford et al. (2010) Bolitoglossa striatula N145 Nicaragua: Bosawas + + This study Bolitoglossa synoria JHT2900 Honduras: Cerro El Pital + + This study JHT2901 Honduras: Cerro El Pital + + This study Cryptotriton alvarezdeltoroi MVZ 158942 México: Chiapas AF199196 — García-París & Wake (2000) Cryptotriton sierraminensis MVZ 160907 Guatemala: Sierra Las Minas AF199198 — García-París & Wake (2000) Cryptotriton veraepacis GAR059 Guatemala: Baja Verapaz + + This study (E.N. Smith, UTA) UTA A51396 Guatemala: Baja Verapaz + + This study (E.N. Smith, UTA) MVZ 215913 Guatemala: Baja Verapaz AF199197 — García-París & Wake (2000) Dendrotriton rabbi UTA A51086 Guatemala: Uspantán AF199232 — García-París & Wake (2000) Nototriton abscondens UCR 12071 Costa Rica: Cascada La Paz AF199199 — García-París & Wake (2000) Nototriton barbouri JHT2420 Honduras: Macuzal + + This study Nototriton ―barbouri‖ (N. sp. A) USNM 339712 Honduras: Pico Bonito AF199201 — García-París & Wake (2000) Nototriton ―barbouri‖ (N. sp. B) JHT3159 Honduras: Texíguat (Atlántida) + + This study Nototriton brodiei UTA A51490 Guatemala: Sierra de Caral AF199202 + 16S: García-París & Wake (2000); COI: This study (E.N. Smith, UTA) MVZ(FN252407) Honduras: Sierra de Omoa + — This study (S.M. Rovito, MVZ) Nototriton gamezi MVZ 207122 Costa Rica: Monteverde AF199200 — García-París & Wake (2000) Nototriton guanacaste MVZ 207106 Costa Rica: Volcán Cacao AF199203 — García-París & Wake (2000) Nototriton lignicola USNM 497540 Honduras: La Muralla AF199204 — García-París & Wake (2000) (=N. sp. inq. 4) JHT2122 Honduras: Cataguana + + This study Nototriton limnospectator None Honduras: Santa Bárbara + — S.M. Rovito (MVZ) (=N. sp. inq. 5) IRL035 Honduras: Azul Meámbar + — This study IRL070 Honduras: Azul Meámbar + + This study IRL076 Honduras: Azul Meámbar + + This study IRL088 Honduras: Azul Meámbar + + This study

231

Table 4-1. Continued. Nototriton picadoi MVZ 225899 Costa Rica: Tapantí AF199205 — García-París & Wake (2000) Nototriton richardi UCR 12057 Costa Rica: Cascajal de Las AF199206 — García-París & Wake (2000) Nubes Nototriton saslaya N650 Nicaragua: Cerro Saslaya + — This study SMF collection Nicaragua: Cerro Saslaya + — This study Nototriton sp. inq. 1 JHT2437 Honduras: Texíguat (Yoro) + + This study (= N. tomamorum) Nototriton sp. inq. 2 JHT3192 Honduras: Cerro La Picucha + + This study (= N. picucha) JHT3193 Honduras: Cerro La Picucha + + This study Nototriton sp. inq. 3 JHT3398 Honduras: Sierra de Botaderos + + This study JHT3399 Honduras: Sierra de Botaderos + + This study JHT3400 Honduras: Sierra de Botaderos + + This study JHT3401 Honduras: Sierra de Botaderos + + This study Oedipina alleni MVZ 190857 Costa Rica: Sirena AF199207 — García-París & Wake (2000) MVZ 225903 Costa Rica: Damas AF199208 — García-París & Wake (2000) Oedipina carablanca None Costa Rica: Limón FJ196862 — McCranie et al. (2008) Oedipina collaris SIUC H-08896 Panamá: El Copé FJ196863 — Crawford et al. (2010) Oedipina complex DBW5105 Panamá: Barro Colorado AF199213 — García-París & Wake (2000) DBW5787 Panamá: Cerro Campana AF199212 — García-París & Wake (2000) Oedipina cyclocauda MVZ 138916 Costa Rica: La Selva AF199214 — García-París & Wake (2000) MVZ 293747 Costa Rica: La Selva AF199215 — García-París & Wake (2000) Oedipina elongata UTA A56809 Guatemala + + This study (E.N. Smith, UTA) UTA A56810 Guatemala + — This study (E.N. Smith, UTA) Oedipina gephyra JHT2443 Honduras: Texíguat (Yoro) + + This study JHT2451 Honduras: Texíguat (Yoro) + + This study USNM 530582 Honduras: Texíguat (Yoro) AF199218 — García-París & Wake (2000) Oedipina cf. gephyra USNM 343462 Honduras: Pico Bonito AF199217 — García-París & Wake (2000) (= O. petiola) Oedipina gracilis MVZ 210398 Costa Rica: La Selva AF199219 — García-París & Wake (2000) Oedipina grandis MVZ 225904 Costa Rica: Cerro Pando AF199220 — García-París & Wake (2000) Oedipina cf. ignea USNM 530586 Honduras: Cerro El Pital AF199231 — García-París & Wake (2000) Oedipina kasios MVZ 232825 Honduras: La Muralla FJ196866 — McCranie et al. (2008) (=Oedipina sp. inq. 1) JHT2974 Honduras: above Guaymas + + This study Oedipina parvipes AJC1786 Panamá: El Copé FJ784316 FJ766760 Crawford et al. (2010) MVZ 210404 Panamá: Nusagandi AF199210 — García-París & Wake (2000) MVZ 210405 Panamá: Río Frijoles AF199211 — García-París & Wake (2000) Oedipina poelzi MVZ 163703 Costa Rica: Vara Blanca AF199224 — García-París & Wake (2000) MVZ 207128 Costa Rica: Monteverde AF199225 — García-París & Wake (2000) MVZ 206398 Costa Rica: Braulio Carrillo AF199223 — García-París & Wake (2000) Oedipina quadra MVZ 232824 Honduras: Warunta FJ196865 — McCranie et al. (2008) Oedipina savagei UCR LG961327 Costa Rica: Cerro Zapote AF199209 — García-París & Wake (2000) Oedipina sp. SMF 78738 Nicaragua + — García-París & Wake (2000) JHT2974 Honduras: above Guaymas + + This study

232

Table 4-1. Continued. Oedipina sp. inq. 2 (= O. nica) N567 Nicaragua: Cerro Kilambé + + This study N569 Nicaragua: Cerro Kilambé + — This study N570 Nicaragua: Cerro Kilambé + — This study N964 Nicaragua: Cerro Kilambé + + This study N965 Nicaragua: Cerro Kilambé + + This study N1029 Nicaragua: Peñas Blancas + — This study N1030 Nicaragua: Peñas Blancas + — This study Oedipina sp. inq. 3 (= O. koehleri) N614 Nicaragua: Cerro Saslaya + + This study JS782 Nicaragua: Cerro Musún + — This study Oedipina stenopodia MVZ 163649 Guatemala: San Rafael AF199228 — García-París & Wake (2000) Oedipina taylori UGSC 1134 Guatemala: Zacapa HM068304 — Sunyer et al. (2010) Oedipina tomasi JHT1553 Honduras: Sierra de Omoa + — This study MVZ 258037 Honduras: Sierra de Omoa + — Oedipina uniformis MVZ 190853 Costa Rica: Ciénega Colorado AF199229 — García-París & Wake (2000) MVZ 203751 Costa Rica: Tapantí AF199230 — García-París & Wake (2000) ANURA Bufonidae (14) Crepidophryne chompipe UCR 16075 Costa Rica HM563859 — Mendelson et al. (In press) Incilius coccifer JHT3301 Honduras: Isla El Tigre + + This study JS1016 Nicaragua: Ometepe + — This study JS1058 Nicaragua: Ometepe + — This study JS1150 Nicaragua: Atlántico Norte + — This study SDSNH AEH-013 Nicaragua: Ometepe AY927857 — Mendelson et al. (2005) KU 290030 El Salvador: Morazán AY927856 — Mendelson et al. (2005) USNM 547980 Honduras: Rus Rus AY929301 — Mendelson et al. (2005) Incilius coniferus MVUP 1820 Panamá: El Copé FJ784379 FJ766768 Crawford et al. (2010) USNM 572086 Panamá: El Copé FJ784382 FJ766767 Crawford et al. (2010) USNM 572087 Panamá: El Copé FJ784444 FJ766766 Crawford et al. (2010) USNM 572092 Panamá: El Copé FJ784586 FJ766765 Crawford et al. (2010) toe 140 Panamá: El Copé FJ784595 FJ766764 Crawford et al. (2010) toe 141 Ocon Panamá: El Copé FJ784597 FJ766732 Crawford et al. (2010) toe 144 Ocon Panamá: El Copé FJ784599 FJ766762 Crawford et al. (2010) toe 151 Ocon Panamá: El Copé FJ784601 FJ766761 Crawford et al. (2010) Incilius cf. coniferus N416 Nicaragua: Bosawas + + This study Incilius cycladen UTA-JRM 4607 Mexico: Guerrero AY927858 — Mendelson et al. (2005) Incilius ibarrai JHT2205 Honduras: Cerro Zarciadero + + This study JHT2604 Honduras: Guajiquiro + + This study JHT2605 Honduras: Guajiquiro + + This study JHT2763 Honduras: Intibucá + + This study JHT2765 Honduras: Intibucá + + This study JHT2874 Honduras: Intibucá + — This study JHT2906 Honduras: Intibucá + + This study IRL002 Honduras: Comayagua + + This study

233

Table 4-1. Continued. IRL004 Honduras: Comayagua + + This study IRL005 Honduras: Comayagua + + This study Incilius karenlipsae UTA A-59522 Panamá: El Copé GU552454 — Mendelson & Mulcahy (2010) Incilius leucomyos JHT3034 Honduras: Texíguat (Atlántida) + + This study JHT3242 Honduras: Texíguat (Atlántida) — + This study CAC013 Honduras: Pico Bonito + + This study CAC059 Honduras: Pico Bonito + + This study Incilius luetkenii JS853 Nicaragua: Las Nubes + + This study Incilius porteri JHT2228 Honduras: Cerro Uyuca + + This study JHT3302 Honduras: Isla El Tigre + + This study CAC044 Honduras: Comayagua + + This study JHT2249 Honduras: Francisco Morazán HM563882 — Mendelson et al. (In press) Incilius valliceps JHT2013 Honduras: Marale + + This study JHT2271 Nicaragua: Selva Negra + + This study JHT2428 Honduras: Texíguat (Yoro) + + This study JHT2519 Honduras: Azul Meámbar + + This study JHT2807 Honduras: Quebrada Grande + + This study JHT3175 Honduras: Texiguat (Atlántida) + — This study IRL047 Honduras: Azul Meámbar + + This study LK005 Honduras: Cerro Santa Bárbara + + This study N089 Nicaragua: Bosawas + + This study USNM 509524 Honduras: Pico Bonito AY008231 — Mulcahy & Mendelson (2000) Rhaebo haematiticus MVUP 1842 Panamá: El Copé FJ784439 FJ766816 Crawford et al. (2010) USNM 572094 Panamá: El Copé FJ784404 FJ766818 Crawford et al. (2010) USNM 572095 Panamá: El Copé FJ784426 FJ766817 Crawford et al. (2010) USNM 572096 Panamá: El Copé FJ784452 FJ766815 Crawford et al. (2010) USNM 572097 Panamá: El Copé FJ784546 FJ766814 Crawford et al. (2010) USNM 572098 Panamá: El Copé FJ784560 FJ766813 Crawford et al. (2010) Toe 120 Panamá: El Copé FJ784593 FJ766812 Crawford et al. (2010) Rhaebo cf. haematiticus N063 Nicaragua: Bosawas + + This study N137 Nicaragua: Bosawas + + This study N594 Nicaragua: Bosawas + + This study Rhinella marina MVUP 1802 Panamá: El Copé FJ784357 FJ766819 Crawford et al. (2010) Craugastoridae (11) Craugastor angelicus MVZ 149762 Costa Rica: Volcán Barba EU186681 — Hedges et al. (2008) Craugastor aurilegulus JHT2993 Honduras: Pico Bonito + + This study JHT3015 Honduras: Lancetilla + + This study JHT3016 Honduras: Lancetilla + + This study JHT3017 Honduras: Lancetilla + + This study JHT3018 Honduras: Lancetilla + + This study JHT3241 Honduras: Texíguat (Atlántida) + + This study C007 Honduras: Pico Bonito + + This study C024 Honduras: Pico Bonito + + This study

234

Table 4-1. Continued. Craugastor cf. azueroensis KRL0680 Panamá: El Copé FJ784332 FJ766637 Crawford et al. (2010 USNM 572219 Panamá: El Copé FJ784393 FJ766636 Crawford et al. (2010 USNM 572278 Panamá: El Copé FJ784324 FJ766675 Crawford et al. (2010 USNM 572279 Panamá: El Copé FJ784325 FJ766674 Crawford et al. (2010 Craugastor bransfordii USNM 572220 Panamá: El Copé FJ784339 FJ766631 Crawford et al. (2010) MVUP 1803 Panamá: El Copé FJ784358 FJ766630 Crawford et al. (2010) USNM 572221 Panamá: El Copé FJ784376 FJ766629 Crawford et al. (2010) MVUP 1841 Panamá: El Copé FJ784427 FJ766628 Crawford et al. (2010) USNM 572222 Panamá: El Copé FJ784481 FJ766627 Crawford et al. (2010) USNM 572223 Panamá: El Copé FJ784496 FJ766626 Crawford et al. (2010) Craugastor charadra JHT1820 Honduras: El Paraiso Valley + + This study JHT1826 Honduras: El Paraiso Valley + + This study Craugastor fitzingeri N060 Nicaragua: Bosawas + + This study KRL 0693 Panamá: El Copé FJ784337 — Crawford et al. (2010) USNM 572256 Panamá: El Copé FJ784344 — Crawford et al. (2010) MVUP 1798 Panamá: El Copé FJ784356 — Crawford et al. (2010) Craugastor laevissimus JHT1824 Honduras: El Paraiso Valley — + This study JHT2489 Honduras: Azul Meámbar + + This study JHT2501 Honduras: Azul Meámbar + + This study JHT2510 Honduras: Azul Meámbar + + This study JHT2517 Honduras: Azul Meámbar + + This study JHT2529 Honduras: Comayagua — + This study JHT2539 Honduras: Comayagua — + This study JHT2552 Honduras: Comayagua — + This study JHT2559 Honduras: Comayagua — + This study JHT2779 Honduras: Pico Pijol — + This study JHT2978 Honduras: Azul Meámbar + + This study JHT2979 Honduras: Azul Meámbar + + This study JHT3000 Honduras: Cerro Santa Bárbara + + This study JHT3004 Honduras: Los Naranjos + + This study IRL023 Honduras: Azul Meámbar + + This study N556 Nicaragua: Cerro Kilambé — + This study N557 Nicaragua: Cerro Kilambé — + This study N639 Nicaragua: Cerro Saslaya — + This study N950 Nicaragua: Cerro Kilambé — + This study N961 Nicaragua: Cerro Kilambé — + This study Craugastor punctariolus MVUP 1784 Panamá: El Copé FJ784333 FJ766673 Crawford et al. (2010) USNM 572281 Panamá: El Copé FJ784411 FJ766672 Crawford et al. (2010) USNM 572282 Panamá: El Copé FJ784417 FJ766671 Crawford et al. (2010) USNM 572283 Panamá: El Copé FJ784418 FJ766670 Crawford et al. (2010) Craugastor sandersoni UTA-A49803 Guatemala: Sierra de Santa Cruz EF493712 — Heinicke et al. (2007) Craugastor tabasarae KRL0706 Panamá: El Copé FJ784342 — Crawford et al. (2010) KRL1373 Panamá: El Copé FJ784512 — Crawford et al. (2010)

235

Table 4-1. Continued. KRL1387 Panamá: El Copé FJ784515 — Crawford et al. (2010) Craugastor underwoodi UCR 16315 EF562362 — Streicher et al. (2009) USNM 561403 EF562361 — Streicher et al. (2009) Eleutherodactylidae (3) Diasporus aff. diastema MVUP 1783 Panamá: El Copé FJ784338 — Crawford et al. (2010) USNM 572442 Panamá: El Copé FJ784425 — Crawford et al. (2010) MVUP 1830 Panamá: El Copé FJ784395 — Crawford et al. (2010) Diasporus cf. diastema N546 Nicaragua: Cerro Kilambé + + This study N928 Nicaragua: Cerro Kilambé + + This study N934 Nicaragua: Cerro Kilambé + + This study Hylidae (15) Duellmanohyla soralia JHT1585 Honduras: Sierra de Omoa + + This study UTA A50812 Guatemala: Sierra de Caral AY843584 — Faivovich et al. (2005) Exerodonta catracha JHT2209 Honduras: Cerro Zarciadero + + This study JHT2211 Honduras: Cerro Zarciadero + + This study JHT2315 Honduras: Intibucá + — This study JHT2316 Honduras: Intibucá + + This study JHT2647 Honduras: Guajiquiro + + This study JHT2648 Honduras: Guajiquiro + + This study JHT2649 Honduras: Guajiquiro + + This study JHT2650 Honduras: Guajiquiro + + This study JHT2651 Honduras: Guajiquiro + + This study JHT2652 Honduras: Guajiquiro + — This study JHT2653 Honduras: Guajiquiro + + This study JHT2654 Honduras: Guajiquiro — — This study JHT2655 Honduras: Guajiquiro + + This study JHT2656 Honduras: Guajiquiro + + This study JHT2657 Honduras: Guajiquiro + — This study JHT2658 Honduras: Guajiquiro + + This study JHT2659 Honduras: Guajiquiro + + This study JHT2660 Honduras: Guajiquiro + + This study JHT2661 Honduras: Guajiquiro + + This study JHT2861 Honduras: Guajiquiro + + This study JHT2862 Honduras: Guajiquiro + + This study JHT2863 Honduras: Guajiquiro — — This study JHT2864 Honduras: Guajiquiro + + This study JHT2870 Honduras: Guajiquiro + + This study JHT2871 Honduras: Guajiquiro + + This study JHT2872 Honduras: Guajiquiro + — This study JHT2896 Honduras: Intibucá + + This study JHT2897 Honduras: Intibucá + + This study JHT2898 Honduras: Intibucá + + This study JHT2899 Honduras: Intibucá + + This study

236

Table 4-1. Continued. JHT2907 Honduras: Intibucá + — This study Plectrohyla chrysopleura JHT3077 Honduras: Texíguat (Atlántida) + + This study JHT3081 Honduras: Texíguat (Atlántida) + + This study JHT3142 Honduras: Texíguat (Atlántida) + + This study JHT3166 Honduras: Texíguat (Atlántida) + + This study JHT3167 Honduras: Texíguat (Atlántida) + + This study JHT3169 Honduras: Texíguat (Atlántida) + + This study Plectrohyla dasypus JHT1624 Honduras: Cusuco + + This study JHT2988 Honduras: Cusuco + + This study Plectrohyla exquisita JHT1600 Honduras: Cusuco + + This study JHT1601 Honduras: Cusuco + + This study JHT1623 Honduras: Cusuco + + This study Plectrohyla guatemalensis JHT2058 Honduras: Cataguana + + This study JHT2059 Honduras: Cataguana + + This study JHT2060 Honduras: Cataguana + + This study Plectrohyla psiloderma JHT2746 Honduras: Cerro Celaque + + This study Ptychohyla euthysanota JS1244 Guatemala: Los Tarrales — + This study UTA A-54786 México: Chiapas AY843744 — Faivovich et al. (2005) Ptychohyla hypomykter JHT1622 Honduras: Cusuco + + This study JHT2272 Nicaragua: Selva Negra + — This study JHT2273 Nicaragua: Selva Negra + + This study JHT2274 Nicaragua: Selva Negra + + This study JHT2276 Nicaragua: Selva Negra + + This study JHT2300 Honduras: Comayagua + + This study JHT2494 Honduras: Azul Meámbar + + This study JHT2530 Honduras: Comayagua — + This study JHT2540 Honduras: Comayagua + + This study JHT2541 Honduras: Comayagua + + This study JHT2548 Honduras: Comayagua + + This study JHT2855 Honduras: Quebrada Grande + + This study JHT2916 Honduras: Cusuco + + This study JHT2930 Honduras: Pico Pijol — + This study JHT2931 Honduras: Pico Pijol + + This study JHT2933 Honduras: Pico Pijol + + This study JHT2934 Honduras: Pico Pijol + + This study JHT2935 Honduras: Pico Pijol + + This study JHT2936 Honduras: Pico Pijol + + This study JHT2945 Honduras: Pico Pijol + + This study JHT2946 Honduras: Pico Pijol + + This study JHT2949 Honduras: Pico Pijol + + This study JHT2950 Honduras: Pico Pijol + + This study JHT2951 Honduras: Pico Pijol + + This study JHT2954 Honduras: Pico Pijol + — This study

237

Table 4-1. Continued. JHT2998 Honduras: Cerro Santa Bárbara + + This study JHT2999 Honduras: Cerro Santa Bárbara + + This study IRL067 Honduras: Azul Meámbar + + This study IRL091 Honduras: Azul Meámbar + + This study IRL092 Honduras: Azul Meámbar + + This study IRL093 Honduras: Azul Meámbar + + This study IRL094 Honduras: Azul Meámbar + + This study IRL095 Honduras: Azul Meámbar + + This study N293 Nicaragua: Cerro Saslaya + + This study N528 Nicaragua: Cerro Kilambé + + This study N547 Nicaragua: Cerro Kilambé + + This study N993 Nicaragua: Cerro Peñas Blancas + + This study N1000 Nicaragua: Cerro Peñas Blancas + + This study N1004 Nicaragua: Cerro Peñas Blancas + + This study N1028 Nicaragua: Cerro Peñas Blancas + + This study Ptychohyla cf. hypomykter ENS8486 Guatemala: Izabal AY843745 Faivovich et al. (2005) Ptychohyla salvadorensis JHT2262 Honduras: Cerro Uyuca + + This study Ptychohyla spinipollex USNM 514381 Honduras: Pico Bonito AY843748 — Faivovich et al. (2005) Ptychohyla cf. spinipollex JHT3041 Honduras: Texíguat (Atlántida) + + This study JHT3042 Honduras: Texíguat (Atlántida) + + This study JHT3055 Honduras: Texíguat (Atlántida) + + This study JHT3056 Honduras: Texíguat (Atlántida) — + This study JHT3057 Honduras: Texíguat (Atlántida) + + This study JHT3058 Honduras: Texíguat (Atlántida) + + This study JHT3059 Honduras: Texíguat (Atlántida) + + This study JHT3070 Honduras: Texíguat (Atlántida) + + This study JHT3071 Honduras: Texíguat (Atlántida) + + This study JHT3072 Honduras: Texíguat (Atlántida) + + This study JHT3073 Honduras: Texíguat (Atlántida) + + This study JHT3114 Honduras: Texíguat (Atlántida) — + This study JHT3115 Honduras: Texíguat (Atlántida) + + This study JHT3116 Honduras: Texíguat (Atlántida) + + This study JHT3154 Honduras: Texíguat (Atlántida) + + This study JHT3170 Honduras: Texíguat (Atlántida) + + This study JHT3233 Honduras: Texíguat (Atlántida) + + This study JHT3234 Honduras: Texíguat (Atlántida) + + This study JHT3235 Honduras: Texíguat (Atlántida) + + This study JHT3236 Honduras: Texíguat (Atlántida) + + This study Smilisca cf. baudinii JHT2277 Nicaragua: Selva Negra + + This study JHT2592 Honduras: Los Naranjos + + This study JHT2593 Honduras: Los Naranjos + — This study JHT2938 Honduras: Pico Pijol + + This study N781 Nicaragua: Bosawas + + This study

238

Table 4-1. Continued. N1019 Nicaragua: Peñas Blancas + + This study Smilisca phaeota USNM 572702 Panamá: El Copé FJ784413 FJ766835 Crawford et al. (2010) USNM 572703 Panamá: El Copé FJ784433 FJ766834 Crawford et al. (2010) Smilisca sila USNM 572707 Panamá: El Copé FJ784578 FJ766837 Crawford et al. (2010) USNM 572708 Panamá: El Copé FJ784579 FJ766836 Crawford et al. (2010) Smilisca sordida N030 Nicaragua: Bosawas + + This study N031 Nicaragua: Bosawas + + This study N114 Nicaragua: Bosawas + + This study N115 Nicaragua: Bosawas — + This study N819 Nicaragua: Bosawas + + This study Tlalocohyla loquax JHT2247 Honduras: Cerro Uyuca + + This study JHT2268 Nicaragua: Selva Negra + + This study JHT2269 Nicaragua: Selva Negra — + This study Leptodactylidae (1) Leptodactylus fragilis JHT3238 Honduras: Texíguat (Atlántida) + + This study N462 Nicaragua: Bosawas + + This study N653 Nicaragua: Bosawas + + This study N841 Nicaragua: Bosawas + + This study Leptodactylus cf. fragilis USNM 572722 Panamá: El Copé FJ784331 FJ766745 Crawford et al. (2010) USNM 572725 Panamá: El Copé FJ784416 FJ766744 Crawford et al. (2010) MVUP 1836 Panamá: El Copé FJ784437 FJ766743 Crawford et al. (2010) USNM 572727 Panamá: El Copé FJ784453 FJ766742 Crawford et al. (2010) Leptodactylus savagei MVUP 1828 Panamá: El Copé FJ784394 FJ766748 Crawford et al. (2010) Leptodactylus poecilochilus KRL 0118 Panamá: El Copé FJ784321 — Crawford et al. (2010) Microhylidae (1) Hypopachus barberi JHT2630 Honduras: Guajiquiro + + This study JHT2634 Honduras: Guajiquiro + + This study Ranidae (4) Lithobates areolata KU 204370 USA: Kansas AY779229 — Hillis & Wilcox (2005) Lithobates berlandieri JSF1136 USA: Texas AY779235 — Hillis & Wilcox (2005) Lithobates brownorum JHT2329 Honduras: Los Naranjos + — This study JHT2349 Honduras: Los Naranjos + — This study JHT2350 Honduras: Los Naranjos + — This study JHT2351 Honduras: Los Naranjos + — This study JHT2466 Honduras: Los Naranjos + — This study JHT2525 Honduras: Comayagua + — This study JHT2560 Honduras: Comayagua + — This study JHT2561 Honduras: Comayagua + — This study JHT2562 Honduras: Comayagua + — This study JHT2563 Honduras: Comayagua + — This study JHT2564 Honduras: Comayagua + — This study JHT2565 Honduras: Comayagua + — This study JHT2566 Honduras: Comayagua + — This study

