phylogenetic systematics of dart poison frogs

PHYLOGENETIC SYSTEMATICS

OF DART-POISON FROGS AND THEIR

RELATIVES (AMPHIBIA:

ATHESPHATANURA: DENDROBATIDAE)

TARAN GRANT,1,2 DARREL R. FROST,1 JANALEE P. CALDWELL,3

RON GAGLIARDO,4 CELIO F. B. HADDAD,5 PHILIPPE J. R. KOK,6

D. BRUCE MEANS,7 BRICE P. NOONAN,8 WALTER E. SCHARGEL,9

AND WARD C. WHEELER10

1Division of Vertebrate Zoology (Herpetology), American Museum of Natural History

(TG: [email protected]; DRF: [email protected]). 2Department of Ecology, Evolution,

and Environmental Biology, Columbia University, New York, NY 10027.3Sam Noble Oklahoma Museum of Natural History and Department of Zoology,

University of Oklahoma, Norman, OK 73072 ([email protected]).4Curator of Tropical Collections, The Dorothy C. Fuqua Conservatory,

Atlanta Botanical Garden, 1345 Piedmont Avenue, Atlanta,

GA 30309 ([email protected]).5Departamento de Zoologia, I.B., Universidade Estadual Paulista (UNESP),

Caixa Postal 199, 13.506-900 Rio Claro, SP, Brazil ([email protected]).6Vertebrates Department (Herpetology), Royal Belgian Institute of Natural Sciences,

Rue Vautier 29, B-1000 Brussels, Belgium ([email protected]).7Coastal Plains Institute and Land Conservancy, 1313 Milton Street,

Tallahassee, FL 32303; and Department of Biological Sciences,

Florida State University, Tallahassee, FL 32306 ([email protected]).8Brigham Young University, Department of Integrative Biology, Provo,

UT 84602 ([email protected]).9Department of Biology, The University of Texas at Arlington,

TX 76019 ([email protected]).10Division of Invertebrate Zoology,

American Museum of Natural History([email protected]).

BULLETIN OF THE AMERICAN MUSEUM OF NATURAL HISTORY

CENTRAL PARK WEST AT 79TH STREET, NEW YORK, NY 10024

Number 299, 262 pp., 79 figures, 37 tables, 8 appendices

Issued August 15, 2006

Copyright E American Museum of Natural History 2006 ISSN 0003-0090

Frontispiece. Adelphobates castaneoticus (Caldwell and Myers, 1990), the type species of Adelphobatesn.gen., named for Charles W. Myers and John W. Daly in recognition of the enormity of their contributionto the scientific knowledge of dart-poison frogs.

CONTENTS

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6History of Dendrobatid Systematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Part I: 1797–1926, Early History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Part II: 1926–Present, Relationships among Dendrobatids . . . . . . . . . . . . . . . . . . . . 13Part III: 1926–Present, Relationships Between Dendrobatidae and Other Frogs . . . . . 30Summary of Historical Review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Phylogenetic Placement of Dendrobatids and Outgroup Sampling . . . . . . . . . . . . . . . . 39Theoretical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Empirical Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Outgroup Sampling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44Conventions and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44General Analytical Approach: Theoretical Considerations . . . . . . . . . . . . . . . . . . . . 45

Choice of Phylogenetic Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45Sources of Evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46Nucleotide Homology and the Treatment of Indels . . . . . . . . . . . . . . . . . . . . . . . 47Total Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

General Analytical Approach: Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49Taxon Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49Phenotypic Character Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49Genotypic Character Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53Laboratory Protocols. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55Molecular Sequence Formatting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56Total Evidence Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56Heuristic Character Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58Heuristic Tree Searching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58Heuristic Data Exploration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60Species Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Phenotypic Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

General Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117Dendrobatid Monophyly and Outgroup Relationships . . . . . . . . . . . . . . . . . . . . . . 118Relationships among Dendrobatids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120Summary of Relationships among Dendrobatids . . . . . . . . . . . . . . . . . . . . . . . . . . 145

A Monophyletic Taxonomy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146Preliminary Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146The Higher-Level Taxonomy of Athesphatanura Frost et al., 2006 . . . . . . . . . . . . . 150

Centrolenidae Taylor, 1951 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151Cruciabatrachia new taxon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

Leptodactylidae Werner, 1896 (1838) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151Chthonobatrachia Frost et al., 2006. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152

Ceratophryidae Tschudi, 1838 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152Batrachylinae Gallardo, 1965 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152Ceratophryinae Tschudi, 1838 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152Telmatobiinae Fitzinger, 1843 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Hesticobatrachia Frost et al., 2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153Cycloramphidae Bonaparte, 1850 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153Calamitophrynia new taxon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

Leiuperidae Bonaparte, 1850 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154Agastorophrynia Frost et al., 2006 . . . . . . . . . . . . . . . . . . . . . . . . . 154

4

Bufonidae Gray, 1825 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154Nobleobatia new taxon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Hylodidae Gunther, 1858 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156Dendrobatoidea Cope, 1865 . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

The Taxonomy Dendrobatoidea Cope, 1865. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157Aromobatidae new family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Anomaloglossinae new subfamily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158Anomaloglossus new genus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158Rheobates new genus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

Aromobatinae new subfamily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160Aromobates Myers, Paolillo, and Daly, 1991 . . . . . . . . . . . . . . . . . . . . . . . . . 160Mannophryne La Marca, 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Subfamily: Allobatinae new subfamily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161Allobates Zimmermann and Zimmermann, 1988. . . . . . . . . . . . . . . . . . . . . . . 162

Dendrobatidae Cope, 1865 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163Colostethinae Cope, 1867 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

Ameerega Bauer, 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163Colostethus Cope, 1867 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165Epipedobates Myers, 1987. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166Silverstoneia new genus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

Hyloxalinae new subfamily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168Hyloxalus Jimenez de la Espada, 1871 ‘‘1870’’ . . . . . . . . . . . . . . . . . . . . . . . . 168

Dendrobatinae Cope, 1865. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169Phyllobates Dumeril and Bibron, 1841. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170Minyobates Myers, 1987. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170Ranitomeya Bauer, 1988. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171Adelphobates new genus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172Oophaga Bauer, 1994 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172Dendrobates Wagler, 1830 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Incertae Sedis and Nomina Dubia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173Discussion: Character Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

Adult Habitat Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175Reproductive Amplexus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176Sex of Nurse Frogs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176Larval Habitat and Diet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184Appendix 1: Chronology of Available Species Names Proposed or Currently

in Dendrobatoidea, with Original, Current, and Proposed Generic Placement. . . . . . 204Appendix 2: Chronology of Available Dendrobatoid Genus-Group Names . . . . . . . . . 210Appendix 3: Chronology of Dendrobatoid Family-Group Names . . . . . . . . . . . . . . . . 211Appendix 4: Numbers and References for Sequences Obtained from Genbank. . . . . . . 211Appendix 5: Tissue and Sequence Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214Appendix 6: Specimens Examined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241Appendix 7: Phenotypic Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247Appendix 8: Diagnostic DNA Sequence Transformations for Named Clades . . . . . . . . 252

2006 GRANT ET AL.: PHYLOGENETICS OF DART-POISON FROGS 5

ABSTRACT

The known diversity of dart-poison frog species has grown from 70 in the 1960s to 247 atpresent, with no sign that the discovery of new species will wane in the foreseeable future.Although this growth in knowledge of the diversity of this group has been accompanied bydetailed investigations of many aspects of the biology of dendrobatids, their phylogeneticrelationships remain poorly understood. This study was designed to test hypotheses ofdendrobatid diversification by combining new and prior genotypic and phenotypic evidence ina total evidence analysis. DNA sequences were sampled for five mitochondrial and six nuclearloci (approximately 6,100 base pairs [bp]; x53,740 bp per terminal; total dataset composed ofapproximately 1.55 million bp), and 174 phenotypic characters were scored from adult andlarval morphology, alkaloid profiles, and behavior. These data were combined with relevantpublished DNA sequences. Ingroup sampling targeted several previously unsampled species,including Aromobates nocturnus, which was hypothesized previously to be the sister of all otherdendrobatids. Undescribed and problematic species were sampled from multiple localities whenpossible. The final dataset consisted of 414 terminals: 367 ingroup terminals of 156 species and47 outgroup terminals of 46 species.

Direct optimization parsimony analysis of the equally weighted evidence resulted in 25,872optimal trees. Forty nodes collapse in the strict consensus, with all conflict restricted toconspecific terminals. Dendrobatids were recovered as monophyletic, and their sister groupconsisted of Crossodactylus, Hylodes, and Megaelosia, recognized herein as Hylodidae. Amongoutgroup taxa, Centrolenidae was found to be the sister group of all athesphatanurans exceptHylidae, Leptodactyidae was polyphyletic, Thoropa was nested within Cycloramphidae, andCeratophryinae was paraphyletic with respect to Telmatobiinae. Among dendrobatids, themonophyly and content of Mannophryne and Phyllobates were corroborated. Aromobatesnocturnus and Colostethus saltuensis were found to be nested within Nephelobates, andMinyobates was paraphyletic and nested within Dendrobates. Colostethus was shown to berampantly nonmonophyletic, with most species falling into two unrelated cis- and trans-Andeanclades. A morphologically and behaviorally diverse clade of median lingual process-possessingspecies was discovered.

In light of these findings and the growth in knowledge of the diversity of this large clade overthe past 40 years, we propose a new, monophyletic taxonomy for dendrobatids, recognizing theinclusive clade as a superfamily (Dendrobatoidea) composed of two families (one of which isnew), six subfamilies (three new), and 16 genera (four new). Although poisonous frogs did notform a monophyletic group, the three poisonous lineages are all confined to the revised familyDendrobatidae, in keeping with the traditional application of this name. We also proposechanges to achieve a monophyletic higher-level taxonomy for the athesphatanuran outgrouptaxa.

Analysis of character evolution revealed multiple origins of phytotelm-breeding, parentalprovisioning of nutritive oocytes for larval consumption (larval oophagy), and endotrophy.Available evidence indicates that transport of tadpoles on the dorsum of parent nurse frogs—a dendrobatid synapomorphy—is carried out primitively by male nurse frogs, with threeindependent origins of female transport and five independent origins of biparental transport.Reproductive amplexus is optimally explained as having been lost in the most recent commonancestor of Dendrobatoidea, with cephalic amplexus arising independently three times.

INTRODUCTION

The past four decades have witnesseda dramatic increase in scientific knowledgeof dendrobatid frogs, known commonly asdart-poison frogs. Extensive field and collec-tion studies have more than tripled thenumber of recognized species from 70 in

1960 to 247 at the time of manuscriptcompletion. Dendrobatid species occupystreams, dense forests, open fields, lowlandrainforests, cloud forests, paramos, andaquatic, terrestrial, and arboreal habitatsfrom Nicaragua to Bolivia and the Atlanticforest of Brazil and from the Pacific coast ofSouth America to Martinique in the French

6

Antilles. All species but one are diurnal.Insfar as is known, all dendrobatids layterrestrial eggs, either on the ground or inphytotelmata, and many are characterized byelaborate reproductive behaviors, includingtransport of tadpoles on the dorsum ofparent frogs and provisioning of nutritiveoocytes for larval consumption.

Approximately one-third of the knownspecies of dendrobatids secrete powerfultoxins from dermal granular (poison) glands.Three of these poisonous species were usedtraditionally by the Embera people of theChoco region of western Colombia to poisontheir blow-gun darts for hunting (Myers etal., 1978), earning the group its commonname. The pioneering work initiated by JohnW. Daly and Charles W. Myers more than30 years ago has led to the discovery indendrobatids of over 450 lipophilic alkaloidsof at least 24 major structural classes (Daly etal., 1999), with novel alkaloids being discov-ered continuously. Many of these so-calleddendrobatid alkaloids have proven to beinvaluable research tools outside systematics.For example, batrachotoxins are used exten-sively in research on sodium channels,epibatidine is a powerful tool in the studyof nicotinic receptors and functions, andhistrionicotoxins are important for studyingthe neuromuscular subtype of nicotinic re-ceptors (Daly et al., 1997, 2000; Daly, 1998).It has become clear that some kind ofsequestration mechanism is responsible forobtaining alkaloids from the diet and in-corporating them into the skin (Daly et al.,1994a), but the details of the mechanism areunknown, as are the dietary sources of thevast majority of dendrobatid alkaloids. For-micine ants, a siphonotid millipede, melyridbeetles, and scheloribatid mites have beenidentified as likely dietary sources for certainalkaloids (Saporito et al., 2003; Dumbacheret al., 2004; Saporito et al., 2004; Takada etal., 2005), but the remaining alkaloids are stillunknown elsewhere in nature. The hydro-philic alkaloid tetrodotoxin has also beendetected in one species of dendrobatid (Dalyet al., 1994b), and it is unknown if itsoccurrence is of symbiotic or dietary origin.Dendrobatid toxicology continues to bea highly active area of investigation.

In addition to studies of dendrobatidtoxins, the conspicuous, diurnal activity ofmany species of dendrobatids has given riseto a large and growing literature in manyareas of evolutionary biology. Among thediverse studies are many investigations ofbreeding biology and territoriality (e.g.,Silverstone, 1973; Wells, 1978, 1980a,1980b, 1980c; Weygoldt, 1987; Zimmermannand Zimmermann, 1988; Summers, 1989;Aichinger, 1991; Caldwell, 1997; Fandino etal., 1997; Junca, 1998; Caldwell and deOliveira, 1999; Summers et al., 1999a,1999b; Luddecke, 2000 ‘‘1999’’; Bourne etal., 2001; Prohl and Berke, 2001; Prohl, 2003;Narins et al., 2003, 2005; Summers andMcKeon, 2004), diet specialization (Silver-stone, 1975a, 1976; Toft, 1980,1995; Don-nelly, 1991; Caldwell, 1996; Parmelee, 1999;Darst et al., 2005), predation (Test et al.,1966), resource use and partitioning (Crump,1971; Donnelly, 1989b; Caldwell, 1993; Limaand Moreira, 1993; Lima and Magnusson,1998; Wild, 1996), learning (Luddecke, 2003),population dynamics (e.g., Toft et al., 1982;Aichinger, 1987; Donnelly, 1989a, 1989c;Duellman, 1995), phonotaxis (Gerhardt andRheinlaender, 1980), energetics (Navas,1996a, 1996b), and correlates of ecologyand physiology (Pough and Taigen, 1990).Similarly, investigations in comparative anddevelopmental morphology have revealedbizarre and fascinating structures (Haas,1995; Grant et al., 1997; de Sa, 1998; Myersand Donnelly, 2001). Ongoing research inthese and related fields continues to generatenovel discoveries with far-reaching implica-tions in evolutionary biology.

In contrast to the major advances achievedin many aspects of their biology, the phylog-eny of dendrobatid frogs remains poorlyunderstood. Detailed knowledge of phyloge-ny is necessary to explain the evolutionaryorigins of the behaviors and other featuresthat have been studied and provides anessential predictive framework to guide fu-ture research. Some progress has been madein recent years, as several workers haveincorporated phylogenetic analysis into theirresearch programs (e.g., Summers et al.,1999b; Santos et al., 2003; Vences et al.,2003a; Graham et al., 2004; Darst et al.,2005), but they have looked at only a small


portion of the diversity of dendrobatids andhave not incorporated all available evidence.Many questions remain unaddressed or un-satisfactorily answered because of a lack ofunderstanding of dendrobatid phylogeny.

Dendrobatid monophyly has been upheldconsistently (e.g., Myers and Ford, 1986;Ford and Cannatella, 1993; Haas, 2003;Vences et al., 2003a) since it was firstproposed by Noble (1926; see Grant et al.,1997), but the relationships among dendro-batids remain largely unresolved. Recently,studies of DNA sequences (e.g., Clough andSummers, 2000; Vences et al., 2000, 2003a;Santos et al., 2003) have provided someinsights, but limitations in both taxon andcharacter sampling have restricted theirimpact on the understanding of dendrobatidphylogeny, and few taxonomic changes haveresulted. Generally, dendrobatid systematicshas been based on few characters, fewrigorous tests, and no comprehensive analysisof available evidence.

Difficulties in understanding the phyloge-ny of dendrobatid frogs are compounded bythe taxonomic problems that surround manynominal species and lack of appreciation ofspecies diversity (Grant and Rodrıguez,2001). Sixty-six recognized species werenamed over the decade 1996–2005, 51 ofwhich are currently referred to Colostethus.Many nominal species throughout Dendro-batidae are likely composed of multiplecryptic species awaiting diagnosis (e.g., Cald-well and Myers, 1990; Grant and Ardila-Robayo, 2002), but the rapid increase inrecognized diversity is not unaccompanied byerror, and critical evaluation of the limits ofnominal taxa will undoubtedly result in somenumber of these being placed in synonymy(e.g., Coloma, 1995; Grant, 2004).

The most generally accepted view ofdendrobatid systematics is based primarilyon the work of Savage (1968), Silverstone(1975a, 1976), and Myers and colleagues(e.g., Myers and Daly, 1976b; Myers et al.,1978, 1991; Myers, 1982, 1987; Myers andFord, 1986), with additional taxonomiccontributions by Zimmermann and Zimmer-mann (1988), La Marca (1992, 1994), andKaplan (1997). Approximately two-thirds ofthe species of dendrobatids are assigned toa ‘‘basal’’ grade of brown, nontoxic frogs

(including Aromobates, Colostethus, Manno-phryne, and Nephelobates), whereas the re-maining third is hypothesized to form a cladeof putatively aposematic frogs (includingAllobates, Ameerega, Dendrobates, Epipedo-bates, Minyobates, Oophaga, Phobobates,Phyllobates, and Ranitomeya). Compellingevidence for that split is lacking, however, assome of the putatively aposematic taxa havebeen shown experimentally to be unable tosequester significant amounts of alkaloids(e.g., Daly, 1998), alkaloid profiles for mostdendrobatids remain unexamined, and sever-al of the species assigned to the ‘‘basal’’ gradeare no less brightly colored than several ofthe species assigned to the aposematic clade(e.g., Colostethus abditaurantius and C. im-bricolus). Furthermore, recent molecularstudies (e.g., Clough and Summers, 2000;Vences et al., 2000, 2003a; Santos et al., 2003)have found several putatively aposematictaxa to be more closely related to species ofColostethus than to other toxic species.

Compelling evidence for the monophyly ofmost genera, especially the ‘‘basal’’ taxa, isalso lacking. The nonmonophyly of Colos-tethus has been recognized for decades(Lynch, 1982), and the naming of Ameerega,Aromobates, Epipedobates, Mannophryne,and Nephelobates has merely exacerbatedthe problem (Kaiser et al., 1994; Coloma,1995; Meinhardt and Parmelee, 1996; Grantet al., 1997; Grant and Castro-Herrera,1998). Colostethus is also the most diversegenus of dendrobatids, with 138 namedspecies recognized currently. Generally, Co-lostethus is regarded as a group of conve-nience for all dendrobatids that cannot bereferred to one of the other genera (e.g.,Grant and Rodrıguez, 2001). Detailed in-vestigations of several new species of Colos-tethus have led to the discovery of novelmorphological characters that help elucidatephylogeny (Coloma, 1995; Grant et al., 1997;Grant and Castro-Herrera, 1998; Grant andRodrıguez, 2001; Myers and Donnelly, 2001;Caldwell et al., 2002a), and molecular studiesare accumulating data rapidly (e.g., Santos etal., 2003; Vences et al., 2003a), but to dateprogress has been limited. Molecular evi-dence for the monophyly of Mannophryneand Nephelobates was presented by La Marcaet al. (2002) and Vences et al. (2003a), but the

8 BULLETIN AMERICAN MUSEUM OF NATURAL HISTORY NO. 299

relationship of those genera to other den-drobatids is unclear. Aromobates has beenhypothesized to be the monotypic sistergroup of all other dendrobatids (Myers etal., 1991), but synapomorphies shared withMannophryne and Nephelobates, also fromthe northern Andes of Venezuela, cast doubton that claim; no molecular evidence hasbeen presented for this taxon.

Among the ‘‘aposematic’’ taxa, only Phyl-lobates is strongly corroborated (Myers et al.,1978; Myers, 1987; Clough and Summers,2000; Vences et al., 2000; Widmer et al.,2000). No synapomorphy is known forAmeerega or Epipedobates, and they arelikely para- or polyphyletic with respect toeach other and/or Allobates, Colostethus,Cryptophyllobates, and Phobobates. Schulte(1989) and Myers et al. (1991) rejectedAllobates and Phobobates on the basis oferrors in the analysis of behavior, lack ofevidence, unaccounted character conflict, in-correct character coding, and creation ofparaphyly in Epipedobates (as also found byClough and Summers, 2000; Vences et al.,2000, 2003; Santos et al., 2003), but manyauthors continue to recognize those genera.Conflicting results have been obtained forsome genera. Phobobates was found to bemonophyletic by Vences et al. (2000), butparaphyletic by Clough and Summers (2000).Minyobates may or may not be nested withinDendrobates (Silverstone, 1975a; Myers,1982, 1987; Jungfer et al., 1996a, 2000;Clough and Summers, 2000). Likewise, al-though neither study recognized Minyobates,it was found to be monophyletic by Santos etal. (2003) but polyphyletic by Vences et al.(2003a). Cryptophyllobates is the most re-cently named genus, but it contains only twospecies and its relationship to other dendro-batids is unclear.

Although the recent studies have demon-strated unambiguously the inadequacies ofthe current state of dendrobatid systematics,they have generated many more questionsthan decisive answers. To a certain extent,this means that this is an exciting time indendrobatid systematics. New light is beingshed on old problems, which is causingscientists to reconsider their prior beliefs(e.g., regarding the single origin of aposema-tism; Santos et al., 2003; Vences et al.,

2003a). However, much of the currentconfusion is due to unreconciled conflictamong datasets analyzed in isolation (e.g.,regarding the monophyly of Minyobates),limited taxon sampling, and failure to includeprior evidence (e.g., morphology) in the newanalyses. This is not surprising, as moststudies to date have been designed to addressparticular questions in evolutionary biologyrather than to resolve dendrobatid phylogenyper se (e.g., Santos et al., 2003). The twokinds of problems are inextricably linked,and more thorough phylogenetic studies mayhave important consequences for the pro-posed evolutionary scenarios, but their em-pirical and analytical requirements differ.

The present study was designed to testcurrent hypotheses of dendrobatid diversifi-cation as severely as possible by combiningnew and prior genotypic and phenotypicevidence in a total evidence analysis. Weincluded as many species of dendrobatids aspossible through our own fieldwork, collea-gues’ ongoing fieldwork, and existing naturalhistory collections. In light of the manyoutstanding problems in species taxonomy,we included numerous undescribed speciesand samples of taxonomically problematicspecies from multiple localities. The primaryaim of this paper is to address dendrobatidsystematics, and to that end we used theoptimal phylogenetic hypothesis to constructa monophyletic taxonomy. Finally, we ana-lyzed the evolution of several charactersystems given the new hypothesis of relation-ships.

HISTORY OFDENDROBATID SYSTEMATICS

Scientific knowledge of dendrobatid frogsbegan in 1797 when the first species wasnamed by Cuvier as Rana tinctoria (seeSavage et al., in press). Over the next twocenturies the number of available species-group names associated with the familyswelled to 304, of which 247 are currentlyrecognized and included in Dendrobatidae(fig. 1; for data see appendix 1). New speciescontinue to be described at a rapid rate,especially in the taxonomically challenginggenus Colostethus.


In this section we review the developmentof knowledge of the systematics of dendro-batids as background for the present study.Rather than present a strict chronology, thisreview is divided into three parts. The firstpart looks at the early history, ending in 1926when Noble provided the modern content ofthe group. The second and third parts beginin 1926 with a monophyletic Dendrobatidaeand continue to the present, examining therelationships among dendrobatids and be-tween Dendrobatidae and other frogs, re-spectively. Ford (1993) and Grant et al.(1997) summarized many aspects of the earlyhistory and the relationships of Dendrobati-dae to other groups, but we also cover somedetails here.

This review does not treat every paperpublished on dendrobatids. First, we includeonly systematics papers (and only the mostrelevant of these; i.e., species descriptions andsynonymies are not detailed), and second, weinclude only papers published for a scientificaudience. Because of the elaborate behaviors,brilliant coloration, diurnal activity, andoccurrence of skin toxins in some species,large ecological, ethological, biochemical,and hobbyist literatures exist, and reviewingthem all lies beyond the scope and purpose ofthe present paper. Also, we include onlyauthorship and date of publication of scien-tific names where relevant; authorship and

date of species-, genus-, and family-groupnames (including new taxa proposed below)are included in appendices 1–3, respectively.We do not address nomenclatural problems.Grant et al. (2006; see also Grant, 2004) andSavage et al. (in press) have pending petitionsto the Commission on Zoological Nomen-clature regarding the use of the species-groupname panamensis and family-group nameDendrobatidae, respectively, and we directthe reader to those papers (especially thelatter) for nomenclatural discussion.Throughout, ‘‘dendrobatid frogs’’ and ‘‘den-drobatids’’ refer to species contained in themodern Dendrobatidae, and formal taxo-nomic names are used as by the author inquestion.

The most thorough study of amphibianrelationships to appear in recent years is thatof Frost et al. (2006). Owing to the dispro-portionate importance of that paper indesigning the present study (e.g., in samplingoutgroup species), we address it in a separatesection (Phylogenetic Placement of Dendro-batidae and Outgroup Sampling).

PART I: 1797–1926, EARLY HISTORY

Although scientific study of dendrobatidshad begun roughly 40 years earlier (Cuvier,1797), little progress was achieved untilDumeril and Bibron’s (1841) publication.

Fig. 1. Accumulation of dendrobatid species, 1797–2005. Only named species currently recognized andincluded in Dendrobatidae are counted. For data see appendix 1.


They delimited major groups of frogs basedon the occurrence of teeth (vomerine[‘‘palate’’] and jaw) and the tongue, butthey also employed characters from thetympanum and middle ear, parotoid glands,number of digits, webbing, hand and foottubercles, vertebrae, and vocal sac todistinguish and group species. Additionally,they employed the relative length of the firstfinger as a character to arrange the threerecognized species of Dendrobates (Dumeriland Bibron, 1841: 651). All dendrobatidswere placed in Phaneroglossa, but they wereallocated to different families based on thepresence and absence of maxillary teeth.Dumeril and Bibron named Phyllobatesbicolor as a new genus and species, and theyconsidered it to be the last hylaeform genus,grouped with either Crossodactylus andElosia (p. 637) or Hylodes and Phyllomedusa(p. 502). Dendrobates was the first bufoni-form genus, grouped with Hylaedactylus (5Hyladactylus, currently a junior synonym ofthe microhylid Kaloula; p. 645). Althoughthe dendrobatid genera were placed indifferent families, Dumeril and Bibron(p. 638; translated freely from the French)actually saw them as being much closer toeach other than many subsequent workerswould appreciate:

This genus [Phyllobates], by the whole of itsstructure, makes obvious the passage of the lastHylaeformes to the first species [those ofDendrobates] of the following family, that ofBufoniformes, in which there are no longerteeth on the whole of the upper jaw and whichalmost always lack them on the palate.

That is, in the transitional, ‘‘grade-thinking’’of the time (as opposed to the ‘‘clade-thinking’’ of the present), Dendrobates andPhyllobates were adjacent genera.

Fitzinger (1843: 32; see also Fitzinger,1860) also recognized the resemblance ofdendrobatid species. He grouped Dendro-bates and Phyllobates in his family Phylloba-tae, but he included Crossodactylus andScinacodes (5 Hylodes) as well.

Gunther (1858) placed all dendrobatids inOpisthoglossa Platydactyla, but Phyllobateswas in Hylodidae with Crossodactylus, Hy-lodes, and Platymantis (now in Ceratobatra-chidae; Frost et al., 2006), while Hylaplesia

(5 Hysaplesia 5 Dendrobates) was in its ownfamily, Hylaplesidae.

Cope (1865) named the family Dendroba-tidae for Dendrobates and placed it inBufoniformia. The remaining dendrobatidswere placed in Arcifera in the heterogeneousfamily Cystignathidae. As discussed in detailby Grant et al. (1997: 30), within two years,Cope (1867; see also Cope, 1871) had begunto see the problems with separating dendro-batids solely on the basis of teeth, but he stillrefused to group them together. All dendro-batids were placed in Raniformia, butColostethidae (containing Colostethus) wasin Ranoid Raniformia, whereas Dendrobati-dae (containing Dendrobates) was in Bufo-noid Raniformia (he did not address Phyllo-bates). In his description of Prostherapis,Cope (1868: 137) argued that, althoughProstherapis was closest in general appear-ance to Phyllobates, it was most closelyrelated to Colostethus, and he placed bothin his Colostethidae. He also stated thatLimnocharis (now a synonym of Crossodac-tylus) was most closely related to Phyllobates.Subsequently, Cope (1875) restricted Rani-formia to the ranoids and applied the nameFirmisternia to the bufonoid taxa. Thisarrangement was based on novel charactersof the pectoral girdle and the number of lobesof the liver, as well as the traditional onesdating to Dumeril and Bibron (1841).

Boulenger (1882) simplified Cope’s schemesomewhat, grouping all dendrobatids inFirmisternia, but he placed Hyloxalus (mis-spelled Hylixalus), Prostherapis, Phyllodro-mus, and Colostethus in Ranidae, Dendro-bates and Mantella in the separate familyDendrobatidae, and Phyllobates in Cy-stignathidae. Gadow (1901) divided Ranidaeinto three subfamilies (Ceratobatrachinae,Dendrobatinae, and Raninae), with thetoothed dendrobatids (including Phyllobates)in Raninae, and Dendrobates, Mantella, andCardioglossa in Dendrobatinae. Gadow wasuncomfortable with this arrangement, how-ever, noting (1901: 272):

This mere loss of teeth, and the geographicaldistribution suggest that these frogs do not forma natural group, but have been developedindependently from other Ranidae, the Neo-tropical Dendrobates from some likewise Neo-tropical genus like Prostherapis, the Malagasy


Mantella from an African form like Megalix-

alus.

Boulenger (1910) eliminated Dendrobati-nae altogether and placed all dendrobatids inRanidae. However, although he did notformally recant, it seems that he was notentirely convinced that Dendrobatidae wasnot a valid group, given that Ruthven (1915:3) acknowledged Boulenger ‘‘for assistance indiagnosing the form’’ Geobatrachus walkerias a new species and genus of Dendrobatidae,and further specified that ‘‘the form fallsunder Boulenger’s definition of the familyDendrobatidae’’ (Ruthven, 1915: 1). Giventhat Ruthven only collected the specimens in1913, his interactions with Boulenger musthave occurred after the publication of LesBatraciens in 1910.

Nicholls (1916) did away with Arcifera andFirmisternia and proposed instead to dividePhaneroglossa into four groups based on thestructure of the vertebral column, particular-ly the centra, the groups being descriptivelynamed Opisthocoela (sacral vertebra bicon-vex, free from coccyx; presacral vertebraeconvex anteriorly and concave posteriorly[5opisthocoelous]); Anomocoela (sacral ver-tebra ankylosed to coccyx or articulatingwith single condyle; presacral vertebraeconcave anteriorly and convex posteriorly[5procoelous] or rarely opisthocoelous);Procoela (sacral vertebra free and articulat-ing with double condyle; presacral vertebraeprocoelous); and Diplasiocoela (sacral verte-bra biconvex; eighth presacral vertebra bi-concave, other seven presacrals procoelous).Insofar as he believed the diplasiocoelouscondition to occur in all firmisternal taxa,this new arrangement did not affect theplacement of dendrobatids.

In a series of four papers, G. K. Noblesynthesized published information with hisown research on the development andstructure of vertebrae, pectoral girdles, thighmusculature, and external morphology toprovide the framework for the modernunderstanding of Dendrobatidae. First, Bar-bour and Noble (1920) carried out a majortaxonomic revision. They followed Peracca(1904: 17) in referring Phyllodromus toProstherapis, but they went on to includeboth Prostherapis and Colostethus (the latter

based largely on a letter from Boulenger toBarbour) as junior synonyms of Phyllobates.Next, Noble (1922) argued against the closerelationship of Dendrobates and Mantella

and explicitly endorsed Boulenger’s (1910)elimination of Dendrobatidae (p. 8), disput-ed Nicholls’s (1916) claim that all firmister-nal species are diplasiocoelous (describinga numberof dendrobatid species as procoe-lous and transferring them to Procoela[pp. 14–15]), and gathered together Brachy-

cephalus, Atelopus, Rhinoderma, Sminthillus

(now a synonym of the brachycephalidEleutherodactylus), Geobatrachus, Oreophry-

nella, Phyllobates, Hyloxalus, Chilixalus (nowa synonym of Rana), and Dendrobates inBrachycephalidae (pp. 68–69). Noble (1923)subsequently diagnosed Hyloxalus fromPhyllobates by the presence of webbing(contra Savage, 1968, who attributed thedefinition of Hyloxalus as toothed dendro-batids with webbed toes to Dunn, 1931).Finally, on the basis of the occurrence of‘‘leathery scutes on the upper surface of eachdigit tip’’, Noble (1926: 7) united Phyllobates,Hyloxalus, and the toothless Dendrobates ina single, exclusive group, the first time suchan arrangement had been proposed (Grant etal., 1997).

Noble (1926) was not only the first to unitethe dendrobatids into an exclusive group, buthe also provided the hypothesis of family-level phylogeny that has guided thinking eversince by proposing that (p. 9)

Crossodactylus gave rise to Hyloxalus by merely

a fusion of the coracoid cartilages. … Hyloxalus

gave rise to Phyllobates by a reduction in its

digital webs. The latter genus evolved and is

evolving directly into Dendrobates by a loss of

its maxillary teeth.

That is, although most of the theoreticalviews Noble held are no longer embraced,such as the notion of group or stockevolution and nonmonophyletic yet naturalgroups (see below and Grant et al., 1997: 31,footnote 18), the scheme of the webbed, moreaquatic species being primitive to the un-webbed, more terrestrial species, and thesebeing primitive to the terrestrial, toothlessspecies has yet to be seriously questioned—ortested.


PART II: 1926–PRESENT, RELATIONSHIPS

AMONG DENDROBATIDS

Having grouped dendrobatids togetherwholly on the basis of anatomical characters,Noble (1927: 103) noted that his conclusion‘‘receives an eloquent support from lifehistory data’’ as well. He pointed out thatmales of species of Dendrobates and Phyllo-bates transport tadpoles to pools, and,further, that ‘‘[n]o other Salientia havebreeding habits exactly like Dendrobates andPhyllobates’’ (p. 104).

Noble (1931: 507) formally recognized thegroup of Phyllobates, Hyloxalus, and Den-drobates as Dendrobatinae, a subfamily ofthe procoelan Brachycephalidae, and hereiterated that the group evolved from Cross-odactylus.