239

Table 4-1. Continued. JHT2567 Honduras: Comayagua + — This study JHT2569 Honduras: Comayagua + — This study JHT2760 Honduras: Intibucá + — This study JHT2761 Honduras: Intibucá + — This study JHT2814 Honduras: Quebrada Grande + — This study JHT2815 Honduras: Quebrada Grande + — This study JHT2818 Honduras: Quebrada Grande + — This study JHT2955 Honduras: above Guaymas + — This study JHT2956 Honduras: above Guaymas + — This study JHT2957 Honduras: above Guaymas + — This study JHT2958 Honduras: above Guaymas + — This study IRL051 Honduras: Azul Meámbar + — This study LK002 Honduras: Cerro Santa Bárbara + — This study Lithobates brownorum X forreri JHT2138 Honduras: Cerro Uyuca + + This study JHT2139 Honduras: Cerro Uyuca + + This study JHT2140 Honduras: Cerro Uyuca — + This study JHT2141 Honduras: Cerro Uyuca + + This study JHT2142 Honduras: Cerro Uyuca + + This study JHT2143 Honduras: Cerro Uyuca + + This study JHT2144 Honduras: Cerro Uyuca + + This study JHT2145 Honduras: Cerro Uyuca + + This study JHT2153 Honduras: Cerro Uyuca + + This study JHT2308 Honduras: Intibucá + + This study JHT2310 Honduras: Intibucá + + This study JHT2314 Honduras: Intibucá + + This study JHT2327 Honduras: Intibucá + + This study JHT2328 Honduras: Intibucá + + This study JHT2772 Honduras: Intibucá + + This study JHT2883 Honduras: Intibucá + + This study Lithobates capito TNHC 60195 USA: Florida AY779231 — Hillis & Wilcox (2005) Lithobates catesbeiana DMH 84-R2 USA: Kansas AY779206 — Hillis & Wilcox (2005) None No data GBX12841 — Hillis & Wilcox (2005) Lithobates chiricahuensis KU 194442 México: Durango AY779225 — Hillis & Wilcox (2005) KU 194419 USA: Arizona AY779226 — Hillis & Wilcox (2005) Lithobates clamitans JSF1118 USA Missouri AY779204 — Hillis & Wilcox (2005) Lithobates forreri KU 194581 México: Sinaloa AY779233 — Hillis & Wilcox (2005) Lithobates grylio MVZ 175945 USA: Florida AY779201 — Hillis & Wilcox (2005) Lithobates heckscheri MVZ 164908 USA: Florida AY779205 — Hillis & Wilcox (2005) Lithobates ―macroglossa‖ KU 195138 México: Chiapas AY779242 — Hillis & Wilcox (2005) UTA A-17185 Guatemala: Sololá AY779243 — Hillis & Wilcox (2005) Lithobates maculatus KU 195258 México: Oaxaca AY779207 — Hillis & Wilcox (2005) USNM 559483 Honduras: Pico Bonito DQ283303 — Frost et al. (2006) Lithobates maculatus (―Anura sp.‖) None México — B:AAE6665 BOLD Database

240

Table 4-1. Continued. Lithobates maculatus complex JHT2007 Honduras: Marale + + This study JHT2028 Honduras: Los Planes + + This study JHT2029 Honduras: Los Planes + + This study JHT2117 Honduras: Cataguana + + This study JHT2136 Honduras: Cerro Uyuca + + This study JHT2137 Honduras: Cerro Uyuca + + This study JHT2204 Honduras: Cerro Zarciadero + + This study JHT2244 Honduras: Cerro Uyuca + + This study JHT2439 Honduras: Texíguat (Yoro) + + This study JHT2492 Honduras: Azul Meámbar + + This study JHT2528 Honduras: Comayagua + + This study JHT2571 Honduras: Comayagua + + This study JHT2594 Honduras: Los Naranjos + + This study JHT2617 Honduras: Guajiquiro + + This study JHT2851 Honduras: Quebrada Grande + + This study JHT2852 Honduras: Quebrada Grande + + This study JHT2853 Honduras: Quebrada Grande + + This study JHT2939 Honduras: Pico Pijol + + This study JHT2940 Honduras: Pico Pijol + + This study IRL015 Honduras: Azul Meámbar + + This study IRL016 Honduras: Azul Meámbar + + This study IRL017 Honduras: Azul Meámbar + + This study IRL018 Honduras: Azul Meámbar + + This study IRL019 Honduras: Azul Meámbar + + This study IRL020 Honduras: Azul Meámbar + — This study IRL052 Honduras: Azul Meámbar + + This study IRL053 Honduras: Azul Meámbar + + This study IRL057 Honduras: Azul Meámbar + + This study IRL058 Honduras: Azul Meámbar + + This study IRL087 Honduras: Azul Meámbar + — This study N285 Nicaragua: Cerro Saslaya + + This study N371 Nicaragua: Bosawas + + This study N610 Nicaragua: Cerro Saslaya + + This study N638 Nicaragua: Cerro Saslaya + — This study N649 Nicaragua: Cerro Saslaya + + This study N920 Nicaragua: Cerro Kilambé + + This study N1009 Nicaragua: Peñas Blancas + + This study N1021 Nicaragua: Peñas Blancas + + This study Lithobates magnaocularis KU 194592 México: Sonora AY779239 — Hillis & Wilcox (2005) Lithobates montezumae KU 195251 México: Morelos AY779223 — Hillis & Wilcox (2005) Lithobates neovolcanica KU 194536 México: Michoacan AY779236 — Hillis & Wilcox (2005) Lithobates okaloosae None (toe clip) USA: Florida AY779203 — Hillis & Wilcox (2005) Lithobates omiltemana KU 195179 México: Guerrero AY779238 — Hillis & Wilcox (2005)

241

Table 4-1. Continued. Lithobates palmipes KU 202896 Ecuador: Napo AY779211 — Hillis & Wilcox (2005) Lithobates palustris KU 204425 USA: Indiana AY779228 — Hillis & Wilcox (2005) Lithobates pipiens JSF1119 USA: Ohio AY779221 — Hillis & Wilcox (2005) None No data GBX12841 — Hillis & Wilcox (2005) Lithobates septentrionalis TNHC (no #) Canada: Ontario AY779200 — Hillis & Wilcox (2005) Lithobates sevosa TNHC 60194 USA: Mississippi AY779230 — Hillis & Wilcox (2005) Lithobates spectabilis KU 195186 México: Hidalgo AY779232 — Hillis & Wilcox (2005) Lithobates sphenocephalus JSF845 USA: Kansas AY779251 — Hillis & Wilcox (2005) USC7448 USA: Florida AY779252 — Hillis & Wilcox (2005) Lithobates subaquavocalis TNHC (no #) USA: Arizona AY779227 — Hillis & Wilcox (2005) Lithobates sylvatica MVZ 137426 USA: New York AY779198 — Hillis & Wilcox (2005) Lithobates taylori JHT2263 Nicaragua: Selva Negra + + This study JHT2264 Nicaragua: Selva Negra + + This study JHT2265 Nicaragua: Selva Negra — + This study N918 Nicaragua: Cerro Kilambé + + This study N919 Nicaragua: Cerro Kilambé + + This study N997 Nicaragua: Peñas Blancas + + This study N1020 Nicaragua: Peñas Blancas + + This study TCWC 55963 Nicaragua: 2.5 mi NW Rama AY779244 — Hillis & Wilcox (2005) Lithobates tlaloci KU 194436 México: Districto Federal AY779234 — Hillis & Wilcox (2005) Lithobates vibicarius MVZ 149033 Costa Rica: San José AY779208 — Hillis & Wilcox (2005) Lithobates warszewitschii USNM 572770 Panamá: El Copé FJ784454 FJ766752 Crawford et al. (2010) USNM 572779 Panamá: El Copé FJ784552 FJ766751 Crawford et al. (2010) USNM 572780 Panamá: El Copé FJ784558 FJ766750 Crawford et al. (2010) Lithobates ―warszewitschii‖ JSF1127 Panamá: Coclé AY779209 — Hillis & Wilcox (2005) Lithobates aff. warszewitschii USNM 572787 Panamá: El Copé FJ784384 FJ766749 Crawford et al. (2010) Lithobates cf. warszewitschii N334 Nicaragua: Bosawas — + This study N599 Nicaragua: Bosawas + + This study Lithobates spp. (―Anura sp.‖) None México — B:AAB6680 BOLD Database None México — B:AAB6680 BOLD Database None México — B:AAB6680 BOLD Database None México — B:AAB6680 BOLD Database None México — B:AAB6680 BOLD Database None México — B:AAB6680 BOLD Database None México — B:AAB6680 BOLD Database None México — B:AAB6680 BOLD Database None México — B:AAB6679 BOLD Database None México — B:AAB6679 BOLD Database None México — B:AAB6678 BOLD Database None México — B:AAB6681 BOLD Database None México — B:AAB6682 BOLD Database None México — B:AAB6682 BOLD Database None México — B:AAB6682 BOLD Database

242

Table 4-1. Continued. None México — B:AAE6666 BOLD Database None México — B:AAK8798 BOLD Database None México — B:AAC6120 BOLD Database None México — B:AAC6120 BOLD Database Lithobates sp. 1 QCAZ 13219 Ecuador: Esmeraldas AY779213 — Hillis & Wilcox (2005) Lithobates sp. 2 KU 2044420 México: San Luis Potosí AY779224 — Hillis & Wilcox (2005) Lithobates sp. 3 KU 194559 México: Michoacan AY779250 — Hillis & Wilcox (2005) Lithobates sp. 4 AMNH 124167 Panama: Chiriquí AY779245 — Hillis & Wilcox (2005) Lithobates sp. 5 LACM 146764 Costa Rica: Heredia AY779246 — Hillis & Wilcox (2005) Lithobates sp. 6 LACM 146810 Costa Rica: Puntarenas AY779247 — Hillis & Wilcox (2005) Lithobates sp. 7 KU 194492 México: Jalisco AY779241 — Hillis & Wilcox (2005) Lithobates sp. 8 KU 195346 México: Puebla AY779248 — Hillis & Wilcox (2005) Strabomantidae (2) Pristimantis caryophyllaceus MVUP 1925 Panamá: El Copé FJ784473 FJ766776 Crawford et al. (2010) USNM 572343 Panamá: El Copé FJ784375 FJ766772 Crawford et al. (2010) Pristimantis cerasinus N121 Nicaragua: Bosawas + — This study Pristimantis ridens A USNM 572416 Panamá: El Copé FJ784388 FJ766808 Crawford et al. (2010) USNM 572417 Panamá: El Copé FJ784389 FJ766807 Crawford et al. (2010) Pristimantis ridens B MVUP 1829 Panamá: El Copé FJ784398 FJ766806 Crawford et al. (2010) USNM 572415 Panamá: El Copé FJ784399 FJ766805 Crawford et al. (2010) Pristimantis cf. ridens N090 Nicaragua: Bosawas + + This study N143 Nicaragua: Bosawas + + This study N144 Nicaragua: Bosawas + + This study N201 Nicaragua: Bosawas + + This study N263 Nicaragua: Cerro Saslaya + + This study N541 Nicaragua: Cerro Kilambé — + This study N998 Nicaragua: Cerro Peñas Blancas + + This study N1008 Nicaragua: Cerro Peñas Blancas + + This study

243

Table 4-2. Average nucleotide compositions of various taxonomic groups Taxon # of sequences Avg. Length A G C T COI Amphibia 390 636.3 23.9% 18.0% 27.6% 30.5% Anura 256 629.3 23.3% 18.1% 28.0% 30.6% Caudata 144 649.5 25.1% 17.9% 26.7% 30.2% 16S Amphibia 436 537.0 31.0% 20.1% 23.2% 25.7% Anura 277 553.6 30.0% 20.2% 23.8% 26.1% Caudata 269 512.5 33.2% 19.9% 22.3% 24.7%

244

Table 4-3. Potential candidate species identified through this study; UCS = Unconfirmed Candidate Species, CCS = Confirmed Candidate Species; DCL = Deep Conspecific Lineage; for IUCN Red List categories, CR = Critically Endangered, EN = Endangered, and VU = Vulnerable. Interspecific divergences are measured at the genus- level. Potential A priori IUCN Intraspecific taxonomic assignment Current status Red List Divergence Interspecific Divergence Notes Plethodontidae (16) Bolitoglossa sp. inq. 1 Bolitoglossa cataguana CR 16S =0.0% 16S =2.3–8.7%, Described by Townsend et al. (2009a) COI =0.7% COI =11.6–20.9% (0.0% from B. decora for 16S) Bolitoglossa sp. inq. 2 Bolitoglossa oresbia CR 16S =0.2% 16S =3.0–9.7% Conspecific with B. oresbia and to COI =0.1% COI =10.0–22.1% represent a range extension (Townsend et al. 2011b) Bolitoglossa celaque Bolitoglossa sp. 1 DCL EN 16S =0.8% 16S =1.8–10.0% Population from Sierra de Puca- COI =0.3% COI =4.5–21.7% Opalaca; under study by author (within B. celaque complex: 16S =1.0–1.5%, COI =1.7– 3.0%) Bolitoglossa celaque Bolitoglossa sp. 2 DCL EN 16S =0.3% 16S =2.0–10.3% Population from Montañas de la COI =0.2% COI =5.2–21.5% Sierra; under study by author (within B. celaque complex: 16S =1.3–1.5%, COI =1.7– 3.6%) Bolitoglossa conanti Bolitoglossa sp. 3 DCL EN 16S =0.0% 16S =3.1–9.3% Population from Sierra de Omoa; COI =0.0% COI =9.6–21.4% under study by author (within B. conanti complex: 16S =0.6–1.6%, COI =2.7– 3.1%) Bolitoglossa conanti Bolitoglossa sp. 4 DCL EN 16S =0.0% 16S =3.5–9.4% Population from Sierra del Merendón; COI =0.0% COI =9.6–20.4% under study by author (within B. conanti complex: 16S =0.8–1.6%, COI =2.5– 3.7%) Bolitoglossa Bolitoglossa sp. 5 CCS CR 16S =0.2% 16S =2.0–8.6% Population from Texíguat; under study porrasorum COI =0.7% COI =9.2–23.3% by author Bolitoglossa Bolitoglossa sp. 6 CCS CR 16S =N/A 16S =2.0–8.1% Population from Pico Bonito; under porrasorum COI = N/A COI = N/A study by author Bolitoglossa rufescens Bolitoglossa sp. 7 CCS 16S =0.2% 16S =5.5–10.3% Syntopic in Sierra de Omoa with B.

245

COI =0.2% COI =17.6–22.7% nympha; under study by author

Nototriton sp. inq. 1 Nototriton tomamorum CR 16S = N/A 16S =3.3–5.4% Described by Townsend et al. (2010a) COI = N/A COI =11.3–14.1% Nototriton sp. inq. 2 Nototriton picucha CR 16S =0.0% 16S =1.2–5.2% Described in Chapter 5 and by COI =0.2% COI = 2.9–12.3% Townsend et al. (2011a) Nototriton sp. inq. 3 Nototriton sp. 1 CCS CR 16S =0.0% 16S =1.7–5.8% Population from Sierra de Botaderos COI =0.6% COI =2.9–12.5% Nototriton sp. inq. 4 Nototriton limnospectator N/A 16S =0.3% 16S =1.2–5.2% Conspecific with N. limnospectator and COI =0.2% COI =7.8–12.6% to represent a range extension (Townsend et al. 2011b) Nototriton sp. inq. 5 Nototriton lignicola N/A 16S =0.2% 16S =2.4–5.9% Conspecific with N. lignicola and to COI = N/A COI =10.0–13.9% represent a range extension (Townsend et al. 2011b) Nototriton barbouri Nototriton sp. 2 CCS CR 16S =N/A 16S =1.2–6.0% Population from Pico Bonito; described COI = N/A COI = N/A in Chapter 5 (as N. sp. A) Nototriton barbouri Nototriton sp. 3 CCS CR 16S =N/A 16S =1.7–5.8% Population from Texíguat; described in COI = N/A COI =5.9%–13.6% Chapter 5 (as N. sp. B) Oedipina sp. inq. 1 Oedipina kasios N/A 16S =0.1% 16S =2.6–11.3% Conspecific with O. kasios and to COI = N/A COI =6.6–15.7% represent a range extension (Townsend et al. 2011b) Oedipina sp. inq. 2 Oedipina nica EN 16S =0.1% 16S =2.6–10.2% Described by Sunyer et al. (2010) COI =0.0% COI =6.6–16.4% Oedipina sp. inq. 3 Oedipina koehleri EN 16S =0.0% 16S =3.1–8.4% Described by Sunyer et al. (2011) COI = N/A% COI =14.4–16.2% Oedipina gephyra Oedipina petiola CR 16S = N/A 16S =3.7–10.5% Described by McCranie & Townsend COI = N/A COI = N/A (2011) Bufonidae (2) Incilius coniferus Incilius sp. DCL DD 16S = N/A 16S =3.6–9.3% Population from northern Nicaragua; COI = N/A COI =11.2–17.5% under study by Sunyer, Townsend, and (from Panamanian I. Travers coniferus: 16S =0.7%, COI =3.5%) Rhaebo haematiticus Rhaebo sp. UCS DD 16S =0.0% 16S =10.3–12.4% Population from northern Nicaragua; COI = 0.2% COI =16.1–19.5% under study by Sunyer, Townsend, and (from Panamanian R. Travers haematiticus: 16S =2.2%, COI =5.4%) Craugastoridae (3) Craugastor aurilegulus Craugastor sp. 1 DCL EN 16S =0.1% 16S =4.9–21.0% Population from vicinity of Texíguat;

246

COI =0.6% COI =11.2–25.6% may represent DCL (from Pico Bonito C. aurilegulus: 16S =1.1%, COI =3.9%) Craugastor sp. inq. 1 Craugastor sp. 2 UCS DD 16S = N/A 16S =7.5–21.1% Population from northern Nicaragua; COI = N/A COI =16.8–26.6% field identification Diasporus diastema Craugastor sp. inq. 2 Craugastor sp. 3 UCS DD 16S =0.0% 16S =7.5–21.3% Population from northern Nicaragua COI =0.0% COI =11.4–26.7% Eleutherodactylidae (1) Diasporus diastema Diasporus sp. UCS DD 16S = N/A 16S =7.1–21.6% Population from northern Nicaragua COI = N/A COI =15.7–27.2% Hylidae (5) Plectrohyla Plectrohyla sp. UCS CR 16S =0.0% 16S =2.6–13.8% Population from Cordillera de La Flor- guatemalensis COI =0.1% COI =8.8–23.8% La Muralla Ptychohyla Ptychohyla sp. 1 UCS DD 16S =N/A 16S =5.5– 14.7% Population from eastern Guatemala hypomykter COI = N/A COI = N/A Ptychohyla spinipollex Ptychohyla sp. 2 UCS CR 16S =0.2% 16S =4.8–13.4% Population from vicinity of Texíguat; COI =0.5% COI =15.1–22.6% may represent DCL Smilisca baudinii Smilisca sp. 1 UCS LC 16S =0.0% 16S =1.6–15.0 % Populations from northern Nicaragua COI =0.0% COI =6.6–22.7% Smilisca baudinii Smilisca sp. 2 UCS LC 16S =0.0% 16S =1.6–14.4% Populations from Honduras COI =0.1% COI =6.6–22.7% Leptodactylidae (1) Leptodactylus fragilis Leptodactylus sp. UCS LC 16S =0.0% 16S =3.6–21.4% Populations from Panama (Crawford et COI =0.3% COI =15.7–26.2% al. 2010); Chortís Block samples considered conspecific with nominal form Ranidae (7) Lithobates brownorum Lithobates sp. 1 CCS EN 16S =0.8% 16S =2.2–16.5% Populations from Southern Cordillera X forreri COI =0.5% COI =9.7–22.4% of Serranía; being described by author Lithobates maculatus Lithobates sp. 2 UCS EN 16S =0.0% 16S =1.6–12.2% Populations from highlands in north- COI =0.1% COI =2.4–20.5% central Nicaragua; under study by author Lithobates maculatus Lithobates sp. 3 UCS EN 16S =0.4% 16S =1.6–12.8% Populations from low to moderate COI =0.5% COI =2.4–21.1% elevations in northeastern Nicaragua; under study by author Lithobates maculatus Lithobates sp. 4 UCS EN 16S =0.0% 16S =1.8–12.8% Populations from Texíguat and Sierra COI =0.0% COI =5.7–20.7% de Sulaco; under study by author

247

Lithobates maculatus Lithobates sp. 5 UCS EN 16S =0.6% 16S =2.9–12.3% Populations from throughout Honduran COI =1.2% COI =7.2–21.5% Serranía; under study by author Lithobates maculatus Lithobates sp. 6 UCS EN 16S =N/A 16S =2.6–12.6% Population from vicinity of Pico Bonito; COI =N/A COI = N/A under study by author Lithobates Lithobates sp. 7 UCS DD 16S =0.6% 16S =8.2–17.2 Populations from northern Nicaragua warszewitschii COI = N/A COI =10.1–22.8% Strabomantidae (1) Pristimantis ridens Pristimantis sp. UCS DD 16S =0.3% 16S =4.8–22.2% Populations from northern Nicaragua COI =0.1% COI =9.1–27.0%

248

Table 4-4. Results of BLASTN searches of the NCBI database for 16S consensus sequences representing 10 populations of taxonomically-unassigned salamanders. Max Max Query Sample Matching sequence (GenBank accession #) ident score coverage Bolitoglossa sp. inq. 1 1. Bolitoglossa decora (GU725449) 99% 904 100% Cataguana 2. Bolitoglossa morio (GU725452) 97% 830 100% 3. Bolitoglossa porrasorum (AF526151) 96% 824 100% 4. Bolitoglossa morio (AF218495) 96% 819 100% 5. Bolitoglossa flavimembris (GU725449) 96% 808 100% Bolitoglossa sp. inq. 2 1. Bolitoglossa flavimembris (GU725449) 96% 843 100% IRL 081 2. Bolitoglossa diaphora (GU725447) 95% 833 100% 3. Bolitoglossa morio (AF526144) 95% 821 100% 4. Bolitoglossa morio (GU725452) 95% 815 100% 5. Bolitoglossa dunni (GU725446) 95% 811 100% Nototriton sp. inq. 1 1. Nototriton lignicola (AF199204) 95% 780 100% Texiguat 2. Nototriton barbouri (AF199201) 94% 760 100% 3. Nototriton richardi (AF199206) 94% 758 100% 4. Nototriton abscondens (AF199199) 94% 758 100% 5. Nototriton brodiei (AF199202) 94% 754 100% Nototriton sp. inq. 2 1. Nototriton lignicola (AF199204) 97% 837 100% Agalta 2. Nototriton brodiei (AF199202) 96% 804 100% 3. Nototriton barbouri (AF199201) 96% 804 100% 4. Nototriton richardi (AF199206) 95% 785 100% 5. Nototriton saslaya (GU981761) 95% 776 100% Nototriton sp. inq. 3 1. Nototriton lignicola (AF199204) 96% 813 99% Botaderos 2. Nototriton barbouri (AF199201) 96% 802 99% 3. Nototriton richardi (AF199206) 95% 784 99% 4. Nototriton brodiei (AF199202) 95% 780 99% 5. Nototriton saslaya (GU981761) 94% 758 99% Nototriton sp. inq. 4* 1. Nototriton lignicola (AF199204) 99% 898 100% Cataguana 2. Nototriton brodiei (AF199202) 96% 798 100% 3. Nototriton barbouri (AF199201) 96% 798 100% 4. Nototriton richardi (AF199206) 95% 780 100% 5. Nototriton saslaya (GU981761) 94% 743 100% Nototriton sp. inq. 5 1. Nototriton lignicola (AF199204) 96% 821 100% Meámbar 2. Nototriton brodiei (AF199202) 96% 804 100% 3. Nototriton barbouri (AF199201) 96% 804 100% 4. Nototriton richardi (AF199206) 96% 797 100% 5. Nototriton saslaya (GU981761) 94% 765 100% Oedipina sp. inq. 1* 1. Oedipina kasios (FJ196866) 99% 893 100% Mta de Yoro 2. Oedipina kasios (FJ196867) 99% 893 100% 3. Oedipina quadra (FJ196865) 95% 769 99% 4. Oedipina uniformis (AF199230) 91% 682 100% 5. Oedipina grandis (FJ196864) 91% 676 100% Oedipina sp. inq. 2 1. Oedipina kasios (FJ196866) 97% 837 99% nica 2. Oedipina kasios (FJ196867) 97% 835 99% 3. Oedipina quadra (FJ196865) 94% 765 100% 4. Oedipina poelzi (AF199223) 92% 691 99% 5. Oedipina grandis (FJ196864) 91% 680 99% Oedipina sp. inq. 3 1. Oedipina cyclocauda (AF199214) 96% 811 100% 2. Oedipina pseudouniformis (AF199227) 96% 808 100% 3. Oedipina cyclocauda (AF199215) 96% 806 100% 4. Oedipina uniformis (AF199230) 95% 782 100% 5. Oedipina uniformis (AF199229) 95% 776 100%

249

CHAPTER 5 CRYPTIC DIVERSITY AND REVISIONARY SYSTEMATICS OF CHORTÍS HIGHLAND MOSS SALAMANDERS (CAUDATA: PLETHODONTIDAE)

Inconspicuous form, obscure or secretive behavior, and challenging habitat associations are principal reasons why cryptozoic1 diversity historically has been overlooked and understudied (Pfenninger & Schwenk 2007). Molecular phylogenetic analyses, facilitated through techological advances like PCR (Mullis 1990), Sanger sequencing (Sanger et al. 1977), and next generation DNA sequencing (Schuster

2008), have led to the discovery of numerous genetically divergent but morphologically cryptic lineages (Bickford et al. 2007). Many tropical lungless salamanders (Amphibia:

Caudata: Plethodontidae) are characterized by their small size, conserved morphology, and secretive habits, and present some of the most challenging subjects for systematic study (e.g. Wake & Elias 1983; Highton 1995; Jockusch et al. 2001).

As revealed in the preceding chapters, lungless salamanders (Caudata:

Plethodontidae) represent the most diverse (in both named and candidate species) and the most endangered group of vertebrates in the Chortís Block. In this chapter, I evaluate cryptic diversity and provide a taxonomic revision of an endemic clade of cryptozoic salamanders: the genus Nototriton, or moss salamanders.

Salamanders as Models for Evolutionary Study

Wake (2009) provided an excellent review of the advances in the biological sciences that have resulted in taxon-based research focusing on salamanders. In brief

(Wake 2009:333):

The clade [Caudata] is widespread and diverse, yet sufficiently small that one can keep all of the species in mind. This facilitates research from

1 Cryptozoic – Refers to organisms of small size, cryptic coloration, and that exhibit secretive behavior

250

diverse perspectives: systematics and phylogenetics, morphology, development, ecology, neurobiology, behavior, and physiology. Different avenues of research offer unique perspectives on how a relatively old vertebrate clade has diversified. An integrated, hierarchically organized, multidimensional program of research on a taxon illuminates many general principles and processes. Among these are the nature of species and homology, adaptation and adaptive radiations, size and shape in relation to issues in organismal integration, ontogeny and development in relation to phylogeny, the ubiquity of homoplasy, ecological niche conservation, species formation, biodiversity, and conservation.

Specific to the use of salamanders as a study group in the Chortís Block, I recognize a number of advantages. The majority of living salamanders (613 species;

AmphibiaWeb, 2 September 2011) are members of the family Plethodontidae (417 species, 68% of all living Caudata), and the majority of plethodontids are members of

Neotropical-restricted genera (265 Neotropical species, 63.6% of all Plethodontidae,

43% of all living Caudata), indicating that an ancient order underwent extraordinary

Cenozoic diversification in tropical America. The Chortís Block is demonstrated to have a diverse endemic salamander fauna, with the majority of endemism found in the highlands (McCranie & Wilson 2002; Wilson & McCranie 2004b). In another highland area of high ecophysiographic heterogeneity, the comparatively well-studied

Appalachian Mountains of eastern North America, plethodontids are highly diverse and demonstrate a wide variety of evolutionary patterns and processes, including both adaptive and non-adaptive radiations, parallel evolution, and ecological niche conservatism accompanied with morphological conservatism (e.g. Tilley & Mahoney

1996; Kozak & Wiens 2006, 2010; Kozak et al. 2006; Crespi et al. 2010).

Among the taxa known from the Chortís Highlands, a number of taxonomic problems exist, and numerous areas of presumably suitable habitat remain unsampled.

The Chortís Highlands salamander fauna is highly threatened by loss of habitat,

251

potentially due to climate change (as demonstrated in the Appalachian Mountains by

Milanovich et al. 2010), and by both direct and indirect effects from wider declines in highland amphibian communities (Rovito et al. 2009). Of the 36 species of salamanders recorded from the Chortís Highlands (reviewed in Appendix 1), 11 are Critically

Endangered, 12 are Endangered, 2 are Vulnerable, 1 is Near Threatened, 6 are Least

Concern, and 4 are Data Deficient (AmphibiaWeb 2011, IUCN 2011), meaning that close to 70% of all known salamander species in the Chortís Highlands are in the

IUCN’s three highest threat categories (over 80% when poorly-known, Data Deficient species are included).

Salamander Diversity in the Chortís Block

Salamanders originated and initially diversified in Laurasia (Zhang and Wake

2009), with a single family, Plethodontidae, dispersing into the Neotropics beginning in the late Cretaceous Period (Hanken and Wake 1982; Vieites et al. 2007). Plethodontid salamanders are hypothesized to have dispersed from north-to-south as the Central

American land bridge formed, possibly colonizing the Chortís Block prior to the divergence of Cryptotriton from remaining neotropical plethodontids (Appendix 3 of

Wiens et al. 2007).

Neotropical salamanders can be characterized by their high degree of diversification and widespread morphological conservatism and homoplasy (Wake

1991; Parra-Olea & Wake 2001). Diverse radiations of plethodontids are well documented in western Nuclear Central America (southern México and Guatemala;

Wake 1987) and southern Central America (Costa Rica and western Panama; Wake

2005); however, much of this cryptic diversity has been revealed only recently by molecular studies (e.g., Garcia-París et al. 2000; Hanken et al. 2005). Perhaps the best

252

example of this taxonomic uncertainty is seen in the suite of small salamanders currently divided into the genera Chiropterotriton, Cryptotriton, Dendrotriton, and

Nototriton (García-París & Wake 2000). As recently as 1983, these morphologically conserved genera were considered part of a single wide-ranging genus, Chiropterotriton

(Wake & Elias 1983). The application of molecular techniques coupled with increased sampling and rigorous studies of external and internal morphology have led to the current taxonomy, which continues to be refined as new populations and taxa are discovered and described (e.g., McCranie et al. 2008; Townsend et al 2010a). Even among well-studied taxa and regions, such as Bolitoglossa in Costa Rica, morphologically conserved taxa have presented long-standing problems that can best be resolved using an integrative approach that maximizes use of both molecular and morphological evidence (e.g., Hanken et al. 2005).

Like western Nuclear and southern Central America, the Chortís Highlands are a site of remarkable plethodontid diversity (McCranie and Wilson 2002). The Bolitoglossa

(Magnadigita) dunni species group, a Chortís Highlands endemic radiation of 12 nominal species (McCranie et al. 2005), has the highest estimated rate of diversification of any major clade of Mesoamerican salamanders (Wiens et al. 2007).

Prior to initiation of this study, there were 13 recognized species of moss salamanders, genus Nototriton, distributed across Central American highland forests

(García-París & Wake 2000). These salamanders are difficult to differentiate, given their small adult size (all species < 40 mm standard length) and highly conserved morphology (Good & Wake 1993; García-París & Wake 2000). Their utilization of moss mats, deep leaf litter, and rotten logs as habitat makes them particularly difficult to

253

detect (Good & Wake 1993). The genus is subdivided into three well-supported species groups: the N. barbouri group, the N. picadoi group, and the N. richardi group

(Papenfuss & Wake 1987, Savage 2002). Each morphologically-defined species group corresponds to a clade supported by data from the mitochondrial genes 16S and cytochrome b (García-París & Wake 2000, Wiens et al. 2007).

The Nototriton barbouri group is the only clade of Nototriton restricted to the

Chortís Highlands (García-París & Wake 2000). This group contains five nominal species inhabiting highland forests in eastern Guatemala and northern and central

Honduras. The five putative species of the N. barbouri group are each endemic to isolated cloud forest localities in the Chortís Highlands (Figure 5-1): N. barbouri

(Schmidt 1936), N. brodiei Campbell & Smith 1998, N. lignicola McCranie & Wilson

1996 [1997], N. limnospectator McCranie, Wilson & Polisar 1998 and N. stuarti Wake &

Campbell 2000. Another species of Nototriton found in the Chortís Highlands, Nototriton saslaya, is extralimital to the Nototriton barbouri group, representing the northernmost representative of the N. picadoi group, endemic to a single highland forest area at the southern edge of the Nuclear Central American piedmont in north-central Nicaragua.

Nototriton saslaya aside, the remaining five species of Chortís Highland Nototriton all occur in small areas of highland forest in a belt from easternmost Guatemala to north- central Honduras, corresponding to the Northern and Central Cordilleras of the Chortís

Highlands (Figure 5-1).

While the salamander fauna of Nuclear Central America has been the focus of considerable study (Dunn 1924; Schmidt 1933, 1936; Wake 1987; McCranie & Wilson

1993, 1997; Campbell & Smith 1998), new species continue to be discovered and

254

Figure 5-1. Distribution of Nototriton in the Chortís Highlands. Shaded areas >1000 m elevation; open circle = N. picucha; open triangles = N. brodiei; closed triangles = N. limnospectator; open square = N. sp. A; solid square = N. barbouri; open star = N. tomamorum, N. sp. B; solid star = N. stuarti; open diamonds = N. lignicola; solid diamond = N. saslaya; N. sp A. and N. sp. B refer to populations currently assigned to N. barbouri. described, often from areas that are considered relatively well known in terms of their amphibian diversity (Campbell et al. 2010; McCranie et al. 2005, 2008; Rovito et al.

2010; Townsend et al. 2009a, 2010a; Vásquez-Almazán et al. 2009). This can be attributed in part to the small size and cryptic nature of many tropical salamanders, which, in the case of Nototriton, results in the majority of species being known from 10 or fewer specimens (Good & Wake 1993; McCranie & Wilson 2002). Indeed, the five species of the N. barbouri group have been represented previously in published phylogenetic studies by only seven samples from four putative taxa (García-París &

Wake 2000; Wiens et al. 2007; Adams et al. 2009).

255

Intensive sampling in the Chortís Highlands over the past six years has led to the collection of additional samples of nominal taxa and the discovery of new populations of

Nototriton from isolated localities (Chapters 3 and 4). In order to evaluate the evolutionary relationships among Chortís Highland Nototriton in light of their currently accepted taxonomy, I employed two strategies for analysis of a three gene mtDNA dataset, which supplements the 16S/COI dataset used in Chapter 4 with the a fragment of the gene cytochrome b (cyt b). First, I used a DNA barcoding approach to identify potential species boundaries and candidate species by evaluating intra- and interspecific sequence divergence and the effectiveness of each gene for species delimitation using uncorrected distance-based methods. Second, I inferred phylogenetic relationships using Bayesian and maximum likelihood (ML) analysis with evolutionary models partitioned by gene and codon. Finally, I applied the distance-based and phylogenetic results, supplemented by comparative data from external morphology and osteology, to redefine N. barbouri and describe three new species of Nototriton endemic to the Chortís Highlands of Honduras.

Methods and Materials

Sampling

Taxa and samples used in this study, along with their associated locality data and museum and GenBank accession numbers, are presented in Table 5-1. All nominal taxa in the N. barbouri group were included, with the exception of N. stuarti, which is only known from a single specimen collected in 1991 (Wake & Campbell 2000).

Nototriton saslaya, the only representative of the N. picadoi group found in the Chortís

Highlands, was used as an outgroup. Institutional abbreviations follow those standardized by the American Society of Ichthyologists and Herpetologists

256

(http://www.asih.org/codons.pdf), except for IRL (field series of Ileana R. Luque-Montes) used for a sample donated to the Florida Museum of Natural History in May 2009 that remains uncataloged. Forest formations follow Holdridge (1967) as applied by McCranie

& Wilson (2002).

In this chapter, I present genetic evidence for two undescribed species heretofore considered synonymous with N. barbouri. However, I do not include the morphological diagnoses and descriptions associated with those species, as they have not yet been published at the time of completing this dissertation.

DNA Extraction, PCR Amplification, and Sequencing

Template DNA was extracted from muscle tissue using the Qiagen PureGene

DNA Isolation Kit following manufacturer’s instructions. Fragments of the mitochondrial genes 16S large subunit RNA (16S), cytochrome b (cyt b), and cytochrome oxidase subunit I (COI) were amplified using the primers 16Sar-L and 16Sbr-H for 16S (Palumbi et al. 1991), MVZ15L and MVZ16H for cyt b (Moritz et al. 1992), and dgLCO-1490 and dgHCO-2198 for COI (Meyer 2003). PCR reactions were typically 20L in total volume, containing ~25ng of DNA template, 4 L 5X PCR buffer, 1.2 L MgCl2 (25mM), 0.09 L dNTPs (10 mM), 0.8 L of each primer (10M), 0.2 L GoTaq Flexi polymerase

(Promega, Madison, WI, USA), and 11.91 L H2O. Amplification profiles were as follows: for 16S, an initial denaturation for 3 minutes at 94°C, 35 cycles of denaturation at 94°C for 45 seconds, annealing at 50°C for 45 seconds, and extension at 72°C for 45 seconds, with a final elongation at 72°C for 5 minutes; for cyt b, an initial denaturation for 3 minutes at 94°C, followed by 38 cycles of 94°C for 30 seconds, 48°C for 1 minute, and 1 minute for 45 seconds, final elongation at 72°C for 5 minutes; and for COI,

257

denaturation for 1.5 minutes, 37 cycles of 94°C for 40 seconds, 45°C for 40 seconds and 72°C for 40 seconds, with a final elongation at 72°C for 5 minutes. All PCR products were verified using electrophoresis on a 1.5% agarose gel stained with ethidium bromide. Unincorporated nucleotides were removed from PCR products using

1 uL of ExoSAP-IT (USB, Santa Clara, CA, USA) per 10uL of PCR product. I cycle sequenced both forward and reverse stands using the BigDye Terminator 3.1 Cycle

Sequencing kit, followed by spin column filtration through Sephadex before electrophoresing the products on an ABI 3130xl (Applied Biosystems, Inc).

Sequence Alignment and Model Selection

A dataset containing all available sequences of Nototriton was generated from a combination of our newly generated sequence data (Table 5-1) and published data available from NCBI (http://www.ncbi.nlm.nih.gov/). Sequences were aligned using

ClustalW (Thompson et al. 1994) within the program package MEGA5 (Tamura et al.

2011) using the default parameters. The dataset was partitioned by gene (16S, which codes for RNA) and by codon position (1st, 2nd, 3rd) for cyt b and COI (both protein- coding genes), and selected the best fit model of nucleotide evolution for each gene and each partition with the program jModeltest v0.1 (Posada 2008), which uses PhyML 3.0

(Guindon & Gascuel 2003) to estimate models under a likelihood framework. The number of substitution schemes was set to three to limit the number of models tested to

24, corresponding to the number of different models that can be implemented in

MrBayes 3.1.2 (Huelsenbeck & Ronquist 2001). Models selected for each partition are summarized in Table 5-2.

258

Sequence Analyses

A DNA barcoding approach was utilized in order to determine whether monophyletic clusters of samples corresponding to named taxa, as well as to identify candidate species for further taxonomic evaluation (Vences et al. 2005; Smith et al.

2008). Uncorrected (p-distance) pairwise sequence divergence was calculated for all samples and for each gene, and uncorrected neighbor-joining (NJ) distance trees

(10,000 bootstrap pseudoreplicates) were estimated to provide graphical representation of intra- and interspecific variation. Sequence divergence estimation and NJ analyses were performed in MEGA5 (Tamura et al. 2011).

Two of the three target genes are protein-coding, making them potentially susceptible to substitution saturation at the third codon position that, when excessive, can render the locus phylogenetically uninformative (Halanych & Robinson 1999; Farias et al. 2004; Parra-Olea et al. 2004). Substitution saturation was evaluated using an entropy-based index calculated in the program package DAMBE 5.2.34 (Xia & Xie

2001), which uses two simulated topologies (one perfectly symmetrical and one extremely asymmetrical) based on the supplied sequence data to calculate an Index of

Substitution Saturation (ISS). High substitution saturation is indicated by an ISS significantly greater than or not significantly different than the Critical ISS (ISS.C) for both the perfectly symmetrical topology and an extremely asymmetrical topology (Xia et al.

2003; Xia & Lemey 2009) determined with two-tailed t-test. Prior to measuring substitution saturation, the proportion of invariant sites was estimated for each gene in

DAMBE 5.2.34 (Xia & Xie 2001) using a goodness-of-fit test with a Poisson+Invariant distribution, and the proportion of invariant sites was used as a parameter to optimize estimation of substitution saturation.

259

Bayesian Inference and Maximum Likelihood Phylogenetic Analyses

Bayesian inference (BI) was performed using MrBayes 3.1.2 (Huelsenbeck &

Ronquist 2001), and consisted of two parallel runs of four Markov chains (three heated, one cold) run for 10 x 106 generations and sampled every 1,000 generations, with a random starting tree and the first 2 x 106 generations discarded as burnin. Maximum likelihood analysis was carried out in RAxML v7.2.8 (Stamatakis 2006), with 1000 bootstrap pseudoreplicates under the default GTR-GAMMA substitution model (the simplest model implemented in RAxML); the dataset was partitioned by gene for 16S and by codon position for cyt b and COI.

Comparative Morphology

Morphological measurements were taken with precision digital calipers and a stereo microscope with an optical micrometer, with measurements rounded to the nearest 0.1 mm. Abbreviations used for morphological measurements are as follows: snout to posterior edge of vent, SL; axilla–groin length, AG; trunk width at midbody, TW; head length from tip of snout to gular fold, HL; head width taken at maximum, HW; tail length, TL; hind limb length, HLL; forelimb length, FLL; combined forelimb and hind limb lengths, CLL; hind foot length, HFL; hind foot width, HFW; and nares length, NL. To allow for comparative studies across taxa, most measurements are standardized by SL; morphological comparisons of the aforementioned characters are presented in Table 5-

4. Comparative data for other taxa not examined by the authors is taken from Good &

Wake (1993), Campbell & Smith (1998), Lynch & Wake (1978), McCranie et al. (1998),

Ehmcke & Clemen (2000), Köhler (2002), McCranie & Wilson (2002), Savage (2002), and Wake & Campbell (2000).

260

Results

Best fit nucleotide substitution models varied by gene and codon position, supporting use of a gene and codon based partitioning strategy (Table 5-2). For the 489 bp of 16S, 10.8% of nucleotides were variable. Variability increased to 27.4% and

27.8% for COI (658 total bp) and cyt b (702 total bp) respectively. Distance-based analyses of each gene yielded unambiguous results in terms of delimiting species-level lineages and clusters, displaying a bimodal pattern of pairwise sequence divergence

(i.e., a ―barcoding gap‖) for each locus that can be represented graphically and easily interpreted using distance-scaled NJ trees (Figure 5-2). The smallest range of intraspecific sequence divergence in our data was for COI (0.002–0.003%), followed by

16S (0.0–0.6%) and cyt b (0.0–1.9%). Interspecific divergence ranged from 1.2–6.4% for 16S, the most conservative gene, to 6.1–13.9% for COI and 5.2–14.7% for cyt b

(Table 5-3). For all three genes, the bimodal distribution of sequence divergence allows for simple identification between the ranges of intra- and interspecific divergence

(Figure 5-2), with 16S having the narrowest gap (0.6%), followed by cyt b (3.3%) and

COI (5.8%).

Substitution saturation was not a factor for 16S, with the ISS value (0.2197) being highly significantly less than the critical ISS+C value for both the symmetrical

(ISS+C(SYM)=0.7112, P=0.0000) and asymmetrical (ISS+C(ASYM)=0.4960, P=0.0000) topologies. Likewise, COI did not exhibit evidence of substitution saturation, with an ISS value (0.3667) significantly less than the critical ISS+C value for both the symmetrical and asymmetrical topologies (ISS+C (SYM)=0.7320, P=0.0000; ISS+C (ASYM)=0.5482, P=0.0000).

For cyt b, however, ISS (0.7722) was significantly larger than ISS+C for both the symmetrical (ISS+C(SYM)=0.7337, P=0.0005) and asymmetrical (ISS+C(ASYM)=0.5254,

261

Figure 5-2. Comparison of ―barcoding gaps‖ for three mitochondrial genes used to delimit species boundaries in Nototriton. Unrooted neighbor-joining trees (10,000 bootstrap replicates) scaled by uncorrected p-distance, with shaded area representing gap between upper limit of intraspecific variation and lower limit of interspecific variation. Note that the trees for COI and cyt b are scaled to 0.01, while the 16S tree is scaled to 0.005 due to lower overall divergence. Numbers following taxon labels refer to those in Table 5-1.

262

263

P=0.0000) topologies, indicating that excessive substitution saturation was present and the data, therefore, were poorly suited for use phylogenetic study (Xia et al. 2003; Xia &

Lemey 2009). To further investigate, I tested for substitution saturation at each codon position on cyt b, and found the third position to have a higher but not significantly different ISS value (0.6981) on the perfectly symmetrical topology (ISS+C(SYM)=0.6797,

P=0.4553) and to be significantly higher than the asymmetrical topology

(ISS+C(ASYM)=0.4554, P=0.0000), while the 1st and 2nd codon positions did not show evidence of saturation (1st codon position: ISS=0.3721; ISS+C (SYM)=0.6797, P=0.0000;

ISS+C (ASYM)=0.5482, P=0.0509; 2nd codon position: ISS=0.3438; ISS+C (SYM)=0.6797,

P=0.0000; ISS+C (ASYM)=0.5482, P=0.0094).

Phylogenetic analyses were performed both on the complete combined mitochondrial dataset, as well the combined dataset with the 3rd codon position of cyt b excluded, in order to account for saturation (Figure 5-3). Both analyses recovered a strongly supported ―northern‖ clade (bs=100/100; pp=1.0/1.0) consisting of N. brodiei and two unnamed candidate species (N. sp. A and B) currently referred to as N. barbouri sensu lato. Also strongly supported by both analyses is the sister relationship between N. barbouri sensu stricto and N. limnospectator (bs=82/87; pp=1.0/1.0), a

―central‖ clade of the N. barbouri group.

The two datasets yielded different topologies in terms of the placement of N. tomamorum and the N. sp. inquirenda 2 with respect to the remaining members of the

N. barbouri group. The phylogeny excluding 3rd position recovers a topology congruent with recently published phylogenies of Nototriton (Townsend et al. 2010a) supplemented by additional sampling, including two samples from a newly discovered

264

Figure 5-3. Bayesian phylograms showing discordance between combined mtDNA phylogenies due to saturation at the third codon position for cytochrome b. Note the alternative placement of Nototriton tomamorum and N. picucha sp. nov. as sister to the remaining N. barbouri group. Bootstrap scores from maximum likelihood analyses shown above branches; posterior probability values from Bayesian inference shown below branches. population in eastern Honduras. The new species is recovered as sister to the remaining N. barbouri group (bs=82; pp=1.0), with N. tomamorum as sister to the entire clade (bs=100; pp=1.0), including the new species (Figure 5-3). The phylogeny including 3rd position recovers N. tomamorum as sister to N. lignicola (bs=72; pp=0.84), and, instead, places the new species as sister to a N. barbouri group (bs=100; pp=1.0).

265

Discussion

DNA Barcode Identification of Chortís Highland Moss Salamanders

Sequence divergence data from 16S, cyt b, and COI all exhibit a bimodal distribution of uncorrected pairwise comparisons, each demonstrating a ―barcoding gap‖ indicative of clear differentiation between intra- and interspecific variation (Figure 5-2).

Results for all three loci were congruent, identifying the same sample groupings and unambiguously separating putative species-level lineages and revealing three potential candidate species: a newly discovered population from the Sierra de Agalta, N. barbouri sensu lato from Parque Nacional Pico Bonito (N. sp. A), and N. barbouri sensu lato

Reserva de Vida Silvestre Texíguat (N. sp. B). Samples representing two recently discovered allopatric populations of Nototriton were also included in this study: one from the vicinity of Cataguana in Parque Nacional Montaña de Yoro and the other from

Parque Nacional Cerro Azul Meámbar on the eastern side of Lago de Yojoa. All three gene loci clearly indicated that these samples were conspecific with N. lignicola and N. limnospectator, respectively, and, therefore, confirm that these samples represent new populations of two species previously considered to be highly threatened single-site endemics (Townsend et al. 2011b).

In the case of Chortís Highland Nototriton, analysis of mtDNA sequence data using a barcoding approach is shown here to be an effective means of identifying potential candidate species and newly discovered populations of known taxa, but in and of itself is not grounds for proposing new taxa or making other taxonomic changes.

Rather, I view the barcoding approach as an effective and increasingly necessary first step in an integrative process that leads to correct taxonomic allocation of newly collected samples, particularly in difficult and cryptic groups like Nototriton.

266

Influence of Substitution Saturation in Cytochrome B Dataset

Our analyses of sequence data from cyt b revealed excessive substitution saturation at the third codon position, a phenomenon that has been shown to lead to bias in phylogenetic inference in studies of deep relationships among salamanders

(Zhang & Wake 2009), including plethodontids (Mueller et al. 2004; Macey 2005; Vieites et al. 2011). This study provides an example of excessive saturation biasing phylogenetic results at the interspecific level in salamanders, and its influence appears to be limited, at least in this case, to two species-level lineages. That the saturation problem has not been previously identified in Nototriton can be attributed to use of a longer fragment of cyt b (702 bp) than had been used in previous phylogenetic analyses of the genus (385 bp; García-París & Wake 2000; Townsend et al. 2010a). When our dataset is trimmed to 385 bp to match the length of most cyt b sequences available from

NCBI (http://www.ncbi.nlm.nih.gov/) for Nototriton, substitution saturation is not a factor at any codon position, with all ISS values being significantly less than the critical ISS+C value for both the symmetrical topologies (1st codon position: ISS=0.2396,

ISS+C(SYM)=0.6639, P=0.0000; 2nd codon position: ISS=0.0.989, ISS+C(SYM)=0.6630,

P=0.0000; 3rd codon position: ISS=0.3061, ISS+C(SYM)=0.6690, P=0.0000) and asymmetrical topologies (1st codon position: ISS = ISS+C(ASYM)=0.4113, P=0.0037; 2nd codon position: ISS+C(ASYM)=0.4101, P=0.0004; 3rd codon position: ISS+C(ASYM)=0.4575,

P=0.0000). Phylogenetic analysis using the 385 bp cyt b fragment (not shown) produced a topology completely congruent with that of the dataset that excluded the 3rd codon position of cyt b (468 bp), suggesting that the 385 bp fragment is short enough to mitigate bias from 3rd position saturation.

267

Phylogenetic Systematics and Candidate Species

As indicated here, paraphyly exists among populations currently referred to as

Nototriton barbouri (Schmidt 1936). Populations from Reserva de Vida Silvestre

Texíguat and Parque Nacional Pico Bonito in northern Honduras referred to N. barbouri

(McCranie & Wilson 1995; McCranie 1996a; McCranie & Wilson 2002; McCranie &

Castañeda 2007) are paraphyletic with respect to a sample from Montaña de Macuzal, the presumed type locality of N. barbouri (McCranie and Wilson 2002: 145). Based on our phylogenetic results (Figure 5-3), N. barbouri sensu stricto is the sister species to N. limnospectator, which together form a ―central‖ clade distributed in highland forests from

Montaña de Santa Bárbara eastward through the Sierra de Sulaco to Montaña de

Macuzal.

García-París & Wake (2000:49) suggested that samples of N. barbouri sensu lato from Parque Nacional Pico Bonito (USNM 339712, 497552) and Reserva de Vida

Silvestre Texíguat (USNM 509333) might represent distinct taxa, an assertion well supported by our results. Populations of N. barbouri sensu lato from Parque Nacional

Pico Bonito (N. sp. A) and Reserva de Vida Silvestre Texíguat (N. sp. B) form a well- supported clade with N. brodiei of eastern Guatemala and northwestern Honduras, together forming the ―northern‖ clade of the N. barbouri group (Figure 5-3). Somewhat surprisingly, given their geographic distributions, and contrary to the previous cyt b- based phylogenetic hypothesis (García-París & Wake 2000), both combined phylogenies strongly support N. sp. B, from north-central Honduras, as sister to a N. brodiei/N. sp. A clade (bs=100; pp=1.0). These results provide phylogenetic support for recognition of both N. sp. A and N. sp. B as species-level taxa; however one known

268

population of N. barbouri sensu lato remains to be evaluated phylogenetically (Pico Pijol in Departamento de Yoro).

Two samples from a newly discovered population of Nototriton in Parque Nacional

Sierra de Agalta in eastern Honduras are shown to be conspecific with each other and a relatively divergent lineage within the group (Figure 5-3). This undescribed species appears to be a typical member of the N. barbouri group in terms of external morphology, with relatively small nostrils and free, well-differentiated toes, which leads us to conclude that the phylogeny excluding 3rd codon position on cyt b, which includes the new species as part of an ingroup that is sister to N. tomamorum, most accurately represents the evolutionary history of these salamanders. The alternative phylogenetic hypothesis, obtained using the concatenated mtDNA dataset with cyt b fragment including the saturated 3rd codon position, implies that the distinctive morphology of N. tomamorum, such as enlarged nostrils and syndactylous feet, are derived characters that evolved as the result of the divergence of N. tomamorum and N. lignicola from a common ancestor. Future analyses that include additional samples and nuclear gene loci should provide some measure of resolution of the phylogenetic position of N. tomamorum.