The same year, Dunn (1931) namedPhyllobates flotator, a new species witha swollen third finger in males and anumbelliform oral disc, reduced rows ofkeratodonts, and scattered median papillaein tadpoles. Dunn (1924) had previouslyobserved the same third finger morphologyin P. nubicola, also from Panama, and hepostulated that these two species formeda group within Phyllobates. In error, Dunn(1924) had attributed these characteristics toP. talamancae, and he later stated (Dunn,1931) that in his 1924 paper he had mis-takenly referred specimens of his new P.flotator to P. talamancae. (However, his[Dunn, 1924: 7] description that ‘‘[t]he throatof the male is black’’ indicates that thespecimens mistakenly identified as P. tala-mancae were P. nubicola, not P. flotator; butsee also Savage, 1968.)

Dunn (1931: 389) explicitly followed No-ble’s (1926) evolutionary scenario but furtherpartitioned Phyllobates into groups, stating:

The Phyllobates from Panama, Costa Rica, and

Nicaragua that I have seen fall into three

groups; typical Phyllobates, without specialized

tadpoles, or modified male third finger (these

apparently stem from Hyloxalus, which has

webbed toes); Phyllobates which have special-

ized tadpoles and modified third finger (flotator

and nubicola); and Phyllobates which have

markings black and yellow instead of black

and white, and ventral light markings. (These

are close to Dendrobates.)

Dunn (1933: 69) reviewed Hyloxalus andmodified it slightly to include species withboth webbed and fringed toes. He concludedthat six species were attributable to Hylox-alus thus diagnosed, including Hyloxalusfuliginosus, H. bocagei, H. chocoensis, Hylix-alus collaris, H. granuliventris (now a syno-nym of Phyllobates palmatus), and H. pana-mansis (name subsequently emended to H.panamensis by Dunn, 1940; see Grant, 2004;Grant et al., 2006). Dunn (1933) excluded H.huigrae (now a junior synonym of thebrachycephalid ‘‘Eleutherodactylus’’ diaste-ma) and H. beebei—the latter exclusion beingthe only practical consequence of Dunn’s(1933) redelimitation of Hyloxalus. Dunn didnot apply his new diagnosis consistently oversubsequent years, however; on occasion hereturned to Noble’s (1923) diagnosis, that is,without reference to fringes (e.g., Dunn,1941: 89, 1944: 519), but he also applied hisown diagnosis of having both webbing andfringes (e.g., Dunn, 1957: 77 [as Prostherapis,see below]). Dunn (1933) noted that males ofhis new species H. panamensis possesseda swollen third finger, which he had pre-viously observed in Phyllobates nubicola andP. flotator and had used to group themphylogenetically, but he did not attribute anyphylogenetic significance to the present ob-servation.

In his discussion of the relationships ofDendrobates auratus, Dunn (1941: 88–89)recognized a group of species with roundedlight markings, formed by D. auratus, a spe-cies from ‘‘the western part of Colombia …[in which] the light color is red or yellow’’(presumably D. histrionicus), D. pumilio, andD. speciosus. He also recognized a secondgroup of ‘‘typical Dendrobates’’ with ‘‘dorso-lateral light lines like Phyllobates … [but]lacking maxillary teeth’’ for ‘‘tinctorius,trivittatus, etc.’’, as well as ‘‘lugubris, minutus,and shrevei.’’ In total, Dunn (1941) nowrecognized 18 species of Dendrobates, 26Phyllobates, and 8 Hyloxalus.

Prostherapis remained in the synonymy ofPhyllobates, where it had been placed byBarbour and Noble (1920), for over 35 years.The sole exception was Breder (1946: 405)who reported Prostherapis inguinalis fromPanama without commenting on the status ofthe genus. It was Test (1956: 6), acting on the


advice of Dunn, who resurrected the genus asa senior synonym of Hyloxalus. A moredetailed account of this synonymy waspublished after Dunn’s death (Dunn, 1957:77), where Dunn clarified that ‘‘the presenceof webs and fringes on the toes distinguishesProstherapis from Phyllobates which hasn’tgot them.’’

Bhaduri (1953) studied the urinogenitalsystems of diverse amphibians, includingDendrobates auratus, D. tinctorius, and Co-lostethus flotator (as Phyllobates nubicolaflotator). He noted several differences amongthese species, such as the greater posteriorextension of the kidneys in Dendrobates thanin Phyllobates (p. 56), but he nonethelessconcluded that ‘‘[t]he structural similarities ofthe urinogenital organs which I have ob-served in these two genera lend furthersupport to Noble’s view [that Dendrobatesand Phyllobates are closely related]’’ (p. 72).

Rivero (1961) provided accounts for Ven-ezuelan species. In his description of Pros-therapis shrevei, he postulated that it was‘‘perhaps a race’’ of Prostherapis bocagei, buthe concluded that the two were distinct, butpresumably closely related, species. Rivero(1961) suggested that Phyllobates brunneusand Phyllobates marchesianus might prove tobe conspecific, but he did not proposephylogenetic relationships for the otherspecies.

Dunn’s arrangement was followed until1966, by which time Cochran (1966) hadbecome skeptical of the usefulness of toewebbing in diagnosing these groups of frogs.This change was foreshadowed by Cochranand Goin’s (1964) description of a newwebbed dendrobatid with teeth as Phyllo-bates mertensi. Cochran (1966) accepted therecognition of Phyllobates and Dendrobateson the basis of maxillary teeth, but she (p. 61;see also p. 64) argued against the furtherdivision of toothed species because ‘‘[t]hevariation in degree of webbing of the species[of Prostherapis] is so great … that no validreliance can be placed on it to justify sucha separation on that characteristic.’’ Cochranand Goin (1970) employed this taxonomy,even though it had already become outdatedby the time their monograph was published.

Although Cochran (1966) treated only theColombian species, she proposed a number

of novel groups. These included a group forD. trivittatus and an as yet undescribedspecies (D. ingeri), and a second group forD. hahneli and D. lugubris. A third group wasfurther divided into subgroups for D. opistho-melas and D. minutus ventrimaculatus, andfor the subspecies of D. tinctorius: D. t.histrionicus, D. t. wittei, D. t. chocoensis, andD. t. confluens. Among Colombian species ofPhyllobates, Cochran (1966) recognizeda group for P. bicolor, P. mertensi, P.boulengeri, and P. femoralis, with the lattertwo species more closely related. Anothergroup included P. subpunctatus, P. vergeli, P.chocoensis, and another as yet unnamedspecies (presumably P. thorntoni, named byCochran and Goin, 1970). Curiously, a soon-to-be-named subspecies of P. subpunctatus(P. s. walesi) was placed in a different groupwith P. palmatus. Finally, a group containingP. brunneus, P. pratti, P. latinasus, and P.inguinalis was also proposed.

Savage (1968) ushered in the modern eraof dendrobatid research. Although his studyfocused on the Central American taxa, it washighly influential and arguably the mostimportant paper since Noble’s (1926) inestablishing a framework for much of thedendrobatid systematics research of thefollowing decades. In addition to addressinga number of species-level taxonomic prob-lems in Central America, Savage divided theCentral American species into three groups,and to each of these groups he assigned theoldest available name. He also referredspecies outside of Central America to eachgenus, as far as he could, though subsequentauthors would have to provide completeassignments. New characters Savage em-ployed to diagnose his three groups includedpigmentation of the flesh, size of digital discs,and in larvae the oral disc morphology, rowsof keratodonts, and position of the anus.

Savage (1968: 746–747) resurrected Colos-tethus for his Group I, which included fiveCentral American species and ‘‘most speciescalled Phyllobates in South America’’. Savage(1968: 765) clarified that Dendrobates lugu-bris was a toothed species and that recentworkers had mistakenly applied that name toDendrobates truncatus. Consequently, he as-signed Phyllobates to his Group II, composedof P. lugubris in Central America, and P.


bicolor and P. aurotaenia ‘‘among others’’ inSouth America. Dendrobates was assigned tohis remaining Group III, still composed oftoothless dendrobatids, as it always had been.

In the late 1960s, two graduate studentsundertook studies of the systematics ofDendrobatidae. Stephen R. Edwards wrotehis Ph.D. dissertation (Edwards, 1974a) onColostethus (sensu Savage, 1968, with minormodification). He studied 63 species in hisdissertation, including many undescribedspecies, but only two small papers ondendrobatids were published as a result ofthis work (Edwards, 1971, 1974b); the bulk of

Edwards’s dissertation research—includingdescriptions for the unnamed species in hisdissertation and the quantitative pheneticanalysis—were never published (whichprompted the naming of Colostethus exas-

peratus; see Duellman and Lynch, 1988) andwill therefore not be discussed here (but seediscussion below of Rivero, 1990 ‘‘1988’’ andRivero and Serna, 1989 ‘‘1988’’). In the firstof his papers, Edwards (1971) referred 43

nominal species to Colostethus and describedtwo more species as new; he did not discussthe relationships among the species. In hissecond publication, Edwards (1974b) nameda new species and clarified the identities ofanother three. More importantly, he alsoarranged the nominal species into sevengroups. Although Edwards (1974b: 1) wasexplicit that these groups ‘‘do not reflectevolutionary or taxonomic units’’ and that

their sole purpose was to facilitate compar-isons (e.g., C. vertebralis, shown below inbold, was listed in each appropriate group),this was the first arrangement ever providedfor most of these species. The groups were asfollows:

1. C. elachyhistus, C. fraterdanieli, C. kingsburyi,

C. subpunctatus, C. variabilis.

2. C. alagoanus, C. brunneus, C. capixaba, C.

carioca, C marchesianus.

3. C. collaris, C. dunni, C. herminae, C. mer-

idensis, C. riveroi, C. trinitatus [5 trinitatis].

4. C. beebei, C. chocoensis, C. fuliginosus, C.

granuliventris, C. mandelorum, C. mertensi, C.

palmatus, C. shrevei, C. talamancae, C.

vergeli.

5. C. intermedius, C. latinasus.

6. C. nubicola, C. pratti.

7. C. alboguttatus, C. bromelicola, C. infragutta-

tus, C. olfersioides, C. pratti, C. ranoides, C.vertebralis.

8. C. anthracinus, C. infraguttatus, C. lehmanni,

C. ramosi, C. taeniatus, C. vertebralis, C.

whymperi.

Because Edwards’s dissertation was aquantitative phenetic analysis, he focusedlargely on meristic data and reported fewnovel characters. His most lasting contribu-tion in terms of character delimitation was tofocus on and demarcate explicitly the differentpale lateral stripes found in most species.

Philip A. Silverstone carried out his Ph.D.research on the systematics of Dendrobates(Silverstone, 1970). He published two smallpapers (Silverstone, 1971, 1975b) on dendro-batid systematics, but most of Silverstone’sfindings were published in two comprehen-sive monographs; the first (Silverstone,1975a) summarized his dissertation on Den-drobates and included accounts for 16 spe-cies; the second (Silverstone, 1976) reportedhis research on Phyllobates and included 20species.

Silverstone (1975a: 3) did not put muchcredence in the generic taxonomy he em-ployed (which was largely that of Savage,1968). He noted that there were species withmorphology intermediate between the gen-era, and that ‘‘any rigidly applied definitionof more than one genus for dendrobatid frogscould result in unnatural (5 polyphyletic)groups.’’ But rather than place all dendroba-tids into a single genus, Silverstone (1975a: 3)continued ‘‘to recognize the three currentlyaccepted genera as categories of convenience,that is, as taxonomic units convenient tostudy, but not necessarily natural.’’ Althoughhe thought the three genera may grade intoeach other, Silverstone (1975a: 4) implicitlyfollowed Noble’s (1926) evolutionary scenar-io, stating that he was ‘‘concerned more withthe relationship of Phyllobates to Dendro-bates than with that of Phyllobates toColostethus.’’

The generic diagnoses Silverstone usedwere very similar to Savage’s (1968), al-though he did incorporate new characters(occurrence of the palatine, omosternum,vertebral fusion; he also used fusion andsculpturing of the cranium to diagnose


species groups). In terms of content, therewere two major differences. First, Phyllobatessensu Savage was, explicitly at least, a groupof only three, very similar species, whereasPhyllobates sensu Silverstone included 20species, most of which had been implicitlyreferred to Colostethus by Savage. Second,Silverstone went against all previous workersby transferring two toothless species fromDendrobates to Phyllobates. Although allspecimens of P. trivittatus and most of P.pictus lacked teeth, Silverstone (1975a) wasoverwhelmed by evidence from chromosomesand finger morphology that indicated thesespecies should be placed in Phyllobates. Thus,dendrobatid systematics was finally rid of thea priori weighting applied to teeth that hadhindered progress since Dumeril and Bibron(1841).

In his two monographs, Silverstone(1975a, 1976) proposed numerous speciesgroups, many of which he thought werenatural. Within Dendrobates, he proposed thehistrionicus group for D. histrionicus and D.leucomelas. Significantly, Silverstone (1975a:25) clarified that D. histrionicus was nota subspecies of ‘‘the large, striped, Guiananspecies to which D. tinctorius is restricted’’,but he remained ambivalent with regard tothe putative subspecies of D. histrionicus; hedid not separate them formally, but he didattribute diagnostic color patterns to severalof them. His reasons for treating all the colorpatterns as a single species were that they all‘‘lack an omosternum and have the samebreeding call’’ (Silverstone, 1975a: 23). Basedon larval morphology, Silverstone (1975a: 23)surmised that ‘‘the histrionicus group is moreclosely related to the pumilio group than tothe other two groups of Dendrobates.’’

Silverstone’s minutus group contained sixspecies: D. altobueyensis, D. fulguritus, D.minutus, and D. opisthomelas from CentralAmerica and northwestern South Americaand D. quinquevittatus and D. steyermarkifrom the Amazon basin (the former fromlowlands, the latter from 1,200 m on thesummit of the tepui Cerro Yapacana). Withinthis group, Silverstone (1975a: 29) hypothe-sized a close relationship between D. fulgur-itus and D. minutus on the basis of size anddorsal striping; his decision to treat them asdistinct species was due to his having

collected them in sympatry. He also con-jectured that D. minutus and D. opisthomelaswere closely related, as tadpoles of thesespecies were the only ones in the genus withan indented oral disc and dextral anus;Silverstone was not completely convinced ofthe identity of the tadpoles he assigned to D.altobueyensis and D. fulguritus, but they alsohad an indented oral disc and dextral anus.Tadpoles of D. quinquevittatus and D. steyer-marki were unknown to Silverstone, and heassigned those species to the minutus groupon the basis of other characters. He alsohypothesized that D. steyermarki was ‘‘moreclosely related to [the western Andean D.opisthomelas] than to any other species ofDendrobates’’ (Silverstone, 1975a: 36).

Silverstone (1975a) proposed the pumiliogroup for D. granuliferus, D. pumilio, and D.speciosus. Silverstone (1975a: 38) argued thatD. granuliferus and D. pumilio were veryclosely related, perhaps even conspecific, andthat they were ‘‘probably geographically andgenetically continuous before the onset oforogeny and aridity in Costa Rica.’’ Thiswould leave D. speciosus as their sister group.As mentioned above, Silverstone hypothe-sized that the pumilio and histrionicus groupswere sister groups.

The tinctorius group included D. auratus,D. azureus, D. galactonotus, D. tinctorius, andD. truncatus. Within this group, Silverstone(1975a) proposed that D. auratus was mostclosely related to D. truncatus. He alsohypothesized that D. azureus had ‘‘arisen byisolation of a population of D. tinctorius inforest islands surrounded by unsuitablehabitat’’ (Silverstone, 1975a: 44).

The 20 species of Phyllobates Silverstone(1976) recognized were arranged into fourgroups, but the relationships among thesefour groups were not addressed. The bicolorgroup was the same as Phyllobates sensuSavage (1968) with the addition of two morespecies. That is, he placed P. aurotaenia, P.bicolor, and P. lugubris in a single group (ashad Savage) together with an as yet unnamedtaxon (later named Dendrobates silverstonei;Silverstone doubted the inclusion of thisspecies in this group but placed it there dueto its superficial resemblance with P. bicolor)and P. vittatus (which Savage considered tobe conspecific with P. lugubris). Silverstone


did not further resolve the relationships ofthis group.

The femoralis group included P. anthonyi,P. boulengeri, P. espinosai, P. femoralis, P.tricolor, and P. zaparo. Within this group,Silverstone (1976) proposed the followingrelationships: (P. tricolor (P. femoralis P.zaparo) (P. anthonyi P. boulengeri P. espino-sai)).

The pictus group contained P. bolivianus,P. ingeri, P. parvulus, P. petersi, P. pictus, P.pulchripectus, and P. smaragdinus. Silver-stone (1976) was doubtful that this groupwas monophyletic, but he did think parts of itwere. He grouped P. pictus and P. parvulustogether based on the shared presence ofa calf spot. He also grouped P. petersi and P.pulchripectus on the basis of similar colorpatterns, and united them with P. bolivianus(although he was more ambivalent about thelatter’s relationship). Silverstone did notplace P. smaragdinus, and he did not proposea scheme of relationships among thesegroups.

Silverstone (1976) was more certain aboutthe naturalness of the trivittatus group, whichcontained only the similarly colored P.bassleri and P. trivittatus. Silverstone didnot publish further studies on dendrobatidfrogs, as he discontinued working in herpe-tology to pursue a scientific career in botany.

In 1976, Charles W. Myers and John W.Daly began publishing on the systematicsimplications of their work initiated a decadeearlier (Daly and Myers, 1967). They addedthree new sources of evidence: alkaloidprofiles, vocalizations, and behavior. Modernresearch in dendrobatid alkaloids was initi-ated by Marki and Witkop (1963), and theaccumulated data appeared to have clearsystematics implications. Similarly, audio-spectrographic analysis of vocalizations hadbeen carried out for several groups of frogs(e.g., Bogert, 1960; Martin, 1972), but not fordendrobatids. Numerous workers had pub-lished observations on dendrobatid parentalcare and other behaviors (Wyman, 1859a,1859b [reported as Hylodes lineatus; Dendro-bates trivittatus fide Boulenger, 1888], Ruth-ven and Gaige, 1915; Senfft, 1936; Dunn,1944; Test, 1954; Stebbins and Hendrickson,1959; Duellman, 1966; Goodman, 1971;Crump, 1972; Bunnell, 1973; Silverstone,

1973, 1975a, 1976; Dole and Durant, 1974),and to these were added the extensive fieldand laboratory observations of Myers andDaly, who analyzed the phylogenetic impli-cations of these advances.

Based on these and traditional data, Myersand Daly (1976b) named three new speciesand redescribed D. histrionicus. They alsoadded support to Silverstone’s (1975a) pumi-lio group, and they proposed a group con-sisting of D. histrionicus, D. lehmanni, and D.occultator (they did not mention D. leucome-las, which Silverstone had grouped with D.histrionicus). That same year, Myers andDaly (1976a) named D. abditus and added itand D. viridis to Silverstone’s (1975a) minutusgroup.

Myers et al. (1978) proposed a restrictedapplication of Phyllobates as an explicitlymonophyletic genus (the first in the family).They argued that Phyllobates sensu Silver-stone (1976) had been diagnosed on the basisof symplesiomorphy, whereas the occurrenceof batrachotoxins was a synapomorphy fora group containing P. aurotaenia, P. bicolor,P. lugubris, P. terribilis, and P. vittatus, andthus resembling Phyllobates sensu Savage(1968). In order to avoid coining new nameswithout evidence of monophyly, Myers et al.(1978) referred the rest of Phyllobates sensuSilverstone (1976) to Dendrobates, pendinga comprehensive phylogenetic analysis.

Rivero (1978 ‘‘1976’’) named three speciesof Colostethus and proposed that C. haydeeaeand C. orostoma were closest relatives (laterdubbed the haydeeae group by Rivero, 1980‘‘1978’’: 99). This conjecture was basedlargely on the supposed occurrence of fouranterior and five posterior rows of kerato-donts in larvae, although Rivero did note thepossibility that the larvae were not of thesespecies. Rivero (1978 ‘‘1976’’) speculated thatC. leopardalis was most closely related to C.alboguttatus, C. collaris, and C. meridensisand concluded that ‘‘in spite of the presenceof a collar in C. leopardalis and its absence inC. alboguttatus, these two species are moreclosely related to each other than either is toC. collaris [which has a collar]’’ (p. 334;translated from the Spanish).

Rivero (1979) suggested that the presenceof a dark chest collar delimited a mono-phyletic group of species confined to the


Venezuelan Cordillera de la Costa. Rivero(1979) mentioned the occurrence of similardark spotting on each side of the chest inseveral species from southern Colombia tonorthern Peru, and he (Rivero, 1979: 172)proposed that the collared species werederived from the species with chest spotting.Curiously, Rivero (1984 ‘‘1982’’) later in-cluded C. mandelorum, a species that lacksa dark collar, in this group, and, followingRivero (1979: 173), went on to hypothesizethat the ‘‘ancestral stock of C. trinitatis …gave origin to the other collared forms ofVenezuela and C. mandelorum’’ (p. 12). Theinclusion of this uncollared species in thisgroup was based on the species’s ‘‘affinitywith collared species, its limited altitudinaldistribution, and the absence currently of anyuncollared species similar to it’’ (Rivero, 1984‘‘1982’’: 12).

Myers and Daly (1979) further character-ized the trivittatus group based on vocaliza-tions, and they added to it D. silverstonei.The following year, Myers and Daly (1980)named a new species (D. bombetes), resur-rected D. reticulatus, and assigned both to theminutus group. (They also included an un-named species, finally described 20 yearslater as D. claudiae by Jungfer et al., 2000.)Furthermore, they hypothesized that D.abditus, D. bombetes, and D. opisthomelas,all from the western Andes of Colombia andEcuador, formed a monophyletic groupdelimited by a ‘‘median gap that interruptsthe papillate fringe on the posterior (lower)edge of the oral disc’’ (Myers and Daly, 1980:20).

Based on finger length and color pattern,Rivero (1980 ‘‘1978’’) proposed that C.inflexus was part of the haydeeae group(sensu Rivero 1978 ‘‘1976’’), and that theirclosest relative was C. alboguttatus. Colos-tethus inflexus was later placed in thesynonymy of C. alboguttatus by Rivero(1984 ‘‘1982’’), but he did not address thephylogenetic implications of this change.Although he did not retract his previousclaim that C. haydeeae and C. orostoma hada larval keratodont row formula of 4/5,Rivero (1980 ‘‘1978’’) did seriously questionits veracity, given that no other Colostethuswas known to possess this morphology. LaMarca (1985) subsequently identified Riv-

ero’s C. haydeeae tadpole as Hyla platydac-tyla.

Myers (1982) resurrected and redescribedD. maculatus but clarified that he was ‘‘un-able at this time to demonstrate a closerelationship with any other known dendro-batid’’ (p. 2). Myers (1982: 2) also resurrectedD. fantasticus from synonymy with D.quinquevittatus and placed D. vanzolinii, D.fantasticus, D. quinquevittatus, and D. reticu-latus in a monophyletic quinquevittatus groupdelimited by ‘‘distinctively reticulate limbs’’.Myers (1982) speculated that D. captivus andD. mysteriosus were sister species, but he wasunable to present any synapomorphies tocorroborate this hypothesis.

In 1982, Lynch published two papers onColombian dendrobatids. Lynch (1982)named C. edwardsi and C. ruizi and hypoth-esized that they formed a distinct groupwithin Dendrobatidae, based on the occur-rence of an ‘‘anal sheath’’ and putativelyderived absence of a tarsal fold or tubercle(also known in dendrobatid literature as‘‘tarsal keel’’). He refrained from namingthis group formally to avoid encumberingfuture research; he also observed that nosynapomorphies were known for Colostethusand declared that the genus was paraphyletic(although he did not present evidence to thateffect).

Lynch and Ruiz-Carranza (1982) de-scribed the new genus and species Atopo-phrynus syntomopus as a dendrobatid. Theyreported a number of features unknown inother dendrobatids, but they were unable toelucidate the relationships of this taxon withrespect to other dendrobatids. They (Lynchand Ruiz-Carranza, 1982: 561) explicitlyrejected the absence of teeth as a synapomor-phy ‘‘because it postulates loss of an attri-bute.’’

Rivero (1984) clarified that C. dunni didnot have a throat collar (contra Edwards,1974a, 1974b) and provided a name, C.oblitteratus, for the MCZ material Edwardshad seen.

Myers et al. (1984) combined what hadbeen the pumilio and histrionicus groups intoa new, expressly monophyletic histrionicusgroup delimited by the synapomorphic oc-currence of a ‘‘chirp call’’. This groupcontained D. arboreus, D. granuliferus, D.


histrionicus, D. lehmanni, D. occultator, D.pumilio, D. speciosus, and an unnamedspecies.

Maxson and Myers (1985) employedmicrocomplement fixation to compare theserum albumin of several dendrobatids. Theyconcluded that recognition of Phyllobates asa separate group was warranted, and that the‘‘[s]peciation events leading to the livingspecies of true dart-poison frogs (Phyllobates)appear to have occurred within the last fivemillion years’’ (Maxson and Myers, 1985:50). They also found that the species ofDendrobates they studied were much moredivergent than the species of Phyllobates, andthat this was ‘‘consistent with accumulatingevidence that Dendrobates is a polyphyleticassemblage’’ (Maxson and Myers, 1985: 50).They suggested that at least four majorlineages were represented, and that initialdivergence dated back some 60 million years.

Pefaur (1985) described two new species ofColostethus from Venezuela, but he did notdiscuss their phylogenetic relationships. LaMarca (1985: 4) claimed that his new speciesC. molinarii was ‘‘a member of the C.alboguttatus group, a monophyletic assem-blage’’ comprised additionally of C. albogut-tatus, C. dunni, C. haydeeae, C. leopardalis,C. mayorgai, C. meridensis, and C. orostoma.However, La Marca (1985) did not offer anyevidence in support of this conjecture.

Dixon and Rivero-Blanco (1985) namedColostethus guatopoensis (placed in the syn-onymy of Colostethus oblitterata by Rivero,1990 ‘‘1988’’) and grouped it with C. riveroion the basis of the shared absence of theouter metatarsal tubercle. This synapomor-phy was disputed by La Marca (1996‘‘1994’’), who reported the occurrence of theouter metatarsal tubercle in both species (andconsidered both species to be valid).

In a series of privately published butnomenclaturally valid (according to ICZN,1999) papers, Bauer (1986, 1988, 1994)named several genera and speculated on theirrelationships. Bauer’s proposals were basedon reinterpretations of Silverstone (1975a,1976) and Myers and colleagues (mainlyMyers et al., 1978, 1984; Myers and Bur-rowes, 1987) augmented with limited obser-vations of a few species in captivity. Bauer’spublications were overlooked by all workers

except Wells (1994), and as a result theliterature is now quite confusing; for thatreason we break from chronological order tosummarize Bauer’s contributions. Bauer(1986) named Ameerega (type specie: Hylatrivittata) for the species of Phyllobates sensuSilverstone (1976) that were not placed inPhyllobates sensu Myers et al. (1978). Bauer(1988) named Ranitomeya (type species:Dendrobates reticulatus) for Dendrobates cap-tivus, D. fantasticus, D. imitator, D. myster-iosus, D. quinquevittatus, D. reticulatus, andD. vanzolinii. Bauer (1988) attributed thename to ‘‘Bauer, 1985’’, and it was alsoemployed by Bauer (1986); however, thoseprior uses do not constitute nomenclaturalactions because (1) the 1985 use was ina publication that did not specify authorship(Anonymous, 1985) and (2) the 1986 use didnot specify a type species. Only Bauer’s 1988use was sufficient to make Ranitomeya anavailable name. In that paper, Bauer alsonamed Pseudendrobates, but that is anobjective synonym of Phobobates Zimmer-mann and Zimmermann, 1988 (see below)because it was published later and specifiedthe same type species (Dendrobates silversto-nei). Bauer (1994: 1) stated that ‘‘Phobobatesshould be considered a synonym’’, but ofwhat he did not say, and he did not provideevidence to substantiate his view. Bauer(1994) proposed the name Oophaga (typespecies: Dendrobates pumilio) for the histrio-nicus group of Myers et al. (1984), namely,Dendrobates arboreus, D. granuliferus, D.histrionicus, D. lehmanni, D. occultator, D.pumilio, D. speciosus, and D. sylvaticus.Although Oophaga was never placed in thesynonymy of Dendrobates, it was also neverused again. Finally, Bauer (1994) namedParuwrobates as a monotypic genus toaccommodate D. andinus; Bauer did notaddress the placement of D. erythromos,although Myers and Burrowes (1987) hadgrouped them together (and Bauer claimed tobe basing his new taxonomy on their paper).In that paper, Bauer also resurrected Pros-therapis, but he did not list the content of thegenus and the evidence he cited for distin-guishing Prostherapis inguinalis from Colos-tethus latinasus was his erroneous claim thatthey differ in the occurrence of swelling in thethird finger in adult males (see Grant, 2004).


Bauer (1986, 1988, 1994) was the onlyrecent worker to recognize subfamilies withinDendrobatidae. In the most recent proposal(Bauer, 1994), he recognized Dendrobatinaefor Dendrobates, Oophaga, Ranitomeya, andMinyobates; Phyllobatinae for Phyllobatesand Ameerega; and Colostethinae for Aro-mobates, Colostethus, and Epipedobates.(Note that Bauer’s use of Epipedobates wasrestricted to Silverstone’s femoralis group,and he applied Ameerega to the bulk ofPhyllobates sensu Silverstone.) Bauer wasapparently unaware of Mannophryne LaMarca, 1992. Subfamily diagnoses employeddifferences in chromosome number, colora-tion, occurrence of maxillary teeth, skintoxins, webbing, length of first finger, musclecoloration, clutch size, breeding biology, andtadpole specialization. He believed Dendro-batinae and Phyllobatinae to be monophy-letic, but thought that Colostethinae wasparaphyletic (though he did not say withrespect to what); he did not otherwisepropose relationships among the subfamilies.

Meanwhile, Myers and Ford (1986) exam-ined Lynch and Ruiz-Carranza’s (1982)assertion that Atopophrynus was a dendroba-tid. They could find no support for Lynchand Ruiz-Carranza’s claim, given that speci-mens they examined showed major differ-ences from dendrobatids in external mor-phology, jaw musculature, thigh muscu-lature, skull, finger structure, and hyoidstructure, and shared no particular synapo-morphy. Consequently, they removed thegenus from Dendrobatidae and placed itin Leptodactylidae (transferred with othereleutherodactylines to Brachycephalidae byFrost et al., 2006).

Myers (1987) proposed a major taxonomicrearrangement aimed to better reflect hy-potheses of monophyly, whereby ‘‘[d]end-robatids that produce lipophilic alkaloids area monophyletic group that is now partitionedamong four genera’’ (p. 304). Epipedobates(type species: Prostherapis tricolor) wasnamed to accommodate most species ofPhyllobates sensu Silverstone (1976) minusthe species Myers et al. (1978) had placed intheir restricted Phyllobates. AlthoughMyers’s intention was the same as Bauer’s(discussed above), his designation of a differ-ent type species means that the two names

may be applied to different groups. Dendro-bates was redefined as a monophyletic groupdelimited by a suite of synapomorphies fromlarval, adult, behavioral, and alkaloid char-acters. Dendrobates included the quinquevit-tatus group of Myers (1982), which had beenpart of the minutus group (Silverstone, 1975a;Myers and Daly, 1976a, 1980; Myers, 1982).The remainder of the minutus group wastransferred to Minyobates, which retained theplesiomorphic states not found in Dendro-bates. Dendrobates and Phyllobates wereclaimed to be sister groups based on ‘‘theloss of cephalic amplexus (cephalic embracesometimes retained in an aggressive context),loss of the primitive oblique lateral line, andfirst appearance of 3,5-disubstituted indolizi-dine alkaloids’’ (Myers, 1987: 305).

Myers and Burrowes (1987) named Epi-

pedobates andinus and postulated that its

nearest relative was E. erythromos based on

‘‘a few similarities of the color patterns’’ and

‘‘an overall morphological similarity’’ (Myers

and Burrowes, 1987: 16). They followed Vigle

and Miyata (1980) in tentatively placing these

species in Silverstone’s (1976) pictus group.

Given their placement in this group, indirect

evidence for the close relationship of E.

andinus and E. erythromos not cited by Myers

and Burrowes is given by their occurrence on

the Pacific slopes in contrast to the cis-

Andean distribution of the remainder of the

pictus group. Myers and Burrowes (1987)

also transferred Phyllobates azureiventris to

Epipedobates, also in the pictus group.

Zimmermann and Zimmermann (1988)performed a phenetic analysis of 62 char-

acteristics (mostly behavioral, but also in-cluding vocalizations and larval morphology)

for 32 species. Their analysis resulted in ninegroups of decreasing similarity:

N Colostethus group: C. inguinalis, C. collaris, C.

trinitatis, C. palmatus

N Epipedobates pictus group: E. pulchripectus, E.

pictus, E. parvulus

N Epipedobates tricolor group: E. anthonyi, E.

boulengeri, E. espinosai, E. tricolor

N Epipedobates silverstonei group: E. bassleri, E.

silvestonei, E. trivittatus

N Epipedobates femoralis group: E. femoralis

N Phyllobates terribilis group: P. lugubris, P.

terribilis, P. vittatus


N Dendrobates leucomelas group: D. auratus, D.azureus, D. leucomelas, D. tinctorius, D.truncatus

N Dendrobates quinquevittatus group: D. fantas-ticus, D. imitator, D. quinquevittatus, D.reticulatus, D. variabilis

N Dendrobates histrionicus group: D. granuli-ferus, D. histrionicus, D. lehmanni, D. pumilio,D. speciosus

Furthermore, Zimmermann and Zimmer-mann (1988) proposed Phobobates for theirsilverstonei group (viz., Dendrobates bassleri,D. silverstonei, and Hyla trivittata) andAllobates for the monotypic femoralis group.However, Schulte (1989: 41) and Myers et al.(1991: 18) rejected those genera, citing errorsin the analysis of behavior, lack of evidence,unaccounted character conflict, incorrectcharacter coding, and creation of paraphyly.

In 1989, the Colostethus collaris group,delimited by ‘‘a dark band present on theposterior part of the throat and anterior partof the chest in all members’’, was proposedby La Marca (1989: 175) for C. collaris, C.oblitteratus (as C. guatopoensis), C. herminae,C. neblina, C. olmonae, C. riveroi, C. trinita-tis, and C. yustizi.