Systematics

A Divergent New Lineage from Refugio de Vida Silvestre Texíguat

During my first visit to Refugio de Vida Silvestre Texíguat in April 2008, malacologist John Slapcinsky collected a small salamander in leaf litter packed into a crevice along the stream below the La Fortuna campsite. This single specimen possesses a series of morphological characteristics unique among Honduran salamanders. Comparison of morphological and mitochondrial DNA sequence data

269

confirms that this specimen represents a distinctive and undescribed species, and a relict lineage sister to the clade corresponding to the Nototriton barbouri group (Figure

5-4).

Nototriton tomamorum Townsend, Butler, Wilson, & Austin 2010a

Figure 5-4

Nototriton sp. inq. 1: Chapter 3.

Holotype. UF 155477, a female from 2.5 km north-northeast of La Fortuna, 1,550 m elevation, Refugio de Vida Silvestre Texíguat, Departamento de Yoro, Honduras.

GenBank accession numbers GU971731 (16S), GU971732 (cyt b), JN377407 (COI).

Diagnosis. A small member of the genus Nototriton (SVL=26.9 mm; Table 5-4) based on having 13 costal grooves (>16 costal grooves in Oedipina), hands and feet longer than broad (hands and feet broader than long in Bolitoglossa), and nares that are smaller than most Cryptotriton and Dendrotriton (Figure 5-4A, B; 0.018 NL/SVL; 0.020–

0.029 NL/SVL in Cryptotriton [except some individuals of C. veraepacis] and

Dendrotriton). Cryptotriton veraepacis has nares ranging from 0.017–0.027 NL/SVL

(mean 0.022), and can be differentiated from N. tomamorum by having a uniformly dark gray ventral surface (ventral surface pale with gray flecks in N. tomamorum; Figure 5-

4B). Generic placement in Nototriton is also strongly supported by sequence data from the mitochondrial genes 16S, COI, and cyt b (Table 5-3; Figures 5-2, 5-3). This new species is distinguished from all other Nototriton, except N. richardi and N. tapanti, by having syndactylous hands and feet (Figure 5-4D, E; hands and feet with free, differentiated toes in all other species) and relatively large nares (Figure 5-3C, B; 0.018

NL/SVL versus 0.010–0.016 in N. picadoi, 0.003–0.014 in N. abscondens, 0.012 in N. stuarti, 0.005–

270

Figure 5-4. Nototriton tomamorum. A) Dorsal and ventral views of the holotype of N. tomamorum (UF 155377), standard length 26.9 mm. B) Dorsal view of the head of the holotype of N. tomamorum. C) Lateral view of the head of the holotype of N. tomamorum. D) Dorsal view and E) ventral view of right hind foot of the holotype of N. tomamorum, showing the lack of separation or differentiation in the toes, and absence of subdigital pads. Photos © J.H. Townsend.

271

0.011 in N. barbouri, 0.006–0.009 in N. lignicola, 0.004–0.009 in N. guanacaste, 0.004–

0.005 in N. brodiei, 0.003 in N. limnospectator, 0.003 in N. major, and 0.002–0.003 in N. saslaya). Nototriton tomamorum can be further differentiated from members of the N. barbouri group by having a broader head (0.145 HW/SVL versus 0.138 in N. stuarti,

0.104–0.132 in N. barbouri, 0.120 in N. brodiei, 0.103–0.118 in N. lignicola, and 0.095–

0.118 in N. limnospectator) and fewer maxillary teeth (26, versus 36 in N. stuarti, 41–54 in N. barbouri, 42–55 in N. limnospectator, 46–54 in N. lignicola, and 60–62 in N. brodiei), from members of the N. picadoi group by having a relatively shorter tail (0.911

TL/SVL, versus 1.441 in N. major, 1.123–1.344 in N. picadoi, 1.013–1.365 in N. abscondens, 1.210–1.337 in N. guanacaste, and 1.10–1.30 in N. gamezi) and narrower feet (0.037 HFW/SVL, versus 0.058–0.071 in N. abscondens, 0.059 in N. major, 0.060–

0.070 in N. picadoi, and 0.066–0.072 in N. guanacaste), and from N. saslaya by having shorter forelimbs (0.160 FLL/SVL, versus 0.194–0.210 in N. saslaya) and hind limbs

(0.197 HLL/SVL, versus 0.217–0.244 in N. saslaya) and narrower feet (0.037

HFW/SVL, versus 0.075–0.091 in N. saslaya). The new species also differs from N. richardi and N. tapanti in having a pale ventral surface mottled with gray chromatophores (ventral surface brown with dark flecks in N. richardi and dark brown in

N. tapanti), by having a tail that is shorter than the snout-vent length (0.91 TL/SVL, versus 1.072–1.482 in N. richardi and 1.205 in N. tapanti), longer forelimbs (0.160

FLL/SVL, versus 0.140–0.146 in N. richardi and 0.147 in N. tapanti), longer hind limbs

(0.197 HLL/SVL, versus 0.174–0.187 in N. richardi and 0.174 in N. tapanti), and narrower feet (0.037 HFW/SVL, versus 0.044–0.050 in N. richardi and 0.041 in N. tapanti). This new species is also well differentiated from all other species of Nototriton

272

based on mitochondrial sequence data, and is 3.5–5.7% divergent on 16S, 11.5–13.9% divergent on COI, and 9.5–13.1% divergent on cyt b from other Chortís Highland

Nototriton (Table 5-3).

Description of holotype. Nototriton tomamorum is known only from a single, presumably female (mental gland and cloacal papillae absent) specimen, preserved with its mouth open and tongue extended, and is a relatively small member of the genus

(SVL=26.9 mm, total length=51.4 mm) with a slender body and reduced limbs. Its head is rounded and slightly broader than the body, and the nostrils are relatively large

(NL/SVL=0.018), and the snout is rounded and of moderate length (Figure 5-4B, C).

The nasolabial protuberances are apparent but not well developed, and barely extend below the upper lip line. The eyes are relatively large and protuberant, and the parotoid glands appear large but not well defined. The teeth are exceedingly small; there are approximately 26 maxillaries, 4 premaxillaries set slightly forward from the maxillary teeth, and 11 vomerines; vomerine teeth are arranged in two short medially-positioned arches. The limbs are short (CLL/SVL=0.36), with the adpressed limbs being separated by approximately 5.5 costal grooves. The hands and feet are narrow and have poorly- developed, poorly-differentiated digits that are fused and lack subdigital pads (Figure 5-

4D, E). The free tips of digits III on the hands and III and IV on the feet are pointed, and digits I, II and IV on the hands and I, II, and V on the feet are very short and essentially completely fused, being demarcated by shallow grooves on the dorsal side of the feet.

The relative length of the digits is I

The tail is slightly shorter than the body (TL/SVL=0.91), with a slight basal constriction

273

most apparent on the ventral side, and is slightly compressed laterally (tail depth 1.17 times tail width at level of basal constriction).

Measurements of holotype (in mm). SVL 26.9; AG 15.2; TW 3.6; HL 4.8; HW

3.9; TL 24.5; HLL 5.3; FLL 4.3; CLL 9.6; HFL 1.7; HFW 1.2; NL 0.5; eye length 1.6; eye width 1.2; interorbital distance 1.4; anterior rim of orbit to snout 1.1; distance separating internal nares 0.8; distance separating external nares 1.8; snout projection beyond mandible 0.6; tip of snout to axilla 7.7; distance from axilla to groin 15.2; snout to anterior edge of vent 24.9; tail depth at basal constriction 2.8; tail width at basal constriction 2.4.

Coloration of holotype. Dorsal surfaces of head, body, and tail medium grayish brown, with profuse pale chromatophores laterally, becoming less abundant dorsolaterally. The head has some pale mottling on the top of the snout, and two irregular lines of pale chromatophores extending from the lateral region above the forelimbs onto the posterior portion of the head and parotoid glands. There is a very thin, pale middorsal stripe, with a herringbone pattern with lines extending from the middorsal stripe posteriorly. There is an indistinct dark dorsolateral stripe starting about one-third the way down the trunk and extending onto the proximal one-third of the tail.

Ventral surface of head, body, and tail cream, mottled with dark gray chromatophores, becoming somewhat more profuse toward the distal end of the tail.

Etymology. The specific name ―tomamorum‖ means ―belonging to the Tomams.‖

Tomams are the highest level of deities recognized in the belief system of the indigenous Tolupan of Honduras, of which there are four: Tomam Pones Popawai

(Grandfather Tomam), his wife Tomam Pones Namawai (mother of all that exists), and

274

their children Tomam the Elder and Tomam the Younger (Chapman 1992). This name is given in recognition that the Tolupan are the traditional inhabitants of this area and that this new species is known only from the Cordillera Nombre de Dios, or ―Name of

God Mountains,‖ a name which, ironically, was given by 15th century Spanish explorers.

Distribution and Natural History. The single known specimen was collected during the daytime from leaf litter packed onto a rock ledge alongside a small creek at about 1,550 m elevation in the Lower Montane Wet Forest formation (Townsend et al.

2010a). This species is presumably endemic to the vicinity of the type locality, and is sympatric with a congener, N. sp. A, and three other plethodontids: Bolitoglossa dofleini,

Bolitoglossa cf. porrasorum, and Oedipina gephyra.

Conservation Status. Critically Endangered B1ab(iii)+2ab(iii).

A New Species from the Sierra de Agalta

Nototriton picucha Townsend, Medina-Flores, Murillo, and Austin 2011a

Figure 5-5

Nototriton sp. inq. 2: Chapter 3.

Holotype: USNM 578299 (Figures 5-5A, 5-5B), a female, from the northwestern slope of Cerro La Picucha (14.9733°N, 85.9279°W), 1,890 m, Parque Nacional Sierra de Agalta, Departamento de Olancho, Honduras, collected 17 July 2010 by M. Medina

F., J. L. Murillo, I. Zúniga, and J. M. Soto. Original field number JHT 3192; Genbank accession numbers: JN377388 (16S), JN377392 (cyt b), JN377404 (COI).

Paratype: USNM 578298 (Figure 5-5C), same collection data as holotype, except from 1,920 m elevation. Genbank accession numbers: JN377389 (16S), JN377393 (cyt b), JN377405 (COI).

275

Diagnosis. A member of the genus Nototriton (SL=27.9 mm) possessing 13 costal grooves (>16 costal grooves in Oedipina), hands and feet longer than broad (hands and feet broader than long in Bolitoglossa), and small nares (0.007 NL/SL; 0.017–0.029

NL/SL in Cryptotriton and Dendrotriton). Placement of the new species in the Nototriton barbouri species group is supported by phylogenetic analysis of the mitochondrial genes 16S and cyt b (Figure 5-2). In addition to molecular characteristics, the new species is distinguished from other members of the N. barbouri group by having a broader head (HW/SL≥0.140), and is further differentiated from N. barbouri, N. brodiei,

N. lignicola, and N. stuarti by having relatively longer front limbs (FLL/SL≥0.179, versus

0.142–0.174 in N. barbouri, 0.148–0.151 in N. brodiei, 0.137–0.160 in N. lignicola, and

0.172 in N. stuarti) and relatively longer hind limbs (HLL/SL≥0.204–0.218, versus

0.153–0.200 in N. barbouri, 0.166–0.180 in N. brodiei, 0.158–0.181 in N. lignicola, and

0.178 in N. stuarti), and from N. limnospectator by having narrower hind feet

(HFW/SL≤0.043, versus 0.048–0.061) and larger nares (NL/SL≥0.007, versus 0.003).

Nototriton tomamorum can be distinguished from N. picucha by having enlarged nares

(NL/SL=0.018, versus 0.007–0.008 in N. picucha), syndactylous feet lacking subdigital pads (toes well differentiated with subdigital pads in N. picucha), and fewer maxillary and vomerine teeth (MT=26 and VT=11, versus 41 and 16–19 and N. picucha).

Nototriton saslaya, a member of the N. picadoi group, is known from a single highland forest site in northern Nicaragua, and differs from the new species in having fewer maxillary teeth (17–23, versus 41 in N. picucha) and vomerines (3–11, versus 16–19 in

N. picucha), relatively longer front limbs (FLL/SL 0.194–0.210, versus 0.179–0.191 in N. picucha) and relatively wider hind feet (HFW/SL 0.075–0.091, versus 0.042–0.043 in N.

276

Figure 5-5. Nototriton picucha. A) Dorsolateral and ventrolateral views of the holotype of N. picucha (USNM 578299), standard length 27.9 mm (photo © J.H. Townsend). B) X-ray radiograph of the holotype of N. picucha (photo © J.H. Townsend). C) Paratype of N. picucha (USNM 578298) in life, standard length 25.7 mm (photo © Melissa Medina-Flores). picucha). Comparative morphological data for Chortís Highland Nototriton are summarized in Table 5-4.

Description of holotype. A relatively small female (SL=27.9 mm, total length=59.8 mm) with a slender body and reduced limbs. The head is rounded, slightly broader than the body; nostrils are relatively small (NL/SL=0.007), and the snout is acutely rounded and of moderate length. The nasolabial groove is apparent, but nasolabial protuberances are not well developed and barely visible. The eyes are relatively large and protuberant, and the parotoid glands are apparent, but relatively flat and not well defined. There are 41 maxillary teeth, 2 slightly enlarged premaxillary teeth

277

in line with the maxillaries, and 19 vomerine teeth. The limbs are short (CLL/SL=0.38), with the adpressed limbs being separated by approximately 3.5 costal grooves. The hands and feet are narrow with well-developed digits that bear subdigital pads. The relative length of the digits is I

Measurements of holotype (in mm). SL 27.9; AG 16.8; TW 3.9; HL 5.5; HW 3.4;

TL 31.9; HLL 5.7; FLL 5.0; CLL 10.7; FFL 1.4; FFW 0.8; HFL 1.7; HFW 1.2; NL 0.2; eye length 1.7; eye width 0.8; interorbital distance 1.2; anterior rim of orbit to snout 1.1; distance separating internal nares 1.1; distance separating external nares 1.5; snout projection beyond mandible 0.4; tip of snout to axilla 7.3; snout to anterior edge of vent

25.3; tail depth at basal constriction 2.6; tail width at basal constriction 2.4.

Coloration of holotype. Dorsal surfaces of head, body, and tail with medium to dark reddish brown ground color; abundant pale yellowish-brown blotching , present on dorsal surface of snout and eyes, between eyes, and on paratoid glands, becoming more profuse posteriorly until becoming solid and fully encircling the distal one-third of tail. Pale yellowish-brown herringbone pattern on dorsal surface of body, with pale yellowish brown dorsolateral stripes arising anterior to insertion of hind limbs and continuing onto tail, where pattern grades into solid yellowish brown towards tip of tail; lower edge of dorsolateral stripes bordered by solid dark brown coloration. Irregular pale chromatophores scattered on anterior dorsal surface of head and lateral surfaces of body. Dorsal and ventral coloration sharply demarcated laterally by an irregular line of

278

pale chromatophores separating dark reddish-brown dorsal and dorsolateral ground color from dark blackish gray ventral ground color; venter with some sparse whitish gray spotting primarily on the chin and around ventral midline; yellowish brown coloration extends down from the dorsolateral surfaces around the basal portion of the tail to surround the vent with yellowish brown mottling. Proximal one-half to two-thirds of forelimbs and two-thirds to three-quarters of hind limbs mottled yellowish brown dorsally and ventrally. Ventral surface of head, body, and tail cream, mottled with dark gray chromatophores, becoming somewhat more profuse toward the distal end of the tail.

See Figure 5-4b for color photographs of the dorsal and ventral surfaces of the holotype.

Variation in paratype. The female paratype (Figure 5-5c; USNM 578298,

SL=25.7, tail missing) agrees almost completely with the holotype in most characteristics and standardized ratios, with the only notable differences being more premaxillaries (6, same size as and in line with the maxillaries), fewer vomerines (16), and in having a limb interval of approximately 4.5. Coloration is also similar to that seen in the holotype.

Osteology. Based on examination of digital radiographs for both the holotype

(Figure 5-5b) and paratype, Nototriton picucha is a typical member of the genus (Wake

& Elias 1983) possessing a single cervical vertebra, 14 trunk vertebrae (the anterior 13 of which bear ribs), 2 caudosacral vertebrae; a complete skull roof formed through contact of the parietal bones; frontal processes of premaxilla fused at point of origin and separate immediately dorsoposterior to origin; septomaxillary bones absent; preorbital vomerine processes well-developed into elongate arches bearing a single row of

279

numerous teeth; columella absent; no tarsal spur evident; phalangeal formulae 1-2-3-2 and 1-2-3-3-2; penultimate phalanges reduced, exceeded in length by terminal phalanges on digits II, III, and IV of the forelimbs and digits II, III, IV, and V of hind limbs; terminal phalanges tapered at distal tip, or slightly expanded in digit III of forelimbs and digit III of the hindlimbs; mesopodial elements not mineralized.

Etymology. The specific epithet references the type locality and tallest mountain in the Sierra de Agalta, Cerro La Picucha, whose high elevation elfin forests make it one of the natural wonders of Honduras. The type series was collected on the northwestern slope of Cerro La Picucha, above the highest basecamp used to access the summit.

Natural history. The type specimens of Nototriton picucha were collected at 1,890 and 1,920 meters in the Lower Montane Wet Forest formation on the north-northwest slope of Cerro La Picucha (2,354 m) in the Sierra de Agalta. The holotype was collected near the highest base camp used for accessing the elfin forest atop Cerro La Picucha, and the paratype was found farther up the ridgeline trail to the peak before the transition to elfin forest. Both specimens were collected while active underneath leaf litter on the ground between 21:00 and 00:00 hours on a rainy night. Nototriton picucha is sympatric with another endemic plethodontid salamander, Bolitoglossa longissima, which was also found along the same trail.

Conservation Status. Critically Endangered B1ab(iii).

Remarks. Parque Nacional Sierra de Agalta contains the easternmost fragment of cloud forest (Lower Montane Wet Forest formation) in Nuclear Central America, being located in Dept. Olancho approximately 180 km to the northeast of Tegucigalpa,

Honduras. The park was established in 1987 to protect an area of approximately of

280

73,924 ha under the management of the National Institute of Conservation and Forest

Development, Protected Areas and Wildlife (ICF). There are 75 species of amphibians and reptiles recorded from Parque Nacional Sierra de Agalta (Castañeda 2006;

McCranie & Cruz 2010); however to date only three species (Bolitoglossa longissima,

Nototriton picucha, and Omoadiphas cannula) are considered to have geographic distributions completely restricted to this mountain range.

Restriction of the taxon Nototriton barbouri (Schmidt, 1936)

The precise type locality of Nototriton barbouri is a matter of some uncertainty.

The taxon was described by Schmidt (1936: 43) based on material collected by R.E.

Stadelman from ―Portillo Grande, Yoro, Honduras.‖ The type series was collected ―from bromeliads, and from altitudes between 5000 [=1,524 m] feet (the type), and 6000

[=1,828 m] feet.‖ While there are numerous communities in Yoro named ―El Portillo‖ and at least one ―Cerro Portillo Grande‖ (maximum elevation approximately 1,020 m), the only locality with that name and local access to elevations above 1,800 m is El Portillo

(15.095323°N, 87.343902°W), a small community sitting at around 1,300 m elevation on the northern side of Montaña Macuzal (maximum elevation approximately 1,945 m;

Chapter 3, Figures 3-7G, 3-7H), an isolated karstic mountain supporting a small fragment of broadleaf cloud forest on its summit. McCranie and Wilson (1995: 137) suggested that Bolitoglossa porrasorum collected from Portillo Grande by Stadelman were ―probably from Montaña Macuzal south of this town,‖ and I agree that Montaña

Macuzal indeed represents the only known suitable habitat for cloud forest salamanders in the vicinity of El Portillo, and, therefore, consider it to be the type locality of N. barbouri.

281

A single specimen (AMNH 54949) from a locality near Lago de Yojoa (―El Volcán,‖ north slopes of Cerro Azul Meámbar, 860 m elevation [McCranie & Wilson 2002:576]) is also assigned to this taxon. Given the confirmation that populations of Nototriton from nearby Cerro Azul Meámbar represent N. limnospectator (Townsend et al. 2011b;

Chapter 4), I consider this single specimen to be representative of N. limnospectator, and not a distjunct and lower elevation locality for N. barbouri sensu stricto.

Nototriton barbouri (Schmidt 1936)

Oedipus barbouri Schmidt 1936: 43.

Holotype. MCZ 21247, an adult male from Portillo Grande, 1,524 m elevation,

Departamento de Yoro, Honduras; collected 9 May 1934 by R.E. Stadelman.

Paratypes. All females from the vicinity of the holotype, 1,524–1,828 m elevation;

MCZ 21248–50, FMNH 21866–67.

Genetype. UF 156538, an adult female from Montaña Macuzal (15.079455°N,

87.352985° W), 1,760 m elevation, above El Panal to the west of Yorito, Departamento de Yoro, Honduras; collected 9 April 2008, by J.M. Butler, L. Ketzler, J. Slapcinsky,

N.M. Stewart, J.H. Townsend, and L.D. Wilson; original field number JHT 2420;

GenBank accession numbers GU971733 (16S), GU971734 (cyt b), JN377401 (COI).

Referred specimens. 15; ANSP 28981–85, 28200, Montaña Macuzal S of Pueblo

Viejo; USNM 339700–08, from the eastern slope of Pico Pijol (15.175°N, 87.558°W),

1,920 m elevation, Parque Nacional Pico Pijol, Departamento de Yoro, Honduras.

Distribution and natural history. I consider this taxon to be restricted to the

Sierra de Sulaco in southwestern Departamento de Yoro, where it is known to occur in the vicinity of the type locality in an unprotected patch of remant cloud forest on the top of Montaña Macuzal (maximum elevation approximately 1,945 m) at the eastern

282

Figure 5-6. Nototriton barbouri sensu stricto. A and B) Genetype of Nototriton barbouri sensu stricto (UF 156538), from Montaña Macuzal, Departamento de Yoro, Honduras. Photos © J.H. Townsend. terminus of the range and in the cloud forests of PN Pico Pijol (maximum elevation approximately 2,282 m) at the western end of the range. Nototriton barbouri is known to occur in the Lower Montane Wet Forest formation from 1,520–1,920 m elevation.

Conservation status. Endangered B1ab(iii).

283

Description of unassigned populations from the Cordillera Nombre de Dios

Restriction of the taxon Nototriton barbouri to populations inhabiting the highlands of southern Yoro (Montaña de Pijol and Montaña de Macuzal) leaves two allopatric populations from the Cordillera Nombre de Dios without assignment to an existing taxon. With no referable name present in the synonymy of N. barbouri for these populations, putatively endemic to highland forests at opposite ends of the Cordillera

Nombre de Dios in Parque Nacional Pico Bonito and Refugio de Vida Silvestre Texiguat

(Figure 5-1), I herein describe these two cryptic species as new taxa. Together, they form a clade with N. brodiei, a species residing in the Sierra de Omoa and Cordillera de

Merendón in northwestern Honduras and adjacent Guatemala (Figure 5-1).

Nototriton sp. A, sp. nov.

Nototriton barbouri: McCranie 1996a: 28.

Holotype. USNM 497552, an adult female from the south slope of Cerro Búfalo

(15.66°N, 86.79°W), 1,540 m elevation, Parque Nacional Pico Bonito, Departamento de

Atlántida, Honduras; collected 30 May 1996 by S. Gotte and J.R. McCranie; original field number LDW 10724; GenBank accession number AF199137 (cyt b).

Paratype. USNM 339712, an adult female from Quebrada de Oro (15.64°N,

86.80°W), 1,210 m elevation, Parque Nacional Pico Bonito, Departamento de Atlántida,

Honduras; collected 13 February 1995 by J.R. McCranie & J.C. Rindfleish; original field number LDW 10378; GenBank accession numbers AF199201 (16S), AF199136 (cyt b).

Measurements of holotype (in mm). SL 33.7; AG 19.5; TW 4.9; HL 6.0; HW 4.1;

TL 39.4; HLL 6.4; FLL 6.0; CLL 12.4; FFW 1.0; HFW 1.9; NL 0.3; eye length 1.7; eye width 1.2; interorbital distance 1.3; anterior rim of orbit to snout 1.5; distance separating internal nares 1.1; distance separating external nares 1.6.

284

Distribution and natural history. This species is known only from the vicinity of

Cerro Búfalo in Parque Nacional Pico Bonito, Honduras, 1,210–1,540 m elevation.

Conservation status. Critically Endangered B1ab(iii).

Nototriton sp. B, sp. nov.

Holotype. USNM 578300, an adult male from Cerro El Chino above La Liberación

(15.525394°N, 87.278672°W), 1,420 m elevation, Refugio de Vida Silvestre Texiguat,

Departamento de Atlántida, Honduras; collected 19 June 2010 by E. Aguilar, B.K.

Atkinson, C.A. Cerrato M., A. Contreras, A. Portillo, J.H. Townsend, and L.D. Wilson; original field number JHT 3159; GenBank accession numbers JN377387 (16S),

JN377391 (cyt b), JN377403 (COI).

Paratypes. USNM 339709–10, about 2.5 km (airline) NNE of La Fortuna, Refugio de Vida Silvestre Texiguat, Departamento de Yoro, 1,690 m elevation; USNM 339711, about 2.5 km (airline) NNE of La Fortuna, Refugio de Vida Silvestre Texiguat,

Departamento de Yoro, 1,800 m elevation; USNM 509333, about 2.5 km (airline) NNE of La Fortuna, Refugio de Vida Silvestre Texiguat, Departamento de Yoro, 1,600 m elevation.

Measurements of holotype (in mm). SL 31.9; AG 19.1; TW 3.8; HL 6.9; HW 4.2;

TL 41.0; HLL 7.2; FLL 6.4; CLL 13.6; FFW 1.1; HFW 1.5; NL 0.2; eye length 1.8; eye width 0.9; interorbital distance 1.6; anterior rim of orbit to snout 1.5; distance separating internal nares 1.4; distance separating external nares 1.8; tip of snout to axilla 9.6.

Distribution and natural history. This species is known only from highland forest within Refugio de Vida Silvestre Texíguat, 1,420–1,800 m. The holotype was collected as it crawled out of an arboreal bromeliad approximately 1.5 m above the ground on a rainy night.