Over 60 years after the only previousspecimen had been collected, Schulte (1990)rediscovered Dendrobates mysteriosus fromAmazonian Peru. Despite some similarities,Schulte (1990: 66) determined that it wasnecessary to exclude D. mysteriosus from thequinquevittatus group (sensu Silverstone,1975a, presumably), and he further stipulatedthat it was not closely related to D. captivusas proposed by Myers (1982). Rather,Schulte (1990: 67) believed D. mysteriosus tobe most closely related to D. histrionicus fromthe lowlands of Pacific Ecuador and Colom-bia. He based this on shared size, absence ofomosternum, occurrence of round spots ona dark background, reproductive behavior,an elevated number of small ova, anda similar fundamental frequency of the call,although no audiospectrographic analysiswas performed. None of these characters isunique to the histrionicus group, and severalother reported character states conflict withthis relationship (e.g., larval mouth parts).

Rivero (1990 ‘‘1988’’) selectively extracteddata from Edwards’s unpublished disserta-tion (1974a) and arranged the species of

Colostethus into eight groups, which weresoon expanded to nine by Rivero and Serna(1989 ‘‘1988’’). Numerous species were notplaced in any group because of apparentcharacter conflict and other concerns. Al-though these groups were putatively based onderived characters and were hypothesized tobe monophyletic, ‘‘characteristics shared bythe majority of members’’ and geographicdistribution were attributed evidential signif-icance (Rivero, 1990 ‘‘1988’’: 4). The contentof the groups (as modified by Rivero andSerna, 1989 ‘‘1988’’ and augmented byRivero and Granados-Dıaz, 1990 ‘‘1989’’;Rivero, 1991a ,1991b; Rivero and Almen-dariz, 1991; Rivero and Serna, 1991, 2000‘‘1995’’; La Marca, 1998 ‘‘1996’’) was asfollows:

N Group I (vertebralis group): C. elachyhistus,C. exasperatus, C. idiomelus, C. infraguttatus,C. mittermeieri, C. peculiaris, C. shuar, C.

sylvaticus, C. vertebralis

N Group II (brunneus group): C. brunneus, C.

intermedius [5 C. kingsburyi fide Coloma,1995], C. kingsburyi, C. marchesianus, C.

olfersioides, C. peruvianus, C. talamancae, C.

trilineatus

N Group III (alagoanus group): C. alagoanus, C.

capixaba, C. carioca

N Group IV (inguinalis group): C. agilis, C.

alacris, C. brachistriatus [as C. brachystriatus],C. cacerensis [5 inguinalis fide Grant, 2004],C. dysprosium, C. erasmios, C. fallax, C.

fraterdanieli, C. inguinalis, C. latinasus, C.

mertensi, C. nubicola, C. paradoxus [5 Epipe-

dobates tricolor fide Coloma, 1995], C. pratti

N Group V (edwardsi group): C. edwardsi, C.

ruizi

N Group VI (fuliginosus group sensu stricto; i.e.,sensu Rivero and Serna, 1989 ‘‘1988’’): C.

abditaurantius, C. betancuri, C. chocoensis, C.

excisus, C. faciopunctulatus, C. fuliginosus, C.

furviventris, C. maculosus [5 C. bocagei fideColoma, 1995], C. nexipus, C. palmatus, C.

pseudopalmatus, C. ramirezi (?), C. shrevei, C.

thorntoni, C. vergeli

N Group VII (trinitatis group): C. collaris, C.

neblina, C. oblitteratus, C. olmonae, C. riveroi,C. trinitatis

N Group VIII (alboguttatus group): C. albogut-

tatus, C. duranti, C. haydeeae, C. mayorgai, C.

molinarii, C. orostoma, C. saltuensis, C.

serranus

N Group IX (subpunctatus group): C. anthraci-

nus, C. borjai, C. cevallosi, C. citreicola [5 C.


nexipus fide Coloma, 1995], C. degranvillei, C.

festae, C. jacobuspetersi, C. lehmanni, C.

marmoreoventris, C. mystax, C. parcus [5 C.

exasperatus fide Coloma, 1995], C. pinguis, C.

poecilonotus, C. pumilus, C. ramirezi (?), C.

ramosi, C. ranoides, C. sauli, C. subpunctatus,C. taeniatus [5 C. pulchellus fide Coloma,1995], C. tergogranularis [5 C. pulchellus fideColoma, 1995], C. torrenticola [5 C. jacobus-

petersi fide Coloma, 1995], C. whymperi, C.

yaguara

Among these groups, Rivero (1990‘‘1988’’: 26) hypothesized that the brunneusgroup formed (or was close to) the ‘‘ancestralstock’’ from which the other Colostethus werederived. On the same page, he also hypoth-esized that the brunneus group gave rise to theinguinalis group (see also Rivero, 1991a: 23).He postulated that the fuliginosus group(sensu lato; fuliginosus + subpunctatus groupsof Rivero and Serna, 1989 ‘‘1988’’) wasderived from the inguinalis group, and thatthe members of the fuliginosus group thatlack toe webbing ‘‘could be close to theancestral stock that gave rise to [the verte-bralis group].’’ The edwardsi group wasconjectured to have arisen from the fuligino-sus group (sensu lato), and the alboguttatusgroup was believed to have arisen from thesame ancestral stock as the edwardsi group.However, Rivero (1990 ‘‘1988’’) also specu-lated that the trinitatis group (which wasidentical to La Marca’s, 1989, collaris group)may have given rise to the alboguttatus group(which differed only slightly from La Mar-ca’s, 1985, alboguttatus group), citing puta-tive intermediate forms as evidence. Rivero(1990 ‘‘1988’’) was more ambivalent withregard to the relationships of the trinitatisgroup than he had been previously (Rivero,1979). He now concluded that the trinitatisgroup may have arisen from the vertebralisgroup (as he had argued in 1979), or thatboth the trinitatis and vertebralis groups mayhave arisen from the fuliginosus group (sensulato). Besides Rivero and his colleagues, fewauthors have recognized these groups (seeColoma, 1995).

Caldwell and Myers (1990) further eluci-dated the systematics of the Dendrobatesquinquevittatus group, which had been re-vised previously by Myers (1982). In theprocess, they proposed that D. quinquevitta-

tus sensu stricto was sister to D. castaneoti-cus, united by the synapomorphic absence ofthe inner metacarpal tubercle, as well asa number of character states of moreambiguous polarity. As a working hypothe-sis, they further proposed that this group wassister to a clade united by the synapomorphyof pale limb reticulation (i.e., D. fantasticus,D. quinquevittatus, D. reticulatus, D. vanzoli-nii), but they were unable to propose anysynapomorphies to support this arrange-ment.

Myers et al. (1991) named a new genus andspecies, Aromobates nocturnus. They arguedthat this was the sister of all other dendro-batids on the basis of (1) nocturnal and (2)aquatic behavior, (3) large size, and (4)presence of m. adductor mandibulae externussuperficialis in many specimens. They alsoproposed an informal redefinition of Colos-tethus based on the occurrence of the swollenthird finger in adult males; they were explicitthat they were not proposing formal nomen-clatural changes. Their Colostethus sensustricto corresponded with Rivero’s (1990‘‘1988’’) and Rivero and Serna’s (1989‘‘1988’’) inguinalis group with the additionof Phyllobates flotator and Colostethus im-bricolus. The remaining species of Colostethussensu lato were assigned to Hyloxalus (withinwhich was included Phyllodromus), althoughno synapomorphies or diagnostic characters(besides the lack of the swollen third finger)were proposed. Almost immediately, Myers(1991; see also Myers and Donnelly, 1997:25) retreated from this arrangement, giventhat the swollen third finger also occurs insome species of Epipedobates. Myers et al.(1991) provided a cladogram summarizingtheir views on the relationships of thedendrobatid frogs, reproduced here as fig-ure 2.

In comparing Aromobates nocturnus toother dendrobatids, Myers et al. (1991)speculated that it may be most closely relatedto the collared species of Venezuelan Colos-tethus. They listed 10 species (1 undescribed)as definitely possessing a collar and 2 more aspossibly having one. They did not definea group for these species, and their list ofcollared species differed from La Marca’s(1989) collaris group (5 trinitatis group ofRivero, 1990 ‘‘1988’’ and Rivero and Serna,


1989 ‘‘1988’’) by including species of LaMarca’s alboguttatus group.

Also in 1991, Myers named Colostethuslacrimosus and, based on several similarities(but no clear synapomorphies), speculatedthat it may be closely related to C. chocoensis.He also suggested that they, in turn, wererelated to C. fuliginosus.

In a series of papers in the 1990s, LaMarca proposed a number of novel relation-ships and taxonomic changes. In 1992 heformally named the collaris group as Manno-phryne and later (La Marca, 1995) presenteda hypothesis of relationships based on fivecharacters from morphology and behavior,shown here as figure 3.

Although this study purported to bea quantitative cladistic analysis, few charac-ters were used and some characters discussedby the author were ignored, not all characterswere scored based on observations (i.e., some

states were merely assumed), and monophylyand character polarity were assumed (i.e., nooutgroup species were included). In additionto the species originally placed in Manno-phryne, the genus currently includes M.caquetio, M. cordilleriana, M. larandina, andM. lamarcai (Mijares-Urrutia and Arends R.,1999).

In discussing the systematics of Colostethusmandelorum (about which he only concludedthat the species is not closely related to eitherMannophryne or the C. alboguttatus group),La Marca (1993) considered Aromobatesnocturnus to be most closely related to theC. alboguttatus group of La Marca (1985).Regardless, instead of transferring the albo-guttatus group into Aromobates, La Marca(1994) named it Nephelobates. The group wasdelimited by the occurrence of elongate teeth(also reported for Aromobates; see Myers etal., 1991; La Marca, 1993) and a dermal

Fig. 2. Hypothesized phylogeny of dendrobatids, redrawn from Myers et al. (1991: 29, fig. 20). Allevidence is shown on the cladogram. In this scenario, Aromobates nocturnus is postulated to be the sisterspecies of all other dendrobatids. All of the unquestioned synapomorphies listed for Dendrobatidae applyonly to A. nocturnus and are unknown in any other dendrobatid.


covering of the cloaca (also reported for theedwardsi group; Lynch, 1982), and it includedN. alboguttatus, N. haydeeae, N. leopardalis,N. mayorgai, N. meridensis, N. molinarii, andN. orostoma; Mijares-Urrutia and La Marca(1997) subsequently included N. duranti andN. serranus. La Marca (1994) did not includeC. saltuensis, which had been included inRivero’s alboguttatus group (Rivero, 1990‘‘1988’’), but he did not state his reasons forits exclusion. No explicit hypothesis ofrelationships has been proposed for thespecies of Nephelobates, but Mijares-Urrutiaand La Marca (1997) reported several larvalcharacter-states of unclear polarity, as well asthe occurrence of ‘‘reduced nasal bones’’(p. 134) as a synapomorphy for the genus.

Kaiser et al. (1994) described Colostethuschalcopis from Martinique in the FrenchAntilles. Although they were skeptical ofthe monophyly of Mannophryne, they con-jectured that C. chalcopis could be the sisterspecies to that assemblage on the basis of theshared occurrence of a dark throat collar.

Although Coloma (1995) did not intend toprovide a phylogenetic hypothesis for Colos-tethus, the taxonomic changes he made hadnumerous phylogenetic implications. Forexample, some of the species he consideredto be synonymous had been placed indifferent and presumably distantly relatedgroups by Rivero (e.g., Rivero and Almen-dariz, 1991, placed C. nexipus in the fuligino-sus group, whereas its junior synonym C.citreicola was placed in the subpunctatusgroup), which called into question thephylogenetic validity (or even taxonomicutility) of those groups. Coloma (1995: 58)also summarized the recognized speciesgroups of Colostethus, arguing that ‘‘mostof the character states given by Rivero [1990‘‘1988’’] and Rivero and Serna [1989 ‘‘1988’’]seem to be plesiomorphic at the level used.’’Although he concluded that ‘‘the phyloge-netic relationships within ‘Colostethus’ (sensulato) constitute an enormous polytomy’’(Coloma, 1995: 60), Coloma tentativelysupported the following relationships:

Fig. 3. Hypothesized phylogeny of Mannophryne, redrawn from La Marca (1995: 70, fig. 11).Synapomorphies are: A1, narrow collar, uniformly colored; B1, tadpoles with small papillae (presumablyB1 on the cladogram is B0 from the text on p. 53); B2, tadpoles with large papillae (B2 is undefined in thetext on p. 53; presumably it refers to B1); C1, uniformly colored dorsum; A2, wide collar withoutconspicuous pale markings; D1, posteroventral dark band present; A3, wide collar with pale flecks orspots; E1, bright throat coloration reduced, melanophores on anterior part of throat; A4, wide collar withlarge pale dots.


N Some species of Colostethus may be more

closely related to some species of Epipedobates

(based on the shared occurrence of a swollen

third finger in adult males) than to otherspecies of Colostethus.

N Species in Aromobates, Mannophryne, and thevertebralis and fuliginosus groups may be

basal within Colostethus.

N The edwardsi group is monophyletic.

N A novel group composed of Aromobates

nocturnus, Colostethus awa, C. bocagei, C.

nexipus, and C. riveroi may be monophyletic

on the basis of shared (albeit facultative)nocturnal behavior. Myers et al.’s (1991)

claim that nocturnal activity is plesiomorphicwas not addressed.

Grant et al. (1997) reviewed the distribu-tion of the median lingual process in den-drobatids and other frogs. The occurrence ofthe median lingual process in a putative sistergroup (see Interfamilial Relationships, be-low) led them to interpret it tentatively assymplesiomorphic and, consequently, theydid not use it to delimit a group withinDendrobatidae.

Kaplan (1997) followed Silverstone(1975a) in studying the distribution of thepalatine (neopalatine of Trueb, 1993), and heused these data to further resolve Myers etal.’s (1991) hypothesis of relationships (andhe explicitly incorporated Mannophryne andNephelobates). He concluded that the absenceof the palatine delimits a clade composed ofpart of Hyloxalus sensu Myers et al. (1991),Colostethus sensu stricto, and the aposematicdendrobatids. Kaplan (1997) presenteda cladogram, shown here as figure 4. Theseparate treatment of Epipedobates was dueto the presence of a swollen third finger insome species of that genus (Myers, 1991).

La Marca (1998 ‘‘1996’’) reviewed thespecies of Guayanan Colostethus and as-signed C. ayarzaguenai, C. guanayensis, C.murisipanensis, C. parimae, C. parkerae, C.praderioi, C. roraima, C. sanmartini, C.shrevei, and C. tepuyensis to the fuliginosusgroup sensu lato (i.e., sensu Rivero, 1990‘‘1988’’). He did not note the occurrence ofthe median lingual process, although it ispresent in several of these species.

Grant and Castro (1998) proposed theColostethus ramosi group based on theoccurrence of a patch of black, apparently

glandular tissue on the ventral and medialsurfaces of the distal extreme of the upperarm, just proximal to the elbow (referred toby Grant and Castro as the black arm band).This group presently includes C. cevallosi, C.exasperatus, C. fascianiger, C. lehmanni, C.ramosi, and C. saltuarius (Grant and Ardila-Robayo, 2002), but this character-state alsooccurs in C. anthracinus and an undescribedspecies from the slopes of the Magdalenavalley, Colombia (T. Grant, personal obs.).

Schulte’s (1999) book on Peruvian Den-drobates and Epipedobates included a numberof novel phylogenetic arrangements, many ofwhich involved non-Peruvian species as well.Lotters and Vences (2000) strongly criticizedmany of Schulte’s (1999) taxonomic conclu-sions, and below we exclude the nomina nudaand taxa they placed in synonymy. Schulte(1999) proposed eight groups of Dendrobatesand six groups of Epipedobates, and heprovided branching diagrams depicting therelationships of each (Schulte, 1999: 24–25,160–161). The groups he proposed are asfollows:

Dendrobates:

N Group 1 (amazonicus): D. amazonicus, D.

duellmani, D. fantasticus, D. variabilis

N Group 2 (quinquevittatus): D. quinquevittatus,D. castaneoticus, D. flavovittatus

N Group 3 (imitator): D. imitator

N Group 4 (vanzolinii): D. biolat, D. lamasi, D.

vanzolinii

N Group 5 (ventrimaculatus): D. ventrimaculatus

N Group 6: D. reticulatus, D. rubrocephalus, D.

sirensis, D. steyermarki, D. (M.) virolinensis

[sic]

N Group 7: D. captivus

N Group 8 (histrionicus): D. histrionicus, D.

lehmanni, D. mysteriosus

Epipedobates:

N Group 1 (giant types [‘‘Riesenarten’’]): E.

bassleri, E. planipaleae, E. silverstonei, E.

trivittatus

N Group 2 (petersi/pictus): petersi subgroup: E.

cainarachi, E. labialis, E. macero, E. petersi, E.

pongoensis, E. smaragdinus, E. zaparo; pictus

subgroup: E. bolivianus, E. hahneli, E. pictus,E. rubriventris

N Group 3 (azureiventris): E. azureiventris,Phyllobates [i.e., Phyllobates sensu Myers etal., 1978]


N Group 4 (femoralis): E. femoralis, E. ingeri, E.labialis, E. myersi, E. zaparo

N Group 5 (parvulus): E. espinosai, E. parvu-lus

N Group 6 (tricolor): E. anthonyi, E. espinosai,E. parvulus, E. subpunctatus, E. tricolor

Rather than detail exhaustively the rela-tionships Schulte (1999) proposed, we limitourselves to pointing out a few of his moreheterodox hypotheses. Without comment hetransferred Prostherapis subpunctatus fromColostethus (where it had been placed byEdwards, 1971) to Epipedobates as sister

species to E. anthonyi and E. tricolor. Alsowithout comment, he referred Dendrobatessteyermarki and Minyobates virolinensis—both of which had been in Minyobates(Myers, 1987; Ruiz-Carranza and Ramırez-Pinilla, 1992)—to Dendrobates, but he didnot discuss the relationships of the remainingspecies of Minyobates. Further, according tohis own diagrams he rendered Epipedobatesparaphyletic by grouping E. azureiventriswith species of Phyllobates. Schulte redefinedthe histrionicus group to include D. myster-iosus, but he excluded most of the species

Fig. 4. Hypothesized phylogeny of dendrobatids, redrawn from Kaplan (1997: 373, fig. 3). Numberedsynapomorphies are: (1) tympanum posterodorsally tilted under anterior edge of massive superficial slip ofm. depressor mandibulae, (2) mercaptanlike defensive odor, (3) diurnal activity, (4) riparian–terrestrialhabitat preference, (5) smaller size (,50 mm SVL), (6) m. adductor mandibulae externus superficialisabsent (‘‘s’’ pattern), (7) neopalatines absent, (8) finger three of males swollen, and (9) lipophilicalkaloids present.


Myers et al. (1984)—and even Silverstone(1975a) and Myers and Daly (1976b)—hadreferred to that group, and he once againplaced D. leucomelas in that group (as hadSilverstone, 1975a). Relationships amongmost groups were not specified, but somegroups (e.g., Groups 1 and 7) were para-phyletic in Schulte’s own diagrams, and somespecies (e.g., D. labialis and D. zaparo; E.parvulus and E. espinosai) were included inmultiple groups, with their relationships toeach other being different in each group. Nonew character systems were added in thisstudy, and, although Schulte provided limit-ed group diagnoses and details on naturalhistory, behavior, coloration and color pat-terns, and external morphology, no explicitsynapomorphies were provided for any of hisgroups.

Grant (1998) named Colostethus lynchi andargued that it was part of the C. edwardsigroup on the basis of the occurrence ofa cloacal tube (he did not address theoccurrence of this character in Nephelobates).More specifically, he argued that C. lynchiwas the sister species to the group of C.edwardsi + C. ruizi.

The first attempt to address phylogeneticrelationships among dendrobatids with DNAsequence data was published by Summers etal. (1997), although that paper only includedthe distantly related Dendrobates pumilio,

Dendrobates claudiae (as Minyobates sp.),and Phyllobates lugubris (plus C. talamancae,used as the root). Since 1999, nearly a dozenphylogenetic studies of differing scales,scopes, and datasets have appeared (Sum-mers et al., 1999b; Clough and Summers,2000; Vences et al., 2000, 2003a; Widmer etal., 2000; Symula et al., 2001, 2003; La Marcaet al., 2002; Santos et al., 2003). Thecladograms that resulted from those studiesare redrawn in figures 5–13. Interpretation ofthese studies is complicated by their use ofdifferent methods, nonoverlapping taxonsamples, and heterogeneous datasets, buttheir findings can be summarized as follows:

N Colostethus: Found to be either para- or

polyphyletic by all authors who tested its

monophyly.

N Epipedobates: Found to be monophyletic by

Clough and Summers (2000) (with femoralis

placed outside in Allobates) but polyphyletic

by Vences et al. (2000, 2003a; see also Santos

et al., 2003).

Fig. 5. Hypothesized phylogeny of Dendro-bates, redrawn from Summers et al. (1999b: 261,fig. 1), based on parsimony analysis of cytochromeoxidase I, cytochrome b, and 16S DNA sequencesaligned with Clustal W (Thompson et al., 1994)(parameters not specified) and modified manually,excluding ambiguously aligned regions. Numbersare bootstrap frequencies (unlabeled nodes presentin fewer than 50% of replicates).

Fig. 6. Hypothesized phylogeny of dendroba-tids, redrawn from Clough and Summers (2000:342, fig. 1), based on parsimony analysis of 12S,16S, and cytochrome b DNA sequences alignedwith Clustal W (Thompson et al., 1994) (param-eters not specified) and modified by eye andexcluding ambiguously aligned regions. Numbersare bootstrap frequencies (unlabeled nodes presentin fewer than 50% of replicates).


N Phobobates: Found to be monophyletic by

Vences et al. (2000) but paraphyletic by

Clough and Summers (2000), Santos et al.

(2003), and Vences et al. (2003a).

N Allobates: This small genus fell out in a clade

with species of Colostethus in Vences et al.

(2000, 2003a) and Santos (2003). (Jungfer and

Bohme, 2004 added the enigmatic Dendro-

bates rufulus to Allobates, but that species has

not been included in any analysis.)

N Phyllobates: Without exception, this genus

was found to be monophyletic. The optimal

topology found by Widmer et al. (2000) was

((vittatus lugubris) (aurotaenia (bicolor terribi-

lis))) (outgroup taxa were Epipedobates azur-

eiventris and Dendrobates sylvaticus, and the

tree was rooted on E. azureiventris). In their

more inclusive study, Vences et al. (2003a)

found P. aurotaenia to be the sister of the

remainder, and P. bicolor to be sister to the

Central American species, giving the topology

(aurotaenia (terribilis (bicolor (lugubris vitta-

tus)))).

N Minyobates: Both Clough and Summers

(2000) and Vences et al. (2000) found

Minyobates to be nested within Dendrobates.

Because each analysis used only one species of

Minyobates, they did not test the monophyly

Fig. 7. Hypothesized phylogeny of dendrobatids, redrawn from Vences et al. (2000: 37, fig. 1), basedon neighbor-joining analysis of 16S DNA sequences aligned manually and excluding highly variableregions. Numbers are bootstrap frequencies (unlabeled nodes present in fewer than 50% of replicates).


of Minyobates itself. Vences et al. (2003a)included M. steyermarki (type species), M.

minutus, and M. fulguritus and found Minyo-

bates to be paraphyletic with respect to allother Dendrobates. Santos et al. (2003) in-cluded M. minutus and M. fulguritus andfound them to be the monophyletic sister tothe D. quinquevittatus group (i.e., they re-covered a monophyletic minutus group sensuSilverstone, 1975b).

N The Dendrobates histrionicus group is mono-phyletic in all studies that test its monophyly.

N The Dendrobates quinquevittatus group ispotentially monophyletic. Although the treepresented by Clough and Summers (2000:524) indicates that Minyobates minutus is thesister species of a monophyletic D. quinque-

vittatus group, there is in fact no evidence tosupport this assertion, given that these nodescollapse in the strict consensus. Symula et al.(2003) found Dendrobates leucomelas to besister to part of the D. quinquevittatus group,with a D. quinquevittatus + D. castaneoticus

clade in a basal trichotomy (they rooted thenetwork with D. histrionicus, so it is unknownfrom their results if D. quinquevittatus + D.

castaneoticus or D. histrionicus is more closelyrelated to the D. leucomelas + other D.

quinquevittatus group clade).

N Nephelobates and Mannophryne were bothfound to be monophyletic by La Marca et al.(2002) and Vences et al. (2003a).

Lotters et al. (2000) erected the new genusCryptophyllobates for Phyllobates azureiven-tris (which was placed in Epipedobates byMyers and Burrowes, 1987). The justificationfor this monotypic genus is somewhat con-

voluted. On pp. 235–236, the authors statedthat ‘‘from the genetic point of view, it isapparent that azureiventris is more closelyrelated to Epipedobates than to Phyllobates’’,but that ‘‘the species is not a member ofEpipedobates, from which it differs by at leastone apomorphy.’’ However, they also as-serted that ‘‘[i]t shares more—but not all—characters with Phyllobates from which itappears genetically well separated.’’ Similar-ly, although Vences et al. (2000) found thisspecies to be the sister of Colostethus bocagei,Lotters et al. (2000) stated that they ‘‘negatethat both species are representatives of thesame genus for C. bocagei is crypticallycoloured, lacking dorsal stripes at all, andpossesses webbed feet.’’ Insofar as thischange did not fix the nonmonophyly ofEpipedobates, the creation of this monotypicgenus did little to improve matters. Recently,Caldwell (2005) described a new species andreferred it to this genus based on the sister-species relationship between the new speciesand azureiventris recovered in an independentphylogenetic study, despite noting that thesespecies ‘‘were nested in a clade of Ecuadorianand Peruvian Colostethus.’’

Morales (2002 ‘‘2000’’) combined Rivero’sGroups II (brunneus) and III (alagoanus) intoa newly defined trilineatus group (whichexcluded C. kingsburyi and C. peruvianus)

Fig. 8. Hypothesized phylogeny of dendroba-tids, redrawn from Widmer et al. (2000: 561, fig.2), based on parsimony analysis of cytochromeb sequences aligned with Clustal W (Thompson etal., 1994) (parameters not specified). Numbers areparsimony/maximum likelihood/neighbor-joiningbootstrap frequencies.

Fig. 9. Cladogram summarizing species levelphylogeny of Dendrobates (reduced from 30terminals) proposed by Symula et al. (2001: 2419,fig. 3), based on maximum likelihood analysis(under the GTR + C + I model) of cytochrome b,cytochrome oxidase I, 12S, and 16S DNAsequences aligned with ClustalX (Thompson etal., 1997) (parameters not specified) and ‘‘a fewregions of ambiguous alignment … removed fromthe analysis’’. Numbers are parsimony boot-strap frequencies.


on the basis of an analysis of 12 characters.However, in addition to problems of charac-ter individuation (e.g., characters 6, ‘‘lınealateral oblicua’’, and 10, ‘‘lista oblicua ante-roinguinal’’, are logically dependent; seeGrant and Rodrıguez, 2001), the monophylyof the group was assumed (the cladogramwas rooted on an unspecified ‘‘Hylodes’’ andno other dendrobatids were included), andthe putatively derived states of all 12characters are well known to occur in manydendrobatids.

In the most recent contribution to den-drobatid phylogenetics, Graham et al. (2004)added 12S, tRNAval, and 16S mtDNAsequences from a specimen collected nearthe type locality of Epipedobates tricolor andanalyzed them with the data from Santos etal. (2003). Graham et al. reported that the E.tricolor sample from southern Ecuador wasmore closely related to Colostethus machalillathan to true E. tricolor, and, as such, theyresurrected E. anthonyi from synonymy withE. tricolor. However, the Bremer supportvalue1 reported for the critical node is 0,meaning that this relationship is not re-covered in other, equally parsimonious solu-tions.

PART III: 1926–PRESENT, RELATIONSHIPS BE-

TWEEN DENDROBATIDAE AND OTHER FROGS

Noble (1931) summarized his research onthe evolutionary relationships of anurans. Heconsidered the three genera of dendrobatids,which he had grouped together in his earlierpaper (Noble, 1926), to be a subfamily ofBrachycephalidae. The other subfamilieswere Rhinodermatinae (Geobatrachus, Smin-thillus, and Rhinoderma) and Brachycephali-nae (Atelopus, Brachycephalus, Dendrophry-niscus, and Oreophrynella). Noble (1931: 505;see Grant et al., 1997: 31, footnote 18)maintained his curious view that indepen-dently derived groups may constitute naturalassemblages:

Each subfamily has arisen from a different

stock of bufonids, but as all the ancestral stocks

were bufonids residing in the same general

region, Brachycephalidae may be considered

natural, even though a composite, family.

Fig. 10. Hypothesized phylogeny of dendrobatids, redrawn from La Marca et al. (2002: 239, fig. 4),based on maximum likelihood analysis (model not specified) of 16S DNA sequences (alignment methodnot stated), excluding hypervariable regions. Numbers are maximum likelihood/parsimony/neighbor-joining bootstrap frequencies.

1Graham et al. (2004) did not define the numberson the nodes in their cladogram, but C. H. Graham(personal commun.) informed us that they arebootstrap frequencies (above) and Bremer values(below).


Particularly, Noble (1931; see also Noble,1926) reiterated that Dendrobatinae evolvedfrom the elosiine bufonid genus Crossodacty-lus. Brachycephalidae was included in thesuborder Procoela, which also containedBufonidae and Hylidae, as well as the extinctPalaeobatrachidae.

Noble was aware that his placement ofBrachycephalidae in Procoela instead ofDiplasiocoela could be viewed as problemat-ic. He (Noble, 1931: 514) pointed out that thefrogs he referred to Diplasiocoela ‘‘differstrikingly from most other Salientia exceptBrachycephalidae’’, but he reasoned that‘‘[t]he latter are purely neotropical, and asthe genera of Brachycephalidae are welldefined, they should not be confused withthe Diplasiocoela.’’ He also observed thatboth Dendrobatinae and the African ranidPetropedetinae (nested well within Diplasio-coela) had ‘‘apparently identical’’ (Noble,1931: 520) dermal scutes on the upper surfaceof each digit, but he explained away thissimilarity as adding ‘‘one more to the manycases of parallel evolution in the Salientia’’.

Although Noble’s general scheme waswidely accepted as the standard for decades(e.g., Dunlap, 1960), it attracted extensivecriticism almost immediately. For example,Trewavas (1933: 517) concluded that thehyolaryngeal apparatus provided ‘‘little sup-port for the inclusion of Dendrobates in thefamily [Brachycephalidae]’’ and recom-mended that the relationships of the familybe reconsidered. Davis (1935: 91) criticizedNoble’s belief that independently derivedtaxa could be grouped naturally, and heraised each of Noble’s (1931) brachycephalidsubfamilies (i.e., Brachycephalinae, Dendro-batinae, and Rhinodermatinae) to familyrank. Laurent (1942: 18; translated, italicsas in original) concluded that the similaritiesin the initial phases of parental care of larvaein dendrobatids (tadpoles are transported onthe male’s back) and rhinodermatids (tad-poles are transported in the male’s mouth)‘‘constituted a weighty argument in favor ofthe common ancestry of the Rhinodermatinaeand the Dendrobatinae’’, and he includedboth in Dendrobatidae. Orton (1957; see alsoOrton, 1953) was highly critical of Noble’ssystem because it conflicted with larvalmorphology; but, beyond her rejection of

suborder rank within Salientia, dendrobatidswere unaffected. Likewise, Reig (1958) in-corporated evidence from a variety of pre-vious studies (e.g., Trewavas, 1933; Davis,1935; Walker, 1938; Taylor, 1951; Griffiths,1954) and his own fossil work to provide anextensively modified higher taxonomy, butthe placement of Dendrobatidae was un-affected (i.e., Reig’s neobatrachian ‘‘Super-family A’’ [now Hyloidea, 5 Bufonoideaauctorum] was identical to Noble’s Procoelawith the exclusion of Palaeobatrachidae).

Griffiths (1959, 1963) provided the firstmajor challenge to Noble’s placement ofDendrobatidae. Griffiths (1959) reviewedNoble’s (1922, 1926, 1931) evidence thatdendrobatids were part of Procoela andrelated to Crossodactylus, and, arguing that(1) ‘‘vertebral pattern has not the exacttaxonomic validity vested in it by Noble’’(p. 481); (2) path of insertion of the m.semitendinosus is ranoid in Hyloxalus; (3)‘‘Noble’s claim that Phyllobates has anarciferal stage cannot be held’’ (p. 482); (4)the bursa angularis oris is found only infirmisternal genera; (5) dermal scutes (whichhe claimed to be ‘‘glandulo-muscular or-gans’’) on the digits occur in petropedetidranids (as well as Crossodactylus); and (6) thebreeding habits of dendrobatids ‘‘are foundin no other Salientia except in the arthrolep-tid ranids’’ (p. 483), proposed ‘‘that theDendrobatinae be redefined as a Neotropicalsubfamily of the Ranidae’’ (p. 483). Sub-sequent reviews either explicitly endorsed(e.g., Hecht, 1963: 31) or did not address(e.g., Tihen, 1965; Inger, 1967; Kluge andFarris, 1969) Griffiths’s hypothesis of therelationships of dendrobatids.

However, Lynch (1971: 164; see alsoLynch, 1973) supported Noble’s hypothesis,arguing that elosiines (including Crossodac-tylus) and dendrobatids ‘‘agree in cranialmorphology, vertebral columns, the T-shaped terminal phalanges, the dermal glan-dular pads on top of the digital pads, and inthe presence, in at least some species of eachgroup, of toxic skin secretions’’ (see fig. 14).

Lynch (1971: 164) also asserted that Cross-odactylus and dendrobatids exhibit the ra-noid pattern of thigh musculature, which‘‘mitigates the importance of one of thecriteria used by Griffiths (1963) to associate


Fig. 11. Hypothesized phylogeny of dendrobatids, redrawn from Santos et al. (2003: 12794, fig. 1),based on unweighted parsimony analysis of the mitochondrial transcription unit H1 (ca. 2,400 bp), alignedwith ClustalX (Thompson et al., 1997) ‘‘under various parameters … and finally adjusted by eye toproduce a parsimonious alignment’’ whereby ‘‘informative sites were minimized’’ (Santos et al., 2003:


the dendrobatids as a Neotropical subfamilythe Ranidae.’’ Interestingly, Lynch (1971:210–211) also indicated that there was‘‘considerable similarity in myology andosteology’’ between the Neotropical lepto-dactyloid Elosiinae and Dendrobatidae andthe African ranid subfamily Petropedetinae.Further, although he cautioned that hisexamination of the African taxa was notexhaustive, he stated that ‘‘[t]he similaritiesare quite striking and probably reflecta community of ancestry rather than paral-lelism.’’

Lynch’s (1971, 1973) resurrected version ofNoble’s (1926) hypothesis stood for 15 years.For example, although Savage (1973)adopted Starrett’s (1973) scheme of higherlevel relationships and did not discuss den-drobatid phylogeny per se, he followedLynch (1971) in considering Dendrobatidaeto be a South American, tropical, leptodac-tyloid derivative. Bogart (1973: 348) conjec-tured that ‘‘Dendrobatidae may be derivedchromosomally from a 26-chromosome an-cestor, such as the leptodactylid Elosia’’(although he did not examine any African

r

12793). Parsimony bootstrap frequencies shown above branches, frequency of clades among trees sampledin Bayesian analysis shown below branches. Stars indicate clades found in $95% of the replicates orsampled trees.

Fig. 12. Hypothesized phylogeny of Dendrobates, redrawn from Symula et al. (2003: 459, fig. 3), basedon maximum likelihood (under the GTR + C model) analysis of cytochrome b and cytochrome oxidase IDNA sequences aligned with ClustalX (Thompson et al., 1997) (parameters not specified). Maximumlikelihood branch lengths shown above branches, parsimony bootstrap frequencies shown below branches(frequencies .75% shown).


ranoid species for comparison). Duellman(1975) included Dendrobatidae in Bufonoi-dea (though not explicitly with Crossodacty-lus). Ardila-Robayo (1979) evaluated 68characters and found two equally parsimo-nious topologies, both of which showedDendrobatidae (‘‘Phyllobatinae’’; see Du-

bois, 1982, and Holthius and Dubois, 1983,for discussion of nomenclature) to be relatedto elosiines. Like Duellman (1975), Laurent(‘‘1979’’ 1980) and Dubois (1984) did notaddress dendrobatid relationships specifical-ly, but they included Dendrobatidae inBufonoidea (except that the latter replaced

r

Fig. 13. Phylogeny of dendrobatids redrawn from Vences et al. (2003a: 219, fig. 3), based on maximumlikelihood analysis (under the GTR + I + C model) of 368 bp of 16S rDNA aligned ‘‘using the clustaloption of Sequence Navigator (Applied Biosystems)’’ (parameters not specified) and excluding ‘‘all regionsof the gene fragments that could not be clearly aligned among all taxa’’ (p. 217). Numbers above branchesare neighbor-joining bootstrap frequencies (frequencies .50% shown); numbers below branches arefrequency of clades among trees sampled in Bayesian analysis (frequencies .70% shown).

Fig. 14. Hypothesized phylogeny of anurans, redrawn from Lynch (1973).


Bufonoidea with the senior synonym Hyloi-dea).

Both the ranoid and hyloid hypotheseshave suffered from mistaken observations.Against Griffiths (1959), Kaplan (1997)confirmed Noble’s (1926) claim that thepectoral girdle of Colostethus subpunctatusoverlaps in ontogeny (which had been deniedby Griffiths), and Silverstone (1975a) andGrant et al. (1997) showed that Griffiths’sclaims regarding dendrobatid thigh muscula-ture were also false. Against Lynch (1971),

the thigh musculature in hylodines conformswith Noble’s (1922) ‘‘bufonid’’ pattern, notthe dendrobatid pattern (Grant et al., 1997:31; see also Dunlap, 1960), and no species ofhylodine tested by Myers and Daly wasfound to possess lipophilic alkaloids (Grantet al., 1997).

Fifteen years after Lynch (1971) resur-rected the hyloid hypothesis, Duellman andTrueb (1986) resurrected the ranoid one.Based on a cladistic analysis of 16 characters,they placed Dendrobatidae in a ranoid polyt-omy, unrelated to South American leptodac-tylids. Myers and Ford (1986) did notaddress the phylogenetic position of dendro-batids, but they listed a number of diagnosticcharacter-states for Dendrobatidae, including(1) the posterodorsal portion of the tympa-num concealed beneath the massive superfi-cial slip of the m. depresssor mandibulae, (2)the alary processs of the premaxilla tiltedanterolaterally, (3) occurrence of a retroartic-ular process on the mandible, (4) absence ofm. adductor mandibulae externus, (5) singleanterior process on hyale, (6) the occurrenceof digital scutes, and (7) the m. semitendino-sus tendon of insertion piercing the tendon ofthe m. gracilis.

Shortly thereafter, Ford (1989) completedher doctoral dissertation on the phylogeneticposition of Dendrobatidae, based on 124osteological characters, and found that themost parsimonious solution placed Dendro-batidae as the sister taxon of the Old Worldranoid family Arthroleptidae. That disserta-tion remains unpublished, but it was sum-marized by Ford and Cannatella (1993; seealso Ford, 1993). They reiterated the den-drobatid synapomorphies given by Myersand Ford (1986) and cited Ford’s dissertationas finding that ‘‘dendrobatids were nestedwithin Ranoidea, close to arthroleptid andpetropedetine ranoids’’ (p. 113; see fig. 15),but they did not list any synapomorphies insupport of that hypothesis.

The phylogenetic position of Dendrobati-dae alternated between the ranoid and hyloidhypotheses through the 1990s. Bogart (1991:251–252) compared karyotypes, averagemeasurements, and idiograms of severalspecies of petropedetids and hylodines withdendrobatids and concluded that ‘‘Hylodesand other hylodine leptodactylids have the

Fig. 15. Hypothesized phylogeny of neobatra-chians, redrawn from Ford and Cannatella (1993).


more similar karyotypes to the dendrobatidfrogs.’’ Blommers-Schlosser’s (1993) rede-fined Ranoidea excluded Dendrobatidae,but she still considered Dendrobatidae to bepart of the more inclusive ‘‘firmisternalfrogs’’ group, which is equivalent to Ranoi-dea sensu lato. However, Blommers-Schlos-ser (1993) also proposed the novel hypothesisthat Dendrobatidae is most closely related tomicrohylids, brevicipitids, and hemisotids inher Microhyloidea group. Hillis et al.’s (1993)combined analysis of the morphological datafrom Duellman and Trueb (1986) and theirown 28S rDNA sequence data indicated thatthe hyloid hypothesis was more parsimoniousthan the ranoid hypothesis. Hedges andMaxson’s (1993) neighbor-joining analysisof 12S mitchondrial DNA (mtDNA) se-quences also placed dendrobatids amonghyloids, as did Hay et al.’s (1995) andRuvinsky and Maxson’s (1996) neighbor-joining analyses 12S and 16S mtDNA data.Haas (1995) described an additional dendro-batid synapomorphy (viz., proximal ends ofCertatobranchialia II and III free, lackingsynchondritic attachment). He failed to findevidence of a ranoid relationship, but discov-ered a number of character-states shared withhyloid taxa; however, these characters are ofuncertain polarity, and no hylodine wasincluded to rigorously test Noble’s hypothe-sis. Grant et al. (1997) discovered thata median lingual process occurs in manyOld World ranoid genera (including thosethought to be most closely related todendrobatids) and several species of dendro-batids but failed to detect it in any nonranoidfrog. Burton (1998a) reported a synapomor-phy in the musculature of the hand (absenceof caput profundum arising from carpals) inDendrobatidae, Hylodes, and Megaelosia,but not the putative ranoid relatives (butnote that this state also occurs in part or allof Ascaphidae, Bombinatoridae, Discoglossi-dae, Heleophrynidae, Hemisotidae, Pipidae,and Sooglosidae).

Additional support for the ranoid hypoth-esis has not been proposed, as most studiesthis decade have found dendrobatids to benested among hyloids, if not directly relatedto hylodines. Vences et al.’s (2000) analysis of12S and 16S mtDNA sequence data showedDendrobatidae to group with hyloids, not

ranoids, as did Emerson et al.’s (2000)analysis of 12S, tRNAval, and 16S mtDNAdata (although the latter authors also foundDendrobatidae to be polyphyletic, broken upby Bufo valliceps and Atelopus chiriquiensis).Haas’s (2001) study of the mandibular archmusculature of larval and postmetamorphicamphibians included Phyllobates bicolor,which was found to possess the neobatra-chian (plus pelobatid) synapomorphy (viz.,presence of functionally differentiated m.levator mandibulae lateralis) and lack thethree ranoid synapomorphies, hence leavingit in a ‘‘hyloid’’ polytomy. In an explicitcladistic analysis, Haas (2003) assembleda dataset composed of mostly larval char-acters (but including most traditionally im-portant characters from adult morphologyand behavior) and found Dendrobatidae tobe sister to his two hylodine species (fig. 16).Vences et al. (2003a) also included two speciesof hylodines in their analysis of 12S and 16SmtDNA sequences, but they found dendro-batids to be sister to Telmatobius simonsi.Darst and Cannatella (2004) analyzed 12S,16S, and tRNAval mtDNA sequences andfound dendrobatids to be nested withinHylidae (parsimony result shown in fig. 17)or sister to a group consisting of cerato-phryines, hemiphractines, and telmatobiines(maximum likelihood; topology not shown).Faivovich et al. (2005) were primarily in-terested in the relationships among hylids, buttheir outgroup sample was extensive; theiranalysis of nuclear and mitochondrial DNAplaced dendrobatids as the sister group to theHylodinae (topology not shown).

SUMMARY OF HISTORICAL REVIEW

The picture that emerges from the reviewof the history of dendrobatid systematics isone of considerable conflict and confusion.There is near universal support for themonophyly of the family, which has notbeen seriously challenged since it was firstproposed by Noble 90 years ago, but thephylogenetic position of Dendrobatidae hasalternated between two predominant hy-potheses: (1) deeply embedded among ra-noids as the sister to petropedetids orarthroleptids or (2) deeply embedded among


hyloids as the sister to hylodines. Recentstudies based on DNA sequences (mostlymtDNA) have favored the hyloid hypothe-sis, but there is extensive conflict in thedetails of both hypotheses. Within Dendro-batidae, the once uncontroversial monophy-ly of the aposematic taxa has been rejectedby mtDNA studies, and there is littleagreement on the monophyly and relation-ships among most genera. The monophyly ofPhyllobates has been universally supported,although the relationships among its fivespecies have not. To date, no study hascombined DNA sequences with evidencefrom morphological, behavioral, and bio-chemical (alkaloid) sources, and all explicitphylogenetic analyses have included a limitedsample of the diversity of dendrobatids.

PHYLOGENETIC PLACEMENT OFDENDROBATIDS AND

OUTGROUP SAMPLING

THEORETICAL BACKGROUND

Although the present study was notdesigned primarily to test the relationshipsbetween dendrobatids and other anurans,that question is key to selecting an adequatesample of outgroup taxa to rigorously testthe relationships (including monophyly) andtransformation series among dendrobatids.That is, the position taken in this study is thatall nondendrobatids constitute ‘‘the out-group,’’ and outgroup taxa are sampled forthe purpose of testing hypothesized patristicand cladistic relationships. Ideally, one wouldcode all nondendrobatids for all includedcharacters; however, given the practicalimpossibility of that ideal, prior knowledgeof phylogeny and character variation must beused to inform sampling of those taxa mostlikely to falsify ingroup hypotheses (includingingroup monophyly), the scope and scale ofoutgroup sampling being limited primarily bypractical limitations of time and resources

(e.g., specimen and tissue availability, labo-ratory resources, computer power and time).The possibility always exists that expansionof the outgroup sample may lead to im-proved phylogenetic explanations—a consid-eration that points the way to increasedtesting in future research cycles.

Although this approach to outgroup sam-pling incorporates prior knowledge, it doesso in an expressly non-Bayesian way. Theeffect of prior knowledge in Bayesian ap-proaches is to constrain hypothesis prefer-ence toward prior beliefs about ingroupevolution. Here, prior knowledge is usedheuristically to maximize the probability offalsifying prior beliefs about ingroup evolu-tion (for discussion of heurism in phyloge-netic inference see Grant and Kluge, 2003).That this ‘‘probability’’ is not frequentist,logical, or personal and is not formallyquantifiable does not deny its relevance.The goal is to test phylogenetic hypothesesas severely as possible, and prior knowledgeis key to that undertaking.

EMPIRICAL BACKGROUND

As summarized above, the phylogeneticplacement of Dendrobatidae is among themost controversial problems in anuran sys-tematics. In part, this is because the twocladistic hypotheses that have emerged as theleading contenders are so radically contra-dictory, effectively placing dendrobatids atopposite extremes of the neobatrachianclade: dendrobatids are placed as sister tohylodine hyloids from South America or areallied to petropedetid or arthroleptid ranoidsfrom Africa. Minimally, evaluation of thesehypotheses would require a phylogeneticanalysis of Neobatrachia, which was beyondthe scope of the present study.

Nevertheless, in a recently completed studyFrost et al. (2006) sampled 532 terminals forthe mitochondrial H-strand transcription

r

Fig. 16. Hypothesized phylogeny of anurans, redrawn from Haas (2003), based on parsimony analysisof larval and adult morphology.


unit 1 (H1; composed of 12S ribosomal,tRNAval, and 16S ribosomal sequences),histone H3 (H3), tyrosinase, rhodopsin,seventh in absentia (SIA), 28S large ribo-somal subunit, and Haas’s (2003) morpho-logical transformation series in a phylogeneticanalysis of living amphibians. That studyincluded approximately 9% of each of the‘‘major’’ amphibian clades (caecilians, sala-manders, and frogs), including eight species(and genera) of dendrobatids and all putativesister groups. Insofar as that study is themost extensive analysis of amphibian phy-logeny undertaken to date, we used thoseresults to inform outgroup sampling for thecurrent study.

The Frost et al. (2006) study resulted infour trees of 126,929 steps, the relevantportion of which is shown in figure 18.

Relevant to the present study, Frost et al.(2006) corroborated the monophyly of Den-drobatidae. Furthermore, dendrobatids werenot found to be closely related to petropede-tids, arthroleptids, or any other ranoid andwere instead nested deeply among hyloidtaxa. Specifically, Dendrobatidae was foundto be sister to Thoropa, those taxa were sisterto Bufonidae, and that inclusive clade wassister to Cycloramphidae (including Cross-odactylus, Hylodes, and Megaelosia as thesister clade of the remaining cycloramphids).Alternative hypotheses of the placement ofDendrobatidae (e.g., placed in a clade withCrossodactylus, Hylodes, and Megaelosia, asfavored by Noble, 1926; Lynch, 1971; Haas,2003; Faivovich, 2005) were tested explicitlyby inputting constraint topologies for di-agnosis and swapping, but they all requiredadditional transformations (breaking up theThoropa + Dendrobatidae clade required atleast 39 extra steps).

Although detailed knowledge of the place-ment of Thoropa did not exist prior to thatanalysis, its placement as the sister ofDendrobatidae is unconventional, to say theleast. No morphological synapomorphieshave been proposed to unite these taxa, andit was expected that Thoropa would be nestedamong cycloramphids. Nevertheless, insofaras this is the most parsimonious solutionfound in the most complete study of am-phibian relationships carried out to date, theFrost et al. (2006) hypothesis provides thestarting point for further testing. Also, theimmediately relevant nodes of the Frost et al.(2006) tree are all well supported (Bremersupport for Dendrobatidae + Thoropa 5 39,Dendrobatidae + Thoropa + Bufonidae 5

30); considering that Thoropa was onlyscored for the mtDNA and H3 loci (i.e.,over 1,500 bp of nuDNA were missing), theBremer value for the Thoropa + Dendroba-tidae clade is remarkably high. Furthermore,the general placement of Dendrobatidaeis reminiscent of (but not identical to)Noble’s (1922) Brachycephalidae, which in-cluded the dendrobatids, Brachycephalus,Atelopus, Rhinoderma, Sminthillus (now a syn-onym of Eleutherodactylus), Geobatrachus,and Oreophrynella (the latter two genera notsampled by Frost et al.). According to Frostet al. (2006), Brachycephalus and Eleuther-odactylus are part of the distantly relatedBrachycephalidae (not shown in fig. 18), butAtelopus and Rhinoderma are placed in thesame general neighborhood as Dendrobati-dae. As such, the results of Frost et al. (2006)provide both an objectively defensible andsubjectively ‘‘reasonable’’ basis for outgroupsampling, and we therefore sampled out-group taxa from among these closely relatedgroups.

r

Fig. 17. Hypothesized phylogeny of anurans, redrawn from Darst and Cannatella (2004: 465, fig. 1),based on parsimony analysis of mitochondrial transcription unit H1 (ca. 2,400 bp), aligned with ClustalX(Thompson et al., 1997) ‘‘under a variety of gap penalty weightings’’, adjusted manually ‘‘to minimizeinformative sites under the parsimony criterion’’, using secondary structure models to help alignambiguous regions, and excluding regions ‘‘for which homology of the sites could not be inferred’’ (Darstand Cannatella, 2004: 463). Numbers are bootstrap frequencies.


OUTGROUP SAMPLING

In light of Frost et al.’s (2006) findings, it isclear that dendrobatids are not closely relatedto the Old World ranoids and are insteadnested among New World hyloids. Despitethe relatively high support for the relevantnodes, the actual sister-group relationship ofDendrobatidae remains controversial, andthe present study aimed to further test thistopology by including relevant morphologi-cal characters, additional molecular data,and additional taxa. Especially relevant isthe large amount of missing data for Thoropaand the relatively low Bremer support (BS)for the monophyly of Cycloramphidae (BS 5

9) and several of the cycloramphid nodes(BS as low as 4). With that in mind, wetargeted the following 46 athesphatanuranoutgroup taxa: Allophryne ruthveni, Alsodesgargola, Atelognathus patagonicus, Atelopusspurrelli, Atelopus zeteki, Batrachyla leptopus,Centrolene geckoideum, Centrolene prosoble-pon, Ceratophrys cranwelli, Chacophrys pier-ottii, Cochranella bejaranoi, Crossodactylusschmidti, Cycloramphus boraceiencis, Dendro-phryniscus minutus, Edalorhina perezi, Eupso-phus calcaratus, Hyalinobatrachium fleisch-manni, Hypsiboas boans, Hyla cinerea,Osteocephalus taurinus, Hylodes phyllodes,Hylorina sylvatica, Lepidobatrachus laevis,Leptodactylus fuscus, Leptodactylus discodac-tylus, Leptodactylus hylaedactyla, Leptodac-tylus lineatus, Leptodactylus ocellatus, Limno-medusa macroglossa, Megaelosia goeldii,Melanophryniscus klappenbachi, Odontophry-nus achalensis, Odontophrynus americanus,Paratelmatobius sp., Physalaemus gracilis,Pleurodema brachyops, Proceratophrys aveli-noi, Pseudopaludicola falcipes, Rhaebo gutta-tus, Rhaebo haematiticus, Rhinoderma darwi-

nii, Scythrophrys sawayae, Telmatobiusjahuira, Telmatobius marmoratus, Telmato-bius sp., and Thoropa miliaris. Hypsiboasboans was designated as the root for analyses.

All but one of these species were the sameones used by Frost et al. (2006), the exceptionbeing Atelopus spurrelli, which we includedbecause (1) sequences proved difficult togenerate for the Atelopus zeteki tissue, soadding an additional species was necessary toensure full coverage of molecular data, and(2) adequate whole specimens of thisChocoan endemic were available at AMNHto allow morphological study.

We included all molecular data from theFrost et al. (2006) analysis for these term-inals. To these we added phenotyic charactersand sequences for cytochrome oxidase csubunit I, cytochrome b, recombinationactivating gene 1, and several fragments thatwere missing from Frost et al. (2006) for 12of those terminals (marked with an asteriskin fig. 18): Atelopus spurrelli, Atelopus zeteki,Crossodactylus schmidti, Cycloramphus bor-aceiencis, Dendrophryniscus minutus, Eupso-phus calcaratus, Hylodes phyllodes, Megaelo-sia goeldii, Melanophryniscus klappenbachi,Rhinoderma darwinii, Telmatobius jahuira,Thoropa miliaris. These terminals were tar-geted for increased sampling because of theirphylogenetic proximity to Dendrobatidae,phenotypic affinities, and the availability ofwhole specimens and other data (e.g., behav-ior, alkaloid profiles) to score phenotypiccharacters.

As discussed in greater detail below, witha single exception character-states werecoded for each ingroup species and werenot extrapolated from other species (e.g., wedid not assume that all species of Colos-tethus lack lipophilic alkaloids and instead

r

Fig. 18. Phylogenetic placement of Dendrobatidae according to Frost et al. (2006). The Frost et al.study sampled 532 terminals (direct optimization parsimony analysis under equal costs for alltransformations; see text for details of dataset), 51 of which are included here to show the placement ofDendrobatidae with respect to its closest relatives. All of the terminals shown were included in the presentstudy, including three representatives of Hylidae and two ‘‘other bufonids’’. Those targeted for additionalgenotypic and phenotypic evidence are marked in the figure with an asterisk (see text for details). Numbersare Bremer support values.


only coded species that have been examinedfor alkaloids); however, we relaxed thatrequirement to incorporate additional in-formation for outgroup taxa. Specifically,for Crossodactylus alkaloid data were de-rived from Crossodactylus sp. from Tereso-polis, Rıo de Janeiro, Brazil (Flier et al.,1980; Grant et al., 1997; J. W. Daly, in litt.09/15/00), chromosome number was assumedto be the same as in the other five speciesthat have been karyotyped (Aguiar et al.,2004), and all other data were coded fromCrossodactylus schmidti. For Cycloramphus,most data were taken from Cycloramphusboraceiensis, but osteological data weretaken from Cycloramphus fuliginosus. ForEupsophus, DNA sequences and larval datawere taken from Eupsophus calcaratus,whereas all other data were taken forEupsophus roseus (for which material wasavailable at AMNH); see Nunez et al. (1999)for discussion of the identity of these twospecies. For Hylodes, most data wereobtained from Hylodes phyllodes, but oste-ology was coded from Hylodes nasus. Final-ly, for Melanophryniscus, DNA sequenceswere taken from Melanophryniscus klappen-bachi, whereas all other data were scoredfrom Melanophryniscus stelzneri (which isbetter known and adequately represented atAMNH). Chromosome data were not avail-able for Megalosia goeldii, and there isvariation in chromosome number withinthe genus (Rosa et al., 2003). Insofar asthere is no clear empirical evidence to allyMegalosia goeldii with any of the threespecies for which chromosome data havebeen reported, we coded Megalosia goeldii aspolymorphic. The osteological data reportedfor Thoropa miliaris were taken from Thor-opa lutzi. We assumed that Telmatobiusjahuira has the same chromosome numberas reported for all other species in the genus(Kuramoto, 1990). All other outgroup datawere taken from single species.

MATERIALS AND METHODS

CONVENTIONS AND ABBREVIATIONS

One of the goals of this study is to proposea monophyletic taxonomy that represents thephylogeny of dendrobatids. The inadequacy

of the current taxonomy is widely recognized,and although the general scheme remainsthat of Myers (1987) and Myers et al. (1991),the recent application of Bauer’s overlookedgeneric names (e.g., Ameerega), the recogni-tion (as well as continued rejection) ofZimmermann and Zimmermann names(e.g., Allobates), the rejection (as well ascontinued recognition) of Minyobates, andthe proposal of a new name (Cryptophyllo-bates) all indicate that dendrobatid taxonomyis currently in flux with no universallyaccepted standard around which to structurediscussion of dendrobatid diversity. To avoidconfusion due to disagreements between thecurrent taxonomies and our proposal fora monophyletic taxonomy, we use binom-inals only in the introductory sections, above,and after proposing the new taxonomy.Elsewhere (e.g., in the character descriptions)we refer to species using only their trivialnames (e.g., fraterdanieli). Currently, 247nominal species of dendrobatids are recog-nized, very few of which have the same trivialnames. Where giving only the trivial namewould engender confusion, we include theauthor, e.g., sylvaticus Barbour and Nobleversus sylvaticus Funkhouser. All species-group names and their original, approximatecurrent, and proposed placements are listedin appendix 1. We also include several speciesthat are undescribed or of unclear identity.For simplicity, we refer to these species bytheir localities as informal names withinquotes (e.g., ‘‘Magdalena’’ in reference toan undescribed species from the Magdalenavalley in Colombia, ‘‘Curua-Una’’ for anundescribed species from the Rio Curua-Unain Brazil) or as they have been reported in theliterature (e.g., Nephelobates sp.). We alsoassigned unique sample identification num-bers to all tissues used in this study, and thesenumbers are reported as unique terminalidentifiers.

Commands used in computer programsare italicized. Sequences incorporated fromGenBank are listed in appendix 4. Data fortissues (including GenBank numbers forDNA sequences) and specimens examinedare listed in appendices 5 and 6, respectively,referenced with the permanent collectionnumber for the voucher specimen or, if thatis unavailable, the tissue collection number,


as follows: AMCC (Ambrose Monell CryoCollection, American Museum of NaturalHistory, New York, USA), AMNH (Amer-ican Museum of Natural History, New York,USA), ARA (Andres Acosta field series;specimens at MUJ), BB (Boris Blotto fieldseries), BMNH (The Natural History Muse-um, London, UK), BPN (Brice P. Noonanfield series; specimens at UTA), CFBH (CelioF. B. Haddad specimen collection, Brazil),CFBH-T (Celio F. B. Haddad tissue collec-tion, Brazil), CH (Coleccion Herpetologica,Panama), CPI (D. Bruce Means field series,to be deposited at USNM), CWM (CharlesW. Myers field series), GB (Godfrey Bournefield series), IAvH (Instituto de Investigacionde Recursos Biologicos Alexander von Hum-boldt, Villa de Leyva, Colombia), ICN(Instituto de Ciencias Naturales, UniversidadNacional de Colombia, Bogota, Colombia),IRSNB-KBIN (Institut Royal des SciencesNaturelles de Belgique/Koninklijk BelgischInstituut voor Natuurwetenschappen, Brus-sels, Belgium), IZUA (Instituto de Zoologıa,Universidad Austral de Chile, Valdivia,Chile), JAC (Jonathan A. Campbell fieldseries), JDL (John D. Lynch field sereis), JF(Julian Faivovich field series), KRL (KarenR. Lips field series), KU (University ofKansas Natural History Museum, Lawrence,USA), LACM (Natural History Museum ofLos Angeles County, Los Angeles, USA), LR(Lily Rodriguez field series), LSUMZ(Louisiana State University Museum ofNatural Science, Baton Rouge, USA),MACN (Museo Argentino de Ciencias Nat-urales Bernardino Rivadavia, Buenos Aires,Argentina), MAD (Maureen A. Donnellyfield series), MAR (Marco Antonio Rada;specimens at MUJ), MB (Marcus Breececaptive collection), MJH (Martin J. Henzlfield series), MLPA (Museo de la Plata,Buenos Aires, Argentina), MHNUC (Museode Historia Natural Universidad del Cauca,Popyan, Colombia), MPEG (Museu Para-ense Emilio Goeldi, Belem, Brazil), MRT(Miguel Rodrigues tissue collection), MUJ(Museo de Historia Natural, UniversidadJaveriana, Bogota, Colombia), MVZ (Muse-um of Vertebrate Zoology, University ofCalifornia at Berkeley, USA), MZUSP(Museu de Zoologia da Universidade deSao Paulo, Brazil), NK (Museo de Historia

Natural Noel Kempff Mercado, Santa Cruz,Bolivia), OMNH (Sam Noble OklahomaMuseum of Natural History, The Universityof Oklahoman, USA), PK (Philippe Kokfield series; specimens at IRSNB-KBIN),RDS (Rafael de Sa tissue collection), RG(Ron Gagliardo, Atlanta Botanical Garden),ROM (Royal Ontario Museum, Toronto,Canada), RWM (Roy W. McDiarmid fieldseries), SIUC (Southern Illinois University atCarbondale, USA), UMFS (University ofMichigan Museum of Zoology, Ann Arbor,Michigan, USA, field series), UMMZ (Uni-versity of Michigan Museum of Zoology,Ann Arbor, Michigan, USA), USNM (Na-tional Museum of Natural History, Smithso-nian Institution, Washington DC, USA),USNM-FS (National Museum of NaturalHistory, Smithsonian Institution, Washing-ton DC, USA, field series), UTA (Universityof Texas at Arlington, USA), UVC (Uni-versidad del Valle, Cali, Colombia), WES(Walter E. Schargel field series), ZSM(Zoologisches Museum, Munchen, Ger-many). Phenotypic data are given in appen-dix 7; formatted files for all data may bedownloaded from http://research.amnh.org/herpetology/downloads.htm.

GENERAL ANALYTICAL APPROACH:THEORETICAL CONSIDERATIONS

CHOICE OF PHYLOGENETIC METHOD

The general goal of phylogenetic system-atics is to explain the diversity of life bydiscovering the evolutionary relationshipsamong species, where inferred transforma-tions from one character-state to anotherprovide the means to choose among compet-ing explanations. That is, phylogenetic hy-potheses are composite explanations consist-ing of both hypotheses of homology (trans-formation series; Hennig, 1966; see Grantand Kluge, 2004) and hypotheses of mono-phyly (topology). Farris (1967) expressed thissuccinctly by analyzing the concept ofevolutionary relationship into its componentparts of patristic relationship and cladisticrelationship.

Operationally, phylogenetic analysis be-gins by decomposing the observed diversity


of living things into its minimal historicalunits: character-states (sensu Grant andKluge, 2004) and species (sensu Kluge,1990; see also Grant, 2002). Although char-acter-states are the evidential basis thatunderlies phylogenetic inference, they areeffectively ‘‘bundled’’ into individual organ-isms, populations, and species, which con-strains the ways in which they evolve andhow they may be explained (e.g., females andmales evolve as parts of the same lineage;defensible phylogenetic explanations aretherefore not permitted to place them inseparate clades). Likewise, species, which arethe historical entities related through phylog-eny (Hennig, 1966), may be decomposed intoindependently heritable (and independentlyvariable) parts, that is, character-states. Thisontological transitivity of taxic and characterevolution is the foundation of phylogeneticinference (cf. Farris, 1967).

Once these minimal units have been in-dividuated, all possible historical relation-ships between character-states and species aredefined by pure logic (Siddall and Kluge,1997; Wheeler, 1998). Phylogenetic analysisproceeds by mapping hypothetical character-state relationships to hypothetical speciesrelationships and evaluating the competingcomposite hypotheses in terms of the numberof character-state transformations they entail.

All phylogenetic methods aim to minimizecharacter transformations. Unweighted(equally weighted) parsimony analysis mini-mizes hypothesized transformations globally,whereas assumptions (expressed as differen-tial probabilities or costs) about the evolutionor importance (e.g., reliability) of differentclasses of transformations employed in max-imum likelihood, Bayesian analysis, andweighted parsimony methods lead to theminimization of certain classes of transfor-mations at the expense of others. Operationalconsiderations aside (e.g., tree-space search-ing capabilities), disagreements between theresults of unweighted parsimony analysis andthe other methods are due to the increasedpatristic distance required to accommodatethe additional assumptions.

Kluge and Grant (2006) reviewed thejustifications for parsimonious phylogeneticinference previously considered sufficient,namely, conviction (Hennig, 1966), descrip-

tive efficiency (Farris, 1979), minimization ofad hoc hypotheses of homoplasy (Farris,1983), and statistical, model-based inference(maximum likelihood, Sober, 1988). Findingsignificant inconsistencies in all of thosejustifications, Kluge and Grant (2006) pro-posed a novel justification for parsimony.Drawing on recent advances in the under-standing of phylogenetics as a strictly ideo-graphic, historical science and parsimoniousinference generally in the philosophy ofscience literature (e.g., Barnes 2000; Baker,2003), they argued that by minimizingglobally the transformation events postulatedto explain the character-states of terminaltaxa, equally weighted parsimony analysismaximizes explanatory power. As such, inthe present study we analyzed the total,equally weighted evidence under the parsi-mony criterion (for additional discussion ofcharacter weighting and total evidence, seeGrant and Kluge, 2003). Given the size andcomplexity of this dataset, a further advan-tage of parsimony algorithms (whetherweighted or unweighted) is that thoroughanalysis could be achieved in reasonabletimes given available hard- and software.

SOURCES OF EVIDENCE

The empirical evidence of phylogeneticsystematics consists of transformation series(i.e., the ideographic character concept ofGrant and Kluge, 2004). Traditionally, trans-formation series were derived exclusivelyfrom such sources as comparative morphol-ogy, molecular biology, and behavior, but astechnological advances have made DNAsequencing simpler and less costly, phyloge-netic studies have come to rely increasinglyon the genotypic evidence of DNA sequencesto test phylogenetic hypotheses. The presentstudy exemplifies this trend. Nevertheless,both kinds of data provide evidence ofphylogeny, and each has its own suite ofstrengths and weaknesses.

An important strength of phenotypic datais that the complexity of observed variationallows the historical identity of each trans-formation series to be tested independently(Grant and Kluge, 2004). By carrying outprogressively more detailed structural anddevelopmental studies, researchers are able to


refine their hypotheses about the homologyof phenotypic variants. However, the pheno-type is determined by both the directlyheritable components of the genotype andthe nonheritable effects of the environment.In contrast, an obvious strength of DNAsequence evidence is that, because DNA isthe physical material of genetic inheritance,the potentially confounding effects of envi-ronmental factors are avoided altogether.

Nevertheless, DNA sequence character-states are maximally reduced to physico-chemically defined classes of nucleotides, ofwhich there are only four (cytosine, guanine,adenine, and thymine). Whereas this simplic-ity is advantageous in many kinds of geneticsstudies, it poses a serious problem forphylogenetics, because no structural or de-velopmental complexity to distinguishes nu-cleotides that share a common evolutionaryhistory (i.e., those that are homologous, theirphysico-chemical identity owing to the sametransformation event) from those thatevolved independently (i.e., those that arehomoplastic, their physico-chemical identityowing to independent transformationevents). For example, in terms of objectproperties, all adenines are physico-chemical-ly identical, regardless of whether or not theyarose through the same or different trans-formation events. Moreover, DNA sequencesevolve through complete substitutions of onenucleotide for another (meaning that thereare no intermediate states from which to inferhistorical identity) or complete insertions anddeletions (meaning that any given nucleotidecould be homologous with any other nucle-otide). Phylogenetic analysis of DNA se-quences must therefore contend with theproblem of discovering both transformationsbetween nucleotides and the insertion anddeletion of nucleotides. To visualize homol-ogous nucleotides, multiple sequence align-ments codify insertions and deletions (indels)as gaps, that is, placeholders that shiftportions of the sequence to align homologousnucleotides into column vectors.

NUCLEOTIDE HOMOLOGY AND THE

TREATMENT OF INDELS

The method of inferring indels and nucle-otide homology (i.e., alignment) and the

subsequent treatment of indels in evaluatingphylogenetic explanations are of criticalimportance in empirical studies because, asis now widely appreciated, a given datasetaligned according to different criteria orunder different indel treatments may resultin strong support for contradictory solutions(e.g., McClure et al., 1994; Wheeler, 1995).Many workers infer indels in order to alignnucleotides but then either treat them asnucleotides of unknown identity by convert-ing gaps to missing data, or they eliminategap-containing column vectors altogether,either because they believe them to beunreliable or because the implementation ofa method of phylogenetic analysis does notallow them (Swofford et al., 1996). Othersargue that indels provide valid evidence ofphylogeny but suggest that sequence align-ment (homology assessment) and tree evalu-ation are logically independent and must beperformed separately (e.g., Simmons andOchoterena, 2000; Simmons, 2004).