285

Figure 5-7. Nototriton sp. B. A) Holotype of Nototriton sp. B (USNM 578300) in life B) X- ray radiograph of the holotype of Nototriton sp. B. C) Immediate vicinity of the type locality of Nototriton sp. B, Cerro El Chino, 1,420 m, above La Liberación in Refugio de Vida Silvestre Texíguat. Photos © J.H. Townsend.

Conservation status. Critically Endangered B1ab(iii)+2ab(iii).

Review of the Remaining Species of Nototriton from the Chortís Highlands

The additions of Nototriton picucha, N. tomamorum, N. sp. A, N. sp. B, and N. sp.

C (identified in Chapter 4 and being described elsewhere) brings the total number of species in the genus to 18, with 10 of these species being endemic to the Chortís

286

Highlands. This is a more than 27% increase in the number of species in this cryptozoic genus, and a nearly two-fold increase in the number of species known from the Chortís

Highlands.

Nototriton brodiei Campbell & Smith 1998

Holotype. UTA A-50000, an adult female from Cerro Pozo de Agua, 1,125 m elevation, Sierra de Caral, Municipio de Morales, Izabal, Guatemala.

Distribution and Natural History. Premontane forests of the Cordillera de

Merendón along the Guatemala/Honduras border, from 875–1,140 m elevation. This species is known from the vicinity of the type locality in the Sierra de Caral, Guatemala, and the northwestern slope of the Sierra de Omoa, Honduras (Campbell & Smith 1998;

Kolby et al. 2009).

Conservation Status. Critically Endangered B1ab(iii) (Acevedo et al. 2008); the discovery of this species in a relatively well-protected area of Honduras should necessitate re-evaluation of this taxon’s conservation status.

Remarks. This species was recently reported from the Sierra de Omoa in

Honduras (Kolby et al. 2009).

Referred Specimens. GUATEMALA: IZABAL: Sierra de Caral, UTA A-50001,

UTA A-51490. HONDURAS: CORTÉS: Parque Nacional Cusuco, MVZ 258034–36.

Nototriton lignicola McCranie & Wilson 1997

Nototriton ―barbouri‖: Espinal et al. 2000: 102.

Holotype. USNM 497539, an adult male from above the Monte Escondido, 1,780 m elevation, Cerro de Enmedio (15°06'N, 86°44'W), Parque Nacional La Muralla,

Olancho, Honduras.

287

Figure 5-8. Nototriton lignicola. A and B) N. lignicola (UF 156543), from Cataguana, 2,020 m, Parque Nacional Montaña de Yoro, Departamento de Francisco Morazán, Honduras. C) Cloud forest habitat at Cataguana. D) Juvenile N. lignicola (UF 156543) from Cataguana, 1,820 m, Parque Nacional Montaña de Yoro, Departamento de Francisco Morazán, Honduras. Photos © J.H. Townsend.

Distribution and Natural History. Lower Montane Wet Forest of northern

Departamento de Francisco Morazán and northwestern Departamento de Olancho,

Honduras, from 1,780–2,020 m elevation.

Conservation Status. Critically Endangered B1ab(iii).

Referred Specimens. HONDURAS: FRANCISCO MORAZÁN: Cataguana, UF

156542–44, OLANCHO: Cerro de Enmedio, USNM 497539–51, 509335 (cleared and stained).

288

Figure 5-9. A) Nototriton limnospectator (UF 156539), from above Río Varsovia, 1,640 m, Parque Nacional Cerro Azul Meámbar, Departamento de Comayagua, Honduras (Photo © J.H. Townsend). B) Juvenile N. limnospectator (IRL 035) from Cataguana, from above Los Pinos, 900 m, Parque Nacional Cerro Azul Meámbar, Departamento de Cortés, Honduras (Photo © I. Luque).

Nototriton limnospectator McCranie, Wilson, & Polisar 1998

Holotype. UF 98460, an adult female from southwest of San Luís de los Planes

(14°56'N, 88°08'W), 1910 m elevation, northwestern side of Montaña de Santa Bárbara,

Parque Nacional Montaña de Santa Bárbara, Santa Bárbara, Honduras.

Distribution and Natural History. This species is found in Premontane Wet

Forest, Lower Montane Wet Forest, and Montane Wet Forest on both sides of Lago de

Yojoa in central Honduras.

Conservation Status. Endangered B1ab(iii).

289

Referred Specimens. COMAYAGUA: UF 156539–41, IRL 035. CORTÉS: El

Volcán, AMNH 54949. SANTA BARBARA: El Ocotillo, MVZ 225866, USNM 509334

(CS); Montaña de Santa Bárbara, SW of San Luís de Los Planes, UF 98460–66.

Nototriton stuarti Wake & Campbell 2000

Holotype. UTA A-33686, an adult male from 11.6 road km west-southwest of

Puerto Santo Tomás, 744 m elevation, Montañas del Mico (15°38'N, 88°40'W), Izabal,

Guatemala.

Distribution and Natural History. This species is known only from a single specimen from premontane forest in the Montañas del Mico in eastern Guatemala.

Conservation Status. Data Deficient (IUCN 2011).

290

Table 5-1. Samples used in sequence divergence and phylogenetic analyses, with GenBank accession and museum voucher numbers; CR = Costa Rica, GT = Guatemala, HN = Honduras, NI = Nicaragua. GenBank Accession Numbers Taxon Locality GenBank voucher 16S cytb COI N. barbouri HN: Yoro: Macuzal UF156538 GU971733 GU971734 JN377401 N. brodiei (1) GT: Izabal: Sierra de Caral UTA A-51490 AF199202 AF199139 JN377402 N. brodiei (2) HN: Cortés: Cusuco MVZ258035 JN377384 – – N. lignicola (1) HN: Olancho: La Muralla USNM497540 AF199204 AF199141 – N. lignicola (2) HN: Olancho: La Muralla USNM497550 – AF199142 – N. lignicola (3) HN: Francisco Morazán: Cataguana UF156543 GU971735 GU971736 JN377408 N. limnospectator (1) HN: Santa Bárbara: El Ocotillo MVZ225866 – AF199143 – N. limnospectator (2) HN: Santa Bárbara: San Luís Planes MVZ263852 JN377383 – – N. limnospectator (3) HN: Comayagua: Azul Meámbar UF156539 GU971737 GU971738 JN377397 N. limnospectator (4) HN: Comayagua: Azul Meámbar UF156540 GU971739 GU971740 JN377398 N. limnospectator (5) HN: Comayagua: Azul Meámbar UF156541 JN377386 JN377396 JN377399 N. limnospectator (6) HN: Cortés: Azul Meámbar UF-IRL035 JN377385 JN377395 JN377400 N. picucha sp. nov. (1) HN: Olancho: Sierra de Agalta USNM578299 JN377388 JN377392 JN377404 N. picucha sp. nov. (2) HN: Olancho: Sierra de Agalta USNM578298 JN377389 JN377393 JN377405 N. sp. A (1) HN: Atlántida: Quebrada de Oro USNM339712 AF199201 AF199136 – N. sp. A (2) HN: Atlántida: Cerro Búfalo USNM497552 – AF199137 – N. sp. B (1) HN: Atlántida: Texíguat USNM578300 JN377387 JN377391 JN377403 N. sp. B (2) HN: Yoro: Texíguat USNM509333 – AF199138 – N. tomamorum HN: Yoro: Texíguat UF155377 GU971731 GU971732 JN377407 N. saslaya (1) NI: Atlantíco Norte: Saslaya MVZ230241 GU981761 – – N. saslaya (2) NI: Atlantíco Norte: Saslaya UF156352 JN377390 JN377394 JN377406

291

Table 5-2. Models of nucleotide substitution chosen for phylogenetic analyses of Chortís Highland taxa using Akaike Information Criterion values. Partition Model AIC lnL 16S GTR+G 2335.9224 -1126.9612 cyt b (1st) HKY+G 1380.1242 -651.0621 cyt b (2nd) HKY+G 1011.5928 -466.7964 cyt b (3rd) GTR+G 2527.2855 -1220.6427 COI (1st) GTR+G 2603.4728 -1270.7364 COI (2nd) K80+I 1010.9648 -481.4824 COI (3rd) HKY 641.4734 -294.7367

Table 5-3. Within and between species sequence divergence (uncorrected p-distance) for Chortís Highland moss salamanders. Intraspecific Interspecific Taxon 16S cyt b COI 16S cyt b COI N. barbouri – – – 0.012– 0.071– 0.080– 0.055 0.123 0.121 N. sp. A – 0.000 – 0.014– 0.052– – 0.059 0.142 N. sp. B – 0.014 – 0.018– 0.052– 0.061– 0.057 0.144 0.126 N. brodiei 0.000 – – 0.014– 0.057– 0.061– 0.061 0.123 0.133 N. lignicola 0.002 0.005 – 0.023– 0.082– 0.105– 0.064 0.120 0.139 N. limnospectator 0.000– 0.000– 0.002– 0.012– 0.071– 0.080– 0.006 0.019 0.003 0.059 0.147 0.126 N. picucha 0.000 0.008 0.002 0.012– 0.093– 0.085– 0.049 0.123 0.121 N. saslaya 0.006 – – 0.047– 0.104– 0.107– 0.064 0.147 0.136 N. tomamorum – – – 0.035– 0.095– 0.115– 0.057 0.131 0.139

292

Table 5-4. Morphological and morphometric comparison of species of Nototriton; see Material and Methods section for explanation of abbreviations; SVL is given in mm. Comparative morphological data for species other than N. picucha are from Campbell & Smith (1998), McCranie et al. (1998), Wake & Campbell (2000), Köhler (2002), McCranie & Wilson (2002), and Townsend et al. (2010). Taxon SL HW/SL TL/SL HLL/SL FLL/SL HFW/SL NL/SL MT VT N. barbouri 30.2– 0.104– 1.031– 0.153– 0.142– 0.037– 0.005– 41–54 12–23 39.9 0.132 1.398 0.200 0.174 0.060 0.011 N. brodiei 33.2– 0.120 1.420– 0.166– 0.148– 0.040– 0.004– 60–62 23–24 34.5 1.440 0.180 0.151 0.060 0.005 N. lignicola 28.3– 0.103– 0.840– 0.158– 0.137– 0.029– 0.006– 46–54 16–24 33.9 0.118 1.059 0.181 0.160 0.040 0.009 N. limnospectator 33.0– 0.095– 1.027– 0.164– 0.156– 0.048– 0.003 42–55 16–26 38.2 0.118 1.297 0.211 0.183 0.061 N. picucha 25.7– 0.140– 1.143 0.204– 0.179– 0.042– 0.007– 41 16–19 27.9 0.148 0.218 0.191 0.043 0.008 N. saslaya 28.1– 0.133– 0.883– 0.217– 0.194– 0.075– 0.002– 17–22 3–11 34.6 0.155 1.255 0.244 0.210 0.091 0.003 N. stuarti 32.6 0.138 1.264 0.178 0.172 0.049 0.012 36 20

N. tomamorum 26.9 0.145 0.911 0.197 0.160 0.037 0.018 26 11

293

CHAPTER 6 INTEGRATING RESEARCH, EDUCATION, AND OUTREACH IN SUPPORT OF CONSERVATION IN THE CHORTÍS HIGHLANDS

This dissertation represents the first step in developing an accurate picture of evolutionary diversification and phylogenetic distinctiveness for the Chortís Block. The region’s amphibian fauna, already known to have a high degree of endemism and to be facing a comparably high risk of extinction, is herein demonstrated to be greatly underestimated in terms of taxonomic diversity. Throughout my research in Honduras and Nicaragua I encountered a series of significant challenges to the continued documentation of Chortís Block biodiversity, and more importantly to the conservation of the ecosystems that support Chortís Block biodiversity.

First, outside of the herpetofaunal research that has taken place since 1967

(Townsend & Wilson 2010), there has been no concerted effort to inventory and document other major biotic components (e.g. mammals, birds, plants) using modern, molecular-based approaches (but see Matamoros et al. [2009] for details of efforts in freshwater ichthyology). The majority of endemic herpetofaunal species in the Chortís

Block are found in the highlands, owing to the extreme ecophysiographic heterogeneity in its 50+ isolated highland forests and interceding dry valleys. Exploration of these geographically limited, highly threatened mountain-top forests continues apace among the best known groups, as evidenced by the continued regular discovery of new amphibians (Chapter 4) and reptiles (Chapter 3). However, integrative taxonomy inventories are desperately needed for other taxa.

Second, opportunities for recruitment, education, and practical training of aspiring

Honduran systematists and conservation biologists, those representing the next generation of ―front-line warriors‖ in the effort to document and conserve their national

294

biodiversity resources, are increasingly rare. The countries of the Chortís Block are among the poorest and most underdeveloped in the Americas, and opportunities for higher education in systematics and related fields are limited to a few small, under- resourced programs in public universities.

Finally, regional awareness of the unique nature of the endemic Chortís Block biota is low. To date, little concerted outreach has been targeted at the largest group of stakeholders in regional conservation: the public. My experience has been that

Hondurans generally are interested in and supportive of conservation of their national environmental patrimony. However, lacking access to information about endemic biodiversity limits the ability of individuals and groups to lead, or even support, conservation efforts.

Below I outline my vision for promoting biodiversity conservation through education, training, and outreach opportunities generated through a strategic program to inventory the biodiversity of the Chortís Block. Given that it contains the majority of the endemism-rich Chortís Highlands and that geopolitical considerations warrant an initial focus at the national level, I am presenting this strategy as I would see it implemented in Honduras.

Taxonomic Inventories in Promotion of Education and Extension

As I became aware of these three principal obstacles to biodiversity conservation in the Chortís Block, I sought avenues to address these issues through my own research initiatives. Biodiversity inventory and systematics research address the first challenge, cataloging the biological diversity of the region; however activities associated with carrying out taxonomic inventory work also offer an excellent opportunity to provide education and training to multiple groups. My experience in Honduras has

295

demonstrated that a virtually untapped talent pool of university students, parataxonomists, park guards, and conservation practitioners are available for participation in all levels of inventory-related research.

Taxonomic inventory projects also present frequent opportunities for scientists to engage with the public across a variety of platforms. In my case, I have had frequent opportunities to engage the Honduran public through the media, promoting local and regional biodiversity awareness and endemic species conservation through newspaper articles and television interviews (Figure 6-1). I was able to provide both planned and impromptu presentations to municipal governmental leaders, community groups, schools, and residents in rural areas, often within the buffer zones of the highland protected areas where I was working (Figure 6-2). Activities such as these should be promoted as part of any truly ―integrative‖ taxonomic inventory project, and field-based systematists have the unique opportunity to simultaneously carry out research, engage the public, and contribute to the development of Honduran capacity in systematic biology.

As evidenced in this dissertation, Honduras is a country of remarkable herpetofaunal diversity, having the highest degree of amphibian endemism (i.e., percentage of the country’s amphibian species that are found only within its’ borders) of all Central American countries: 36.2%, compared to 27.1% for Guatemala and 26.6% for Costa Rica (Wilson & Johnson 2010). With the discovery of more than 30 potentially new species (Chapter 4), it appears that this high degree of endemism is actually greatly underestimated, and numerous undiscovered species undoubtedly await discovery across Honduras.

296

Figure 6-1. Print media coverage of the discovery of new endemic species during 2008 and 2009.

As I have emphasized throughout this dissertation, much of the diversity is attributed to the high number of isolated highland forests that are home to suites of endemic species. Across Honduras, more than a dozen of these cloud forest ―islands‖ remain virtually unexplored biologically, and likely conceal veritable treasure-troves of new species from a variety of taxonomic groups. As a consequence of the unique biogeography of the Chortís Block, Honduras remains a largely unrealized center for biodiversity in the Americas, overshadowed by more thoroughly studied Central

American countries.

297

Opportunities for Training and Education

Currently, the School of Biology at the National Autonomous University of

Honduras (Universidad Nacional Autónoma de Honduras, or UNAH) offers only a single degree, the Licenciatura en Biología, a degree more comprehensive than a traditional

Bachelor of Science degree but lacking the professional preparation and opportunities for research typical of a Master of Science program (in 2011–12 UNAH has plans to launch an Master of Science in Tropical Botany). Biology majors complete a capstone research design course (BI-025 Seminario de Investigación) as their final class, after which they must complete an 800-hour professional practicum which can take on a wide variety of forms. The primary focus of BI-025 is for students to develop a research proposal, and some highly-motivated students develop proposals to carry out under direction of a supervising scientist in fulfillment of their 800-hour requirement. Despite potential synergies between the requirements of BI-025 and the professional practicum,

UNAH faculty do not typically encourage students to attempt linking the two and it is exceedingly challenging for students to propose and find support for independent, or even faculty-directed, research projects.

As a result of my time spent in Honduras, I have had the opportunity to work closely with numerous UNAH biology students, relying on them in innumerable ways while promoting their professional development through participation in remote biodiversity surveys. In further support of a few dedicated students, I served in the supervisory role for their final-year professional practica, assisting in developing proposals for and carrying out rapid herpetological inventories that, among other things, led directly to the discovery of a new species of salamander (Townsend et al. 2011a). I see providing opportunities of this nature for UNAH students to gain practical

298

experience is critical to strengthening the national biological community, and the promotion of UNAH student participation also serves to enhance the Honduran public’s perception of biodiversity conservation and its practitioners. The excitement and pursuit of scientific discovery, centered on the collection, identification, and further study of new and poorly-known species, was and will continue to be the central theme to my efforts to give students the opportunity to be fully involved in leading-edge systematic research with real-world applications in the realm of Honduran biodiversity exploration and conservation.

Pilot Project: Parque Nacional Montaña de Yoro

One of my principal goals in taking an approach that combines biodiversity inventory work and integrative systematics is to refine our understanding of the patterns of diversification exhibited in the Chortís Highlands. My initiation of inventory work in

Parque Nacional Montaña de Yoro (Chapter 3) provided an opportunity to do so, and at the same time develop and practice the educational and outreach-related elements described above. Parque Nacional Montaña de Yoro was virtually unknown to biologists prior to my initial trip there in 2006. As a result of limited baseline biodiversity data, this large area (~50km2) of cloud forest was among the lowest priorities for national level conservation planning and at one point was recommended for a reduction in its protected area status (Vreugdenhil et al. 2002; COHECO 2003). During a single rapid inventory in 2006, we found new species of salamander (Townsend et al. 2009a), anole lizard (Townsend & Wilson 2009), and spikethumb treefrog (Plectrohyla cf. guatemalensis; Chapter 4), with all three qualify as Critically Endangered based on

IUCN Red List criteria due to severe threat to their limited distributions. In contrast, we also found two species of salamanders previously considered to be endemic to Parque

299

Figure 6-2. Examples of public outreach and dissemination of results from taxonomic inventories. A) Television interview about the endemic biodiversity in Honduras (Photo © James Austin). B) Presenting the results of work in La Liberación de Texíguat to the Alcalde (mayor) of the municipality that receives its water from the reserve (Photo © Ileana Luque). C) Community members attending a public presentation of results from La Liberación de Texíguat (Photo © J.H. Townsend). D) Ileana Luque and I presenting educational materials on endemic species to children in Jilamito Nuevo, in the buffer zone of Refugio de Vida Silvestre Texíguat (Photo © Yensi Flores).

300

301

Nacional La Muralla (Nototriton lignicola and Oedipina kasios) in Parque Nacional

Montaña de Yoro (Chapter 4; Townsend et al. 2011b).

Three expeditions were carried out to Parque Nacional Montaña de Yoro in 2006,

2007, and 2008, each in close coordination with national (Honduran National

Department of Protected Areas and Wildlife [ICF]), municipal (Municipality of Marale, collaborating with the alcalde (mayor) and the office of rural development), park administrative (logistics arranged in close collaboration the director of the park), and community management (were accompanied by the local guardabosque, or park ranger, and worked out of local homes) levels. The guardabosques were not simply viewed as local escorts into the park, rather a deliberate attempt was made to engage them in an active interchange of ecological knowledge, sharing scientific knowledge about biodiversity in exchange for locally-derived traditional knowledge of the cloud forest and its inhabitants.

In addition to the participation by guardabosques, expeditions into Montaña de

Yoro included (along with myself and Dr. Larry David Wilson), two graduate students and one undergraduate from North American universities, two biology students from the

National Autonomous University of Honduras (UNAH), and two Peace Corps volunteers. None of the participants had previous experience in remote field-based research in the tropics, but all were able to fully participate and gain experience in a variety of methods, ranging from expedition planning and wilderness survival to preparation of tissue samples and specimens.

Following the discoveries in Parque Nacional Montaña de Yoro, a priority became the public dissemination of information about this virtually unknown area. Scientific

302

results were published in peer-reviewed outlets as is standard practice in the field. In addition, I wrote (with Ileana Luque-Montes) a newspaper article that, with the assistance of the Zamorano Biodiversity Center, received front page coverage in the two largest Honduran newspapers (Figure 6-1). Additional undiscovered species almost certainly occur in this reserve, but intensive survey work is needed immediately to document Montaña de Yoro’s diversity in support of a plan for conserving its remaining forests.

Concluding Statement

In this dissertation, I have sought to usher in a new era of biodiversity research in the Chortís Block region of Central America. By utilizing an integrative framework built around biodiversity inventory and molecular systematics, I have presented a viable model for addressing critical unmet goals for biodiversity conservation, through the taxonomic and molecular inventory of an endemism hotspot, which simultaneously promotes training for the next generation of Honduran (and Honduras-focused) biologists and raising public awareness of endemic-rich areas and related conservation issues through extension and outreach. I am anticipating the opportunity to fully pursue implementation of this model with even greater intensity and at a broader scale in the coming years, with the sincere hope that this initiative will catalyze efforts to bring the

Chortís Block biodiversity hotspot to the forefront of regional and global conservation and to its recognition as one of the planet’s premier regions for localized evolutionary diversification.

303

APPENDIX TAXONOMIC REVIEW OF CAUDATA FROM THE CHORTÍS BLOCK

Below I present a review of the salamander taxa known to occur with the Chortís

Block as of July 2009, the point at which I began writing this dissertation. ―Available genetic data‖ is indicative of data openly available (i.e., via NCBI) as of July 2009, summarized in order to provide the context for newly-generated data.

Dwarf salamanders (genus Cryptotriton García-París & Wake 2000)

The genus Cryptotriton contains the six species formerly referred to as the

Nototriton nasalis group (sensu Papenfuss & Wake 1987), but recognized by García-

París & Wake (2000) to represent a cryptic clade distinct from Nototriton. An undescribed seventh species was identified from the Sierra de Las Minas in Guatemala by García-París & Wake (2000). Three putative species of Cryptotriton are reported to occur in the Chortís Highlands; however the status of C. wakei is disputed, with

McCranie & Wilson (2002: 139) making the case that this species is a junior synonym of

C. nasalis, given a lack of morphological distinction between the two and the proximity of their known distributions. However, genetic material referable to C. wakei is lacking, and until the status of this taxon can be assessed phylogenetically its taxonomic validity will remain uncertain.

Cryptotriton monzoni (Campbell & Smith 1998)

Nototriton monzoni Campbell & Smith 1998: 6.

Cryptotriton monzoni: García-París & Wake 2000: 58.

Type locality. ―Cerro Del Mono, near La Unión, Zacapa, Guatemala, 1570 m elevation (14°58’N, 89°17’W).‖

304

Known distribution. Known only from a single specimen from the type locality in easternmost Guatemala.

Available genetic data. None.

Cryptotriton nasalis (Dunn 1924)

Oedipus nasalis Dunn 1924: 97.

Chiropterotriton nasalis: Meyer & Wilson 1971: 7.

Nototriton nasalis: Wake & Elias 1983: 11.

Cryptotriton nasalis: García-París & Wake 2000: 58.

Type locality. ―Mountains west of San Pedro, Honduras, at 4500 [= 1,371 m] feet altitude.‖

Known distribution. Restricted to the Sierra de Omoa in northwestern Honduras, from 1,220–2,200 m elevation.

Available genetic data. Cyt b (1 sample; García-París & Wake 2000).

Cryptotriton wakei (Campbell & Smith 1998)

Nototriton wakei Campbell & Smith 1998: 5.

Cryptotriton wakei: García-París & Wake 2000: 58.

Type locality. ―West slope of Cerro Pozo de Agua, Sierra de Caral, Municipio de

Morales, Izabal, Guatemala, 1150 m elevation (15°22’N, 88°42’W).‖

Known distribution. Known only from a single specimen from the type locality in easternmost Guatemala, 1,150 m elevation.

Available genetic data. None.

305

Bromeliad salamanders (genus Dendrotriton Wake & Elias 1983)

The genus Dendrotriton was proposed to accommodate the species of the

Chiropterotriton bromeliacia group (sensu Lynch & Wake 1975) of Chiropterotriton beta

(Wake & Elias 1983). Of the six species of Dendrotriton, the single Chortís Highlands species stands as a biogeographic outlier, with the five remaining species restricted to the highlands of western Guatemala and adjacent Chiapas, México (Wake 1998).

Bromeliad-dwelling salamanders have been known from Cerro Santa Bárbara, and isolated karstic mountain in central Honduras, since first collected by XXX in XXX.

These salamanders were referred to as Chiropterotriton (and later Nototriton) nasalis

(Meyer & Wilson 1971; Wake & Elias 1983), a species otherwise known only from the

Sierra de Omoa in northwestern Honduras, until being recognized as a distinct species by McCranie & Wilson ―1996‖ [1997]. Those authors placed the new species in the genus Nototriton based on its previous assignment to N. nasalis, a member of the N. nasalis group (Papenfuss & Wake 1987; later referred to the genus Cryptotriton; García-

París & Wake 2000); however, Wake (1998) placed the species in the genus

Dendrotriton based on external and internal morphology.

Dendrotriton sanctibarbarus (McCranie & Wilson “1996” [1997])

Nototriton sanctibarbarus McCranie & Wilson ―1996‖ [1997]: 111.

Dendrotriton sanctibarbarus: Wake 1998: 88.

Type locality. ―Montaña de Santa Bárbara (14 55’N, 88 07’W), 2,100 m elevation,

Departamento de Santa Bárbara, Honduras.‖

306

Known distribution. Restricted to Montaña de Santa Bárbara, 1,800–2,744 m elevation.

Available genetic data. None.

Moss salamanders (genus Nototriton Wake & Elias 1983)

The genus Nototriton included 13 species of miniature salamanders that are disjunctly distributed across highland forests in Guatemala, Honduras, Nicaragua, and

Costa Rica (AmphibiaWeb, 2 September 2011). This genus was one of two named to accommodate the species of Chiropterotriton beta (Wake & Elias 1983). Nototriton sensu Wake & Elias 1983 itself was later shown to be paraphyletic, resulting in description of the genus Cryptotriton by García-París & Wake (2000). There are three species groups currently recognized within the genus Nototriton (Papenfuss & Wake

1987; Savage 2002): the N. barbouri group, the N. picadoi group, and the N. richardi group. These three groups exhibit two discrete patterns of geographic distribution. The

N. barbouri group (five species) is restricted to the Chortís Highlands of Honduras and

Guatemala, and the N. picadoi group (six species) and N. richardi group (two species) are found in the highlands of Costa Rica, with a single species in the N. picadoi group,

N. saslaya, endemic to a cloud forest in northern Nicaragua (Köhler 2002). This genus is dealt with in detail in Chapters 4 and 5.

Nototriton barbouri (Schmidt 1936)

Oedipus barbouri Schmidt 1936: 43.

?Pseudoeurycea barbouri: Taylor 1944: 209.

307

Chiropterotriton barbouri: Meyer 1969: 106.

Nototriton barbouri: Wake & Elias 1983: 11.

Type locality. ―Portillo Grande, Yoro, Honduras... from altitudes between 5000 [=

1524 m] feet (the type), and 6000 [= 1828 m] feet.‖

Known distribution. Populations assigned to this taxon are known from the vicinity of Montaña Macuzal and Pico Pijol in Yoro, from Texíguat along the southwestern border between Atlántida and Yoro, and from Pico Bonito in Atlántida; all localities

1,550–1,990 m elevation. A single specimen (AMNH 54949) from a locality near Lago de Yojoa at 860 m elevation is also assigned to this taxon.

Available genetic data. 16S (1 sample; García-París & Wake 2000), cyt b (3 samples; García-París & Wake 2000).

Nototriton brodiei Campbell & Smith 1998

Nototriton brodiei Campbell & Smith 1998: 3.

Type locality. ―West slope of Cerro Pozo de Agua, Sierra de Caral, Municipio de

Morales, Izabal, Guatemala, 1125 m elevation (15°22’N, 88°42’W).‖

Known distribution. This species is known only from the Sierra de Caral in easternmost Guatemala, 875–1,140 m elevation.

Available genetic data. 16S (1 sample; García-París & Wake 2000), cyt b (1 sample; García-París & Wake 2000).

Nototriton lignicola McCranie & Wilson 1997

Nototriton lignicola McCranie & Wilson 1997: 369.

308

Nototriton ―barbouri‖: Espinal et al. 2001: 105.

Type locality. ―Cerro de Enmedio (15°06'N, 86°44'W) along the trail above the

Monte Escondido campground, Parque Nacional La Muralla, 1780 m elev.,

Departamento de Olancho, Honduras.‖

Known distribution. Known only from the vicinity of the type locality in the northwestern corner of Olancho in north-central Honduras, 1,760–1,780 elevation.

Available genetic data. 16S (1 sample; García-París & Wake 2000), cyt b (2 samples; García-París & Wake 2000).

Nototriton limnospectator McCranie, Wilson, & Polisar 1998

Nototriton limnospectator McCranie et al. 1998: 455.

Type locality. ―The northwestern side of Montaña de Santa Bárbara, southwest of

San Luís de los Planes (14°56'N, 88°08'W), 1910 m elevation, Departamento de Santa

Bárbara, Honduras.‖

Known distribution. Apparently restricted to Montaña de Santa Bárbara, 1,640–

1,980 m elevation.

Available genetic data. Cyt b (1 sample; García-París & Wake 2000).

Nototriton saslaya Köhler 2002

Nototriton saslaya Köhler 2002: 205.

Type locality. ―S slope of Cerro Saslaya (13°46.0’N, 85°02.3’W), 1,371 m elevation, Región Autónoma Atlántico Norte, Nicaragua.‖

309

Known distribution. Known only from the vicinity of the type locality in north-central

Nicaragua, 1,280–1,370 m elevation.

Available genetic data. None.

Nototriton stuarti Wake & Campbell 2000

Nototriton stuarti Wake & Campbell 2000: 817.

Type locality. ―11.6 km (road) WSW Puerto Santo Tomás, Montañas del Mico,

Depto. Izabal, Guatemala, 88°40'W, 15°38'N, 744 m elev.‖

Known distribution. Known only from a single specimen from the type locality in easternmost Guatemala.

Available genetic data. None.

Worm salamanders (genus Oedipina Keferstein 1868)

The taxon Oedipina accommodates some of the mostly morphologically divergent

Neotropical salamanders, a monophyletic clade that is sister to Nototriton (García-París

& Wake 2000). These elongate, attenuate species are the only Neotropical salamanders to have lengthened their bodies as a result of an increased number of vertebrae, and along with having greatly reduced limbs and elongate tails (which can be more than two times the body length in many species) are well adapted for life inside root masses, leaf litter, rotten logs, and other fossorial microhabitats (Brame 1968).

Oedipina represents an excellent example of a non-adaptive radiation (highly conserved morphology across three distinct clades) that is adaptive at the clade level (the drastic change in body form in order to exploit fossorial microhabitats). Eleven species of

310

Oedipina are known from the Chortís Highlands, representing all three subgenera:

Oedipina, Oeditriton, and Oedopinola (García-París & Wake 2000, McCranie et al.

2008).

Subgenus Oedipina

This subgenus is represented by six putative species in the Chortís Highlands, two of which have distributions that extend outside the region, leaving the remaining four as endemics.

Oedipina cyclocauda Taylor 1952

Oedipina cyclocauda Taylor 1952: 764.

Type locality. ―Los Diamantes (1 mile south of Gúapiles), Costa Rica.‖

Known distribution. Caribbean lowlands of Costa Rica and southern Nicaragua,

60–500 m elevation.

Available genetic data. 16S (2 samples; García-París & Wake 2000), cyt b (2 samples; García-París & Wake 2000).

Oedipina ignea Stuart 1952

Oedipina ignea Stuart 1952: 1.

Type locality. ―Along the Río Las Brisas, just south of Yepocapa, Department of

Chimaltenango, Guatemala. Elevation, about 1450 meters.‖

Known distribution. Pacific versant of south-central Guatemala to southwestern

Honduras, 1,340–1,750 m elevation.

311

Available genetic data. As ―O. sp. B,‖ 16S (1 sample; García-París & Wake 2000), cyt b (1 sample; García-París & Wake 2000).

Oedipina leptopoda McCranie, Vieites, & Wake 2008

Oedipina cyclocauda (in part): Brame 1968: 30.

Oedipina leptopoda McCranie et al. 2008: 13.

Type locality. ―32 km (road) W of Yoro on road to Morazán, 15.267480 N,

87.434820 W, Dept. Yoro, Honduras.‖

Known distribution. Known from three localities in Departamento de Yoro,

Honduras, 700–1,300 m elevation.

Available genetic data. As ―O. sp. C,‖ cyt b (1 sample; García-París & Wake 2000).

Oedipina pseudouniformis Brame 1968

Oedipina uniformis (in part): Taylor 1952: 350.

Oedipina pseudouniformis Brame 1968: 25.

Type locality. ―Cienaga Colorado approximately three kilometers by road east of

Juan Viñas and 6.3 km by road west of Turrialba, Canton de Turrialba, Provincia de

Cartago, Costa Rica, elevation 1035 m (3400 ft).‖

Known distribution. The type series includes a series of eight specimens from

Hacienda La Cumplida, Matagalpa, Nicaragua, 731 m elevation, and is otherwise known from 19–1,253 m elevation in northern and central Costa Rica.

Available genetic data. 16S (1 sample; García-París & Wake 2000), cyt b (3 samples; García-París & Wake 2000).

312

Oedipina stuarti Brame 1968

Oedipina stuarti Brame 1968: 47.

Type locality. ―Amapala, Isla Tigre, in the Golfo de Fonseca, Departamento de

Valle, Honduras.‖

Known distribution. This enigmatic taxon is known from three specimens, two from

Isla El Tigre in the Pacific Ocean, and the third reportedly from Tegucigalpa, 975 m elevation.

Available genetic data. None.

Oedipina taylori Stuart 1952

Oedipina taylori Stuart 1952: 2.

Type locality. ―4 kilometers east of Hacienda La Trinidad (23 air-line kilometers southeast of Chiquimulilla, Department of Jutiapa, Guatemala. Elevation, about 100 meters.‖ [Where is the terminal parenthesis supposed to be placed?]

Known distribution. Pacific versant of southeastern Guatemala, El Salvador, and

Honduras, 140–1,140 m elevation. This species also is known from an isolated locality in the upper Río Motagua Valley in eastern Guatemala.

Available genetic data. None.

Subgenus Oeditriton

Oedipina kasios McCranie, Vieites, & Wake 2008

Oedipina cyclocauda: Espinal et al. 2001: 103.

Oedipina kasios McCranie et al. 2008: 11.

313

Type locality. ―Near Quebrada Pinol, 15°07’N, 86°44’W, Parque Nacional La

Muralla, 1,190 m a.s.l., Dept. Olancho, Honduras.‖

Known distribution. Known only from the vicinity of the type locality in the northwestern corner of Olancho in north-central Honduras, 950–1,780 m elevation.

Available genetic data. 16S (2 samples; McCranie et al. 2008), cyt b (2 samples;

McCranie et al. 2008).

Oedipina quadra McCranie, Vieites, & Wake 2008

Oedipina cyclocauda (in part): Brame 1968: 30.

Oedipina quadra McCranie et al. 2008: 6.

Type locality. ―Urus Tingni Kiamp, 14°55’N, 84°41’W, tributary of upper portion

of Río Warunta, 160 m above sea level (a.s.l.), Dept. Gracias A Dios, Honduras.‖

Known distribution. Considered relatively widespread in the lowlands and peripheral areas in north-central and eastern Honduras, 70–540 m elevation.

Available genetic data. 16S (1 sample; McCranie et al. 2008), cyt b (1 sample;

McCranie et al. 2008).

Subgenus Oedopinola

Oedipina elongata (Schmidt 1936)

Oedipus elongatus Schmidt 1936: 165.

Oedipina elongata: Taylor 1944: 226.

Type locality. ―Escobas, the site of the water supply for Puerto Barrios, Izabal,

Guatemala.‖

314

Known distribution. Caribbean versant from near the Isthmus of Tehuantepec,

México, to northwestern Honduras, 10–770 m elevation.

Available genetic data. 16S (1 sample; García-París & Wake 2000), cyt b (1 sample; García-París & Wake 2000).

Oedipina gephyra McCranie, Wilson, & Williams 1993

Oedipina gephyra McCranie et al. 1993: 385.

Type locality. ―2.5 airline km NNE La Fortuna (15°26’N, 87°18‖W), 1690 m elev.,

Cordillera Nombre de Dios, Departamento de Yoro, Honduras.‖

Known distribution. Known from the vicinity of the type locality around Cerro

Texíguat at the western end of the Cordillera Nombre de Dios and from Cerro Búfalo in the central portion of the same range in northern Honduras, 1,580–1,810 m elevation.

Available genetic data. 16S (2 samples; García-París & Wake 2000), cyt b (2 samples; García-París & Wake 2000).

Oedipina tomasi McCranie 2006

Oedipina sp.: Townsend et al. 2006: 31.

Oedipina tomasi McCranie 2006: 291.

Type locality. ―Near Sendero de Cantiles, Parque Nacional El Cusuco, 15°30’N,

88°14’W, 1800m elevation, Departamento de Cortés, Honduras.‖

Known distribution. Known only from the vicinity of the type locality in the Sierra de

Omoa in northwestern Honduras, 1,780–1,800 m elevation.

Available genetic data. None.

315

Mushroom-tongued salamanders (genus Bolitoglossa Duméril, Bibron, & Duméril

1854)

The most species-rich genus of salamanders at 117 described species and counting (AmphibiaWeb, 2 September 2011), Bolitoglossa also has the broadest distribution of any salamander genus, ranging from San Luis Potosí in central México south through Middle America and into South America to the Amazon Basin and

Bolivian highlands (Parra-Olea et al. 2004). There are 15 species of Bolitoglossa in the

Chortís Highlands, which are spread across four phylogenetically delimited subgenera

(after Parra-Olea et al. 2004): Bolitoglossa, Magnadigita, Nanotriton, and Pachymandra.

The subgenera Bolitoglossa and Nanotriton both are found generally in the lowlands and peripherally in the highlands, and do not contain any species heretofore considered endemic to the Chortís Highlands. Pachymandra is represented by a single, giant species in the Chortís Highlands, B. dofleini, which inhabits premontane elevations and peripheral areas. Magnadigita represents a diverse highland restricted radiation that is largely endemic to the Chortís Highlands and includes the morphologically-defined B. dunni species group (McCranie & Wilson 1993).

Subgenus Bolitoglossa Duméril, Bibron, & Duméril 1854

This diverse and widespread subgenus includes two species in the Chortís

Highlands, both of which are found in the lowlands formations peripheral to the highlands of the serranía, with one species, B. mexicana, widely occurring at premontane elevations.

316

Bolitoglossa mexicana Duméril, Bibron, & Duméril 1854

Bolitoglossa mexicana Duméril, Bibron, & Duméril 1854: 93.

Type locality. ―Dolorès peten.‖

Known distribution. Atlantic versant from southern Veracruz, México, to eastern

Honduras, near sea level to 1,400 m elevation.

Available genetic data. 16S (17 samples; García-París et al. 2000a), cyt b (6 samples; García-París et al. 2000a).

Bolitoglossa striatula (Noble 1918)

Oedipus striatulus Noble 1918: 344.

Bolitoglossa striatula: Taylor 1941: 147.

Type locality. ―Cukra, Eastern Nicaragua.‖

Known distribution. Atlantic versant from eastern Honduras to central Costa Rica, near sea level to 1,050 m elevation.

Available genetic data. 16S (1 sample; García-París et al. 2000a), cyt b (1 sample;

García-París et al. 2000a).

Subgenus Magnadigita Taylor 1944

This subgenus includes the constituent members of the B. dunni species group, the B. franklini species group, and the B. rostrata species group (Parra-Olea et al.

2004). Only the B. dunni group is found in the region, with 11 putative species endemic to the Chortís Highlands.

317

Bolitoglossa carri McCranie & Wilson 1993

Bolitoglossa carri McCranie & Wilson 1993: 9.

B. (Magnadigita) carri: Parra-Olea et al. 2004: 336.

Type locality. ―Cerro Cantagallo, near Lepaterique (14°06’N, 87°28’W), 1840 m elevation, Departamento de Francisco Morazán, Honduras.‖

Known distribution. Known only from the vicinity of the type locality in the mountains west of Tegucigalpa, Honduras, 1,840–2,070 m elevation.

Available genetic data. 16S (2 samples; Parra-Olea et al. 2004), cyt b (2 samples;

Parra-Olea et al. 2004).

Bolitoglossa celaque McCranie & Wilson 1993

Bolitoglossa dunni (in part): Hidalgo 1983: 6.

Bolitoglossa celaque McCranie & Wilson 1993: 11.

B. (Magnadigita) celaque: Parra-Olea et al. 2004: 336.

Type locality. ―Near the Rio Arcáqual, eastern side of Cerro Celaque, Cordillera de

Celaque (14°32’N‖, 88°40’W), 2480 m elevation, Departamento de Lempira, Honduras.‖

Known distribution. Highland forests across the departments of Lempira, Intibucá, and La Paz in southwestern Honduras, 1,900–2,620 m elevation.

Available genetic data. 16S (2 samples; Parra-Olea et al. 2004), cyt b (2 samples;

Parra-Olea et al. 2004).

Bolitoglossa conanti McCranie & Wilson 1993

Oedipus morio (in part): Dunn 1926: 387.

318

Oedipus dunni (in part). Schmidt 1933: 16.

Magnadigita dunni (in part): Taylor 1944: 218.

Bolitoglossa dunni (in part): Wake & Brame 1963: 386.

Bolitoglossa conanti McCranie & Wilson 1993: 4.

B. (Magnadigita) conanti: Parra-Olea et al. 2004: 336.

Type locality. ―Quebrada Grande (15°05’N, 88°55’W), 1370 m elevation,

Departamento de Copán, Honduras.‖

Known distribution. Several isolated localities in western Honduras along the

Guatemalan and Salvadoran borders, 1,370–2,000 m elevation.

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

Bolitoglossa decora McCranie & Wilson 1997

Bolitoglossa decora McCranie & Wilson 1997: 367.

B. (Magnadigita) decora: Parra-Olea et al. 2004: 336.

Type locality. ―Along the trail to Cerro de Enmedio along the trail above the Monte

Escondido campground (15°05'N, 86°44'W), Parque Nacional La Muralla, 1440 m elev.,

Departamento de Olancho, Honduras.‖

Known distribution. Known only from the vicinity of the type locality in the northwestern corner of Olancho in north-central Honduras, 1,430–1,550 m elevation.

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

319

Bolitoglossa diaphora McCranie & Wilson 1995

Bolitoglossa sp. 1: McCranie & Wilson 1994: 147.

Bolitoglossa diaphora McCranie & Wilson 1995: 448.

B. (Magnadigita) diaphora: Parra-Olea et al. 2004: 336.

Type locality. ―Above the visitors center of Parque Nacional El Cusuco, Cerro

Cusuco (15°31’N, 88°12‖W), 5.6 km WSW Buenos Aires, 1550 m elevation, Sierra de

Omoa, Departamento de Cortés, Honduras.‖

Known distribution. Restricted to the Sierra de Omoa in northwestern Honduras, from 1,470–2,200 m elevation.

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

Bolitoglossa dunni (Schmidt 1933)

Oedipus morio (in part): Dunn 1926: 387.

Oedipus dunni Schmidt 1933: 16 (note: the type series was later shown to contain more than one species by McCranie & Wilson 1993).

Magnadigita dunni (in part): Taylor 1944: 218.

Bolitoglossa dunni (in part): Wake & Brame 1963: 386.

B. (Magnadigita) dunni: Parra-Olea et al. 2004: 336.

Type locality. ―Mountains west of San Pedro, Honduras. Altitude 4500 feet [=

1,371 m].‖

Known distribution. Several isolated localities in western Honduras along the

Guatemalan border, 1,200–1,600 m elevation.

320

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

Bolitoglossa heiroreias Greenbaum 2004

Magnadigita engelhardti: Mertens, 1952: 20.

Bolitoglossa dunni (in part): Wake and Lynch, 1976: 13.

Bolitoglossa engelhardti (in part): Villa et al., 1988: 3.

Bolitoglossa conanti (in part): McCranie and Wilson, 1993: 8.

Bolitoglossa cf. conanti: Leenders and Watkins-Colwell, 2004: 5.

Bolitoglossa sp. 3: Parra-Olea et al., 2004: 336.

Bolitoglossa heiroreias Greenbaum 2004: 412.

Type locality. ―Camel Cigarettes Field Station at base of Cerro Montecristo, Depto.

Santa Ana, El Salvador, elevation 1880 m… coordinates 14°24’04‖N, 89°21’02‖W.‖

Known distribution. Known from the vicinity of Cerro Montecristo in El Salvador,

Guatemala, and Honduras, 1,840–2,300 m elevation.

Available genetic data. 16S (2 samples; Parra-Olea et al. 2004), cyt b (2 samples;

Parra-Olea et al. 2004).

Bolitoglossa longissima McCranie & Cruz 1996

Bolitoglossa longissima McCranie & Cruz 1996: 195.

B. (Magnadigita) longissima: Parra-Olea et al. 2004: 336.

321

Type locality. ―Along the trail to Pico La Picucha, Sierra de Agalta (14°58’N,

88°55‖W), ca. 10 airline km NNW Catacamas, 1900 m elevation, Departamento de

Olancho, Honduras.‖

Known distribution. Known only from the vicinity of the type locality in the Sierra de

Agalta in eastern Honduras, 1,840–2,240 m elevation.

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

Bolitoglossa oresbia McCranie, Espinal, & Wilson 2005

Bolitoglossa oresbia McCranie et al. 2005: 108.

Type locality. ―Cerro El Zarciadero, 14°43.662’N, 87°53.925’W, 1880 m elevation,

Departamento de Comayagua, Honduras.‖

Known distribution. Known only from the vicinity of the type locality, a 1 hectare patch of forest on top of an isolated peak in central Honduras, 1,880 m elevation.

Available genetic data. None.

Bolitoglossa porrasorum McCranie & Wilson 1995

Bolitoglossa dunni (in part): Meyer 1969: 95.

Bolitoglossa conanti (in part): McCranie & Wilson 1993: 8.

Bolitoglossa sp.: Holm & Cruz 1994: 20.

Bolitoglossa sp. 2: McCranie & Wilson 1994: 147.

Bolitoglossa porrasorum McCranie & Wilson 1995: 132.

B. (Magnadigita) porrasorum: Parra-Olea et al. 2004: 336.

322

Type locality. ―East slope of Pico Pijol (15°10’N, 87°33'W), Montaña de Pijol northwest of Tegucigalpita, 1860-1900 m elevation, Departamento de Yoro, Honduras.‖

Known distribution. Populations assigned to this taxon are known from the vicinity of Montaña Macuzal and Pico Pijol in Yoro, from Texíguat along the border between

Atlántida and Yoro, and from the vicinity of Pico Bonito in Atlántida; 980–1,920 m elevation

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

Bolitoglossa synoria McCranie & Köhler 1999

Bolitoglossa celaque (in part): McCranie & Wilson 1993: 13.

Bolitoglossa synoria McCranie & Köhler 1999: 226.

B. (Magnadigita) synoria: Parra-Olea et al. 2004: 336.

Type locality. ―Quebrada La Quebradona (14°23.96’N, 89°07.38’W), north slope of

Cerro El Pital, 2150 m elevation, Departamento de Ocotepeque, Honduras.‖

Known distribution. Known only from the vicinity of the type locality on Cerro El

Pital along the border between Honduras and El Salvador, 2,150–2,713 m elevation.

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

Subgenus Nanotriton Parra-Olea, García-París, & Wake 2004

This subgenus accommodates two small species, one of which is currently known to occur in the Chortís Highlands. The second species, B. occidentalis, has been

323

reported previously from a disjunct locality in Honduras, but a recent review of this specimen showed to represent a poorly preserved juvenile B. mexicana (T. Papenfuss, pers. comm.).

Bolitoglossa rufescens (Cope 1869) complex

Oedipus rufescens Cope 1869: 104.

Spelerpes (Oedipus) rufescens: Peters 1873: 617.

Geotriton rufescens: Smith 1877: 76.

Spelerpes rufescens: Boulenger 1882: 71.

Bolitoglossa rufescens: Taylor 1941: 145.

B. (Nanotriton) rufescens: Parra-Olea et al. 2004: 335.

Type locality. ―Orizava…Vera Cruz, México.‖

Known distribution. On the Caribbean versant from central México to central

Honduras, from near sea level to over 1,400 m elevation, although typically in the lowlands.

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

Subgenus Pachymandra Parra-Olea, García-París, & Wake 2004

This subgenus was erected to contain two very large species of terrestrial salamanders, one of which occurs in the Chortís Highlands.

Bolitoglossa dofleini (Werner 1903)

324

Spelerpes dofleini Werner 1903: 352.

Oedipus schmidti Dunn 1924: 96.

Bolitoglossa doffleini [lapsus]: Taylor 1944: 219.

B. (Pachymandra) dofleini: Parra-Olea et al. 2004: 337.

Type locality. ―Guatemala.‖

Known distribution. The Caribbean versant of eastern Guatemala, southern Belize, and northwestern and north-central Honduras, from 650–1,370 m elevation.

Available genetic data. 16S (1 sample; Parra-Olea et al. 2004), cyt b (1 sample;

Parra-Olea et al. 2004).

325

REFERENCES

Adalsteinnsson SA, Branch WR, Trape S, Vitt LJ, Hedges SB (2009) Molecular phylogeny, classification, and biogeography of snakes of the family Leptotyphlopidae (Reptilia, Squamata). Zootaxa 2244, 1–50.

Adams DC, Berns CM, Kozak KH, Wiens JJ (2009) Are rates of species diversification correlated with rates of morphological evolution? Proceedings of the Royal Society B 276, 2729–2738.

Agüdelo-C N (1987) Ecosistemas Terrestres de Honduras. Asociación Hondureña de Ecología, Tegucigalpa, Honduras.

AmphibiaWeb (2011) Information on amphibian biology and conservation. [web application]. Berkeley, California: AmphibiaWeb. Available: (Accessed: Oct 6, 2011: http://amphibiaweb.org/).

Archibold RC, Cave D (2011) Drug wars push deeper into Central America. New York Times, 3/31/2011 (accessed 12 September 2011: http://www.nytimes.com/2011/03/24/world/americas/24drugs.html?_r=1&pagewante d=print).

Berger L, Speare R, Daszak P, Green DE, Cunningham AA, Goggin CL, Slocombe R, Ragan MA, Hyatt AD, McDonald KR, Hines HB, Lips KR, Marantelli G, Parkes H (1998). Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America. Proceedings of the National Academy of Science, USA 95, 9031–9036.

Bickford D, Lohman DJ, Sohdi NS, Ng PKL, Meier R, Winker K, Ingram KK, Das I (2007) Cryptic species as a window on diversity and conservation. Trends in Ecology and Evolution 22, 148–155.

Bird P (2003) An updated digital model of plate boundaries. Geochemicals, Geophysics, and Geosystems 4, 1027 (doi:10.1029/2001GC000252).

Bocourt MF (1883) Études sur les Reptiles. In: Mission Scientifique au Mexique et dans l’Amérique Centrale. Recherches Zoologiques pour servir à l’Histoire de la Fauna de l’Amérique Centrale et du Mexique (1870–1909). Troisième Partie.—1re Section. (eds Duméril AHA, Bocourt MF, Mocquard F). Imprimerie Nationale, Paris.

Borisenko AV, Sones JE, Hebert PDN (2009) The front-end logistics of DNA barcoding: challenges and prospects. Molecular Ecology Resources 9 (supplement 1), 27–34.

Brame Jr. AH (1968) Systematics and evolution of the Mesoamerican salamander genus Oedipina. Journal of Herpetology 2, 1–64.

326

Brocchi P (1877) Sur quelques batraciens raniformes et bufoniformes de l’Amérique Centrale. Bulletin de la Société Philomatique de Paris (series 7), 1, 175–197.

Buhay JE (2009) ―COI-like‖ sequences are becoming problematic in molecular systematic and DNA barcoding studies. Journal of Crustacean Biology 29, 96–110.

Campbell JA (1998) Comments on the identities of certain Tantilla (Squamata: Colubridae) from Guatemala, with the descriptions of two new species. Scientific Papers, Natural History Museum, University of Kansas 7, 1–14.

Campbell JA (1999) Distribution patterns of amphibians in Middle America. In: Patterns of Distribution of Amphibians: a global perspective. (ed Duellman WE), pp. 111– 210. Johns Hopkins University Press, Baltimore, Maryland.

Campbell JA, Savage JM (2000) Taxonomic reconsideration of Middle American frogs of the Eleutherodactylus rugulosus group (Anura: Leptodactylidae): a reconnaissance of subtle nuances among frogs. Herpetological Monographs 14, 186–292.