The position we take here is that indels areevidentially equivalent to any other classes oftransformation events and, as such, are anindispensable component of the explanationof the DNA sequence diversity. Furthermore,because nucleotides lack the structural and/ordevelopmental complexity necessary to testtheir homology separately, hypotheses ofnucleotide homology can only be evaluatedin reference to a topology (Grant and Kluge,2004; see also Wheeler, 1994; Phillips et al.,2000; Frost et al., 2001). In recognition ofthese considerations, we assessed nucleotidehomology dynamically by optimizing ob-served sequences directly onto topologies(Sankoff, 1975; Sankoff et al., 1976) andheuristically evaluating competing hypothe-ses by searching tree space (Wheeler, 1996).This is achieved using Direct Optimizationtechniques (Wheeler, 1996, 2003a, 2003b,2003c), as implemented in the computerprogram POY (Wheeler et al., 1996–2003).

In this approach, determination of nucleo-tide homology is treated as an optimizationproblem in which the preferred scheme ofnucleotide homologies for a given topology isthat which requires the fewest transformationevents when optimized onto that topology,that is, that which minimizes patristic dis-tance, thus providing the most parsimonious


explanation of the observed diversity. De-termining the optimal alignment for a giventopology is NP-complete2 (Wang and Jiang,1994). For even a miniscule number ofsequences, the number of possible alignmentsis staggering (Slowinski, 1998), making exactsolutions impossible for any contemporarydataset, and heuristic algorithms are requiredto render this problem tractable. Likewise,finding the optimal topology for a givenalignment is also an NP-complete problem(Garey and Johnson, 1977; Garey et al., 1977).

Phylogenetic analysis under Direct Opti-mization therefore consists of two nested NP-complete problems. POY searches simulta-neously for the optimal homology/topologycombination, and search strategies must takeinto consideration the severity and effective-ness of the heuristic shortcuts applied at bothlevels. In any heuristic analysis, a balance issought whereby the heuristic shortcuts willspeed up analysis enough to permit a suffi-ciently large and diverse sampling of topol-ogies and homologies to discover the globaloptimum during final refinement, but not sosevere that the sample is so sparse ormisdirected that the global optimum is notwithin reach during final refinement. Ideally,indicators of search adequacy (e.g., multipleindependent minimum-length hits, stableconsensus; see Goloboff, 1999; Goloboffand Farris, 2001) should be employed tojudge the adequacy of analysis, as is nowreasonable in analysis of large datasets usingprealigned data (e.g., in TNT; Goloboff etal., 2003). However, current hard- andsoftware limitations make those indicatorsunreachable in reasonable amounts of time

for the present dataset analyzed under DirectOptimization, and the adequacy of ouranalysis may only be judged intuitively inlight of the computational effort and strategicuse of multiple algorithms designed for largedatasets (see below for details).

TOTAL EVIDENCE

The majority of phylogenetic studies, eventhose legitimately considered ‘‘total evi-dence’’ (Kluge, 1989), examine either higherlevel or lower level problems. The former aredesigned to address relationships betweenputative clades of multiple species (usuallydiscussed as species groups, genera, families,etc.) by targeting exemplars from each ofthose units and sampling relatively invariablecharacter systems. The latter are designed toaddress species limits and relationshipsamong closely related species (and oftenphylogeographic questions also), and char-acter sampling focuses on more variablesystems.

The nestedness of phylogenetic problemsboth enables and weakens this divide-and-conquer approach. Assuming the monophylyof a group, the relationships within that cladehave no bearing on the relationships of thatclade to other clades. And assuming thesister-group relationships between the in-group and outgroup taxa, the relationshipswithin the ingroup clade are independent ofthe relationships between that clade andmore distant relatives. Nevertheless, althoughthis is a defensible (provided the assumedmonophyly and placement of the ingroup aresupported by evidence) and presently neces-sary strategy, these assumptions may ulti-mately be found to be false, and theirelimination allows hypotheses at both levelsto be more severely tested and may lead tomore globally parsimonious explanations.Furthermore, the ripple effects that cladisti-cally distant optimizations may havethroughout the topology are unpredictable,so that the inclusion of distant terminals thatare not immediately relevant to a problemmay affect local topology. The ultimate goalof total evidence is to analyze all evidencefrom all sources and all terminals at all levelssimultaneously.

2This is a technical term referring to thecomputational complexity of problems (Cormen etal., 2001). Computational problems are classifiedaccording to the time complexity of their algorithmicsolutions. Computationally easy problems can besolved in polynomial time (P) relative to the size ofthe problem. Nondeterministic polynomial (NP)problems cannot be solved in polynomial time, buta solution can be verified in polynomial time. Nopolynomial-time algorithm is known for NP-com-plete (NPC) problems, but it has not been provedthat such an algorithm is impossible; if a polynomial-time algorithm is discovered for any NPC problem, itwould apply to the entire class of problems.


There are many obstacles, computationaland otherwise, that prevent this ideal frombeing achieved in the foreseeable future. Apotential criticism of this study is that we didnot include all of the data from Frost et al.(2006). We restricted the outgroup sample toAthesphatanura to prevent the already com-putationally challenging problem of analyz-ing 414 terminals from becoming even moreintractable. However, once hard- and soft-ware improvements permit thorough analysisof larger datasets, studies that extend out-group sampling beyond Athesphatanura willoffer a test of our results. Insofar as thisstudy was designed to test both the specieslimits of problematic taxa and the relation-ships among dendrobatid clades, and tofurther test the relationships between den-drobatids and other anurans by combiningthe data from Frost et al. (2006) with newdata from key outgroup taxa, it is intended tobe a step toward the total evidence ideal.That this study aimed to simultaneouslyaddress problems of such different hierarchiclevels had important consequences in taxonsampling, character sampling, and the ana-lytical strategy that was undertaken.

GENERAL ANALYTICAL

APPROACH: IMPLEMENTATION

TAXON SAMPLING

Outgroup taxa and the rationale for theirselection are discussed above (PhylogeneticPlacement of Dendrobatids and OutgroupSampling). Selection of ingroup terminalswas governed by three considerations: (1)relevance to testing prior phylogenetic claims,(2) availability of tissues (or sequences onGenBank), and (3) availability of specimensfor morphological study. In light of the manyproblems in species-level taxonomy, we alsosought to sample as many localities aspossible for problematic species.

To facilitate taxonomic changes, everyeffort was made to include type species ofall dendrobatid genera. Both genotypic andphenotypic data were included for typespecies of as many genera as possible,including (genus name in parentheses): azur-eiventris (Cryptophyllobates), bicolor (Phyllo-

bates), femoralis (Allobates), inguinalis (Pros-therapis), nocturnus (Aromobates), pulchellus(Phyllodromus), pumilio (Oophaga), reticula-tus (Ranitomeya), silverstonei (Phobobates),steyermarki (Minyobates), tinctorius (Dendro-bates), tricolor (Epipedobates), and trivittatus(Ameerega). We did not include the typespecies alboguttatus (Nephelobates), fuligino-sus (Hyloxalus), latinasus (Colostethus), oryustizi (Mannophryne), because adequatedata were not available to allow their in-clusion in the present study. Nevertheless, weincluded numerous, presumably closely re-lated representatives of these genera andmade taxonomic changes accordingly.

PHENOTYPIC CHARACTER SAMPLING

We anticipate that a criticism of thepresent study will be that we were toocatholic in the inclusion of phenotypiccharacters. It is common for morphologicalsystematists to seek characters that areconservative at their level of interest, underthe assumption that they are more informa-tive or reliable indicators of relationship,either explicitly (e.g., Kluge, 1993) or, muchmore commonly, implicitly. As a result, muchof the systematics literature—especially theprecladistic literature—consists of specialpleading for the validity (or not) of char-acters as ‘‘higher-level’’, ‘‘family-level’’, ‘‘ge-nus-level’’, ‘‘species-level’’ or some otherrank-specific indicator. Some characters(e.g., presence or absence of teeth, pectoralgirdle architecture, skull morphology), it hasbeen argued, are ‘‘good’’ genus- or family-level characters, others (e.g., external mor-phology, soft-anatomy) are ‘‘good’’ only atthe level of species, and still others areunreliable and should be excluded in theirentirety. We disagree.

For evolution to occur, all charactertransformation events must take place at(or below) the species level, and it is onlysubsequent cladogenetic events that effective-ly push character transformations back inhistory and bring them to delimit clades;there can be no natural law regulatingvariation of characters among clades. Thehistorical debate over the phylogenetic rele-vance of anuran teeth illustrates the futility ofthat approach to systematics: maxillary teeth


are absent in all species of Bufonidae—whichwould make this a conservative, phylogenet-ically informative character at the familylevel—but vary intraspecifically in somespecies of dendrobatids—making this a com-pletely uninformative character according tothat view. Given the conceptual definition ofcharacters as transformation series (Grantand Kluge, 2004), all characters have thesame evidential status in terms of their abilityto test phylogenetic hypotheses. Argumentsover the rank-specific relevance or reliabilityof characters depend on the reification ofranks and lead ultimately to the ad hocdismissal or overlooking of evidence, andsuch procedures should be eliminated fromsystematics.

In addition to the novel characters andcharacter-states discovered in the course ofthis study, our goal was to include allcharacters that have figured in debates onthe monophyly, placement, and internalrelationships of Dendrobatidae. However,because this study aims primarily to testrelationships within Dendrobatidae and notthe position of Dendrobatidae among otherfrogs, the sample of characters is stronglybiased to reflect variation among dendroba-tid terminals.

Numerous characters date to the 19thcentury (mainly Dumeril and Bibron, Cope,Boulenger), and we do not always cite theoriginal sources for these traditional char-acters. However, we do cite more recentpapers that have addressed them in thecontext of dendrobatid systematics, and wecite original sources for all more recentcharacters. All phenotypic character-statesfor caeruleodactylus, humilis, and nidicolawere coded from the literature (Lima andCaldwell, 2001; Caldwell et al., 2002a; LaMarca et al., 2002; Caldwell and Lima, 2003).Other sources for phenotypic data are citedin the relevant sections below. For thepurposes of discussion, phenotypic trans-formation series are classified broadly asmorphological, larval, behavioral, and bio-chemical, the latter referring to alkaloidprofiles. Specific problems or concerns re-garding particular characters or charactersystems are discussed below independently.In anticipation of the expansion of thepresent dataset, we list states and show

illustrations for taxa not included in thepresent analysis.

Comparative anatomical study aimed todelimit transformation series and not todescribe dendrobatid (or outgroup) anatomyper se. We have illustrated either photo-graphically or in line drawings those charac-ter-states we believe may cause confusion,and character names and descriptions wereintended only to be sufficiently precise toallow hypotheses of homology to be tested.With the exception of characters related tothe median lingual process, we coded ana-tomical characters only from gross dissectionunder a dissecting microscope. This is a lim-itation of the present study, as greater insightinto character-state identity and homologywould undoubtedly be gained from histology(e.g., consider the remarkable insights intopectoral girdle architecture attained by Ka-plan, 2004). Osteological character-stateswere coded from dried or cleared and stained(alcian blue and alizarin red) skeletons. Weconsidered tissue with alizarin red-positivecrystals to be calcified and uniformly alizarinred-positive tissue to be ossified.

We applied either Lugol’s solution oralcian blue to facilitate coding of musclecharacters. All muscles are bound by fibrousconnective tissue, so the distinction betweentendinous and fleshy origins and insertions isone of degree: Tendinous insertions andorigins have a confluence of muscle fiberson a distinct segment of fibrous connectivetissue, whereas those that are fleshy appear toinsert or originate directly on the adjacentstructure. We consider a slip to be a distinctbundle of fasciculi isolated from adjacentfasciculi of the same muscle by epimysium.

We examined the histology of the tonguesof several species to individuate characters ofthe median lingual process (Grant et al.,1997). Tissues were embedded in paraffin,sectioned at 6–10 mm, and stained usingeither hematoxylin and eosin (H&E) ora trichrome stain consisting of Alcian Blue,Periodic Acid and Schiff’s reagent (PAS), andH&E. Specifically, we examined histologicalsections of baeobatrachus, tepuyensis, pana-mensis, and auratus (the latter two lacking theMLP). For comparison we examined thehistology of Arthroleptis variabilis, Mantidac-tylus femoralis, Phrynobatachus natalensis, P.


petropedetoides, Platymantis dorsalis, andStaurois natator, although none of thesespecies were coded for the present study.We also performed detailed dissections of thetongues of atopoglossus, Arthroleptis steno-dactylus, Discodeles bufoniformis, and Dis-codeles opisthodon.

In addition to the phenotypic charactersindividuated for this study, other sources ofvariation will undoubtedly yield novel char-acters. For example, spermatozoa ultrastruc-ture is promising but has been examined intoo few species to warrant inclusion in thepresent study. Garda et al. (2002) examinedthe spermatozoa of flavopictus, Aguiar et al.(2003) studied femoralis and an undescribedspecies referred by them to Colostethus(OMNH 37001–37002), and Aguiar et al.(2002) looked at the spermatozoa of hahneliand trivittatus. (For a recent review ofspermatozoa in nondendrobatids, see Schel-tinga and Jamieson, 2003).

Relevant to the placement of Dendrobati-dae, Haas (2003) presented an impressivematrix of detailed morphological evidencescored across the diversity of anurans, muchof which was derived from studies of larvalanatomy. The evidential value of such data ismanifest, but adequate samples were unavail-able for most species included in this study.Haas found (see fig. 16, above) that the fourincluded dendrobatids were monophyletic,and that their sister group was Hylodes +Crossodactylus (Megaelosia was not includedin that study).

Bhaduri (1953) studied the urinogenitalsystems of diverse amphibians, includingauratus, tinctorius, and flotator (as Phyllo-bates nubicola flotator). He noted severaldifferences among these species, such as thegreater posterior extension of the kidneys inDendrobates than in Phyllobates (p. 56), buthe nonetheless concluded that ‘‘[t]he struc-tural similarities of the urinogenital organswhich we have observed in these two generalend further support to Noble’s view [thatDendrobates and Phyllobates are closely re-lated]’’ (p. 72). Although we scored somevisceral characters (e.g., pigmentation of thetestes, pigmentation of the large intestine), inlight of time constraints and the fact thatspecific characters used by Bhaduri have notbeen used since and have therefore not

played an important role in dendrobatidsystematics, we did not study this system indetail.

Burton (1998a) argued that hand muscu-lature supports a relationship between Den-drobatidae and Hylodinae, ‘‘as the unusualcondition of lacking any fibrous connectionto the tendo superficialis or the adjacentaponeurosis is almost restricted to thehylodine genera Hylodes and Megaelosia,and Dendrobatidae’’ (p. 8). However, thephylogenetic implications of this characterare not clear-cut; assuming hylodine mono-phyly and a sister-taxon relationship withDendrobatidae, as implied by Burton, theoccurrence of this character-state wouldoptimize as either independently evolved inDendrobatidae and Hylodes + Megaelosia oras a synapomorphy of the inclusive cladewith subsequent loss in Crossodactylus. Wedid not examine hand musculature in thisstudy.

Trewavas (1933) included tinctorius3 in herstudy of the anuran hyoid and larynx. Weexamined the osteology of this system but didnot examine its musculature. Our experiencewith other groups suggests that hyoid mus-culature may be a rich source of characters,and these characters will be evaluated in thenear future.

There are also several morphologicalvariants that have been claimed as charactersin the literature that we reject in the presentstudy. First, La Marca (1994, 2004) claimedthe occurrence of enlarged, fanglike maxillaryteeth as a synapomorphy for Nephelobates,and they also have been reported for Megae-losia (e.g., Lynch, 1971) and Aromobates(Myers et al., 1991), among others. Althoughwe agree with La Marca and Myers et al. thatdendrobatid tooth morphology varies andthat the teeth of Aromobates and Nephelo-bates seem strikingly elongate and recurved,we were unable to individuate transforma-tions series for several reasons, as follows (seefig. 19).

3Given the taxonomic problems that plagued thisspecies prior to Silverstone (1975a), and the givenrange as ‘‘South America’’, the identity of the‘‘Dendrobates tinctorius’’ specimen(s) examined byTrewavas (1933) is unclear.


(1) No appropriate reference point to assessrelative tooth size has been proposed, andwithout this it is impossible to compareobjectively the size of teeth in specimens ofdifferent species and varying body sizes andmaxilla sizes and shapes (especially the shapeand depth of the facial process). (2) Toothsize varies along the maxilla, and it is unclearwhich teeth should serve as the basis ofcomparison. (3) Superficial assessment oftooth size in cleared and stained specimensof a number of species suggested thatvariation is continuous, which must beaccounted for when individuating transfor-mations series. (4) All well-developed maxil-lary teeth (i.e., those that protrude beyondthe edge of the maxilla) are recurved, at least

in dendrobatids, and comparison of digitalimages (which eliminates the effect of relativesize) shows the curvature of the so-calledfanglike teeth of species referred to Nephelo-bates and Aromobates is no greater thanthose referred to Colostethus. In light of theseconsiderations, we coded the presence andabsence of maxillary teeth, as well as theirstructure, but not variation in size and shape.

Similarly, Lynch (1982) characterized ed-wardsi and ruizi as possessing a conspicuouslylarge and elongate cloacal sheath (vent tube,anal sheath, embudo cloacal), and Rivero(1990 ‘‘1988’’) subsequently referred to thesespecies as the edwardsi group of Colostethus.Later, La Marca (1994) also claimed thepresence of a cloacal sheath as a synapomor-phy for Nephelobates, although he made noreference to that structure in the edwardsigroup. The cloacal sheath has now beenincluded in numbered diagnoses in speciesdescriptions (e.g., Lotters et al., 2003a), andGrant (1998) cited its synapomorphic occur-rence as the basis for including Colostethuslynchi in the edwardsi group. More recently,Grant (2004) noted, without further com-ment, that ‘‘examination of extensive materi-al of most species of dendrobatids has causedme to doubt the validity of that character.’’

The reason for that doubt is that, as shownin figure 20, only the two species originallyplaced in the edwardsi group (exemplifiedhere by edwardsi) possess a conspicuouslymodified vent (cloacal sheath). Variationsamong other species of dendrobatids (in-cluding lynchi) are minor and cannot bedistinguished from artifacts of preservation.Specimens that are positioned differently forfixation (whether floated in formalin or laidout in a fixing tray) vary in apparent ventmorphology. For example, when a frogspecimen is positioned in a fixing tray, theflaccid thigh muscles and loose skin may bepushed posterodorsally, causing the vent andadjacent tissue to ‘‘bunch up’’, or anteroven-trally, causing the vent and adjacent tissue tobe drawn downward, both of which alter theapparent prominence, length, and shape ofthe vent. Desiccation also affects vent prom-inence. In light of these observations, thecloacal sheath is restricted to the two knownspecies of the edwardsi group. We did notinclude the cloacal sheath, thus delimited, in

Fig. 19. Examples of variation in dendrobatidmaxillary teeth. A, B: lateral (A) and lingual (B)views of pictus (UMMZ 184099). Note that theteeth do not protrude beyond the edge of themaxilla. C: lateral view of riveroi (AMNH 134144).D: lateral view of subpunctatus (UMMZ 221159).E: lateral view of undulatus (AMNH 159142). F:lateral view of molinarii (UMMZ 176207). G:lateral view of dunni (UMMZ 167131). H: lateralview of nocturnus (AMNH 129940).


the character matrix because we did notinclude edwardsi or ruizi due to inadequatematerial.

Finally, Savage (1968) was followed bySilverstone (1975a) in identifying dark pig-mentation of the flesh as a synapomorphy ofDendrobates and Phyllobates. We paid con-siderable attention to this character, thankslargely to the numerous large series of skinnedspecimens collected by C. W. Myers andcolleagues and deposited (cataloged anduncataloged) at AMNH. The variation weobserved is much more complicated than thesimple pigmented/unpigmented of Savageand Silverstone. Pigmentation occurs indiffuse, irregular patches and varies continu-ously in intensity from being entirely lackingto a few black specks or intense dark gray orblack. We were unable to delimit transforma-tion series objectively, and therefore excludedpigmentation of the flesh from this study.

GENOTYPIC CHARACTER SAMPLING

In light of the vastly different levels ofdiversity included in this study (from

within localities to among families), wesought to sample genes of differing degreesof variability. We targeted the mitochon-drial H-strand transcription unit 1 (H1),which includes 12S ribosomal, tRNAval,and 16S ribosomal sequence, yieldingapproximately 2,400 bp generated in 5–7overlapping fragments. We also targeteda 385-bp fragment of cytochrome b anda 658-bp fragment of cytochrome oxidase csubunit I (COI). In addition to those fivemitochondrial genes, we targeted the nuclearprotein coding genes histone H3 (328 bp),rhodopsin (316 bp), tyrosinase (532 bp), re-combination activating gene 1 (RAG1,435 bp), and seventh in absentia (SIA,397 bp), and the nuclear 28S ribosomalgene (ca. 700 bp), giving a total of approxi-mately 6,100 bp of nuclear and mito-chondrial DNA. Primers used in PCRamplification and cycle sequencing reactions(and their citations) are given in table 1.Included in this study is a novel primer pair(RAG1 TG1F and TG1R) designed toamplify the RAG1 product using the web-based program Primer3 (Rozen and Ska-

Fig. 20. Posterior view of several species of dendrobatids, showing variation in morphology of thevent. A: edwardsi (ICN 21936). Contrary to the other species depicted, the vent of edwardsi isconspicuously enlarged and elongated relative to that of other anurans. B: molinarii (UMMZ 176222,paratype), a species referred to Nephelobates by La Marca (1994). The vertical folds vary as an artifact ofpreservation. C: alboguttatus (AMNH 10503), the type species of Nephelobates. D: trinitatis (AMNH125796), a species referred to Mannophryne by La Marca (1992). E: petersi (AMNH 42546). F: petersi(AMNH 42506). This species has never been claimed to be part of or closely related to Nephelobates. Notethe differing prominence and apparent shape and size of the vent as an artifact of preservation.


letsky, 2000), available at http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi. Asdiscussed below, the amount of sequencedata analyzed per terminal varied (fig. 21),ranging from 426 bp (subpunctatus obtainedfrom GenBank) to 6,245 bp (chlorocras-pedus 385), with a mean of 3,740 bp perterminal.

As noted above, we targeted loci thatvaried to differing degrees to test hypothesesof relationships at all levels, and we includedmultiple samples from the same and differ-ent localities of the same species in an effortto address problems in alpha taxonomy. Weattempted to sequence all loci for at leastone sample from every locality, but we did

TABLE 1PCR Primers Used in This Studya

Gene region Primer name Direction Primer sequence (59 to 39) Source

16S rDNA AR Forward CGCCTGTTTATCAAAAACAT Palumbi et al., 1991

BR Reverse CCGGTCTGAACTCAGATCACGT Palumbi et al., 1991

Wilkinson2 Reverse GACCTGGATTACTCCGGTCTGA Wilkinson et al., 1996

L2A Forward CCAAACGAGCCTAGTGATAGCTGGTT Hedges, 1994

H10 Reverse TGATTACGCTACCTTTGCACGGT Hedges, 1994

MVZ59 Forward ATAGCACTGAAAAYGCTDAGATG Graybeal, 1997

MVZ50 Reverse TYTCGGTGTAAGYGARAKGCTT Graybeal, 1997

12s A-L Forward AAACTGGGATTAGATACCCCACTAT Goebel et al., 1999

12s F-H Reverse CTTGGCTCGTAGTTCCCTGGCG Goebel et al., 1999

12s L1 Forward AAAAGCTTCAAACTGGGATTAGATACCCCACTAT Feller and Hedges, 1998

12S rDNA L13 Forward TTAGAAGAGGCAAGTCGTAACATGGTA Feller and Hedges, 1998

Titus I Reverse GGTGGCTGCTTTTAGGCC Titus and Larson, 1996

tRNAval tRNAval-H Reverse GGTGTAAGCGARAGGCTTTKGTTAAG Goebel et al., 1999

cytochrome

oxidase c

subunit I

LCO1490 Forward GGTCAACAAATCATAAAGATATTGG Folmer et al., 1994

HCO2198 Reverse TAAACTTCAGGGACCAAAAAATCA Folmer et al., 1994

cytochrome b MVZ 15-L Forward GAACTAATGGCCCACACWWTACGNAA Moritz et al., 1992

H15149 Reverse AAACTGCAGCCCCTCAGAAATGATATTTGTCCTCA Kocher et al., 1989

rhodopsin

exon 1

Rhod1A Forward ACCATGAACGGAACAGAAGGYCC Bossuyt and Milinkovitch,

2000

Rhod1C Reverse CCAAGGGTAGCGAAGAARCCTTC Bossuyt and Milinkovitch,

2000

Rhod1D Reverse GTAGCGGAAGAARCCTTCAAMGTA Bossuyt and Milinkovitch,

2000

tyrosinase

exon 1

TyrC Forward GGCAGAGGAWCRTGCCAAGATGT Bossuyt and Milinkovitch,

2000

TyrG Reverse TGCTGGCRTCTCTCCARTCCCA Bossuyt and Milinkovitch,

2000

histone H3 H3F Forward ATGGCTCGTACCAAGCAGACVGC Colgan et al., 1999

H3R Reverse ATATCCTTRGGCATRATRGTGAC Colgan et al., 1999

28S rDNA 28sV Forward AAGGTAGCCAAATGCCTCATC Hillis and Dixon, 1991

28SJJ Reverse AGTAGGGTAAAACTAACCT Hillis and Dixon, 1991

recombi-

nation

activating

gene 1

RAG1 TG1F Forward CCAGCTGGAAATAGGAGAAGTCTA This study

RAG1 TG1R Reverse CTGAACAGTTTATTACCGGACTCG This study

R1-GFF Forward GAGAAGTCTACAAAAAVGGCAAAG Faivovich et al., 2005

R1-GFR Reverse GAAGCGCCTGAACAGTTTATTAC Faivovich et al., 2005

seventh in

absentiab

SIA1 (T3) Forward TCGAGTGCCCCGTGTGYTTYGAYTA Bonacum et al., 2001

SIA2 (T7) Reverse GAAGTGGAAGCCGAAGCAGSWYTGCATCAT Bonacum et al., 2001

a The gray line separates mitochondrial (above) and nuclear (below) loci. See text for PCR and cycle sequencing

protocols.b These primers were used with the universal T3 and T7 primers following Bonacum et al. (2001).


not sequence nuclear loci for all samples.We chose this strategy because early workon this project showed the nuclear loci to

be generally less variable and usuallyidentical in all samples from a given local-ity.

We augmented our own data with se-quences from GenBank, listed in appen-dix 4, in order to include otherwise un-sampled ingroup species and additional

localities for taxonomically problematicspecies, that is, the dataset analyzed includesall species on GenBank as well as samples ofsome species from multiple localities. Nev-ertheless, we did not include all GenBankdata. First, we only included loci for whichwe also generated data. For example, we didnot include Widmer et al.’s (2000) cyto-chrome b data because their fragment did notoverlap with ours. Second, although we

included multiple samples to address taxo-nomic problems, we did not include allsamples from population-level studies (e.g.,Symula et al., 2003), as such dense intraspe-cific sampling was not required and wouldhave impeded analysis by unnecessarily ex-panding the dataset.

LABORATORY PROTOCOLS

Whole cellular DNA was extracted fromfrozen and ethanol-preserved tissues (liver ormuscle) using the Qiagen DNeasy kit follow-ing the manufacturer’s guidelines. PCR am-plification was carried out in 25-ml reactionsusing puRe Taq Ready-To-Go Beads (Amer-sham Biosciences). The standard PCR pro-gram consisted of an initial denaturing step of3 min at 94uC, 35–40 cycles of 1 min at 94uC,1 min at 45–62uC, and 1–1.25 min at 72uC,followed by a final extension step of 6 min at72uC. PCR-amplified products were cleanedand desalted using either the ARRAYIT kit(TeleChem International) on a BeckmanCoulter Biomek 2000 robot or AMPure(Agencourt Biosciences Corporation). Cycle-sequencing using BigDye Terminators v. 3.0(Applied Biosystems) was run in 8-ml reac-tions, and products were cleaned and desaltedby standard isopropanol-ethanol precipita-tion or using cleanSEQ (Agencourt Bio-sciences Corporation). Sequencing was doneon either an ABI 3700 or ABI 3730XLautomated DNA sequencer. Contigs (sets ofoverlapping sequences) were assembled andedited using Sequencher (Gene Codes).

Fig. 21. Number of DNA base-pairs analyzed per terminal. For specific terminal data see appendices 4and 5.


MOLECULAR SEQUENCE FORMATTING

To enable integration of incomplete se-quence fragments (particularly those fromGenBank; see Taxon Sampling Strategy andCharacter Sampling Strategy, above), accel-erate cladogram diagnosis, and reduce mem-ory requirements under Iterative Pass Opti-mization, we broke complete sequences intocontiguous fragments. (This also improvesthe performance of POY’s implementation ofthe parsimony ratchet; see Heuristic TreeSearching, below.) We did so sparingly,however, as these breaks constrain homologyassessment by prohibiting nucleotide com-parisons across fragments, that is, it isassumed that no nucleotides from fragmentX are homologous with any nucleotides fromfragment Y. As the number of breaksincreases, so too does the risk of overlyconstraining the analysis and failing todiscover the globally optimal solution.

We therefore inserted as few breaks aswere necessary to maximize the amount ofsequence data included, minimize the in-troduction of nucleotides of unknown iden-tity into the sequences (see CharacterSampling Strategy, above), and attain max-imum length fragments of around 500 bases(see table 2). Breaks were placed exclusivelyin highly conserved regions (many of whichcorrespond to commonly used PCR pri-mers), as recovery of such highly invariableregions is generally alignment-method in-dependent (unpublished data) and thereforedoes not prevent discovery of global opti-ma. These highly conserved regions wereidentified via preliminary ClustalX (Thomp-son et al., 1997) alignments under defaultparameters and examination using BioEdit(Hall, 1999). Except for their usefulness inplacing fragments derived from differentPCR primers and detecting errors, thesepreliminary alignments were used solely forthe purpose of identifying conserved re-gions; they did not otherwise inform orconstrain our phylogenetic analysis. Onceappropriate conserved regions were identi-fied, fragments were separated by insertingpound signs (#) at break points. Thus, themultiple fragments of the mitochondrial H1unit remain in the same file and order, forexample.

TOTAL EVIDENCE ANALYSIS

We did not discriminate between classes ofevidence in the phylogenetic analyses. Toallow the molecular data to have bearing onproblems of species taxonomy, we treatedevery specimen sequenced as a separateterminal, that is, we did not fuse putativelyconspecific specimens into a single polymor-phic terminal, which would prevent themolecular data from addressing alpha taxo-nomic problems and require that all decisionson species identity be made prior to phylo-genetic analysis. Loci not sequenced forparticular terminals—either because the pri-mers failed or because other syntopic con-specifics were sequenced instead—were treat-ed as missing for those terminals.

There are three possible methods of in-corporating phenotypic evidence for speci-mens judged to be conspecific but codedseparately for genotypic data.

1. Phenotypic characters may be coded for each

specimen separately. The shortcomings of this

method are numerous: (a) This approach

excludes background knowledge that informs

but is not explicitly encoded in the character

matrix, such as mating behavior and ontoge-

ny. This could result in males, females, and

juveniles being grouped in separate clades. (b)

Similarly, tissue samples usually are not avail-

able for specimens representing all relevant

TABLE 2Summary of DNA Sequence Dataa

Sequence

Approx. no.

basepairs

No.

fragments

No.

terminals

Mitochondrial

H-strand

transcription unit 1

2400 16 414

Cytochrome b 385 3 318

Cytochrome c

oxidase I

658 2 234

Recombination

activating gene 1

435 2 128

28S 700 2 136

Histone H3 328 1 169

Rhodopsin 316 1 154

Seventh in absentia 397 2 135

Tyrosinase 532 2 54

a Approximate number of base pairs refers to complete

sequences.


semaphoronts. As such, many semaphoront-specific characters would be excluded fromanalysis or have to be coded without beingmatched with the molecular evidence. (c) Forthis approach to be applied consistently,evidence obtained from specimens in otherstudies would also have to be rejected, such asalkaloid profiles and vocalizations and otherbehaviors, or also scored separately for eachindividual. Strict application of this approachis clearly infeasible and would result in theexclusion of extensive evidence.

2. The phenotypic data for the species as a wholecan be duplicated for each molecular terminal.

3. The phenotypic data for the species as a wholecan be entered for a single molecular terminal,with those characters treated as missing forother (putatively) conspecific terminals.

The latter two options offer more de-fensible approaches. The second method hasthe advantage of minimizing ambiguousoptimizations due to missing entries, whichmay be crucial in examining the evolution ofsome of the most interesting phenotypiccharacters (e.g., behaviors). The third ap-proach appears to have the advantage ofmaximizing the severity of the molecular testof species identity, that is, terminals judgedconspecific on phenotypic grounds could notbe held together on those grounds alone inphylogenetic analysis. Although we see somevalidity to this argument, given the relativesizes of the phenotypic and genotypic parti-tions (ca. 170 characters vs. ca. 6,100 un-aligned base pairs), we see no a priori reasonto expect morphology to overwhelm theDNA sequence data. Moreover, the identicalentries that would potentially hold thosespecimens together in the face of molecularevidence are, in fact, apomorphies for thespecies, and total evidence analysis demandsthat they be considered as such. That is, thegoal in total evidence analysis is not to testthe results of one data partition againstanother, but to allow all evidence to interactsimultaneously to identify the hypothesis thatbest explains all the evidence.

As such, we opted to duplicate themorphological entries coded for the species,that is, each conspecific terminal was givenidentical entries in the phenotypic matrix.Phenotypic characters not expressed in thesequenced semaphoront (e.g., testis color infemale specimens) were scored and species-

level phenotypic polymorphisms were codedas ambiguities. A caveat is that we did notassociate GenBank sequences with pheno-typic data unless we lacked our own geno-typic data for the taxon (e.g., sauli), and thenonly for one sample if more than one was onGenBank (e.g., we associated the phenotypicentries for kingsburyi with AY364549 only).