Campbell JA, Smith EN. (1998) New species of Nototriton (Caudata: Plethodontidae) from eastern Guatemala. Scientific Papers of the Natural History Museum of the University of Kansas, 6, 1–8.

Campbell JA, Smith EN, Streicher J, Acevedo ME, Brodie Jr ED (2010) New salamanders (Caudata: Plethodontidae) from Guatemala, with miscellaneous notes on known species. Miscellaneous Publications of the Museum of Zoology, University of Michigan 200, i–v, 1–60.

Carr AF (1950) Outline for a classification of animal habitats in Honduras. Bulletin of the American Museum of Natural History 94, 567–594.

Castañeda FE (2006) Herpetofauna del Parque Nacional Sierra de Agalta, Honduras. International Resources Group, United States Agency for International Development, Washington, D.C.

Castoe TC, Sasa M, Parkinson CL (2005) Modeling nucleotide evolution at the mesoscale: The phylogeny of the Neotropical pitvipers of the Porthidium group (Viperidae: Crotalinae). Molecular Phylogenetics & Evolution 37, 881–898.

Castoe TC, et al (2009) Comparative phylogeography of pitvipers suggests a consensus of ancient Middle American highland biogeography. Journal of Biogeography 36, 88–103.

CIPF (Centro de Información y Patrimonio Forestal) (2009) Anuario Estadístico Forestal 2008. Instituto Nacional de Conservación y Desarrollo Forestal, Áreas Protegidas y Vida Silvestre, Comayagüela, Honduras.

327

Chapman A (1992) Masters of Animals. Oral Traditions of the Tolupan Indians, Honduras. Gordon and Breach, Philadelphia, Pennsylvania.

Clare EL, Lim BK, Engstrom MD, Eger JL, Herbert PDN (2007) DNA barcoding of Neotropical bats: species identification and discovery within Guyana. Molecular Ecology Notes 7, 184–190.

COHECO (2003) Plan de Manejo Parque Nacional Montaña de Yoro. Consultoría Hondureña en Ecodesarrollo (COHECO), AFE-COHDEFOR, DAPVS, Regíon Forestal Francisco Morazán, and Regíon Forestal Yoro.

Cope ED (1866) Fourth contribution to the herpetology of tropical America. Proceedings of the Academy Natural Sciences Philadelphia 18, 123–132.

Crawford AJ, Lips KR, Bermingham E (2010) Epidemic disease decimates amphibian abundance, species diversity, and evolutionary history in the highlands of central Panama. Proceedings of the National Academy of Sciences, USA 107, 13777– 13782.

Crespi EJ, Browne RA, Rissler LJ (2010) Taxonomic revisión of Desmognathus wrighti (Caudata: Plethodontidae). Herpetologica 66, 283–295.

Cruz G, Wilson LD, Casteñeda F (2004) Plectrohyla chrysopleura. In: IUCN Red List of Threatened Species. Version 2010.2 (www.iucnredlist.org). Downloaded on 20 July 2010.

Cushman SA (2006) Effects of habitat loss and fragmentation on amphibians: a review and prospectus. Biological Conservation 128, 231–240.

Daudin FM (1803) Histoire Naturelle, Générale et Particulière des Reptiles; Ouvrage faisant suite aux Ouvres de Leclerc de Buffon, et partie du Cours complet d’Histoire naturelle rédigé par C. S. Sonnini, membre de plusieurs Sociétés savantes. Tome Sixième. F. Dufart, Paris.

Daza JM, Smith EN, Páez VP, Parkinson CL (2009) Complex evolution in the Neotropics: the origin and diversification of the widespread genus Leptodeira (Serpentes: Colubridae). Molecular Phylogenetics and Evolution 53, 653–667.

DeMets C, Mattioli G, Jansma P, Rogers R, Tenorio C, Turner HL (2007) Present motion and deformation of the Caribbean plate: constraints from new GPS geodetic measurements from Honduras and Nicaragua. In: Geologic and Tectonic Development of the Caribbean Plate in Northern Central America (ed Mann P), pp. 21–36. Geological Society of America Special Paper 428, Boulder, Colorado, USA.

Dixon JR, Hendricks FS (1979) The wormsnakes (family Typhlopidae) of the Neotropics, exclusive of the Antilles. Zoologische Verhandelingen 173, 1–39.

328

Donnelly TW, Horne G, Finch R, López-Ramos E (1990) Northern Central America: the Maya and Chortís blocks. In: The Geology of North America, Volume 11: The Caribbean Region (eds Dengo G, Case JE), pp. 37–76. Geological Society of America, Boulder, Colorado, USA.

Downs FL (1967) Intrageneric relationships among colubrid snakes of the genus Geophis Wagler. Miscellaneous Publications of the Museum Zoology, University of Michigan 131, 1–193.

Dudgeon D, et al. (2006) Freshwater biodiversity: importance, threats, status and conservation challenges. Biological Review, 81, 163–182.

Duellman WE (1993) Amphibian Species of the World: Additions and Corrections. University of Kansas Museum of Natural History Special Publication 21.

Duellman WE (2001) The Hylid Frogs of Middle America. Society for the Study of Amphibians and Reptiles, Contributions in Herpetology 18.

Duellman WE, Campbell JA (1992) Hylid frogs of the genus Plectrohyla: systematics and phylogenetic relationships. Miscellaneous Publications of the Museum of Zoology, University of Michigan 181, 1–32.

Dunn ER (1924) New salamanders of the genus Oedipus with a synoptical key. Field Museum of Natural History Zoological Series 12, 95–100.

Dunn ER (1926) The Salamanders of the Family Plethodontidae. Smith College, Northhampton, Massachussetts.

Ehmcke J, Clemen G (2000) Teeth and their sex-dependent dimorphic shape in three species of Costa Rican plethodontid salamanders (Amphibia: Urodela). Annals of Anatomy 182, 403–414.

Faivovich J, et al. (2005) Systematic review of the frog family Hylidae, with special reference to Hylinae: phylogenetic analysis and taxonomic revision. Bulletin of the American Museum of Natural History 294, 1–240.

Farias IP, Ortí G, Sampaio I, Schneider H, Meyer A (2004) The cytochrome b gene as a phylogenetic marker: the limits of resolution for analyzing relationships among cichlid fishes. Journal of Molecular Evolution 53, 89–103.

Frost DR, et al. (2006) The amphibian tree of life. Bulletin of the American Museum of Natural History 297, 1–370.

García-París M, Wake DB (2000) Molecular phylogenetic analysis of relationships of the tropical salamander genera Oedipina and Nototriton, with descriptions of a new genus and three new species. Copeia 2000, 42–70.

329

García-París M, Parra-Olea G, Wake DB (2000) Phylogenetic relationships within the lowland tropical salamanders of the Bolitoglossa mexicana complex (Amphibia: Plethodontidae). In: The Biology of Plethodontid Salamanders (eds Bruce RC, Jaeger RG, Houck, LD), pp. 199–214. Kluwer Academic/Plenum Publ., New York.

García-París M, Parra-Olea G, Brame Jr. AH, Wake DB (2002) Systematic revision of the Bolitoglossa mexicana species group (Amphibia: Plethodontidae) with description of a new species from México. Revista Española de Herpetología 16, 43–71.

Good DA, Wake, DB (1993) Systematic studies of the Costa Rican moss salamanders, genus Nototriton, with descriptions of three new species. Herpetological Monographs 7, 131–159.

Goodwin GG (1942) Mammals of Honduras. Bulletin of the American Museum of Natural History 79, 107–195.

Gordon M (1992) Northern Central America (the Chortís block). In: Jurassic of the Circum-Pacific Region: World and Regional Geology v. 3 (ed Westermann G), p. 107-113. Cambridge University Press.

Gordon M, Muehlberger W (1994) Rotation of the Chortis block causes dextral slip on the Guayape fault. Tectonics 13, 858–872.

Gordon M, et al. (2010) Dating tectonic events on the Chortís Block. Geological Society of America Abstracts with Programs 42, 197.

Gose WA (1985) Paleomagnetic results from Honduras and their bearing on Caribbean tectonics. Tectonics 4, 565–585.

Gray MJ, Miller DL, Hoverman JT (2009) Ecology and pathology of amphibian ranaviruses. Diseases of Aquatic Organisms 87, 243–266.

Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52, 696-704.

Hajibabaei M, Singer GAC, Hebert PDN, Hickey DA (2007) DNA barcoding: how it complements taxonomy, molecular phylogenetics, and population genetics. Trends in Genetics 23, 167–172.

Halanych KM, Robinson TJ (1999) Multiple substitutions affect the phylogenetic utility of cytochrome b and 12S rDNA data: examining a rapid radiation in Leporid (Lagomorpha) evolution. Journal of Molecular Evolution 48, 369–379.

Hanken J, Wake DB, Savage JM (2005) A solution to the large black salamander problem (genus Bolitoglossa) in Costa Rica and Panamá. Copeia 2005, 227–245.

330

Hanken J, Wake DB (1982) Genetic differentiation among plethodontid salamanders (genus Bolitoglossa) in Central and South America: implications for the South American invasion. Herpetologica 38, 272–287.

Hasbún CR, Köhler G (2009) New species of Ctenosaura (Squamata, Iguanidae) from southeastern Honduras. Journal of Herpetology 43, 192–204.

Hay WW, DeConto R, Wold CN, Wilson KM, Voigt S, Schulz M, Wold-Rossby A, Dullo WC, Ronov AB, Balukhovsky AN, Soeding E (1999) Alternative global cretaceous paleogeography. In: The Evolution of Cretaceous Ocean/Climate Systems (eds Barrera E, Johnson C), pp. 1-47. Geological Society of America Special Paper 332, Boulder, Colorado, USA.

Hazlett DL (1980) A botanical description of Cerro Azul Meámbar, Honduras. Brenesia 18, 201–206.

Hebert PDN, Cywinska A, Ball SL, deWaard JR (2003) Biological identifications through DNA barcodes. Proceedings of the Royal Society of London, B 270, 313–321.

Hebert PDN, Gregory TR (2005). The promise of DNA barcoding for taxonomy. Systematic Biology 54, 852–859.

Heyer R (2002) Leptodactylus fragilis, the valid name for the Middle American and northern South American white-lipped frog (Amphibia: Leptodactylidae). Proceedings of the Biological Society of Washington 115, 321–322.

Highton R (1995) Speciation in eastern North American salamanders of the genus Plethodon. Annual Review of Ecology and Systematics 26, 579–600.

Hillis DM, de Sá R (1988) Phylogeny and taxonomy of the Rana palmipes group (Salientia: Ranidae). Herpetological Monographs 2, 1–26.

Hillis DM, Wilcox TP (2005) Phylogeny of the New World true frogs. Molecular Phylogenetics and Evolution 34, 299–314.

Holdridge LR (1967) Life Zone Ecology. Revised Edition. Tropical Science Center, San José, Costa Rica.

Holm PA, Cruz-D GA (1994) A new species of Rhadinaea (Colubridae) from a cloud forest in northern Honduras. Herpetologica 50, 15–23.

Honeycutt RL, Hillis DM, Bickman JW (2010) Biodiversity discovery and its importance to conservation. In: Molecular Approaches in Natural Resource Conservation and Management (eds DeWoody JA, et al.), pp. 1–35. Cambridge University Press, New York.

331

Huelsenbeck JP, Ronquist FR (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics (Oxford) 17, 754–755.

IUCN (2001) IUCN Red List Categories and Criteria: Version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland and Cambridge, United Kingdom. (http://www.iucnredlist.org/info/categories_criteria2001).

IUCN (2011) The IUCN Red List of Threatened Species. (http://www.iucnredlist.org/, accessed 1 September 2011).

James KH (2007) Structural geology: from local elements to regional synthesis. In: Central America: Geology, Resources, and Hazards (eds Bundschuh J, Alvarado GE), pp. 277–321. Taylor and Francis, Oxford, United Kingdom.

Janzen D, et al. (2009) Integration of DNA barcoding into an ongoing inventory of complex tropical diversity. Molecular Ecology Resources 9 (supplement 1), 1–26.

Jockusch EL, Yanev KP, Wake DB (2001) Molecular phylogenetic analysis of slender salamanders, genus Batrachoseps (Amphibia: Plethodontidae), from central coastal California with descriptions of four new species. Herpetological Monographs 15, 54– 99.

Johannessen CL (1963) Savannas of Interior Honduras. University of California Press, Berkeley. 160 p.

Johnson JD (1989) A biogeographic analysis of the herpetofauna of northwestern Nuclear Central America. Milwaukee Public Museum Contributions in Biology and Geology 76, 1–66.

Joppa LN, Roberts DL, Pimm SL (2011) The population ecology and social behavior of taxonomists. Trends in Ecology and Evolution 26, 551–553.

Jordan BR, Sigurdsson H, Carey SN (2008) Ignimbrites in Central America and Associated Caribbean Sea Tephra: Correlation and Petrogenesis, VDM Verlag Dr. Müller.

Keppie JD, Morán-Zenteno DJ (2005) Tectonic implications of alternative Cenozoic reconstructions for southern Mexico and the Chortís Block. International Geology Review 47, 473–491.

Kerr KCR, Stoeckle MY, Dove CJ, Weigt LA, Francis CM, Hebert PDN (2007) Comprehensive DNA barcode coverage of Nother American birds. Molecular Ecology Notes 7, 535–543.

332

Ketzler LP, Luque-Montes IR, Cerrato-M CA, Wilson LD, Townsend JH (2011) Natural History Notes. Thamnophis fulvus (Central America Highland Garter Snake). Diet. Herpetological Review 42, 103.

Köhler G (2002) A new species of salamander of the genus Nototriton from Nicaragua (Amphibia: Caudata: Plethodontidae). Herpetologica 58, 205–210.

Köhler G (2008) Reptiles of Central America. 2nd Edition. Herpeton, Offenbach, Germany.

Köhler G (2010) A revision of the Central American species related to Anolis pentaprion with the resurrection of A. beckeri and the description of a new species. Zootaxa 2354, 1–18.

Köhler G (2011) Amphibians of Central America. Herpeton, Offenbach, Germany.

Köhler G, Veselý M (2010) A revision of the Anolis sericeus complex with the resurrection of A. wellbornae and the description of a new species (Squamata: Polychrotidae). Herpetologica 66, 186–207.

Köhler G, McCranie JR, Wilson LD (1999) Two new species of anoles of the Norops crassulus group from Honduras (Reptilia: Sauria: Polychrotidae). Amphibia-Reptilia 20, 279–298.

Köhler G, Quintana AZ, Buitrago F, Diether H (2004) New and noteworthy records of amphibians and reptiles from Nicaragua. Salamandra 40, 15–24.

Köhler J, Vieites DR, Bonett RM, Hita García F, Glaw F, Steinke D, Vences M (2005) New amphibians and global conservation: boost in species discoveries in a highly endangered vertebrate group. BioScience 55, 693–696.

Kolby JE, McCranie JR, Rovito SM (2009) Geographic Distribution: Nototriton brodiei (NCN). Herpetological Review 40, 444.

Kozak KH, Wiens JJ (2006) Does niche conservatism promote speciation? A case study in North American salamanders. Evolution 60, 2604–2621.

Kozak KH, Wiens JJ (2010) Niche conservatism drives elevational diversity patterns in Appalachian salamanders. The American Naturalist 176, 40–54.

Kozak KH, Weisrock DW, Larson A (2006) Rapid lineage accumulation in a non- adaptive radiation: phylogenetic analysis of diversification rates in eastern North American woodland salamanders (Plethodontidae: Plethodon). Proceedings of the Royal Society B, 273, 539–546.

333

Lipscomb D, Platnich N, Wheeler Q (2003) The intellectual content of taxonomy: a comment on DNA taxonomy .Trends in Ecology and Evolution 18, 6 –66.

Luque-Montes IR, Townsend JH (2009) Geographic Distribution. Ninia espinali (Espinal’s Coffee Snake). Herpetological Review 40, 457.

Luque-Montes IR, Ketzler LP, Cerrato-M CA, Wilson LD, Townsend JH (2011) Geographic Distribution. Hypopachus barberi (Montane Sheep Frog). Herpetological Review 42, 384-385.

Lynch JF, Wake DB (1975) Systematics of the Chiropterotriton bromeliacia group (Amphibia: Caudata), with description of two new species from Guatemala. Contributions in Science, Natural History Museum of Los Angeles County 265, 1–45.

Lynch JF, Wake DB (1978) A new species of Chiropterotriton (Amphibia: Caudata) from Baja Verapaz, Guatemala, with comments on relationships among Central American members of the genus. Contributions in Science, Natural History Museum of Los Angeles County 294, 1–22.

Macey JR (2005) Plethodontid salamander mitochondrial genomics: a parsimony evaluation of character conflict and implications for historical biogeography. Cladistics 21, 194–202.

Manton WI (1996) The Grenville of Honduras. Geological Society of America Abstracts with Programs 28, 493.

Mann P, Rogers RD, Gahagan L (2007) Overview of plate tectonic history and its unresolved tectonic problems. In: Central America: Geology, Resources, and Hazards (eds Bundschuh J, Alvarado GE), pp. 206–241. Taylor and Francis, Oxford, United Kingdom.

Marshall JS (2007) The geomorphology and physiographic provinces of Central America. In: Central America: Geology, Resources, and Hazards (eds Bundschuh J, Alvarado GE), pp. 75–122. Taylor and Francis, Oxford, United Kingdom.

Martin, M. (1972) A biogeographic analysis of the freshwater fishes of Honduras. Unpublished Ph.D. Dissertation. University of Southern California.

Matamoros WA, Schaefer JF, Kreiser BR (2009) Annotated checklist of the freshwater fishes of continental and insular Honduras. Zootaxa 2307, 1–38.

Matamoros WA, Schaefer JF (2010) A new species of Profundulus (Cyprinodontiformes: Profundulidae) from the Honduran central highlands. Journal of Fish Biology 76, 1498–1507.

334

McCranie JR (1996a) Geographic Distibution. Nototriton barbouri (NCN). Herpetological Review 27, 28.

McCranie JR (1996b) Geographic Distibution. Oedipina gephyra (NCN). Herpetological Review 27, 29.

McCranie JR (2006) A new species of Oedipina (Amphibia: Urodela) from Parque Nacional El Cusuco, northwestern Honduras. Journal of Herpetology 40, 291–293.

McCranie JR (2009) Amphibians and Reptiles of Honduras. Listas Zoológicas Actualizadas UCR (http://museo.biologia.ucr.ac.cr/Listas/LZAPublicaciones.htm). Museo de Zoología UCR. San Pedro, Costa Rica.

McCranie JR (2011a) The Snakes of Honduras. Systematics, Distribution, and Conservation. Society for the Study of Amphibians and Reptiles, Contributions in Herpetology.

McCranie JR (2011b) A new species of Tantilla of the taeniata species group (Reptilia, Squamata, Colubridae, Colubrinae) from northeastern Honduras. Zootaxa 3037, 37– 44.

McCranie JR, Castañeda FE (2004a) A new species of snake of the genus Omoadiphas (Reptilia: Squamata: Colubridae) from the Cordillera Nombre de Dios in northern Honduras. Proceedings of the Biological Society of Washington 117, 311–316.

McCranie JR, Castañeda FE (2004b): Notes on the second specimens of Geophis damiani Wilson, McCranie, and Williams and Rhadinaea tolpanorum Holm & Cruz-D. (Colubridae). Herpetological Review 35, 341.

McCranie JR, Castañeda FE (2005) The herpetofauna of Parque Nacional Pico Bonito, Honduras. Phyllomedusa 4, 3–16.

McCranie JR, Castañeda FE (2007) Guía de Campo de los Anfibios de Honduras. Bibliomania!, Salt Lake City, Utah.

McCranie JR, Cruz-Diaz GA (2010) A third new species of snake of the genus Omoadiphas (Reptilia, Squamata, Colubridae, Dipsadinae) from Honduras. Zootaxa 2690, 53–58.

McCranie JR, Townsend JH (2011) Description of a new species of worm salamander (Caudata, Plethodontidae, Oedipina) in the subgenus Oedopinola from the central portion of the Cordillera Nombre de Dios, Honduras. Zootaxa 2990, 59–68.

McCranie JR, Wilson LD (1993) A review of the Bolitoglossa dunni group (Amphibia: Caudata) from Honduras with the description of three new species. Herpetologica 49, 1–15.

335

McCranie JR, Wilson LD (1995a) Two new species of colubrid species of the genus Ninia from Central America. Journal of Herpetology 29, 224–232.

McCranie JR, Wilson LD (1995b) A new species of salamander of the Bolitoglossa dunni group (Caudata: Plethodontidae) from northern Honduras. Herpetologica 51, 131–140.

McCranie JR, Wilson LD (1996 [1997]) Two new species of salamanders (Caudata: Plethodontidae) of the genera Bolitoglossa and Nototriton from Parque Nacional La Muralla, Honduras. Proceedings of the Biological Society of Washington 110, 366– 372.

McCranie JR, Wilson LD (1997) Two new species of salamanders (Caudata: Plethodontidae) of the genera Bolitoglossa and Nototriton from Parque Nacional La Muralla, Honduras. Proceedings of the Biological Society of Washington 110, 366– 372.

McCranie JR, Wilson LD (2001) Taxonomic status of Typhlops stadelmani Schmidt (Serpentes: Typhlopidae). Copeia 2001, 820–822.

McCranie JR, Wilson LD (2002) The Amphibians of Honduras. Society for the Study of Amphibians and Reptiles, Contributions in Herpetology.

McCranie JR, Wilson LD, Williams KL (1992) A new species of anole of the Norops crassulus group (Sauria: Polychridae) from northwestern Honduras. Caribbean Journal of Science 28, 208–215.

McCranie JR, Wilson LD, Williams KL (1993) A new species of Oedipina (Amphibia: Caudata: Plethodontidae) from northern Honduras. Proceedings of the Biological Socity of Washington 106, 385–389.

McCranie JR, Wilson LD, Polisar J (1998) Another new montane salamander (Amphibia: Caudata: Plethodontidae) from Parque Nacional Santa Barbara, Honduras. Herpetologica 54, 455–461.

McCranie JR, Wilson LD, Gotte SW (2001) Three new country records for Honduran snakes. Herpetological Review 32, 62–63.

McCranie JR, Espinal MR, Wilson LD (2005) New species of montane salamander of the Bolitoglossa dunni group from northern Comayagua, Honduras (Urodela: Plethodontidae). Journal of Herpetology 39, 108–112.

McCranie JR, Townsend JH, Wilson LD (2006) The Amphibians and Reptiles of the Honduran Mosquitia. Krieger Publishing Company, Malabar, Florida.

336

McCranie JR, Vieites DR, Wake DB (2008) Description of a new divergent lineage and three new species of Honduran salamanders of the genus Oedipina (Caudata, Plethodontidae). Zootaxa 1930, 1–17.

McCranie JR, Valdés-Orellana L, Himes JG (2010) Rediscovery of two Honduran endemic streamside frogs, Craugastor emleni (Dunn) and Craugastor stadelmani (Schmidt). Froglog 94, 12–15.

McDiarmid, R. W., J. A. Campbell, and T. A. Touré. 1999. Snake Species of the World. A Taxonomic and Geographic Reference. Vol. 1. Herpetologist’s League, Washington.

Mejía-Ordóñez & House (2002) Mapa de Ecosistemas Vegetales de Honduras. Preparado para el Proyecto P.A.A.R., Tegucigalpa M.D.C.

Mejía-Valdivieso DA (2001) Honduras. In: Bosques Nublados del Neotrópico (eds Kappelle M, Brown AD), pp. 243–282. Instituto Nacional de Biodiversidad, Santo Domingo de Heredia, Costa Rica.

Mendelson III J (2011) Shifted baselines, forensic taxonomy, and Rabbs’ fringe-limbed treefrog: the changing role of biologists in an era of amphibian declines and extinctions. Herpetological Review 42, 21–25.

Mendelson III J, Mulcahy DG (2010) A new species of toad (Bufonidae: Incilius) from Central Panama. Zootaxa 2396, 61-68.

Mendelson III J, Williams BL, Sheil CA, Mulcahy DG (2005) Systematics of the Bufo coccifer complex (Anura: Bufonidae) of Mesoamerica. Scientific Papers, Natural History Museum of the University of Kansas 38, 1–27.

Mertens R (1952) Die amphibien und reptilien von El Salvador. Abhandlungen der Senckenbergischen Naturforschenden Gesellschaft 487, 1–120.

Meyer C (2003) Molecular systematics of cowries (Gastropoda: Cypraeidae) and diversification patterns in the tropics. Biological Journal of the Linnean Society 79, 401-459.

Meyer, JR (1969) A Biogeographic Study of the Amphibians and Reptiles of Honduras. Unpublished PhD dissertation, University of Southern California.

Meyer JR, Wilson LD (1971) A distributional checklist of the amphibians of Honduras. Los Angeles County Museum Contributions in Science 218, 1–47.

Meyer JR, Wilson LD (1973) A distributional checklist of the turtles, crocodilians, and lizards of Honduras. Natural History Museum of Los Angeles County Contributions in Science 244, 1–39.

337

Milanovich JR, Peterman WE, Nibbelink NP, Maerz JC (2010) Projected loss of a salamander diversity hotspot as a consequence of projected global climate change. PLoS One 5, e12189.

Monaghan MT, et al. (2009) Accelerated species inventory on Madagascar using Coalescent-based models of species delineation. Systematic Biology 58, 298–311.

Monroe Jr BL (1968) A distributional survey of the birds of Honduras. American Ornithological Union, Ornithological Monograph 7, 1–458.

Morán-Zenteno DJ, et al. (2009) Reassessment of the Paleogene position of the Chortís Block relative to southern Mexico: hierarchical ranking of data and features. Revista Mexicana de Ciencias Geológicas 26, 177–188.

Moritz C, Cisero C (2004) DNA barcoding: promises and pitfalls. PLoS Biology 2: 1529– 1531.

Moritz C, Schneider CJ, Wake DB (1992) Evolutionary relationships within the Ensatina eschscholtzii complex confirm the ring species interpretation. Systematic Biology 41, 273–291.

Mueller RL, Macey JR, Jaekel M, Wake DB, Boore JL (2004) Morphological homoplasy, life history evolution, and historical biogeography of plethodontid salamanders inferred from complete mitochondrial genomes. Proceedings of the National Academy of Science, USA 101, 13820–13825.

Mueller RL (2006) Evolutionary rates, divergence dates, and the performance of mitochondrial genes in Bayesian phylogenetic analysis. Systematic Biology 55, 289– 300.

Mueller-Dombois D, Ellenberg H, (1974) Aims and Methods in Vegetation Ecology. J. Wiley & Sons, New York, USA.

Mulcahy DG, Mendelson III JR (2000) Phylogeography and speciation of the morphologically variable, widespread species Bufo valliceps, based on molecular evidence from mtDNA. Molecular Phylogenetics and Evolution 17, 173–189.