With a single exception, the reportedingroup characters were scored for singlespecies (see Outgroup Sampling, above, forcoding of outgroup species). The exceptionwas alagoanus, for which the morphologicalcharacters were scored from specimens ofolfersioides. The two species were named onthe basis of few differences between smallsamples, and study of external morphologyof larger samples from a greater number oflocalities suggests they may be conspecific(V.K. Verdade, personal commun.), althoughadditional data from color in life, vocaliza-tions, and DNA sequences have yet to beexamined. DNA sequences were generatedfor alagoanus, but whole specimens were notavailable for examination. Tissue sampleswere unavailable for olfersioides (which hasnot been observed in Rio de Janeiro sinceapproximately 1980 and may be extinct;C.F.B. Haddad, personal obs.), but numer-ous museum specimens were available formorphological study. As such, we combinedthe data from these two species under theasumption that they represent a unique clade;we refer to the resulting terminal as alagoanus

because most of the evidence was scoredfrom that species.

Simultaneous phylogenetic analysis wasperformed using the program POY (Wheeleret al., 1996–2003) version 3.0.11a and theMPI version 3.0.12a-1109195780.71. AllPOY runs were parallelized across 95 pro-cessors of the AMNH 256-processor Pentium4 Xeon 2.8 GHz cluster or 16–32 processorsof the 560-processor mixed 512 mHz and1 GHz cluster. Results were visualized usingWinclada (Nixon, 1999–2002), and we veri-fied POY results and analyzed impliedalignments using NONA (Goloboff, 1999)spawned from Winclada. Although characterweighting was used heuristically in treesearching (see below), evidence was weightedequally to assess tree optimality.


HEURISTIC CHARACTER OPTIMIZATION

Numerous algorithms of varying exhaus-tiveness have been proposed to optimizeunaligned DNA sequences on a given topol-ogy. Our search strategy employed threeDirect Optimization algorithms; in order ofincreasing exhaustiveness and execution time,these were Fixed-States Optimization(Wheeler, 1999), Optimization Alignment(Wheeler, 1996), and Iterative Pass Optimi-zation (Wheeler, 2003b).

Although Fixed-States Optimization wasproposed as a novel means of conceptualizingDNA sequence homology (Wheeler, 1999),we employed it here simply as a heuristicshortcut. Because Fixed-States is so muchfaster than the Optimization Alignmentalgorithm, it allowed more thorough sam-pling of the universe of trees for subsequentrefinement under more exhaustive optimiza-tion algorithms. Our general strategy wastherefore to examine a large pool of initialcandidate trees quickly under Fixed-Statesand submit those trees as starting points forfurther analysis under Optimization Align-ment. Because the potential exists for theglobally optimal tree (or trees that would leadto the global optimum when swapped undera more exhaustive optimization algorithm) tobe rejected from the pool of candidates underthe heuristic algorithm, we also generateda smaller pool of candidate trees underOptimization Alignment. The resulting opti-mal and near-optimal candidate trees werethen submitted to final evaluation and re-finement under Iterative Pass optimizationusing iterativelowmem to reduce memoryrequirements. (For details on tree-searchingalgorithms, see Heuristic Tree Searching,below.)

We did not employ the exact commandduring most searches, although we did use itin the final stages of analysis to allowaccurate matrix-based length verification(Frost et al., 2001). To verify lengths reportedin POY, we output the implied alignment(Wheeler, 2003a) and binary version of theoptimal topology in Hennig86 format withphastwincladfile and opened the resulting filein Winclada (Nixon, 1999–2002). Becauseeach topology may imply a different optimalalignment, when multiple optimal topologies

were obtained we examined them separatelyby inputting each as a separate file usingtopofile. Examination of the implied align-ments, whether formatted as Hennig files oras standard alignments (impliedalignment),grants another opportunity to detect errorsin formatting or sequencing.

HEURISTIC TREE SEARCHING

Efficient search strategies for large datasetsare, to a certain degree, dataset dependent(Goloboff, 1999), and, as discussed above,common indicators of sufficiency are imprac-tical given current technological limitations.Therefore, rather than apply a simple, pre-defined search strategy (e.g., 100 randomaddition Wagner builds + TBR branchswapping), we employed a variety of treesearching algorithms, spending more time onthose that proved most efficient. Optimaltrees from different searches were pooled fortree-fusing and TBR swapping, all of whichwas followed by refinement under IterativePass Optimization (Wheeler, 2003b). Thesearch strategy is summarized in table 3.

Random addition sequence Wagner builds(RAS) were performed holding one or threetrees. We conducted searches without slop orcheckslop, both of which increase the pool oftrees examined by swapping suboptimal treesfound during the search; although these stepscan be highly effective, initial trials showedthey were too time consumptive for thepresent dataset.

The parsimony ratchet (Nixon, 1999) wasproposed for analysis of fixed matrices. Giventhat under dynamic homology there are noprespecified column vectors to be reweighted,the original approach had to be modified. Inthe current version of POY, the ratchet isprogrammed to reweight randomly selectedDNA fragments. The present dataset wasbroken into 31 fragments (see table 2), soratchetpercent 15 randomly reweighted fivefragments, regardless of their length or relativeposition. We reweighted 15–35% of the frag-ments and applied weights of 2–35 times.

As a complementary approach, we alsoperformed quick searches (few random addi-tion sequence Wagner builds + SPR) underindel-transversion-transition costs of 3-1-1, 1-


3-1, and 1-1-3 and included the resultingtopologies in the pool of trees submitted tofusing and refinement under equal weights,following the general procedure of d’Haese(2003). Reweighting in this method is not donestochastically and therefore differs from bothNixon’s (1999) original version and POY’simplementation of the ratchet and technicallyis not a simulated annealing or Metropolis-Hastings-type strategy like the others; howev-er, because it weights sets of transformationsdrawn from throughout the entire dataset, it islikely to capture different patterns in the dataand may actually be a closer approximation tothe original ratchet than POY’s implementa-tion. Both approaches are effective methods toescape local optima.

We also performed constrained searchesby calculating the strict consensus of treeswithin an arbitrary number of steps of thepresent optimum, saving the topology asa treefile, constructing the group inclusionmatrix (Farris, 1973) in the program Jac-k2Hen, and then employing constraint in thesubsequent searches. To calculate the con-sensus we included trees within 100–150 stepsof the current optimum, the goal being tocollapse enough nodes for swapping to beeffective, but few enough nodes for signifi-cant speedups in RAS + swapping to find

optimal arrangements within the polytomousgroups (see Goloboff, 1999: 420). This iseffectively a manual approximation of Go-loboff’s (1999) consensus-based sectorialsearch procedure, the main difference beingthat we collapsed nodes based only on treelength and not relative fit difference (Golob-off, 1999; Goloboff and Farris, 2001).

Using constraint files generated in thesame way, we also input the current optimumas a starting point for fusing and/or ratchet-ing. This strategy avoids spending time onRAS builds of the unconstrained parts of thetree (which tend to be highly suboptimal) andseeks to escape local optima in the same wayas unconstrained ratcheting, discussed above;however, there is a trade-off in that thearrangements may be less diverse but arelikely to be, on average, closer to optimum,than those examined through RAS.

As a further manual approximation ofsectorial searches, we analyzed subsets oftaxa separately by defining reduced datasetswith terminals files that listed only thetargeted terminals. Rigorous searches (atleast 100 RAS + TBR for each of the reduceddatasets) of these reduced datasets were thenperformed, and the results were then used tospecify starting topologies for additionalsearching of the complete dataset.

TABLE 3Summary of Tree Searching Methods Combined in Overall Search Strategya

Abbreviated name Description

RAS Random addition sequence Wagner builds.

Constrained RAS As above, but constrained to agree with an input group inclusion

matrix derived from the consensus of topologies within 100–150

steps of present optimum.

Subset RAS Separate analysis of subsets of 10–20 taxa. Resulting topologies used

to define starting trees for further analysis of complete dataset.

Ratcheting (fragment reweighting) Ratcheting as programmed in POY, with 15–35% of DNA

fragments selected randomly and weighted 2–83, saving 1 minimum

length tree per replicate.

Ratcheting (transformation reweighting) Ratcheting approximated by applying relative indel–transversion–

transition weights of 3–1–1, 1–3–1, and 1–1–3, saving all minimum

length trees for analysis under equal weights.

Constrained tree fusing and/or ratcheting

(fragment)

As above, but with current optimum input as a starting tree, and

constrained to agree with an input group inclusion matrix derived from

the consensus of topologies within 100–150 steps of present optimum.

Tree fusing Standard tree fusing followed by TBR branch swapping.

Manual rearrangement Manual movement of branches of current optimum.

a Different runs combined multiple procedures, and all runs included SPR and/or TBR refinement. See text for details

and references.


Static matrices may be thoroughly ana-lyzed in a fraction of the time required toperform an equivalent analysis under dynam-ic homology. We therefore output impliedalignments of current optima from POY andran multiple replicates of RAS + 50 rounds ofthe parsimony ratchet using Winclada andNONA. Improvements were not alwaysattained through this procedure, but whenthey were, we then input the optimalcladogram(s) from the static search asa starting point for further analysis in POY.

As a final attempt to discover moreparsimonious solutions, we also rearrangedbranches of current optima manually. Asa general search strategy this would obvious-ly be highly problematic, if for no otherreason than that it would bias analyses.However, we performed this step primarilyto ensure that the ‘‘received wisdom’’ andother arrangements were evaluated explicitlyin the analysis. The procedure was to openthe current optimum in Winclada, target taxawhose placement was strongly incongruentwith current taxonomy, and move them totheir expected positions (or in polytomies,depending on the precision of the expecta-tions). The resulting topologies were saved astreefiles that were read into POY as startingtopologies for diagnosis and refinement (e.g.,tree fusing). In this way we ensured that themore heterodox aspects of our results werenot due to simply failing to evaluate the moreorthodox alternatives during the automatedsearches.

HEURISTIC DATA EXPLORATION

To estimate support (sensu Grant andKluge, 2003), we calculated Bremer (decay)values for all nodes present in the strictconsensus of equally parsimonious solutions(Bremer, 1994). To accomplish this we outputthe implied alignment and optimal trees inHennig86 format using phastwincladfile, con-verted it to NEXUS format in Winclada, andthen generated a NEXUS inverse-constraintsbatch file in PRAP (Muller, 2004), which wasanalyzed in PAUP* 4.0b10 (Swofford, 1998–2002). Bremer analysis consisted of 1 RAS + 5iterations of the parsimony ratchet (reweight-ing 25% of characters by 2) for each clade.More thorough analysis involving more

rigorous tree searches of the unaligned datawould undoubtedly lower the estimates; as isalways the case with heuristic analysis, theBremer values reported are an upper bound.

SPECIES IDENTIFICATION

One of the goals of this study was toaddress problems in determining speciesidentity. Through the course of the studyspecies were identified on phenotypicgrounds. Here we examine the bearing ofevidence from DNA sequences and phyloge-netic analysis on alpha taxonomic problems(see also Total Evidence Analysis, above).

As a means of identifying species limits,the results of phylogenetic analysis should beinterpreted in light of several caveats: (1)Phylogenetic analysis presupposes that thegenealogical relationships among the entitiesanalyzed are phylogenetic (Davis and Nixon,1992); as such, it will impose a hierarchy evenon entities that are related tokogenetically,for example. In such cases, the branchingstructure would be an analytical artifact andfinding that a species is or is not mono-phyletic would be irrelevant. (2) Species arehistorical individuals, and, as such, all partsof a given species need not form a clade(Skinner, 2004; see also Frost and Kluge,1994). Incongruence between the history ofdifferent parts and the whole may be due toany number of natural phenomena, such aslineage sorting and partial/temporary intro-gression, none of which denies the historicalindividuality of the species. (3) Given a clado-gram alone, there is no objective basis foridentifying species limits, that is, there is noway to discriminate intra- from interspecifichierarchic structure without additional in-formation. For example, dividing a pectinatecladogram of N terminals into 1 species, N 2

1 species, and N species are all cladisticallyvalid delimitations. As such, phylogeneticstructure can only disconfirm hypotheses ofspecies identity (but consider points 1 and 2);finding that the parts of a putative speciesform a clade does not deny that the clademay be composed of multiple species.

In spite of the above caveats, phylogeneticanalysis is a valid (if fallible) species discov-ery operation (Frost et al., 1998). Conceptu-ally, species are minimal historical individu-


als, meaning that species boundaries occur atthe point where the properties of contempo-rary individuals dissolve (Kluge, 1990; Grant,2002). Historical individuality may thereforebe apprehended both from ‘‘below’’ bydiscovering the constituent parts that interactand ‘‘above’’ by individuating entities thatare historically distinct. Incongruence be-tween the results of complementary discoveryoperations (those directed from above andbelow, in this case) indicates heuristically thatfurther study is warranted (Grant, 2002).Ultimately, species individuation requiresdiagnostic characters, and phylogenetic anal-ysis facilitates their discovery.

To address the limits of problematicspecies, we considered (1) cladogram topol-ogy (cladistic distance), (2) branch lengths(patristic distance), and (3) uncorrected pair-wise distance4 (uncorrected p, or number of

base mismatches divided by total sequencelength; no length variation was observed forthis locus) of cytochrome b sequences withinand between localities and/or closely relatedspecies. We focused on that sequence because(1) it is sufficiently variable and (2) it isalmost completely represented in our dataset.

Our primary reason for including pairwisedistances in this analysis is that they providea rapid heuristic for species identificationwithout conducting a complete phylogeneticanalysis, in the same way that artificialdichotomous keys are efficient identificationtools (Grant, 2002). We do not advocateusing pairwise distances to delimit species.First, there is no justification for setting somearbitrary distance (e.g., 5%)—phenetic orotherwise—as ‘‘sufficient’’ for granting spe-

Fig. 22. Character 0, dorsal skin texture. A: State 0, smooth (galactonotus, AMNH live exhibit). B:State 1, posteriorly tubercular (fraterdanieli, MHNUC 364). C: State 2, granular (macero, AMNH 129473).D: State 3, spiculate (Dendrophryniscus minutus, AMNH 93856).

4This is usually referred to as sequence divergence.However, divergence is a phylogenetic concept

synonymous with patristic distance (Farris, 1967).These pairwise comparisons are phenetic and arebetter characterized as dissimilarities or pheneticdistances.


cies status. Given variation in evolutionaryrates and sampling density, it is expected thatintraspecific variation may be greater in somespecies than interspecific variation amongothers. Indeed, the inability to distinguishbetween rate variation and artifacts due totaxon sampling (including extinction) castsdoubt on all studies that base conclusions ondegree of divergence or distance. What mattersis the total evidence (including other loci,morphology, behavior, etc.) for the historicalreality of the putative species and clades, forwhich character-state transformations must beidentified to diagnose minimal historical indi-viduals, not degree of similarity (pair- orotherwise). Second, as two-taxon statements,pairwise distances do not distinguish betweensymplesiomorphy and synapomorphy andtherefore fail to explain the observed variation.Third, pairwise distance only discriminatesamong samples, that is, it is a relationalconcept and therefore cannot diagnose anyparticular entity (see Frost, 2000). Neverthe-less, because they do not require extensivesampling or detailed analysis (phylogenetic orotherwise), pairwise comparisons are extreme-ly fast and simple and therefore highlyheuristic, and as such they are a useful startingpoint in examining species identity.

PHENOTYPIC CHARACTERS

0. DORSAL SKIN TEXTURE (fig. 22): smooth5 0; posteriorly granular 5 1; stronglygranular 5 2; spiculate 53. Nonadditive.

Living and well-preserved anuran skinalways has some texture, so even ‘‘smooth’’skin may appear shagreen or faintly granularunder high magnification. In state 0 all dorsalsurfaces lack distinct tubercles or granules(e.g., histrionicus, abditaurantius). In state 1,granules or tubercles are scattered irregularlyover the dorsal surfaces, being more distinctand prominent posteriorly, especially in thesacral region and on the thigh and/or shank,and absent or weaker and sparser anteriorly(e.g., boulengeri, fraterdanieli). These gran-ules or tubercles are often distinctly elevatedand conical. State 2 consists of rounded orflattened granules distributed densely andevenly (e.g., granuliferus, parvulus). Spiculateskin (state 3) is restricted to outgroup species;the skin of Dendropryniscus minutus is

conspicuously spiculate, but in others (e.g.,Atelopus spurrelli) the distinctly spiculate skinis only evident under magnification. Al-though state 1 is intermediate in the amountof granulation, the individual granules ortubercles of states 1 and 2 are qualitativelydifferent, and there is no developmentalevidence to suggest that transformationsbetween states 0 and 2 pass through state 1.

Dorsal skin texture has generally beenused descriptively in species-level taxonomicstudies (e.g., Myers et al., 1995, in distin-guishing between pumilio and granuliferus;Silverstone, 1976, in distinguishing betweenfemoralis and boulengeri). Jungfer (1989)reviewed the ‘‘red-backed granulated’’ Ama-zonian dendrobatids but did not explicitlydelimit them as a group.

Unlike some other anurans (e.g., centrole-nids), skin texture of dendrobatids variesonly minimally (or not at all) in relation toseason and/or reproductive activity, withvariation involving only the protuberance ofgranules or tubercles and not the occurrenceof different states. Nevertheless, care must beexercised in coding this character (and othersinvolving dermal structures) because it isprone to alteration due to preservation.Inadequately fixed or preserved specimenstend to lose granularity or even slough theepidermis. Even well-preserved specimensfixed according to the standard procedureof laying the specimen in a fixing tray prior toimmersion in formalin are often less granularthan they were in life. Conversely, granularitymay be exaggerated in desiccated specimens.As noted by Myers and Daly (1979: 5,footnote 1), the best means of preservingskin texture (as well as other dermal char-acters such as hand and foot tubercles) is tofloat them completely in formalin immedi-ately. (However, we do not mean to endorsethis method generally, as it complicatesexamination of the vast majority of morpho-logical characters.)

Heyer (1983: 322) provides electronmicro-graphs showing the skin texture for Cyclo-ramphus boraceiensis. We coded pulcherrimusaccording to Duellman’s (2004) description.

1. PAIRED DORSAL DIGITAL SCUTES: absent5 0; present 5 1.

All species of dendrobatids have distinctivepaired dermal scutes atop digital discs,


although they may be inconspicuous on firstand last digits and are generally moststrongly expressed on the third finger andfourth toe (i.e., on discs that are mostexpanded). Noble (1926: 7) cited this charac-ter as evidence uniting dendrobatids ina single, exclusive group, and since then ithas been used consistently to diagnosedendrobatids. Noble and Jaeckle (1928)examined the histology of the digital discsand illustrated (but did not discuss) thedigital scutes of a specimen they identifiedas Phyllobates latinasus (actual species un-known but probably not latinasus; see Grant,2004, for discussion of latinasus taxonomy)and Hylodes nasus (as Elosia bufonia). Noble(1931) noted the occurrence of the digitalscutes in his Elosiinae (p. 504) and dendro-batids (p. 507). Although he did not explicitlystate that it was homologous in the twogroups, that was implied by his hypothesisthat dendrobatids arose from ‘‘Crossodacty-lus or a form closely allied to it’’. He furtherobserved (p. 520) that both dendrobatids andthe African ranid Petropedetinae had ‘‘ap-parently identical’’ dermal scutes on theupper surface of each digit (he did notcomment on the shared occurrence of thisstate in his Elosiinae), but he explained awaythis similarity as adding ‘‘one more to themany cases of parallel evolution in theSalientia’’. Liem and Hosmer (1973: 473)also noted that the myobatrachid genusTaudactylus has ‘‘expanded digital discs witha median longitudinal groove dorsally’’.Lynch (1979) illustrated the discs of groupsknown to possess digital scutes or scutelikestructures. Lynch (1979: 7) clarified that thescutes are ‘‘flaplike structures’’, which distin-guishes them from superficially similar digitsof some Eleutherodactylus that exhibit onlya median groove. La Marca (1995: fig. 9)provided scanning electron micrographs ofthe digital scutes of collaris, herminae, ob-litterata, neblina, olmonae, riveroi, trinitatis,yustizi, and an undescribed species. Griffiths(1959: 482) claimed that the scutes are ‘‘reallyglandulo-muscular organs and probablyfunction to facilitate adhesion to foliageetc.’’, but no evidence has been presented insupport of his thesis and their functionalsignificance remains unknown.

2. SUPERNUMERARY TUBERCLES ON HAND:absent 5 0; present 5 1.

Most dendrobatids possess a large, sub-circular palmar tubercle and an ellipticalthenar tubercle. Many nondendrobatids alsopossess distinct supernumerary tubercles scat-tered over the fleshy part of the palm (e.g.,Lynch and Duellman, 1997). As part of theirpolymorphism, some dendrobatids exhibita tiny tubercle-like thickening on the outeredge (not the fleshy part) of the palm, but wedo not consider this to be homologous withthe supernumerary tubercles of other taxa.

3. DISTAL TUBERCLE ON FINGER 4 (fig. 23):absent 5 0; present 5 1.

Most dendrobatids possess both proximaland distal subarticular tubercles on finger IV(state 1). Grant and Rodrıguez (2001) notedthat in some species the distal tubercle onfinger IV is absent (state 0) and that,although this is often associated with re-duction in the length of finger IV (Character4), some species that lack this tubercle showno reduction in finger length (e.g., melano-laemus, pumilio), which demonstrates thetransformational independence of the twocharacters. This independence is furtherreinforced by examination of outgroup taxa,as Thoropa miliaris possesses a long finger IVand lacks the distal subarticular tubercle.

Fig. 23. Character 3, distal subarticular tuber-cle of finger IV. A: State 0, absent (degranvillei,AMNH 90876). B: State 1, present (pictus,AMNH 79209).


4. FINGER IV LENGTH (fig. 24): surpassingdistal subarticular tubercle of finger III 5 0;reaching distal half of distal subarticulartubercle of finger III 5 1; not reaching distalsubarticular tubercle of finger III 5 2.Additive.

The length of finger IV is assessed bypressing it against finger III to determine if itextends well beyond the distal subarticulartubercle (state 0), reaches the distal half of,but does not surpass, the distal subarticulartubercle (state 1), or does not reach the distalsubarticular tubercle (state 2). In the latterstate, finger IV extends to a point approxi-mately midway between the proximal anddistal subarticular tubercles Although it ispossible that we have conflated transforma-tions involving the length of finger III, thefact that finger II reaches the distal half of thedistal subarticular tubercle in all speciessupports indirectly the hypothesis that vari-ation is due exclusively to transformations offinger IV, that is, if the observed variation isdue to changes in the length of finger III,then the same change would also have had toaffect the relative length of finger II. And it isfurther supported by the loss of the distalsubarticular tubercle in species with a rela-tively short finger IV (see Character 3,above). Given the constancy of the lengthof finger II, this character is equivalent to the

traditional taxonomic coding of finger IVversus finger II (i.e., when both are pressedagainst finger III, in state 0 IV is longer thanII, in state 1 fingers IV and II are equal, andin state 2 IV is shorter than II).

5. RELATIVE LENGTHS OF FINGERS I AND

II: I ,, II (II 1.2 or more times longer thanI) 5 0; I , II 5 1; I 5 II 5 2; I . II 5 3.Additive.

Traditionally, the relative lengths of fin-gers I and II have been assessed by pressingthese two fingers together at the point mid-way between the two digits. However, this ishighly dependent on the investigator’s judg-ment of the midway point between the twodigits, that is, bringing finger I further towardfinger II (or vice versa) can affect coding ofthis character. Kaplan (1997) measured thelength of each finger from the same point atthe base of the palmar tubercle to the tip ofeach finger, which is more precise and lessprone to error, and we employed his methodhere. Kaplan’s method also assumes thatthere are no carpal changes that affect thedistance from the palm to finger tips differ-entially (no such variation was detected). Anymethod of measuring finger length requiresthat the fingers be straight; when well-preserved hands were unavailable, digits werestraightened for measurement. In state 0 fin-ger II is at least 20% longer than finger I; instate 1 finger II is less than 15% longer thanfinger I; in state 2 the fingers are subequal inlength; in state 3 finger I is unambiguouslylonger than finger II.

Although developmental data are unavail-able, gross morphology suggests that statetransformations are due to variation in thelength of finger I and not the length of fingerII, that is, the length of finger II relative tofinger III was not observed to vary, as itreaches the midlevel of the distal subarticulartubercle in all taxa. However, it is possiblethat two characters have been conflated, thatis, one involving variation in the length offinger I, the other variation in the length offinger II. We did not attempt to relate thedifferences in relative lengths with the un-derlying osteology, which could also revealthat multiple characters have been conflated.

6. DIGITAL DISCS: absent 5 0; present 5 1.

The differentiation of the digital termina-tions into expanded discs with adhesive pads

Fig. 24. Character 4, length of finger IV. A:State 0, surpassing distal subarticular tubrcle offinger III (histrionicus, AMNH 88259). B: State 1,reaching distal half of distal subarticular tubercle offinger III (tricolor, USNM 286082). Note also thestrong preaxial swelling of finger III. C: State 2, notreaching distal subarticular tubercle of finger III(insperatus, KU 149684). Note also the absence ofthe distal subarticular tubercle of finger IV.


has long been used to infer anuran relation-ships (e.g., Cope, 1867). Numerous authors(e.g., Noble and Jaeckle, 1928; Green, 1979;Emerson and Diehl, 1980; Rivero et al.,‘‘1987’’ 1989; Ba-Omar et al., 2000) haveexamined the structure (and function) of thedisc apparatus in a diversity of frogs andhave found them to differ in only minorstructural details (e.g., number of epidermalcell layers). We are unaware of any anuranthat possesses finger discs but lacks toe discs(or vice versa), or that possesses discs onsome but not all digits (although degree ofexpansion certainly varies among digits; seebelow). We have therefore treated the origin(and loss) of digital discs as a single trans-formation series. All dendrobatids possessdigital discs, but they are absent in several ofthe sampled outgroup taxa.

7–10. EXPANSION OF FINGER DISCS (fig. 25)

In the dendrobatid literature, expansionof finger discs is generally treated as one orat most two characters. Duellman andSimmons’s (1988) standard diagnosis codedonly the disc of finger III, and Dendrobatesspecies descriptions often report the expan-sion of discs I and II–IV separately. Howev-er, there is no logical dependency betweenthe discs of different digits, and the distribu-tion of states in the matrix shows that theexpansion of each digital disc is transforma-tionally independent, and, as such, they aredefensibly coded separately for analysis. Atrend is that the disc of finger I is often (butnot always) less expanded than those of the

other fingers, but this does not violate thetransformational independence of these char-acters.

We detected four discrete states in finger(and three in toe) disc expansion, shownschematically in figure 25. All dendrobatidshave digital discs, so some degree of expan-sion is always detectable, although it may beextremely slight. This is exemplified byelachyhistus and pumilio, in which the discof finger I appears unexpanded or at mostweakly expanded (state 0). States 1 and 2 arefound in most dendrobatids; state 3 is foundin those species with greatly expanded discs(e.g., tinctorius). State 3 was only observedamong fingers II–IV.

Polder (1973: 17) and Silverstone (1975a)claimed that some species of dendrobatidsare sexually dimorphic in the expansion ofthe digital discs, with males possessing largerdiscs than females. Neither author providedquantitative data, however, and when Myersand Daly (1976b: 203) tested the claimquantitatively in histrionicus they found it tobe unsupported. Although we detected (andcoded) polymorphism in disc expansion insome species (including histrionicus), weconcur with Myers and Daly (1976b) that itdoes not reflect differences between sexes.For example, in leucomelas the finger discs ofmale AMNH 137309 are larger than those offemale AMNH 137310, but no more expand-ed than those of female AMNH 46051. Wedid not test the hypothesis that discs of malesare statistically (i.e., on average) larger thanthose of females (Silverstone, 1975a: 8)because that question is unrelated to theproblem of homologizing character-statesand inferring transformation events (Grantand Kluge, 2003, 2004).

7. FINGER DISC I: unexpanded 5 0; weaklyexpanded 5 1; moderately expanded 5 2.Additive.

8. FINGER DISC II: unexpanded 5 0;weakly expanded 5 1; moderately expanded5 2; greatly expanded 5 3. Additive.

9. FINGER DISC III: unexpanded 5 0;weakly expanded 5 1; moderately expanded5 2; greatly expanded 5 3. Additive.

10. FINGER DISC IV: unexpanded 5 0;weakly expanded 5 1; moderately expanded5 2; greatly expanded 5 3. Additive.

Fig. 25. Characters 7–10 and 31–35, schematicillustration of the four states observed in theexpansion of digital discs. The digital shaft isindicated by the inner crosshatched area and theouter edges of the disc are indicated by the heavierouter line.


11–18. FINGER FRINGES (fig. 26)

The occurrence and extent of lateral keelsand/or fringes on fingers and toes has beencited in most alpha taxonomic studies ofdendrobatids for the past several decades atleast (e.g., Edwards, 1971). Duellman andSimmons (1988: 116) noted that ‘‘the de-velopment of fringes on the fingers is vari-able, so standard comparison is made withthe second finger.’’ Nevertheless, as discussedabove in reference to expansion of digitaldiscs, because the fringes on each edge ofeach finger vary independently we codedthem as separate characters.

Although lateral dermal expansions of thedigits are commonly described in the den-drobatid literature, explicit delimitations ofcharacter-states are generally lacking, whichhas led to considerable confusion. They aregenerally referred to as either keels or fringes.As noted by Lynch and Duellman (1997: 33)for species of Eleutherodactylus, ‘‘there isa continuum from keels to fringes, and in

some cases the distinction is arbitrary.’’Although such arbitrariness is relativelyharmless in descriptive taxonomic studies,the cumulative effect of arbitrary delimita-tions can be disastrous in phylogeneticanalyses. Coloma (1995: 6–7) noted thatfringes may be absent, poorly developed, orwell developed. He further clarified that‘‘[w]hen it was difficult to distinguish betweena real fringe and a preservation artifact, Idescribe the dermal modification as a ‘keel’ ’’,which, although explicit, actually engendersgreater confusion because keels are generallyconsidered to be real dermal modifications(e.g., Lynch and Duellman, 1997).

The strength of keeling (extent of dermalthickening) varies extensively, leading LaMarca (1996 ‘‘1994’’: 6) to differentiatebetween keels and fringes as ‘‘very low’’ and‘‘conspicuous but not folding around thetoes’’, respectively. However, although weagree that these descriptors encompass theobserved variation, and despite numerousdissections, we were unable to individuatecharacter-states objectively. Any attempt tosubdivide state 0 into multiple character-states must overcome two difficulties: (1)apparently continuous variation (as sug-gested by external examination and grossdissections) and (2) the fact that these dermalexpansions are highly prone to postmortemmodification, either due to desiccation (asindicated by Coloma, 1995) or simply as anartifact of preservation (and variation inpreservation techniques). It is likely that thegreater precision attained through histologi-cal study could overcome both of theseproblems, but that was beyond the scope ofthe present study. For the purpose ofphylogenetic analysis, we individuated onlytwo character-states: fringes absent (state 0)and fringes present (state 1).

In state 0, the extent of lateral dermalexpansion varies from absent (i.e., the side ofthe digit is smoothly rounded and there is nodetectable dermal thickening along the lateralmargin) to conspicuously keeled. In state 1,the skin that extends from the dorsal surfaceextends ventrad and appears to fold over theside of the digit, which we refer to as a fringe(see fig. 26). In ventral (palmar) view thefolding over can be seen to create a deeplongitudinal crease or groove. We have not

Fig. 26. Characters 11–18, finger fringes. InMegaelosia goeldii (AMNH 103949) fringes arepresent on pre- and postaxial edges of all fingers.


detected evidence that the folding over variesas an artifact of preservation, providinga basis to distinguish this state objectively.This state is approximately equivalent to LaMarca’s (1996 ‘‘1994’’: 6) ‘‘flaps’’, which hediagnosed as ‘‘folding around the toes’’. Thestrength of fringes varies from a weak flap(e.g., toes of degranvillei) to a strong flap thatwraps around much of the ventral surface ofthe digit (the latter condition found only ontoes; see below), but we were unable todelimit distinct states.

Webbing between the fingers does notoccur in any dendrobatid we examined.Donoso-Barros (1965 ‘‘1964’’: 486) described‘‘rudimentary web between 2nd and 3rd

fingers’’ in riveroi, but finger webbing wasnot reported by La Marca (1996 ‘‘1994’’) andis absent in the specimens we examined.Similarly, Coloma (1995) described andillustrated webbing between the fingers inan undescribed species (as Colostethus cho-coensis; see Grant et al., 1997: 24, footnote13), and Grant et al. (1997: 25) mentioned thepossible occurrence of webbing on the handsof atopoglossus. However, closer examinationof the same specimens of ‘‘Colostethuschocoensis’’ and atopoglossus revealed thatthe apparent webbing is due to flattening ofthe loose skin of the hand, as considered byGrant et al. (1997). Lynch (1971: 30) reportedsimilar mistaken reports among ‘‘leptodacty-loids’’.

11. FINGER FRINGE: I preaxial: absent 5 0;present 5 1.

12. FINGER FRINGE: I postaxial: absent 5

0; present 5 1.13. FINGER FRINGE: II preaxial: absent 5

0; present 5 1.14. FINGER FRINGE: II postaxial: absent 5

0; present 5 1.15. FINGER FRINGE: III preaxial: absent 5

0; present 5 1.16. FINGER FRINGE: III postaxial: absent

5 0; present 5 1.17. FINGER FRINGE: IV preaxial: absent 5

0; present 5 1.18. FINGER FRINGE: IV postaxial: absent

5 0; present 5 1.19. METACARPAL RIDGE (fig. 27): absent 5

0; weak 5 1.The metacarpal ridge or fold is a dermal

thickening running from the postaxial edge of

the base of finger IV along the outer edge ofthe palm toward the palmar tubercle. In mostspecies an edge is formed where the relativelyflattened palm meets the rounded side of thehand, but we did not consider this to bea metacarpal ridge unless dermal thickeningcould be detected, either by gross inspectionor by making a transverse incision. Althoughthere is some variation among species in thedegree of expression of the metacarpal ridge,it was minor and we were unable to delimitdiscrete states. As with other dermal char-acters, the metacarpal ridge may be exagger-ated or lost as an artifact of preservation.

20–21. FINGER III SWELLING

Reproductively active males of numerousdendrobatids present swollen third fingers,a condition that is unknown in nondendro-batids. The ‘‘swelling’’ is due to the occur-rence of extensive glandular tissue, the largegranules often being evident in gross dissec-tion or even through the skin. In light of theimportant role this character has played inrecent discussions of dendrobatid systematics(e.g., Myers et al., 1991; but see Myers, 1991),we review its usage here.