Mulcahy DG, Morrill BH, Mendelson III JR. (2006) Historical biogeography of lowland species of toads (Bufo) across the Trans-Mexican Neovolcanic Belt and the Isthmus of Tehuantepec. Journal of Biogeography 33, 1889–1904.

Mullis K. (1990) The unusual origin of the Polymerase Chain Reaction. Scientific American 262, 56.

338

Myers CW (2003) Rare snakes — Five new species from eastern Panama: Reviews of northern Atractus and southern Geophis (Colubridae: Dipsadinae). American Museum Novitates 3391, 1–47.

Myers CW (2011) A new genus and new tribe for Enicognathus melanauchen Jan, 1863, a neglected South American snake (Colubridae: Xenodontinae), with taxonomic notes on some Dipsadinae. American Museum Novitates 3715, 1–33.

Nelson-S CH (2001) Plantas descritas originalmente de Honduras y sus nomenclaturas equivalentes actuales. Ceiba 42, 1–71.

Nelson-S CH (2008) Catálogo de las Plantas Vasculares de Honduras. Espermatofitas. Secretária de Recursos Naturales y Ambiente/Guaymuras, Tegucigalpa, Honduras.

Ortega-Gutiérrez F, et al. (2007) The Maya-Chortís boundary: a tectonostratigraphic approach. International Geological Review 49, 996–1024.

Palumbi SR, Martin AP, Romano S, McMillan WO, Stice L, Grabowski, G (1991) The Simple Fool’s Guide to PCR. Special Publication of the Department of Zoology, University of Hawaii, Honolulu.

Papenfuss TJ, Wake DB (1987) Two new species of plethodontid salamanders (genus Nototriton) from México. Acta Zoologica Mexicana, 21, 1–16.

Parra-Olea G, Wake DB (2001) Extreme morphological and ecological homoplasy in tropical salamanders. Proceedings of the National Academy of Sciences USA, 98, 7888–7891.

Parra-Olea G, García-París M, Wake DB (2004) Molecular diversification of salamanders of the tropical American genus Bolitoglossa (Caudata: Plethodontidae) and its evolutionary and biogeographical implications. Biological Journal of the Linnean Society 81, 325–346.

Parra-Olea G, et al. (2007) Systematics and Conservation. In: Amphibian Conservation Action Plan (eds Gascon C, et al.), pp. 45–48. IUCN/SSC Amphibian Specialist Group, Grand Switzerland.

Parsons JJ (1955) The Miskito pine savanna of Nicaragua and Honduras. Annals of the Association of American Geographers 45, 36–63.

Pfenninger M, Schwenk K (2007) Cryptic animal species are homogeneously distributed among taxa and biogeographical regions. BMC Evolutionary Biology 7, 121.

Posada D (2008) jModelTest: phylogenetic model averaging. Molecular Biology and Evolution 25, 1253–1256.

339

Pyron RA, Wiens JJ (2011) A large-scale phylogeny of Amphibia with over 2,800 species, and a revised classification of extant frogs, salamanders, and caecilians. Molecular Phylogenetics and Evolution 61, 543–583.

Reid F (2009) Field Guide to the Mammals of Central America and Southeast México. Second Edition. Oxford University Press.

Relyea RA, Diecks N (2008) An unforeseen chain of events: Lethal effects of pesticides at sublethal concentrations. Ecological Applications 18, 1728-1742.

Rogers RD (2003) Jurassic-Recent tectonic and stratigraphic history of the Chortís block of Honduras and Nicaragua (northern Central America). Unpublished Ph.D. dissertation, University of Texas at Austin.

Rogers R, Karason H, van der Hilst, R (2002) Epeirogenic uplift above a detached slab in northern Central America. Geology 30, 1031–1034.

Rogers R, Mann P, Emmet PA (2007) Tectonic terranes of the Chortis Block based on integration of regional aeromagnetic and geological data. In: Geologic and Tectonic Development of the Caribbean Plate in Northern Central America (ed Mann P), pp. 65–88. Geological Society of America Special Paper 428, Boulder, Colorado, USA.

Rovito SM, Parra-Olea G, Vásquez-Almazán CR, Papenfuss TJ, Wake DB (2009) Dramatic declines in neotropical salamander populations are an important part of the global amphibian crisis. Proceedings of the National Academy of Sciences USA, 106, 3231–3236.

Rovito SM, Vásquez-Almazán CR, Papenfuss TJ (2010) A new species of Bolitoglossa (Caudata: Plethodontidae) from the Sierra de las Minas, Guatemala. Journal of Herpetology 44, 516–525.

Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating inhibitors. Proceedings of the National Academy of Sciences of the USA 74, 5463– 5467.

Sasa M, Bolaños F (2004) Biodiversity and conservation of Mesoamerican dry-forest herpetofauna. In: Biodiversity Conservation in Costa Rica: Learning the Lessons in a Seasonal Dry Forest (eds Frankie GW, Mata A, Vinson SB), pp. 177–193. University of California Press.

Savage JM (1966) The origins and history of the Central American herpetofauna: dispersal or vicariance? Copeia 1966, 719–766.

Savage JM (1983) The enigma of the Central American herpetofauna: dispersals or vicariance? Annals of the Missouri Botanical Garden 69, 464-547.

340

Savage JM (2002) The Amphibians and Reptiles of Costa Rica: A Herpetofauna Between Two Continents, Between Two Seas. University of Chicago Press. Chicago, Illinois.

Schmidt KP (1933) New reptiles and amphibians from Honduras. Zoological Series of Field Museum of Natural History 20, 1 –22.

Schmidt KP (1936) New amphibians and reptiles from Honduras in the Museum of Comparative Zoology. Proceedings of the Biological Society of Washington 49, 43– 50.

Schuster, SC (2008) Next-generation sequencing transforms today’s biology. Nature Methods 5, 16–18.

Seberg O, et al. (2003) Shortcuts in systematics? A commentary on DNA-based taxonomy. Trends in Ecology and Evolution 18, 63–65.

Seutin G, White BN, Boag PT (1991) Preservation of avian blood and tissue samples for DNA analyses. Canadian Journal of Zoology 69, 82-90.

Silva-Romo G (2008) Guayape-Papalutla fault system: A continuous Cretaceous structure from southern México to the Chortís block? Tectonic implications. Geology 36, 75–78.

Skerratt LF, Berger L, Speare R, Cashins S, McDonald KR, Phillott AD, Hines HB, Kenyon N (2007) Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs. EcoHealth 4, 125–134.

Smith BE, Campbell JA (1996) The systematic status of Guatemalan populations of snakes allied with Ninia maculata (Reptilia: Colubridae). Proceedings of the Biological Society of Washington 109, 749–754.

Smith MA, Poyarkov Jr NA, Herbert PDN (2008) CO1 DNA barcoding amphibians: take the chance, meet the challenge. Molecular Ecology Resources 8, 235–246.

Song H, Buhay JE, Whiting MF, Crandall KA (2008) Many species in one: DNA barcoding overestimates the number of species when nuclear mitochondrial pseudogenes are coamplified. Proceedings of the National Academy of Sciences of the USA 105, 13486–13491.

Stafford PJ (2004) A new species of Tantilla (Serpentes: Colubridae) of the taeniata group from Southern Belize. Journal of Herpetology 38, 43–52.

Stamatakis A (2006) RAxML-VI-HPC: Maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688–2690.

341

Stehli FG, Webb SD (1985) The Great American Biotic Interchange. Plenum Press, New York.

Stuart LC (1954) A description of a subhumid corridor across northern Central America, with comments on its herpetofaunal indicators. Contributions of the Laboratory of Vertebrate Zoology, University of Michigan 65, 1–26.

Stuart LC (1963) A checklist of the herpetofauna of Guatemala. Miscellaneous Publications of the Museum of Zoology, University of Michigan 122, 1-150.

Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues ASL, Fischman DL, Waller RW (2004) Status and trends of amphibian declines and extinctions worldwide. Science 306, 1783–1786.

Sunyer J, Köhler G (2010) Conservation status of the herpetofauna of Nicaragua. In: Conservation of Mesoamerican Amphibians and Reptiles (eds Wilson LD, Townsend JH, Johnson, JD), pp. 488–509. Eagle Mountain Publishing LC, Eagle Mountain, Utah.

Sunyer J, Townsend JH, Wilson LD, Travers SL, Obando L, Paiz G, and Griffith DM. (2009) Three new country records of reptiles from Nicaragua. Salamandra 45, 186– 190.

Sunyer J, Wake DB, Townsend JH, Travers SL, Rovito SM, Papenfuss TJ, Obando LA (2010) A new species of worm salamander (Caudata: Plethodontidae: Oedipina) in the subgenus Oeditriton from the highlands of northern Nicaragua. Zootaxa 2613, 29–39.

Sunyer J Townsend JH, Wake DB, Travers SL, Gonzalez SG, Obando LA, Quintana AZ (2011) A cryptic new species of Oedipina (Caudata: Plethodontidae) from premontane elevations in northern Nicaragua, with comments on the systematic status of the Nicaraguan paratypes of O. pseudouniformis Brame, 1968. Breviora 526, 1–16.

Systematics Agenda 2000 (1994). Systematics Agenda 2000: Charting the Biosphere. New York: Department of Ornithology, American Museum of Natural History.

Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution 28, 2731–2739.

Tautz D, et al. (2002) DNA points the way ahead in taxonomy. Nature 418, 479.

Tautz D, Arctander P, Minelli A, Thomas RH, Vogler AP (2003) A plea for DNA taxonomy. Trends in Ecology and Evolution 18, 70–74.

342

Taylor EH (1944) The genera of plethodont salamanders in México, Part I. University of Kansas Science Bulletin 30, 189–232.

Thomas R, Hedges SB (2007) Eleven new species of snakes of the genus Typhlops (Serpentes: Typhlopidae) from Hispaniola and Cuba. Zootaxa 1400, 1–26.

Thompson JD, Higgins DG, Gibson TJ (1994) CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position- specific gap penalties and weight matrix choice. Nucleic Acids Research 22, 4673- 4680.

Tilley SG, Mahoney MJ (1996) Patterns of genetic differentiation in salamanders of the Desmognathus ochrophaeus complex (Amphibia: Plethodontidae). Herpetological Monographs 10, 1–42.

Townsend JH (2006) Inventory and conservation assessment of the herpetofauna of the Sierra de Omoa, Honduras, with a review of the Geophis (Squamata: Colubridae) of eastern Nuclear Central America. Unpublished Master’s Thesis, University of Florida, Gainesville, Florida, USA. i–xiv, 1–124.

Townsend JH (2009) Morphological variation in Geophis nephodrymus (Squamata: Colubridae), with comments on the conservation of Geophis in eastern Nuclear Central America. Herpetologica 65, 292–302.

Townsend JH, Wilson LD (2008) Guide to the Amphibians and Reptiles of Cusuco National Park, Honduras / Guía de los Anfibios y Reptiles de Parque Nacional Cusuco, Honduras. Bibliomania!, Salt Lake City, Utah.

Townsend JH, Wilson LD (2006) Denizens of the dwarf forest: herpetofauna of the elfin forests of Cusuco National Park, Honduras. Iguana 13, 242–251.

Townsend JH, Wilson LD (2009) New species of cloud forest Anolis (Squamata: Polychrotidae) of the crassulus group from Parque Nacional Montaña de Yoro, Honduras. Copeia 2009, 62–70.

Townsend JH, Wilson LD (2010a) Conservation of the Honduran herpetofauna: issues and imperative. In: Conservation of Mesoamerican Amphibians and Reptiles (eds Wilson LD, Townsend JH, Johnson, JD), pp. 460–487. Eagle Mountain Publishing LC, Eagle Mountain, Utah.

Townsend JH, Wilson LD (2010b) Biogeography and conservation of the Honduran subhumid forest herpetofauna. In: Conservation of Mesoamerican Amphibians and Reptiles (eds Wilson LD, Townsend JH, Johnson, JD), pp. 686–705. Eagle Mountain Publishing LC, Eagle Mountain, Utah.

343

Townsend JH, Wilson LD, Plenderleith TL, Talley BL, Nifong JC (2005) Ninia pavimentata (Squamata: Colubridae): an addition to the snake fauna of Honduras. Caribbean Journal of Science 41, 869–870.

Townsend JH, Wilson LD, Ketzler LP, Luque-Montes IR (2008a) The largest blindsnake in Mesoamerica: a new species of Typhlops (Squamata: Typhlopidae) from an isolated karstic mountain in Honduras. Zootaxa 1932, 18–26.

Townsend JH, Wilson LD, Luque-Montes IR, Ketzler LP (2008b) Redescription of Anolis rubribarbaris (Köhler, McCranie, & Wilson 1999), a poorly known Mesoamerican cloud forest anole (Squamata: Polychrotidae). Zootaxa 1918, 39–44.

Townsend JH, Butler JM, Wilson LD, Austin JD (2009a) A new species of salamander in the Bolitoglossa dunni group (Caudata: Plethodontidae: Bolitoglossinae) from Parque Nacional Montaña de Yoro. Salamandra 45, 95–105.

Townsend JH, Butler JM, Wilson LD, Ketzler LP, Slapcinsky J, Stewart NM (2009b) Significant range extension for the Central American Colubrid snake Ninia pavimentata (Bocourt 1883). Herpetological Bulletin 106, 15–17.

Townsend JH, Butler JM, Wilson LD, Austin JD (2010a) A distinctive new species of moss salamander (Caudata: Plethodontidae: Nototriton) from an imperiled Honduran endemism hotspot. Zootaxa 2434, 1–16.

Townsend JH, Herrera-B LA, Medina-Flores M, Gray LN, Stubbs AL, Wilson LD (2010b) Notes on the second male specimen of the cryptozoic snake Geophis damiani Wilson, McCranie, & Williams 1998 (Squamata: Colubridae: Dipsadinae). Herpetology Notes 3, 305–308.

Townsend JH, Medina-Flores M, Murillo JL, Austin JD (2011a) Cryptic diversity in Chortís Highland moss salamanders (Caudata: Plethodontidae: Nototriton) revealed using mtDNA barcodes and phylogenetics, with a new species from eastern Honduras. Systematics and Biodiversity 9, 275–287.

Townsend JH, Luque-Montes IR, Wilson LD (2011b) Newly-discovered populations of four threatened endemic salamanders (Caudata: Plethodontidae) from the highlands of Honduras. Salamandra 47, 50–54.

Townsend JH, Wilson LD, Cerrato-M CA, Atkinson BK, Herrera-B LA, McKewy-Mejía M (2011c) Discovery of the critically endangered treefrog Plectrohyla chrysopleura in Refugio de Vida Silvestre Texiguat, Honduras. Herpetological Bulletin 115, 22–25.

Townsend JH, Wilson LD, Medina-Flores M, Herrera-B LA, Atkinson BK, Stubbs AL, Cerrato-M CA, McKewy-Mejía M, Gray LN, Austin JD (In press A) Uncovering a hotspot for premontane endemism on the windward side of Refugio de Vida Silvestre Texíguat, Honduras. Salamandra

344

Townsend JH, Wilson LD, Medina-Flores M, Herrera-B LA (In press B) A new species of centipede snake in the Tantilla taeniata group (Squamata: Colubridae) from premontane forest in Refugio de Vida Silvestre Texíguat, Honduras. Journal of Herpetology

Travers SL, Townsend JH, Sunyer J, Obando LA, Wilson LD. 2011. New and noteworthy records of amphibians and reptiles from Reserva de la Biósfera Bosawas, Nicaragua. Herpetological Review 42, 399–403.

Uetz P, et al. (2011) The Reptile Database [website]. (Accessed 6 October 2011: http://www.reptile-database.org).

United Nations Mine Action Service (1998) Nicaragua Landmine Situation Assessment Mission Report. United Nations, 15 December 1998. 6 p.

Vásquez-Almazán CR, Rovito SM, Good DA, Wake DB (2009) A new species of Cryptotriton (Caudata: Plethodontidae) from eastern Guatemala. Copeia 2009, 313– 319.

Vences M, Thomas M, Bonett RM, Vieites DR (2005a) Deciphering amphibian diversity through DNA barcoding: chances and challenges. Philosophical Transactions of the Royal Society B 360, 1859–1868.

Vences M, Thomas M, Meijden Avd, Chiari Y, Vieites DR (2005b) Comparative performance of the 16S rRNA gene in DNA barcoding of amphibians. Frontiers in Zoology 2, 5.

Vieites DR, Min MS, Wake DB (2007) Rapid diversification and dispersal during periods of global warming by plethodontid salamanders. Proceedings of the National Academy of Science, USA 104, 19903–19907.

Vieites DR, et al. (2009) Vast underestimation of Madagascar’s biodiversity evidenced by an integrative amphibian inventory. Proceedings of the National Academy of Science, USA 106, 8267–8272.

Vieites DR, Nieto Román S, Wake MH, Wake DB (2011) A multigenic perspective on phylogenetic relationships in the largest family of salamanders, the Plethodontidae. Molecular Phylogenetic and Evolution 59, 623–635.

Villa J (1972) Anfibios de Nicaragua. Instituto Geográfico Nacional & Banco Central de Nicaragua.

Virgilio M, Backeljau T, Nevado B, De Meyer M (2010) Comparative performance of DNA barcoding across insect orders. BMC Bioinformatics 11, 206.

345

Vreugdenhil D, House P, Cerrato C, Martínez RA, Pereira AC (2002) Racionalizacion del Sistema de Áreas Protegidas del Honduras. Volume I: Main Study. Tegucigalpa, Honduras: PROBAB, World Bank, UNDP, GEF.

Wake DB (1987) Adaptive radiation of salamanders in Middle American cloud forests. Annals of the Missouri Botanical Garden 74, 242–264.

Wake DB (1991) Declining amphibian populations. Science 253, 860.

Wake DB (1998) On the taxonomic status of Nototriton sanctibarbarus McCranie and Wilson (Amphibia: Caudata). The Southwestern Naturalist 43, 88–106.

Wake DB (2009) What salamanders have taught us about evolution. Annual Review of Ecology, Evolution, and Systematics 40, 333–352.

Wake DB, Campbell JA (2000) A new species of diminutive salamander (Amphibia: Caudata: Plethodontidae: Nototriton) from the Montañas del Mico of Guatemala. Proceedings of the Biological Society of Washington 113, 815–819.

Wake DB, Elias P (1983) New genera and a new species of Central American salamanders, with a review of the tropical genera (Amphibia, Caudata, Plethodontidae). Contributions in Science, Natural History Museum of Los Angeles County 345, 1–19.

Wake DB, Vredenburg VT (2008) Are we in the midst of the sixth mass extinction? A view from the world of amphibians. Proceedings of the National Academy of Science, USA 105, 11466–11473.

Wallace, AR (1876) The Geographic Distribution of Animals. With a Study of the Relations of Living and Extinct Faunas as Elucidating the Past Changes of the Earth’s Surface. Two volumes. New York: Harper and Brothers.

Wang, IJ, Crawford, AJ, Bermingham E (2008) Phylogeography of the pygmy rain frog (Pristimantis ridens) across the lowland wet forests of Isthmian Central America. Molecular Phylogenetics and Evolution 47, 992–1004.

Ward RD, Hanner R, Hebert PDN (2009) The campaign to DNA barcode all fishes, FISH-BOL. Journal of Fish Biology 74, 329–356.

Wheeler, QD (2004) Taxonomic triage and the poverty of phylogeny. Philosophical Transactions of the Royal Society B, London 359, 57 –583.

Wiens JJ, Parra-Olea G, García-París M, Wake DB (2007) Phylogenetic history underlies elevational biodiversity patterns in tropical salamanders. Proceedings of the Royal Society B 274, 919–928.

346

Williams H, McBirney AR (1969) Volcanic history of Honduras. University of California Publications in Geological Sciences 85, 1–101.

Williams ST (2007) Safe and legal shipment of tissue samples: does it affect DNA quality? Journal of Molluscan Studies 73, 416–418.

Wilson LD (1999) Checklist and key to the species of the genus Tantilla (Serpentes: Colubridae), with some distributional commentary. Smithsonian Herpetological Information Service 122, 1–34.

Wilson LD, Johnson JD (2010) Distributional patterns of the herpetofauna of Mesoamerica, a biodiversity hotspot. In: Conservation of Mesoamerican Amphibians and Reptiles (eds Wilson LD, Townsend JH, Johnson, JD), pp. 30–235. Eagle Mountain Publishing LC, Eagle Mountain, Utah.

Wilson LD, McCranie JR (1998) The biogeography of the herpetofauna of the subhumid forests of Middle America (Isthmus of Tehuantepec to northwestern Costa Rica). Royal Ontario Museum of Life Sciences Contribution 163, 1–50.

Wilson LD, McCranie JR (1999) The systematic status of Honduran populations of the Tantilla taeniata group (Serpentes: Colubridae), with notes on other populations. Amphibia-Reptilia 20, 326–329.

Wilson LD, McCranie JR (2003) Herpetofaunal indicator species as measures of environmental stability in Honduras. Caribbean Journal of Science 39, 50–67.

Wilson LD, McCranie JR (2004a) The conservation status of the herpetofauna of Honduras. Amphibian and Reptile Conservation 3, 6–33.

Wilson LD, McCranie JR (2004b) The herpetofauna of the cloud forests of Honduras. Amphibian and Reptile Conservation 3, 34–48.

Wilson LD, Meyer JR (1971) A revision of the taeniata group of the colubrid snake genus Tantilla. Herpetologica 27, 11–40.

Wilson LD, Meyer JR (1985) The Snakes of Honduras. 2nd Edition. Milwaukee Public Museum, Milwaukee, Wisconsin.

Wilson LD, Townsend JH (2006) The herpetofauna of the rainforests of Honduras. Caribbean Journal of Science 42, 88–113.

Wilson LD, Townsend JH (2007) Checklist and key to the snakes of the genus Geophis (Squamata: Colubridae), with commentary on distribution and conservation status. Zootaxa 1395, 1–31.

347

Wilson LD, Townsend JH (2010) The herpetofauna of Mesoamerica: biodiversity significance, conservation status, and future challenges. In: Conservation of Mesoamerican Amphibians and Reptiles (eds Wilson LD, Townsend JH, Johnson, JD), pp. 760–812. Eagle Mountain Publishing LC, Eagle Mountain, Utah.

Wilson LD, Cruz-Díaz GA, Villeda E, Flores S (1988) Typhlops costaricensis Jiménez & Savage: an addition to the snake fauna of Honduras. The Southwestern Naturalist 33, 499–500.

Wilson LD, McCranie JR, Cruz-D GA (1994) A new species of Plectrohyla (Anura: Hylidae) from a premontane rainforest in northern Honduras. Proceedings of the Biological Society of Washington 107, 67–78.

Wilson LD, McCranie JR, Williams KL (1998) A new species of Geophis of the sieboldi group (Reptilia: Serpentes: Colubridae) from northern Honduras. Proceedings of the Biological Society of Washington 111, 410–417.

McCranie JR, Wilson LD, Gotte S (2001) Three new country records for Honduran snakes. Herpetological Review 32, 62–63.

Wilson LD, McCranie JR, Espinal MR (2001b) The ecogeography of the Honduran herpetofauna and the design of biotic reserves. In: Mesoamerican Herpetology: Systematics, Zoogeography, and Conservation (eds Johnson JD, Webb RG, Flores- Villela OA), pp. 109–158. Centennial Museum, University of Texas, El Paso, Special Publication 1.

Wilson LD, Townsend JH, Johnson JD (2010) Conservation of Mesoamerican Amphibians and Reptiles (eds). Eagle Mountain Publishing LC, Eagle Mountain, Utah.

Xia Y, et al. (2011) COI is better than 16S rRNA for DNA barcoding Asiatic salamanders (Amphibia: Caudata: Hynobiidae). Molecular Ecology Resources DOI: 10.1111/j.1755-0998.2011.03055.x.

Xia X, Xie Z (2001) DAMBE: Data Analysis In Molecular Biology And Evolution. Journal Of Heredity 92, 371–373.

Xia X, Xie Z, Salemi M, Chen L, Wang Y (2003) An index of substitution saturation and its applications. Molecular Phylogenetics and Evolution 26, 1–7.

Xia X, Lemey P (2009) Assessing substitution saturation with DAMBE. In: The Phylogenetic Handbook: A Practical Approach to DNA and Protein Phylogeny, 2nd edition (eds Lemey P, Salemi M, Vandamme AM), pp. 615-630. Cambridge University Press.

348

Zaldívar-Riverón A, León-Regagnon V, Nieto-Montes de Oca, A (2004) Phylogeny of the Mexican coastal leopard frogs of the Rana berlandieri group based on mtDNA sequences. Molecular Phylogenetics and Evolution 30, 38–49.

Zamora-Villalobos N (2000) Arboles de la Mosquitia Hondureña. CATIE, Serie Técnica, Manual Técnico 43, Turrialba, Costa Rica.

Zhang Z, Schwartz S, Wagner L, Miller W (2000) A greedy algorithm for aligning DNA sequences. Journal of Computational Biology 7, 203–214.

Zhang P, Wake DB (2009) Higher-level salamander relationships and divergence dates inferred from complete mitochondrial genomes. Molecular Phylogenetics and Evolution 53, 492–508.

Zimkus BM, Schink S (2010) Light at the end of the tunnel: insights into the molecular systematics of East African puddle frogs (Anura: Phrynobatrachidae). Systematics and Biodiversity 8, 39–47.

349

BIOGRAPHICAL SKETCH

Josiah (Joe) Townsend received his Ph.D. in Interdisciplinary Ecology from the

University of Florida in Fall 2011, and had previously earned his Master of Arts in Latin

American Studies and Bachelor of Science in Wildlife Ecology and Conservation, both from the University of Florida. Joe was born in Enid, Oklahoma in 1978, the son of

Steve Townsend of Ferriday, Louisiana and Terri Boyd Townsend or Medford,

Oklahoma. After living in Medford for his first year, Joe and his family moved to Kenner,

Louisiana, and then to Sulfur, Louisiana, where they were joined 1981 by his sister,

Katielynn Boyd Townsend. They soon relocated to Bethel Park, a suburb of Pittsburgh,

Pennsylvania, where Joe attended Logan Elementary School. It was here that Joe spent his childhood and developed his fascination with biological diversity in the halls of the Carnegie Museum of Natural History and the creeks and woods of southwestern

Pennsylvania. They relocated briefly to Pembroke Pines, Florida in 5th grade, then to

Hauppauge, New York from 6th through 8th grade, and finally to Miami, Florida in 1992, where they moved the same week that Hurricane Andrew struck. In Miami, Joe attended Southwood Middle School and Miami Palmetto Senior High School, where he graduate in 1996. Joe attended Miami-Dade Community College after high school, while helping to manage a youth hockey league. He began his studies at the University of

Florida in 2000. He married Ileana Rosario Luque-Montes on the 4th of July, 2009, at his parent’s house in Maryland, and they lived happily ever after.

350