Although a number of species possessinga distinctly enlarged third finger in males hadbeen described previously (e.g., trilineatus,the holotype of which is a male), the firstworker to describe and illustrate the swollenthird finger was Dunn (1924: 7–8) fornubicola. Descriptions of ‘‘digital dilatations’’or ‘‘enlargements’’ in the earlier literaturereferred to the expanded digital disc appara-

Fig. 27. Character 19, metacarpal ridge. State1, present (abditaurantius, ICN 9853).


tus (e.g., Cope, 1867: 130, 1887: 55). WhenDunn (1931) named flotator, he grouped itwith nubicola based in part on the sharedoccurrence of the swollen third finger inmales. Dunn (1933) noted that males ofpanamensis possess a swollen third finger,but he did not attribute any phylogeneticsignificance to the observation.

Over the 50 years following Dunn’s firstreport of the swollen third finger in males, thestate of the third finger was mentionedsporadically in diagnoses and descriptions(e.g., among papers that deal with specieswith swollen third fingers in males, it wasmentioned by Dunn, 1931, 1957; Funkhou-ser, 1956; Savage, 1968; Cochran and Goin,1970: 60 [only for their Phyllobates inguinalis,as ‘‘flanges’’ on the third finger of males];Edwards, 1971; and Silverstone, 1971, 1976;but not by Cochran and Goin, 1964;Cochran, 1966; or Silverstone, 1975b), butwas not illustrated again until 1974, whenEdwards provided a schematic representationin his unpublished (but widely distributed;see Myers et al., 1991: 30, footnote 14)dissertation (Edwards, 1974a).

The character has been mentioned fairlyconsistently since 1974, but miscoding iscommon, probably due in part at least tothe inadequacy of Dunn’s (1924) and Ed-wards’s (1974a) illustrations, both of whichdepicted (1) roughly equal expansion on bothsides (preaxial and postaxial) of the digit and(2) distally exaggerated swelling, neither ofwhich is found in all (or even most) of thespecies with swollen third fingers. Similarly,although accurate for a few species, Duell-man and Simmons’s (1988: 117) descriptionthat ‘‘the basal segment of the third finger isdistinctly swollen in males’’ does not apply tomost of the species with clearly swollen thirdfingers in males (and none of the species theyaddressed in their paper). The expansion ofthe third finger can be much more subtle thanDunn’s (1924) and Edwards’s (1974a) illus-trations suggest, and significant variationoccurs in the extent and location of theswelling.

Silverstone (1976: 33) noted in his accountof tricolor that not all adult males of a givensample may express the swollen third finger,a finding that was corroborated more gener-ally by Myers et al. (1991), who speculated

that expression is likely under hormonalcontrol. This and additional difficulties re-lated to the coding of this character werediscussed by Myers et al. (1991, 1998, seeespecially fig. 4), Myers (1991), Myers andDonnelly (1997), and Grant and Rodrıguez(2001).

20. FINGER III SWELLING IN ADULT

MALES: absent 5 0; present 5 1.

This character was scored for awa fromColoma (1995) because no adult males wereincluded in the series we examined.

21. MORPHOLOGY OF SWOLLEN THIRD

FINGER IN MALES (fig. 28): pre- and postaxialswelling 5 0; weak preaxial swelling 5 1;strong preaxial swelling 5 2; swelling extend-ing from wrist, mainly preaxial on digit 5 3.Nonadditive.

22. CARPAL PAD (fig. 29): absent 5 0;present 5 1.

Myers and Donnelly (2001) discovered thecarpal pad in undulatus. It consists ofa conspicuous nonglandular thickening andheavy melanosis of the skin above the wristof males. We did not find this character to bepresent in any other species, but we include ithere in anticipation of future discoveries.

23. MALE EXCRESCENCES ON THUMB:absent 5 0; present 5 1.

Although nuptial excrescences are com-mon among outgroup taxa, most dendroba-tids lack nuptial excrescences (state 0), thesole exception being oblitterata, which wasreported as possessing nuptial excrescences(state 1) by La Marca (1995: 66).

We coded Telmatobius jahuira for thischaracter following Lavilla and Ergueta(1995).

24. MORPHOLOGY OF MALE EXCRESCENCES

ON THUMB: large, cornified spines 5 0; small,uncornified spines 5 1; nonspinous asperities5 2. Additive.

Lavilla and Ergueta (1995: 49, translatedfreely from the Spanish) described the nuptialexcrescences of Telmatobius jahuira as ‘‘fewcornified spines separated by large, unker-atinized spaces’’.

25. FEMALE EXCRESCENCES ON THUMB:absent 5 0; present (large, cornified spines) 5

1.

See Noble (1931: 122, 126) for illustrationsand comments on the large, cornified spines


on the thumb of females of species ofCrossodactylus.

26. THENAR TUBERCLE (fig. 30): absent orsmall, inconspicuous swelling 5 0; large,conspicuous, well-defined tubercle 5 1.

Most dendrobatids have a conspicuous,protuberant, elliptical thenar (inner meta-carpal) tubercle (state 1). Silverstone (1975a)noted that the thenar tubercle is inconspicu-ous or absent in leucomelas (state 0). Like-wise, Caldwell and Myers (1990) illustratedand discussed the absence of the thenartubercle in castaneoticus and quinquevittatus,which they interpreted as a synapomorphyuniting these two species in an exclusive clade(they did not make comparisons with leuco-melas). Other species also exhibit the samemorphology (e.g., pumilio).

Caldwell and Myers (1990: 16) noted somevariation in the expression of the thenartubercle in quinquevittatus; in some specimensit is altogether undetectable, whereas inothers ‘‘possible vestiges of it’’ were detectedas ‘‘possibly represented by slight epidermalthickening’’. Our observations concur withtheirs. Given the propensity for such subtledermal features to be lost as an artifact ofpreservation (due to skin sloughing, desicca-tion, or inadequate fixation, among othercauses), we combined the apparent completeabsence and inconspicuous epidermal thick-ening as state 0. Expression of the thenartubercle is not dependent on overall bodysize; leucomelas is quite large, and the thenartubercles of the small species nubicola andstepheni (roughly the same snout-vent lengthas pumilio) are large and well defined.

27. BLACK ARM GLAND IN ADULT MALES:absent 5 0; present 5 1.

This character was identified, discussed,and illustrated photographically by Grantand Castro-Herrera (1998; see also Grantand Ardila-Robayo, 2002) and used todelimit the ramosi group. It remains unclearif this patch of black, thickened tissue on the

Fig. 28. Character 21, morphology of swollenthird finger in males. A: State 0, pre- and postaxialswelling (mertensi, ICN 43698). B: State 1, weak

r

preaxial swelling (insperatus, KU 149676). C: State2, strong preaxial swelling (nubicola, AMNH114574). D: State 3, swelling extending fromwrist, mainly preaxial on digit (baeobatrachus,AMNH 140650).


ventral and medial surfaces of the distalextreme of the upper arm and often extend-ing onto the inner surface of the lower arm isglandular, but its absence in females andjuveniles and exaggeration in sexually activemales suggests it is involved in amplexus andprobably under hormonal control. In addi-tion to the species listed by Grant and Ardila-Robayo (2002), this character is also presentin anthracinus and the undescribed speciesreferred to herein as ‘‘Ibague’’.

28. TARSAL KEEL: absent 5 0; present 5 1.

The tarsal keel is a dermal structure thatextends obliquely along the plantar surface ofthe tarsus. Regardless of its point of origin(see Character 29), it always terminatesmedially, not on the margin of the tarsus(see Character 30). Silverstone (1975a, 1976)used variation in this structure to diagnosespecies groups in Dendrobates and Phyllo-bates, and Lynch (1982) cited the loss of thetarsal keel in edwardsi and ruizi to delimit theedwardsi group of Colostethus.

Silverstone (1975a: 8) treated the ‘‘tarsalfold’’ and ‘‘tarsal tubercle (at the proximalend of the tarsal fold)’’ as separate charac-ters. He considered the tarsal fold to bepresent in all Dendrobates and the tarsaltubercle to be both present and absent in

Fig. 29. Character 22, male supracarpal pad(undulatus, AMNH 159134).

Fig. 30. Character 26, thenar tubercle. A, B: State 0, absent or small, inconspicuous swelling (pumilio,AMNH 102262). In this specimen, the thenar tubercle appears absent in both palmar aspect and profile. C,D: Another specimen of the same species (pumilio, AMNH 102263). In this specimen, the thenar tubercle isinconspicuous but clearly seen in profile (also scored as state 0). E, F: State 1, large, conspicuous,protuberant tubercle (nubicola, AMNH 114574).


Dendrobates. However, the tarsal fold andtarsal tubercle form a single structure, thetubercle simply being an increased thickeningof the proximal portion of the keel. This isespecially clear in many nonaposematicdendrobatids (which were not the focus ofSilverstone’s work) in which the proximalend of the keel is conspicuously enlarged andmay be described as tubercle-like, but issharply curved to run across the tarsus anddoes not conform to the rounded structuresusually referred to as tubercles.

29. MORPHOLOGY OF TARSAL KEEL

(fig. 31): straight or very weakly curved,extending proximolaterad from preaxial edgeof inner metatarsal tubercle 5 0; tubercle-like(i.e., enlarged) and strongly curved at prox-imal end, extending from metatarsal tubercle

5 1; short, tubercle-like, curved or directedtransversely across tarsus, not extendingfrom metatarsal tubercle 5 2; weak, shortdermal thickening, not extending from meta-tarsal tubercle 5 3. Additive.

30. TARSAL FRINGE (fig. 32): absent 5 0;present 5 1.

The tarsal fringe consists of a conspicuousdermal flap that runs along the entire lengthof the preaxial edge of the tarsus; it iscontinuous with the fringe on toe I. Thetarsal fringe differs from the tarsal keel(characters 28, 29) in that the latter extendsproximolaterad across the tarsus to terminateat roughly the middle of the tarsus on theplantar surface, whereas the former nevercrosses the tarsus and extends along its entirelength.

Fig. 31. Character 29, morphology of tarsal keel. A: State 0, straight or very weakly curved, extendingproximolaterad from preaxial edge of inner metatarsal tubercle (imbricolus, AMNH 102082). B: State 1,tuberclelike and strongly curved at proximal end, extending from metatarsal tubercle (degranvillei, AMNH90876). C: State 2, short, tuberclelike, curved or directed transversely across tarsus, not extending frommetatarsal tubercle (‘‘Neblina species’’, AMNH 118657). D: State 3, weak, short dermal thickening, notextending from metatarsal tubercle (pumilio, AMNH 102261). The hind limb is rotated to view theinconspicuous tarsal keel in profile.


31–35. EXPANSION OF TOE DISCS

Like finger discs, dendrobatid literaturegenerally treats the degree of expansion oftoe discs as a single character. However, asdiscussed above under finger discs, toe discsvary independently of one another and aredefensibly treated as separate characters. Toediscs exhibit three of the four character-statesfound in fingers; the greatest expansionfound in finger discs (finger disc state 3) doesnot occur in toe discs. (Character-states arefigured schematically in fig. 25, above.)

31. TOE DISC I: unexpanded 5 0; weaklyexpanded 5 1; moderately expanded 5 2.Additive.

32. TOE DISC II: unexpanded 5 0; weaklyexpanded 5 1; moderately expanded 5 2.Additive.

33. TOE DISC III: unexpanded 5 0; weaklyexpanded 5 1; moderately expanded 5 2.Additive.

34. TOE DISC IV: unexpanded 5 0; weaklyexpanded 5 1; moderately expanded 5 2.Additive.

35. TOE DISC V: unexpanded 5 0; weaklyexpanded 5 1; moderately expanded 5 2.Additive.

36–45. TOE WEBBING

Webbing has been used consistently indendrobatid systematics since Noble (1923,1926) diagnosed Phyllobates from Hyloxaluson the basis of reduced webbing. Althoughwebbing can be argued to form a single,integrated functional unit (as can the entireorganism), functional independence is atmost secondary to historical independence

in phylogenetic inference (Grant and Kluge,2004), and there is ample evidence that theextent of webbing along each edge of eachdigit varies independently. Coding followsthe nomenclature proposed by Savage andHeyer (1967) and subsequently modified byMyers and Duellman (1982), which quanti-fies webbing in terms of the number of freephalanges, assessed in relation to subarticulartubercles (e.g., in Character 40, state 6, thetwo distal phalanges are free of webbing). Weconsider toe fringes (defined as for fingers,above) to be homologous with webs. We donot consider lateral fringes that meet betweenthe toes to constitute a web unless it isexpanded relative to lateral fringes, that is, ifthe continuous lateral fringes are broader atthe base than along the sides of the digits, weconstrue this as being a web.

Among the sampled outgroup taxa,McDiarmid (1971: 33) noted that the inter-digital webbing of Atelopus and Dendrophry-

niscus ‘‘is not a membrane, as defined byPeters (1964) but rather a thickened integu-mentary connection between digits, similar tothe webbing encountered in many of themore terrestrial anurans, such as toads of thegenus Bufo.’’ This suggests that the inter-digital webbing of these species may not behomologous with that of other anurans.Nevertheless, although the distinction is clearin Dendrophryniscus minutus, it is less so inthe sampled species of Atelopus, and we havetherefore treated webbing as a single trans-formation series and allowed character con-gruence to be the ultimate arbiter (althoughmuch more extensive sampling of relevanttaxa will be required to fully resolve thequestion).

36. WEBBING: TOE I PREAXIAL: absent 5 0;fringe 5 1.

37. WEBBING: TOE I POSTAXIAL: absent 5

0; fringe 5 1; 2 5 2; 1.5 5 3; 1 5 4; 0 5 5.Additive.

Coloma (1995: 51) reported basal webbing(I2–3.5II) for talamancae and toachi, butthere is no trace of webbing in the specimenswe examined in this study.

38. WEBBING: TOE II PREAXIAL: absent 5

0; 2.5 5 1; 2 5 2; 1 5 3; 0 5 4. Additive.

Coloma (1995: 51) reported basal webbing(I2–3.5II) for talamancae and toachi, but

Fig. 32. Character 30, tarsal fringe. State 1,present (Megaelosia goeldii, AMNH 103950).


there is no trace of webbing in the specimenswe examined.

39. WEBBING: TOE II POSTAXIAL: absent 5

0; 2 (without fringe) 5 1; 2 (with fringe) 5 2;1.5 5 3; 1 5 4; 0 5 5. Additive.

40. WEBBING: TOE III PREAXIAL: absent 5

0; fringe 5 1; 3.5 (without fringe) 5 2; 3.5(with fringe) 5 3; 3 5 4; 2.5 5 5; 2 5 6; 1.5 5

7; 1 5 8. Additive.

Coloma (1995: 51) reported more extensivewebbing (equivalent to state 4) for talaman-cae than we observed (state 2).

41. WEBBING: TOE III POSTAXIAL: absent5 0; 3 without fringe 5 1; 3 with fringe 5 2;2.5 5 3; 2 5 4; 1.5 5 5; 1 5 6. Additive.

42. WEBBING: TOE IV PREAXIAL: absent 5

0; 4 without fringe 5 1; 4 with fringe 5 2; 3.55 3; 3 5 4; 2.5 5 5; 2 5 6; 1 5 7. Additive.

43. WEBBING: TOE IV POSTAXIAL: absent 5

0; fringe 5 1; 4 5 2; 3.5 5 3; 3 5 4; 2.5 5 5; 25 6; 1 5 7. Additive.

Coloma (1995: 51) reported basal webbing(IV4.5–3V) in talamancae, but there is notrace of webbing in the specimens weexamined.

44. WEBBING: TOE V PREAXIAL: absent 5

0; fringe 5 1; 2.5 (with fringe) 5 2; 2 5 3; 1.55 4; 1 5 5. Additive.

45. WEBBING: TOE V POSTAXIAL: absent 5

0; fringe 5 1.

This character was coded for insulatus,pulcherrimus, and sylvaticus Barbour andNoble from Duellman (2004), who reportedit as absent in them all.

46. METATARSAL FOLD (fig. 33): absent 5

0; weak 5 1; strong 5 2. Additive.

The metatarsal fold is a dermal thickeningrunning from the postaxial edge of the baseof toe V (often coextensive with the fringe, ifpresent) along the outer edge of the soletoward the outer metatarsal tubercle. In mostspecies an edge is formed where the relativelyflattened sole meets the rounded side of thefoot, but we did not consider this to bea metatarsal ridge or fold unless actualdermal thickening could be detected, eitherby gross inspection or by dissection. A weakmetatarsal fold (state 1) is a ridge; strongdermal folds (state 2) are often folded over orangled relative to the surface of the sole.

47. CLOACAL TUBERCLES: absent 5 0;present 5 1.

Grant et al. (1997) identified and figuredthis pair of tubercles adjacent to the cloacanear the base of the thighs. They alsodiscussed difficulties in scoring this characterdue to postmortem artifacts.

48–66. EXTERNAL COLORATION

Much of the diversity of dendrobatidsinvolves variation in external color andcolor pattern. Among species referred toColostethus, for example, variation in thepattern of lateral stripes and ventral colorserves as one of the main tools for diagnosis.However, color and color pattern are per-haps the most confounding–and thereforeundersampled in this study–sources of vari-ation. Several aposematic dendrobatids (e.g.,pumilio, histrionicus, and tinctorius) are re-nowned for their astonishing intra- andinterpopulational variation in color andcolor pattern, and the difficulties posed bythis immense and often continuous andoverlapping variation can be immediatelyappreciated by glancing at a few pages ofMyers et al.’s (1976b) account of histrionicus.Practically speaking, the three main difficul-ties are (1) detection of objective boundariesbetween different characters and character-

Fig. 33. Character 46, metatarsal fold. A: State1, weak (‘‘Neblina species’’, AMNH 118657). B:State 2, strong (degranvillei, AMNH 90876).


states, (2) requirement of many states percharacter, and (3) distinguishing betweencolor and color pattern. We made everyeffort to incorporate as much of the variationas possible, but much of it was overwhelm-ing. Also, for ease of coding (especiallypreserved specimens) we focused more oncolor pattern than color, but by doing so weundoubtedly conflated characters and char-acter-states. For example, we scored bothauratus and reticulatus as having the thighspale with dark spots, even though the thighs

are different colors. Future studies willundoubtedly advance considerably beyondthe current project by scoring more of thisdiversity of color and color pattern.

48. IRIDESCENT ORANGE OR GOLDEN SPOT

AT DORSAL LIMB INSERTIONS: absent 5 0;present 5 1.

Note that the photo of quinquevittatus inCaldwell and Myers (1990: 11) shows thatthis character is not redundant with ornonindependent of the thigh colorationcharacters.

49. PALE PARACLOACAL MARK (fig. 34):absent 5 0; present 5 1.

This is a pale, elongate mark at the base ofthe thigh. The shape of the spot varies froma straight vertical line to a sickle extending asa pale longitudinal stripe along the posteriorsurface of the thigh. The paracloacal markoriginates adjacent to the vent at the base ofthe thigh, not in the groin or on the top thethigh (as does the pale mark in femoralis, forexample; see character 50).

50. THIGH DORSAL COLOR PATTERN

(fig. 35): pale with dark spots (formingreticulum when spots are close together) 5

0; solid dark 5 1; dark with pale spots/bands5 2; solid pale 5 3; brown with dark brown

Fig. 34. Character 49, pale paracloacal mark.State 1, present (degranvillei, AMNH 90880).

Fig. 35. Character 50, dorsal thigh color pattern. A: State 0, pale with dark spots (quinquivittatus,AMNH 124069). B: State 1, solid dark (petersi, AMNH 111000). Note that the pale spot is confined to theinguinal regions and does not extend onto the dorsal surface of the thigh. C: State 2, dark with pale spots/bands (aurotaenia, AMNH live exhibit). D: State 3, solid pale (terribilis, AMNH live exhibit). E: State 4,brown with dark brown bands/blotches (inguinalis, LACM 42409). F: State 5, dark with pale longitudinalstripe (flavopictus, AMNH 88642).


bands/blotches 5 4; dark with pale longitu-dinal stripe 5 5. Nonadditive.

51. DISCRETE PALE PROXIMOVENTRAL

CALF SPOT (fig. 36): absent 5 0; present 5 1.

Silverstone (1975a, 1975b) used the ab-sence (state 0) and presence (state 1) ofa discrete, pale spot on the proximal portionof the concealed surface of the shank todiagnose species and species groups. In life itis a bright flash mark. A number of species(e.g., fraterdanieli) have bright flash colora-tion on the concealed surface of the shank,but it does not form a discrete spot and wetherefore follow Silverstone in scoring thischaracter as absent for those species.

52–57. PALE LATERAL STRIPES

Edwards (1974a) used the combinations ofpale lateral stripes (or lines) to diagnosespecies of Colostethus, identifying dorsolat-eral, oblique lateral, and ventrolateral stripes.Previous workers (e.g., Savage, 1968) haddrawn attention to these characteristics aswell, but Edwards standardized the distinc-tion between the three stripes and has beenfollowed by most authors. Nevertheless,caution must be employed when consultingthe literature, as terminology varies. Forexample, what is referred to here as theoblique lateral stripe was referred to asa dorsolateral stripe by Edwards (1971) andHaddad and Martins (1994), and consistentlyas the inguinal stripe by La Marca (e.g., 1985,1996 ‘‘1994’’, 1998 ‘‘1996’’; see also Myersand Donnelly, 2001). Duellman and Sim-

mons (1988) discussed these characters as‘‘pale longitudinal stripes’’, and Coloma(1995) followed their usage. Duellman(2004) distinguished between the obliquelateral and dorsolateral stripes in his Sum-mary of Taxonomic Characters but usedthem interchangeably in the text (e.g., idio-melus and sylvaticus are diagnosed as lackingoblique lateral stripes and possessing dorso-lateral stripes, but the converse is true forboth species; e.g., see Duellman’s figs. 5Fand 6F).

Edwards (1974a) was concerned only withthe mostly cryptically colored dendrobatidsthen referred to Colostethus and not the moreconspicuously colored species referred toDendrobates and Phyllobates. The broadersample of the present study showed that thereare (at least) two distinct ‘‘dorsolateral’’stripes, which we have designated A (Char-acter 52) and B (Character 53), the latter alsohaving been confused previously with theoblique lateral stripe.

52. DORSOLATERAL STRIPE A (DOES NOT

DROP TO THIGH; fig. 37): absent 5 0; presentin juveniles only (i.e., lost ontogenetically) 5

1; anterior, narrow, faint 5 2; complete 5 3.Nonadditive.

This dorsolateral stripe runs posteriadfrom the eyelid toward the tip of the urostyle.It does not cross the flank toward the groin(oblique lateral stripe), nor does it drop to thetop of the thigh (dorsolateral stripe B). Myerset al. (1978) reported that in bicolor andterribilis the dorsolateral stripe is present injuveniles and lost ontogenetically (state 1).This ‘‘loss’’ is peculiar, however, as it is dueto the hypertrophy of the bright dorsolateralstripes that expand ontogenetically to coverthe entire dorsum, thus creating a uniformlycolored, stripeless color pattern.5 In state 2,the dorsolateral stripe is short, narrow, andinconspicuous (often more conspicuous injuveniles than adults), running from the

Fig. 36. Character 51, discrete pale proximo-ventral calf spot. State 1, present (imbricolus,AMNH 102082).

5In a captive breeding colony, A. Haas (in litt., 08/08/05) observed that ‘‘yellow coloration appearedbetween the stripes (i.e., not a hypertrophy of thestripes); thus the process was more like filling in thegap between the stripes’’, suggesting that thedorsolateral stripes are retained but concealed by thesurrounding coloration. Further investigation intothis character-state is warranted.


posterior edge of the eye to a point just pastthe insertion of the arm. When present, thedorsolateral stripe of most species is com-plete, reaching or surpassing the level of thesacrum, and persists in adults (state 3).

The ontogenetic loss of the dorsolateralstripe is suggestive of additivity (i.e., absent« present in juveniles only « presentthroughout ontogeny); however, given thepeculiarity of this particular ‘‘loss’’ throughexpansion, the additivity absent « presentthroughout ontogeny « present in juvenilesonly may be more appropriate. Regardless, itis unclear where state 2 would fit into thisseries, as there is no evidence that thedorsolateral stripe extends posteriorlythrough development, nor that state 2 is theresult of reduction from a complete dorso-lateral stripe. We therefore did not specify theadditivity of this transformation series.

53. DORSOLATERAL STRIPE B (DROPS TO

TOP OF THIGH, NOT GROIN; fig. 38): absent5 0; present 5 1.

This dorsolateral stripe extends posteriadfrom the eyelid along the dorsolateral edge ofthe body and turns abruptly ventrad ata position immediately anterior to the thigh.This stripe was considered to be dorsolateralby Silverstone (1975a) and Caldwell andMyers (1990) for quinquevittatus, but obliquelateral (‘‘lateral’’) by Silverstone (1976) forfemoralis. The confusion is understandable,as its path is intermediate between these twocharacters. Unlike the oblique lateral stripe,it does not run diagonally along the flanksbut remains dorsal until almost the level ofthe thigh, but unlike dorsolateral stripe A, itdrops toward the thigh posteriorly.

Fig. 38. Character 53, dorsolateral stripe B.State 1, present (femoralis, AMNH 140646).

Fig. 37. Character 52, dorsolateral stripe A. A,B: State 1, present in juveniles (A), absent in adults(B) (terribilis, A: captive-raised specimen; B:AMNH live exhibit). C: State 2, anterior, narrow,

r

faint (atopoglossus, holotype UVC 12068). D: State3, complete (aurotaenia, AMNH live exhibit).


54. VENTROLATERAL STRIPE (fig. 39): ab-sent 5 0; wavy series of elongate spots 5 1;straight 5 2. Nonadditive.

The ventrolateral stripe (VLS) runs alongthe ventral edge of the flank between thebelly and the usually dark coloration of theflank. It may be present as a wavy series ofelongate, often interconnected spots (state 1)or as a straight line (state 2). The ventrolat-eral stripe can be difficult to detect inpreserved specimens, even those in whichthe ventrolateral stripe was prominent in life,because of the degradation of iridophores,especially in taxa with fairly pale ventralsurfaces. In some of these cases the ventro-lateral stripe can be detected as a lack ofmelanophores. However, the iridophores

break down fairly quickly in preservative,often revealing a deeper layer of underlyingmelanophores invisible in living or freshlypreserved specimens.

Coloma (1995: 47–48) reported that somespecimens of pulchellus have ‘‘an interruptedwhite ventrolateral line’’, but we did notobserve this in the specimens examined.Caldwell and Lima (2003) reported theventrolateral stripe as absent and describedthe holotype of nidicola as having ‘‘irregularwhite blotches, not forming a stripe’’. How-ever, a wavy VLS is evident in the photo-graph shown in their figure 3B (gravidfemale). Among the trivittatus specimensexamined, the ventrolateral stripe is presentin all specimens from Suriname but absent inall but one of the specimens from Peru(AMNH 43204, in which it is a series ofsmall elongate spots on the left and a single,large elongate spot on the right).

55. OBLIQUE LATERAL STRIPE: absent 5 0;present 5 1.

The pale oblique lateral stripe extendsfrom the groin diagonally across the flankstoward the eye.

56. OBLIQUE LATERAL STRIPE LENGTH

(fig. 40): partial 5 0; complete 5 1.Edwards (1974b) distinguished oblique

lateral stripes (OLS) that extend from thegroin part-way to the eye (partial, state 0) orall the way to the eye (complete, state 1).There is some individual variation in theanterior extension of the partial OLS, but itusually terminates prior to and does notextend past the level of the insertion of thearm. There is no evidence that the stripedevelops from one end to the other, which iswhy we did not combine length with pres-ence/absence as an additive multistate char-acter (i.e., absent « partial « complete).

Edwards (1974b: 10) described the OLS ofsauli as incomplete, which is supported byboth his painting (p. 6) and the color plate ofthe same specimen in Coloma (1995: plate1A). However, Coloma (1995) explicitlycompared sauli only to those species havinga complete oblique lateral stripe, and we havealso observed it to be complete. We thereforescored this character as polymorphic.

57. OBLIQUE LATERAL STRIPE STRUCTURE

(fig. 41): solid 5 0; series of spots 5 1; diffuse5 2. Nonadditive.

Fig. 39. Character 54, ventrolateral stripe. A:State 1, wavy series of elongate, interconnectedspots (espinosai, AMNH 104875). In this specimenthe spotting forms a fairly contiguous wavy stripe,but it is common for the elongate spots to beseparated, forming a broken stripe. B: State 2,straight (talamancae, AMNH 69829, photo by R.Zweifel). Note also that the pale dorsolateral stripedoes not drop toward the thigh posteriorly(Character 52, state 3).


The oblique lateral stripe (OLS) of mostspecies consists of a solid line of palepigmentation (e.g., nubicola; state 0). Lynchand Ruiz-Carranza (1985) identified state 1(series of well-defined spots) in agilis, andMyers et al. (1991: 2, 3, figs. 1, 3) illustrated itphotographically for nocturnus. Grant andRodrıguez (2001) discussed variation in thischaracter and described and illustrated pho-tographically state 2. As shown in Grant andRodrıguez (2001: 9, fig. 6), state 2 may alsoinclude spots, but they are smaller, lessdistinct, and arranged irregularly (not ina line).

58. GULAR–CHEST MARKINGS (fig. 42):absent 5 0; present 5 1.

A number of species from the Andes ofsouthern Colombia, Ecuador, and Perupossess highly variable dark spots or blotcheson the posterolateral portion of the gular–chest region. Myers et al. (1991) comparedthese markings with the collars of several

Venezuelan species and considered the possi-bility that they may be homologous. Wecode them as different transformations serieshere, the difference being that the gular–chest markings are always separated mediallyand do not form a continuous transverseband.

Coloma (1995: 10) reported several var-iants in the shape and pattern of the gular–

Fig. 40. Character 56, oblique lateral stripelength. A: State 0, partial (panamensis, AMNH69836, photo by R. Zweifel). B: State 1, complete(fraterdanieli, MHNUC 364).

Fig. 41. Character 57, oblique lateral stripestructure. A: State 0, solid (pulchripectus, AMNH137290). B: State 1, series of spots (mertensi, ICN43698). C: State 2, diffuse (trilineatus, AMNH171974).


chest markings. Much of this variation isintraspecific, and Coloma reported ontoge-netic changes. Consequently, until this vari-ation is better understood, we treated all ofthese variants as homologous and subsumedtheir occurrence within a single character-state. Although Coloma (1995: 10) discussedthem in the same context, the markings onthe mental region and the pair of spots on theposterior chest do not occur in the sameregion, and we did not treat them as part ofthis transformation series.

Coloma (1995) reported the presence ofdiffuse bandlike markings in bocagei, but itwas absent from all the specimens weexamined.

59. DARK DERMAL COLLAR (fig. 43):absent 5 0; present 5 1.

The dermal collar (‘‘chest markings’’ of LaMarca, 1995) is a continuous transverse bandof dark pigmentation that extends across theposterior throat, anterior to the arms. Al-though La Marca (1996 ‘‘1994’’) reportedsexual variation in its occurrence, we ob-served it to be present in adults of both sexesof all species that possess the dermal collar(although we did observe polymorphism inmales of neblina), so we did not code malesand females as separate semaphoronts.

Rivero (1978 ‘‘1976’’: 330; translated fromthe Spanish) noted that ‘‘[i]n almost allspecimens [of leopardalis] a faint dark collarmay be detected, never as clear and welldefined as in C. collaris, and generallyconfined to the sides of the throat.’’ Similar-ly, Myers et al. (1991) noted the occurrenceof faint collarlike pigmentation on the throatof many specimens of nocturnus. Closerexamination and dissection revealed that

Fig. 42. Character 58, markings on gular–chestregion, state 1 (present). A: Diffuse, white-spottedblotches (awa, AMNH 111542). B: Discrete, smalldark spots (vertebralis, USNM 28232). Despitetheir differing shapes and patterns, we treated theoccurrence of these markings as a single character-state.

Fig. 43. Character 59, dermal collar, state 1(present) in trinitatis. A: Male (UMMZ 167474). B:Female (UMMZ 167471). In this species, thedermal collar of males is diffuse and broad, butis clearly distinguished from the fainter graystippling of the adjacent surfaces.


these dark collars are not caused by melano-phores in the skin, as they are in othercollared species (e.g., collaris), but instead bymelanophores in the epimysium of the m.interhyoideus and connective tissue in thehyomandibular sinus (i.e., anterior to thepectoral apparatus) that are visible throughthe semitranslucent skin (fig. 44). The densityof melanophores varies among individuals,with males having greater density (andtherefore a more prominent collar) thanfemales. Dense subdermal pigmentationmay also occur in species with dark dermalpigmentation (e.g., galactonotus; see fig. 44),and individuals with dermal collars may (ormay not) also present extensive subdermalpigmentation. In leopardalis (e.g., UMMZ17170) the subdermal pigmentation is not asconcentrated but still accounts for the faintcollar reported by Rivero. Some degree ofmelanosis of the collar region is widespreadamong dendrobatids. However, as observed

in pigmentation of the flesh generally,variation is continuous from a few melano-phores scattered across the throat to a solidsubdermal collar. As discussed above, wesuspect there are valid transformation serieshere, but we were unable to delimit themobjectively for the present study.

60. DARK LOWER LABIAL STRIPE (fig. 45):absent 5 0; present 5 1.

In fraterdanieli Grant and Castro-Herrera(1998) indicated the occurrence of a distinc-tive dark (black or brown) line along thelower lip and contrasting with the paleadjacent coloration.

61–64. MALE AND FEMALE THROAT AND

ABDOMINAL COLORATION

The color and color pattern of the throatand abdominal regions of adult males andfemales provide some of the most usefulcharacters for discriminating among speciesof dendrobatids. Sexual dimorphism is com-

Fig. 44. Extensive subdermal melanosis of the collar region. A, B: nocturnus (AMNH 130008). C, D:galactonotus (AMNH 128233). Note also the irregular (clumped) stippling or faint, diffuse spotting inA (character 61, state 6; see below).


mon, especially in throat color and colorpattern, but most states occur in both sexes.

In a series of field experiments, Narins etal. (2003) showed that both vocalization andpulsation of the dark vocal sac of femoralisare necessary to elicit aggressive responses,and they have experimented with this systemto reveal the extent of temporal and spatialintegration of these stimuli required for anaggressive response (Narins et al., 2005).Given the extent of species- and sex-specificvariation in throat coloration in manydendrobatids, it is likely that integration ofvisual and vocal cues will be found to bewidespread. To date Narins et al.’s investiga-tions have only examined aggressive male–male interactions in this single species, butapplication of their procedures more gener-ally promises to reveal fascinating insightsinto the role of throat coloration in aggresiveand other interactions between males, fe-males, and juveniles across multiple species.

A few points apply to the coding of all thefollowing characters. First, as discussedabove, we emphasized color pattern overcolor. Second, spotting, marbling, and re-ticulation form a continuous gradient that,though unambiguous in the extremes, wewere unable to delimit objectively. Wetherefore treated these as a single character-state, although we undoubtedly overlookedadditional transformations by doing so.Third, it may be difficult to discriminatebetween pale spotting/reticulation/marblingon a dark background versus dark spotting/reticulation/marbling on a pale background,

as the distinction has to do with adjacentcoloration and the relative concentration ofpale and dark pigmentation. Many speciesare unambiguously one or the other, butothers were either ambiguous or exhibitedboth states and were therefore coded aspolymorphic. Fourth, we also treated irreg-ular stippling (i.e., clumped stippling) and theoccurrence of diffuse dark spotting as a singlestate because we were unable to discriminatetwo states objectively. Finally, the collar andgular–chest markings (characters 58, 59) areindependent of the region referred to as thethroat. For our purposes, throat refers to theregion of the central gular region, that is, thearea of the vocal sac in males.

61. MALE THROAT COLOR (figs. 44A, 46):pale, free or almost free of melanophores 5

0; dark due to absence of iridophores 5 1;evenly stippled 5 2; pale with discrete darkspotting/reticulation/marbling 5 3; soliddark 5 4; dark with discrete pale spotting/reticulation/marbling 5 5; irregular(clumped) stippling or faint, diffuse spotting5 6. Nonadditive.

In state 0 (pale, free or almost free ofmelanophores) the throat appears immacu-late, but closer inspection may reveal sparse,inconspicuous melanophores. State 1 (darkdue to absence of iridophores) is conspicuousin life but may easily be overlooked inpreserved specimens. See Grant and Castro-Herrera (1998) for this character-state in life.The spotting/reticulation/marbling of thevocal sac is often irregular. State 6 (darkwith pale medial stripe) is restricted to onlyboulengeri and espinosai. In both species themedial ‘‘stripe’’ varies from one or moreelongate spots to a solid stripe. Also, theadjacent dark surfaces sometimes includescattered pale spots. States 0–5 are shownin figure 46; state 6 is shown in figure 44A.

62. FEMALE THROAT COLOR (fig. 47): pale,free or almost free of melanophores 5 0;irregular (clumped) stippling or faint, diffusespotting 5 1; solid dark 5 2; dark withdiscrete pale spotting/reticulation/marbling 5

3; pale with discrete dark spotting/reticula-tion/marbling 5 4; dark with pale mediallongitudinal stripe 5 5. Nonadditive.

63. MALE ABDOMEN COLOR (fig. 48): pale,free or almost free of melanophores 5 0; palewith discrete dark spotting/reticulation/mar-

Fig. 45. Character 60, dark lower lip line. State1, present (fraterdanieli, MHNUC 364).


bling 5 1; evenly stippled 5 2; dark withdiscrete pale spotting/reticulation/marbling 5

3; irregular (clumped) stippling or faint,diffuse spotting 5 4; solid dark 5 5. Non-additive.

64. FEMALE ABDOMEN COLOR (fig. 49):pale, free or almost free of melanophores 5

0; pale with discrete dark spotting/reticula-tion/marbling 5 1; solid dark 5 2; dark withdiscrete pale spotting/reticulation/marbling 5

3; irregular (clumped) stippling or faint,diffuse spotting 5 4; evenly stippled 5 5.Nonadditive.

Coloma (1995: 54) described vertebralis ashaving ‘‘dark stippling on abdomen infemales, darker in males’’; however, none ofthe females in the series AMNH 17458,17604–08, 140977–141011 possesses any stip-pling on the abdomen (but all males do).Although we coded riveroi as having theabdominal region evenly stippled, in life it isposteriorly orange (Donoso-Barros, 1965‘‘1964’’).

65. IRIS COLORATION (fig. 50): lackingmetallic pigmentation and pupil ring 5 0;

possessing metallic pigmentation and pupilring 5 1.

Silverstone (1975a: 8) noted that in life theiris of the species he included in Dendrobatesis ‘‘black (or rarely dark brown) and is neverreticulated’’. Similarly Silverstone (1976: 3)stated that in the species he included inPhyllobates ‘‘the iris is black or brown (rarelybronze) and never reticulated’’. We diagnosethis character somewhat more precisely, butwe believe our intentions are the same.

The iris coloration of most dendrobatidsincludes metallic pigmentation (bronze, cop-per, gold, silver) producing a metallic iriswith a black reticulated pattern or a black iriswith metallic flecks. Additionally, a distinctmetallic ring around the pupil invariablyoccurs in irises with metallic pigmentation.A number of the aposematic dendrobatidslack all metallic pigmentation in the iris,giving rise to the solid black or brown irismentioned by Silverstone.

This character can only be coded fromliving specimens. We dissected the eyes ofpreserved specimens of several species but

Fig. 46. Character 61, male throat color. A: State 0, pale, free or almost free of melanophores(‘‘Neblina species’’, AMNH 118689). B: State 1, dark due to absence of iridophores (abditaurantius, ICN9853). This character-state is inconspicuous in preserved specimens but obvious in living or recentlyprepared specimens. C: State 2, evenly stippled gray (infraguttatus, AMNH 104846). Note that the gular–chest markings (character 58) of infraguttatus do not interfere with the even stippling of the throat. D:State 3, pale with dark spots (‘‘nubicola-spC’’, MHNUC 321). E: State 4, solid dark (inguinalis, LACM42329). F: State 5, dark with discrete pale spotting/reticulation/marbling (tricolor, USNM 286082).


failed to detect differences between pigmen-ted and unpigmented irises. We thereforerelied on explicit field notes, personal ob-servations, and high-quality photographs. Inaddition to personal observations and un-published field notes and photographs, thischaracter was scored from the followingpublished accounts: arboreus (Myers et al.,1984); aurotaenia (Silverstone, 1976; Lotterset al., 1997a); azureiventris (Kneller andHenle, 1985; Lotters et al., 2000); awa(Coloma, 1995); baeobatrachus (Lescure andMarty, 2000); bicolor (Myers et al., 1978;Lotters et al., 1997a); bocagei (Coloma,1995); boulengeri (Silverstone, 1976); caeru-leodactylus (Lima and Caldwell, 2001); clau-diae (Jungfer et al., 2000); delatorreae (Co-loma, 1995); degranvillei (Boistel and deMassary, 1999; Lescure and Marty, 2000);elachyhistus (Coloma, 1995; Duellman,2004); flotator (Savage, 2002), Eupsophusroseus ([for E. calcaratus] Nunez et al.,1999); granuliferus (Myers et al., 1995;Savage, 2002); herminae (La Marca, 1996‘‘1994’’); Hylodes phyllodes (Heyer et al.,1990); ideomelus (Duellman, 2004); imitator(Symula et al., 2001); infraguttatus (Coloma,

1995); insperatus (Coloma, 1995); insulatus(Duellman, 2004); kingsburyi (Coloma,1995); lehmanni Myers and Daly (Myersand Daly, 1976b); lugubris (Silverstone,1976; Savage, 2002); macero (Rodrıguez andMyers, 1993; Myers et al., 1998); machalilla(Coloma, 1995); molinarii (La Marca, 1985);nexipus (Frost, 1986; Hoff et al., 1999);nidicola (Caldwell and Lima, 2003); nocturnus(Myers et al., 1991); nubicola (Savage, 2002),parvulus (Silverstone, 1976); pictus (Kohler,2000); petersi (Rodrıguez and Myers, 1993;Myers et al., 1998); pulchellus (Coloma,1995); pulchripectus (Silverstone, 1975a); pul-cherrimus (Duellman, 2004); pumilio (Myerset al., 1995; Savage, 2002); quinquevittatus(Caldwell and Myers, 1990); reticulatus(Myers and Daly, 1983); rubriventris (coverof Herpetofauna 19(110); see also Lotters etal., 1997b); sauli (Coloma, 1995); silverstonei(Myers and Daly, 1979); speciosus (Jungfer,1985); sylvaticus Barbour and Noble (Duell-man, 2004); sylvaticus Funkhouser (Myersand Daly, 1976b [as histrionicus]; Lotters etal., 1999); talamancae (Coloma, 1995); terri-bilis (Myers et al., 1978); toachi (Coloma,1995); trinitatis (Wells, 1980c; La Marca,

Fig. 47. Character 62, female throat color. A: State 0, pale, free or almost free of melanophores(undulatus, AMNH 159128). B: State 1, irregular (clumped) stippling or faint, diffuse spotting (nocturnus,AMNH 130018). C: State 2, solid dark (hahneli, AMNH 96190). D: State 3, dark with discrete palespotting/reticulation/marbling (imbricolus, AMNH 102083). E: State 4, pale with discrete dark spotting/reticulation/marbling (fraterdanieli, AMNH 148021). F: State 5, dark with pale medial longitudinal stripe(boulengeri, USNM 145281).


1996 ‘‘1994’’); trivittatus (Myers and Daly,1979); undulatus (Myers and Donnelly, 2001);vanzolinii (Myers, 1982); ventrimaculatus(Lotters, 1988 [as quinquevittatus]; Lescureand Bechter, 1982 [as quinquevittatus]); ver-tebralis (Coloma, 1995); vicentei (Jungfer etal., 1996b); vittatus (Silverstone, 1976; Sav-age, 2002); zaparo (Silverstone, 1976).

66. LARGE INTESTINE COLOR (fig. 51):unpigmented 5 0; pigmented anteriorly 5

1; pigmented extensively 5 2. Additive.

The large intestine of most species isunpigmented (state 0), being either white or(when distended) translucent. In some spe-cies, heavy melanosis forms a solid blackcoloration extending posteriad from the frontof the large intestine. In state 1 the melanosisis confined to the anterior quarter of the largeintestine; in state 2 it extends beyond themidlevel of the large intestine. The ontogenyof this character invariably progresses fromstate 0 through state 1 to state 2, which weinterpret as evidence for the additivity of thistransformation series.

67. ADULT TESTIS (MESORCHIUM) COLOR

(fig. 52): unpigmented 5 0; pigmented medi-ally only 5 1; entirely pigmented 5 2.Additive.

Testis color is scored from adults only. Inall dendrobatids we have examined, testispigmentation increases ontogenetically, withthe mesorchia of juveniles being invariablyentirely unpigmented white (state 0) andmelanosis beginning medially (state 1) andeventually covering the testis entirely (state2), forming either a dark reticulum or a soliddark color. Ontogenetic series show thischaracter to develop from state 0 to state 1to state 2, which we interpret as evidence ofadditivity.

Polymorphism among adults is rare. Grant(2004) found that of 40 specimens ofpanamensis scored, the left testes of two wereunpigmented whereas the right testes werepigmented brown. Grant (2004) also docu-mented unusual variation in the testis pig-mentation of inguinalis. Testes of all adults

Fig. 48. Character 63, male abdomen color. A:State 0, pale, free or almost free of melanophores(‘‘Neblina species’’, AMNH 118689). B: State 1,pale with discrete dark spotting/reticulation/mar-bling (quinquevittatus, AMNH 124069). C: State 2,evenly stippled (talamancae, AMNH 113893). D:State 3, dark with discrete pale spotting/reticula-tion/marbling (infraguttatus, AMNH 104846). E:State 4, irregular (clumped) stippling or faint,

r

diffuse spotting (nocturnus, AMNH 130012). F:State 5, solid dark (inguinalis, LACM 42329).


had some degree of dark pigmentation, but itvaried from being confined to medial andanterior surfaces to engulfing the entire testis;this variation was not correlated with adult

size, extent of dark ventral pigmentation, ormaturity. It is likely that such variation ishormonally controlled and related to sexualactivity, but no evidence exists to support thisconjecture. Among the included outgrouptaxa, Lotters (1996) reported that, althoughmost species of Atelopus possess permanentlyunpigmented testes, in some species the testesbecome pigmented with the onset of thebreeding season.

68. COLOR OF MATURE OOCYTES (fig. 53):unpigmented (uniformly white or creamyyellow) 5 0; pigmented (animal pole brownor black) 5 1.

The entire oocyte may be white or creamyyellow (state 0) or differential localization ofpigment granules in the animal cortex maygive rise to a dark (brown or black) animal

Fig. 49. Character 64, female abdomen color.A: State 0, pale, free or almost free of melano-phores (undulatus, AMNH 159128). B: State 1,pale with dark spotting/reticulation/marbling (fra-terdanieli, AMNH 39360). C: State 2, solid dark(silverstonei, AMNH 91845). D: State 3, dark withdiscrete pale spotting/reticulation/marbling (infra-guttatus, AMNH 104849). E: State 4, irregular(clumped) stippling or faint, diffuse spotting(nocturnus, AMNH 130018). F: State 5, evenlystippled (riveroi, AMNH 134141).

Fig. 50. Character 65, iris coloration. A: State0, lacking metallic pigmentation and pupil ring(bicolor, AMNH live exhibit). B: State 1, withmetallic pigmentation and pupil ring (subpuncta-tus, uncataloged MUJ specimen from Colombia:Bogota, D.C., campus of Universidad Nacionalde Colombia).


hemisphere and white or creamy yellowvegetal hemisphere (state 1).

Duellman and Trueb (1986) explained eggpigmentation as an adaptation to exposure tosunlight, and they listed a number of anurangroups in support of that hypothesis. How-ever, it is unclear if that adaptive explanationholds among dendrobatids, given that manyspecies with pigmented eggs lay clutches thatare not exposed to sunlight. For example,Myers and Daly (1979) found ‘‘a clutch of 30eggs … on a curled dry leaf that wascompletely concealed by another leaf of thecut-over forest floor’’, yet that species haspigmented eggs. It has also been conjectured(e.g., Duellman and Trueb, 1986) that thismelanosis either raises egg temperature bybetter absorbing ambient heat or providesprotection from exposure to ultraviolet radi-ation. Missing from previous discussions isan evaluation of the polarity of the transfor-mations. Until this is evaluated throughphylogenetic analysis, it is impossible toknow if a particular instance of pigmentation(or lack of pigmentation) is apomorphic (andtherefore a candidate for an explanation ofadaptation) or symplesiomorphic.

Duellman and Trueb (1986: 535) reportedthat Rhinoderma darwinii possesses unpig-mented ova, but specimens examined hadpigmented ova.

69. M. SEMITENDINOSUS INSERTION

(figs. 54, 55): ‘‘bufonid type’’ (ventrad) 5 0;‘‘ranid type’’ (dorsad) 5 1.

Noble’s (1922) seminal work brought thighmusculature to the forefront of studies ofanuran relationships, and since then the pathof insertion of the distal tendon of the m.semitendinosus has played an important rolein discussions of dendrobatid relationships(reviewed by Grant et al., 1997). Nobleidentified two predominant morphologies:(1) the putatively primitive ‘‘bufonid type’’in which the tendon of the m. semitendinosusinserts ventrad to the tendon of insertion ofthe mm. gracilis complex, and (2) theputatively derived ‘‘ranid type’’ in which itinserts dorsad to the mm. gracilis.

Noble also reported a number of ‘‘in-termediate’’ morphologies, including that ofdendrobatids. However, the m. semitendino-sus of dendrobatids clearly inserts dorsal to

Fig. 51. Character 66, large intestine color. A:State 0, unpigmented (‘‘Neblina species’’, AMNH118679). Note that the distended tissue is trans-lucent. B: State 1, anteriorly pigmented (pratti,SIUC 07654). C: State 2, extensively pigmented(beebei, ROM 39631).

Fig. 52. Character 67, adult testis (mesorch-ium) color. State 2, entirely pigmented testes(claudiae, AMNH 124257) in ventral view.

Fig. 53. Character 68, mature ova color. A:State 0, white or yellowish (Atelopus spurrelli,AMNH 50983). B: State 1, pigmented (brown)(‘‘Neblina species’’, AMNH 118679).


Fig. 54. Character 69, m. semitendinosus insertion. Photograph (left) and outline drawing (right) ofventral view of distal thigh of Thoropa miliaris (AMNH 17044), showing state 0, ventrad ‘‘bufonid’’ pathof insertion. Arrow indicates the m. semitendinosus tendon of insertion.

Fig. 55. Character 70, m. semitendinosus binding tendon. State 1, present (aurotaenia, AMNH161109), photograph (left) and outline drawing (right) showing view of the concealed surface of the knee.The mm. gracilis complex is deflected ventrally to reveal the dorsad ‘‘ranid’’ path of the m. semitendinosusand the secondary binding tendon that straps it to the outer edge of the mm. gracilis complex.


the mm. gracilis, the apparent intermediacyowing to a secondary binding tendon (seeCharacter 70, below). Similarly, Noble inter-preted intermediate morphologies as pro-viding evidence for the ‘‘inward migration’’of the tendon of insertion of the m.semitendinosus from the presumptively prim-itive ‘‘bufonid’’ position to the derived‘‘ranid’’ position. However, he relied onphylogenetic evidence to establish characteradditivity, not ontogenetic (or other) evi-dence. The groups Noble considered mostprimitive had ‘‘bufonid type’’ insertion, thosehe thought were most derived had ‘‘ranidtype’’, and variants were treated as interme-diate between the two. Such reasoning isobviously fallacious, as it conflates thepremises of analysis with the conclusions.We are unaware of developmental evidenceof inward migration of the m. semitendinosustendon of insertion from the ‘‘bufonid’’ to the‘‘ranid’’ position.

70. M. SEMITENDINOSUS BINDING TENDON

(fig. 55): absent 5 0; present 5 1.

As first described and illustrated by Noble(1922: 41 and plate XV, fig. 6)6, thedendrobatid thigh has a well-defined bindingtendon7 that straps the m. semitendinosusdistal tendon to the dorsal edge of the innersurface of the mm. gracilis complex (state 1;fig. 55). In some large species, such aspalmatus and nocturnus, this binding tendonis robust and conspicuous, giving the impres-sion that the distal tendon of the m.semitendinosus actually pierces or penetratesthe distal mm. gracilis tendon (e.g., Dunlap,1960: 66). However, even in these species them. semitendinosus tendon does not passthrough the tendinous tissue, but ratherbetween the binding tendon and the gracilismuscle (therefore differing from myobatra-chids; Noble, 1922; Parker, 1940). Thistendinous tissue also forms a secondarytendon that inserts on the inner (posterior)surface of the proximal head of the tibiofib-ula. Another secondary tendon is oftenpresent, arising near the ventral edge of theinner surface of the m. gracilis and leading tothe thick sheet of connective tissue that wrapsaround the knee. Distal to the bindingtendon, the m. semitendinosus tendon ex-pands to insert along the long axis of theventral surface of the tibiofibula.

71. M. LEVATOR MANDIBULAE EXTERNUS

DIVISION: undivided (‘‘s’’) 5 0; divided (‘‘s +e’’) 5 1.

In her dissertation, Starrett (19688) foundthat the jaw adduction musculature ofdendrobatids includes a single muscle origi-nating from the zygomatic ramus of thesquamosal that lies deep (internal, medial) tothe mandibular ramus of the trigeminal nerve(V3), which she interpreted as the presence ofthe m. adductor posterior mandibulae sub-externus and absence of the m. adductormandibulae externus superficialis, or condi-tion ‘‘s’’ in her system (state 0). Silverstone(1975a) found this in all 41 species heexamined, and other workers (Myers et al.,1978, 1991; Myers and Daly, 1979; Myersand Ford, 1986; Grant et al., 1997; Grant,1998) have found this in almost all dendro-

6We examined the thigh musculature of AMNH13472, the palmatus specimen drawn by Noble, andconfirmed that his illustration is accurate in itsdepiction of the path of the m. semitendinosustendon of insertion. However, his illustration iserroneous with regard to the m. gracilis minor andthe insertion of the mm. adductor longus andadductor magnus. The m. gracilis minor is no longerpresent on the left thigh of AMNH 13472, but on theright it is an inconspicuous, narrow, thin muscle thatmerges distally with the m. gracilis major to sharea common tendon of insertion, a morphology thatconforms with all of our previous observations ofdendrobatid thighs; we have never observed the m.gracilis minor to be as thick and broad as indicated byNoble’s illustration. In fact, in many species the m.gracilis minor is completely undetectable distal tomidlength of the thigh. Similarly, although the mm.adductor longus and adductor magnus remain in-dependent along most of the length of the femur, theyfuse distally to share a common insertion in thedendrobatids we have examined, including AMNH13472.

7We follow Noble’s (1922: 41) terminology, exceptthat the appropriate term for connective tissue thatextends from muscle to periosteum (of the tibiofibulaor femur, in this case) is tendon, not ligament.

8Although generally we did not take charactersfrom unpublished sources, the influence of thisdissertation in anuran systematics has been so greatthat it would be inappropriate to ignore this work.


batids (see below). Our observations conformwith the accounts of previous workers, withthe exception that fibers generally originateon the anterior edge of the annulus tympa-nicus as well. In addition to the ‘‘s’’morphology, among other frogs Starrett(1968) reported the presence of both muscles,or ‘‘s + e’’ in her system (state 1), as well asthe absence of the m. adductor posteriormandibulae subexternus and presence of them. adductor mandibulae externus superficia-lis, ‘‘e’’ (not observed in the present study).

Haas (2001) reevaluated the homology ofthese muscles based on detailed developmen-tal studies across the diversity of amphibians.He concluded that the ‘‘subexternus’’ and‘‘externus’’ muscles are actually differentportions of the same muscle (slips, in ourterminology; see Materials and Methods,above)—the m. levator mandibulae externusprofundus and m. levator mandibulae exter-nus superficialis, respectively. As such, Star-rett’s (1968) three conditions (‘‘s’’, ‘‘e’’, and ‘‘s+ e’’) involve two distinct transformationseries: (1) the m. levator externus is constantin both ‘‘s’’ and ‘‘e’’ and the transformationseries is formed by changes in the position ofV3, deep (internal, medial) in the ‘‘s’’ andsuperficial (external, lateral) in the ‘‘e’’; (2) ‘‘s+ e’’ is formed by the division of the m. levatormandibulae externus into profundus andsuperficialis slips, with V3 lying between them.Because we did not observe the ‘‘e’’ conditionin any of the species coded for morphology inthe present study, we scored only the lattertransformation series.

The sole published exception to the un-divided m. levator mandibulae externus (‘‘s’’;state 0) morphology in dendrobatids isnocturnus, in which Myers et al. (1991) foundthe m. levator externus of some specimens (orone side) to form two distinct slips (‘‘s + e’’;state 1). Myers et al. (1991) interpreted thisobservation as possibly indicative of both theranoid origin of dendrobatids and the prim-itiveness of nocturnus within the dendrobatidclade. The only other exception we observedwas a specimen of vanzolinii (AMNH108332) in which V3 pierces the m. levatormandibulae externus. However, we did notcode the superficial and medial fibers asdifferent slips (i.e., ‘‘s + e’’) because the fibersare tightly bound both dorsal and ventral to

the nerve and are not segregated by connec-tive tissue septa (as they are in nocturnus, forexample) and therefore do not form distinctslips (for similar individual variation, seeLynch, 1986). That said, only a single spec-imen was available for dissection, and thesymmetry of this morphology suggests thatfurther study may reveal a phylogeneticallyrelevant change in the path of V3.

Additional intraspecific variation was ob-served in Hylodes phyllodes. Of the 11 frogsin the series AMNH 103885–95, in 2(AMNH 103888, 103890) V3 runs medial toa distinct m. levator mandibulae externussuperficialis (‘‘s + e’’; state 1) on both sides ofthe head, whereas the remaining 9 specimensall lack that slip (‘‘s’’; state 0). As innocturnus, and in contrast to vanzolinii, thelateral fibers form a distinct slip. Indeed, inH. phyllodes all of the lateral fibers appear tooriginate on the rim of the annulus tympa-nicus, whereas the medial slip originates fromthe squamosal. Although this latter consid-eration is suggestive of nonhomology of them. levator mandibulae externus superficialisin these taxa, we coded them as the samestate and subjected that hypothesis to the testof character congruence.

72–75. M. DEPRESSOR MANDIBULAE

Starrett (1968) identified three distinct slipsof the m. depressor mandibulae of dendroba-tids: a massive, superficial slip originating fromthe dorsal fascia overlying the scapula and m.levator posterior longus, a deeper slip origi-nating from the otic ramus of the squamosal,and an additional slip of fibers originating onthe tympanic annulus. The combined mor-phology was codified as DFSQdAT. Lynch(1993: 37) refined the delimitation of thiscondition as ‘‘one in which some number ofsuperficial fibers of the squamosal portion ofthe m. depressor mandibulae extend medial tothe crest of the otic ramus of the squamosaland overlie the fibers of the m. levator posteriorlongus.’’ Silverstone (1975a) confirmed that all41 species of dendrobatids he examined havethis morphology, and this was further con-firmed in additional species by Myers andcoworkers (e.g., Myers et al., 1978, 1991;Myers and Daly, 1979; Myers and Ford,1986).


Lynch (1993) rejected the anatomicalfindings of Starrett (1968), at least as theyapplied to Eleutherodactylus. Of greatest rele-vance to dendrobatids is his finding (pp. 37–38) that the superficial ‘‘DF’’ fibers actuallyare ‘‘bound tightly to deeper fibers … thatoriginate on the lateral face of the oticramus of the squamosal’’. That is, the fibersfrom the two origins are not segregated byconnective tissue septa and therefore do notconstitute distinct slips. Our dissectionsconfirm Lynch’s observations in dendroba-tids and the sampled outgroup taxa as well,leading us to follow him in discardingStarrett’s terminology.

Nevertheless, regardless of whether thedepressor muscle is divided into distinct slipsor not, the variation in fiber origins consti-tutes a valid transformation series. In allspecimens examined, some fibers originatefrom the otic ramus of the squamosal. In alldendrobatids examined, the vast majority offibers originates form the dorsal fasciae.Lynch (1993) referred to the portion of them. depressor mandibulae that originatesmedial to the crest of the squamosal on them. temporalis as a ‘‘dorsal flap’’, and wefollow his terminology here (Character 73).We scored as Character 74 the origin of fibersposterior to the crest of the squamosal. Theoccurrence of fibers originating on theposterior edge of the annulus tympanicus iscoded in Character 75.

Manzano et al. (2003) presented an exten-sive survey of the m. depressor mandibulae inanurans. Among dendrobatids, they exam-ined auratus, pictus, and subpunctatus. Ourobservations and coding conform generallywith theirs, with the following exceptions: (1)Manzano et al. did not recognize the dorsalflap as a separate character. (2) Manzano etal. reported that the superficial ‘‘slip’’ ofauratus is divided into anterior and posterior‘‘slips’’, whereas that of pictus and subpunc-tatus consists of a single, wide, fan-shapedmuscle. We examined 20 uncatalogedAMNH skinned carcasses of auratus fromIsla Tobago, Panama, constituting 40 de-pressor muscles. In that series, variation iscontinuous between an uninterrupted fan-shaped muscle, the occurrence of a slightdivision across the thoracic sinus, and well-defined separate branches, with conspicuous

bilateral asymmetry in some specimens.Given the continuous variation, we wereunable to delimit states objectively. More-over, the degree of individual variationsuggests that the differences are likely non-genetic, although we cannot offer any directevidence to that effect. (3) Manzano et al.reported fibers originating from the annulustympanicus in Rhinoderma darwinii. Howev-er, although we observed fibers to extendtoward the annulus, in the specimens weexamined (AMNH 37849, 58082), the fibersinvariably run along the cartilage and ulti-mately attach to the squamosal.

72. M. DEPRESSOR MANDIBULAE DORSAL

FLAP: absent 5 0; present 5 1.73. M. DEPRESSOR MANDIBULAE ORIGIN

POSTERIOR TO SQUAMOSAL: absent 5 0;present 5 1.

74. M. DEPRESSOR MANDIBULAE ORIGIN

ON ANNULUS TYMPANICUS: no fibers origi-nating from annulus tympanicus 5 0; somefibers originating from annulus tympanicus5 1.

Among dendrobatids, the fibers that at-tach to the annulus tympanicus are generallydeep and easily overlooked, but carefuldissection showed them to be present in alldendrobatids examined.

75. TYMPANUM AND M. DEPRESSOR MANDI-

BULAE RELATION: tympanum superficial to m.depressor mandibulae 5 0; tympanum con-cealed superficially by m. depressor mandi-bulae 5 1.

Myers and Daly (1979: 8) pointed out thatin dendrobatids ‘‘the large superficial slip ofthe depressor mandibulae muscle tends toslightly overlap the tympanic ring and, in anycase, holds the skin away from the rear partof the tympanum, thus accounting for thefact that the tympanum is only partiallyindicated externally’’ (see also Myers andFord, 1986; Myers et al., 1991). Daly et al.(1996) further discussed this character andcompared conditions found in Mantella.

76. VOCAL SAC STRUCTURE: absent 5 0;median, subgular 5 1; paired lateral 5 2.Nonadditive.

This character was coded following Liu(1935). Although we coded the vocal sac forthe sampled species Megaelosia goeldii, inwhich males lack vocal sacs and slits andpresumably do not call (Giaretta et al., 1993),


other species of Megaelosia possess pairedlateral vocal sacs (e.g., M. lutzae; Izecksohnand Gouvea, 1987 ‘‘1985’’). Lynch (1971)reported the state for Eupsophus roseus(coded for E. calcaratus).

77–78. M. INTERMANDIBULARIS SUPPLEMEN-

TARY ELEMENT (fig. 56)

Tyler (1971) described variation in super-ficial throat musculature of hylids and otheranurans, reporting the differentiation of them. intermandibularis to form a supplementa-ry element in two species of Colostethus [asCalostethus], two species of Dendrobates, andtwo species of Phyllobates. He did not list thespecies of dendrobatids he examined ordescribe the dendrobatid condition in anydetail. La Marca (1995: 45) noted theoccurrence of ‘‘supplementary elements ofthe anterolateral type attached to the ventralsurface of the m. submentalis’’. Of relevanceto the current study, Burton (1998b) alsoreviewed the occurrence and variation in

supplementary elements of numerous Neo-tropical hyloid groups.

The treatment of the supplementary ele-ment in phylogenetic analysis is somewhatproblematic, as the homology of the elementsin different taxa is debatable. FollowingTyler’s (1971) terminology, dendrobatidspossess an anterolateral element: Fibersoriginate on the lingual surface of theanterior portion of the mandible and runanteriomediad to insert on the ventral surfaceof the m. submentalis, with the moreposterior fibers underlying (superficial to)the deeper fibers of the m. intermandibularis.Tyler also identified apical and posterolateralelements in other groups of anurans, andthese conditions have been largely supportedby subsequent workers. Tyler (1971) effec-tively treated each of the morphologies asnonhomologous (i.e., the differences inmorphologies was treated as evidence thateach of the supplementary elements wasindependently derived), but other workershave treated them as a homologous entitywith subsequent variation (e.g., Burton,1998b; Mendelson et al., 2000).

The shared origin of the supplementaryelement on the lingual surface of the mandi-ble superficial to the deeper primary sheet ofthe m. intermandibularis and the fact that thedifferent morphologies never co-occur aresufficient evidence to treat the supplementaryelements of different anurans as a homolo-gous structure. We therefore submitted thehypothesis of supplementary slip homologyto the simultaneous test of character congru-ence by coding its occurrence as one charac-ter and the variation in the element asa second character.

77. M. INTERMANDIBULARIS SUPPLEMEN-

TARY ELEMENT OCCURRENCE: absent 5 0;present 5 1.

78. M. INTERMANDIBULARIS SUPPLEMEN-

TARY ELEMENT ORIENTATION: 0 5 anterolat-eral; 1 5 anteromedial.

Among the sampled species that possessthe supplementary element, we observedtwo patterns. In the first (state 0), the fibersradiate anterolaterally from a sagittal ra-phe. In the second, the fibers extendanteromedially from the mandible. Burton(1998b) reported both of these morpholo-gies for a variety of Neotropical ‘‘leptodac-

Fig. 56. Character 79, M. intermandibularissupplementary element morphology. A: State 0,anterolateral (Rhinoderma darwinii, AMNH37849). B: State 1, anteromedial (trinitatis, un-cataloged AMNH specimen, part of series collect-ed with AMNH 87392–93).


tylids’’. However, our coding deviates fromBurton’s (1998b) in that he treated specieswith anterolateral supplementary slips (e.g.,Hylodes spp.) and without supplementaryslips but having all fibers directed mediad oranterolaterad (e.g., Thoropa miliaris) as‘‘variants of the same general pattern’’(pp. 67–68). Burton based this decision on‘‘the fact that two of these variants mayoccur within the same genus (e.g., Cyclor-amphus), or within the same species (C[audi-verbera]. caudiverbera)’’ (p. 67). Co-occur-rence of this nature does not constituteevidence of character-state identity (if itdid, polymorphism would be conceptuallyimpossible), and we scored Hylodes phyllodesand Thoropa miliaris differently.

Manzano and Lavilla (1995) reported anapical supplementary element in Rhinodermadarwinii; according to the terminology em-ployed herein, it is anterolateral (state 0).

79–86. MEDIAN LINGUAL PROCESS

(figs. 57, 58)

Grant et al. (1997) discovered the medianlingual process (MLP; fig. 57) in dendroba-tids and documented its occurrence andvariation throughout Anura (for additionaldendrobatid records, see Myers and Don-nelly, 1997; Grant and Rodrıguez, 2001). Ofgreatest interest was the finding that anapparently homologous modification of thetongue occurs in dendrobatids and several

Old World ranoids (including the putativesister group postulated by Griffiths, 1959)but is entirely lacking among all hyloid taxa.The functional significance of the MLPremains unknown. Variation in the MLPhas been illustrated extensively by Grant etal. (1997) and Myers and Donnelly (1997);here we provide illustrations for novelcharacters.

As a first effort to understand the distri-bution and diversity of the MLP, Grant et al.(1997) allocated the observed variation tofour ‘‘types’’. For phylogenetic analysis itwas necessary to decompose those types intotheir component transformation series. Giventhe relevance of this anatomical structure tothe placement of Dendrobatidae, we exam-ined the histology of the tongues of eightspecies to gain a better understanding of itsstructure. Additional data were gatheredthrough gross dissection. Although severalof the characters we observed do not varyamong the MLP-possessing taxa sampled inthe present study, they vary independently inthe broader context of the evolution of theMLP in anurans, and we therefore score allof these characters here.

To discover differences between the type Cprocesses of Old World and New World taxa,we examined the histology of two species ofthe dendrobatids tepuyensis and baeobatra-chus, and we compared them to Phrynoba-trachus natalensis and Phrynobatrachus pet-ropedetoides. These two species of

Fig. 57. Anterior view of the open mouth ofthe dendrobatid praderioi (CPI 10203) showing theshort, tapered median lingual process (MLP).

Fig. 58. Longitudinal histological section ofthe tongue of baeobatrachus (AMNH 140672)showing that the median lingual process (MLP)is an extension of the m. genioglossus. Note lack ofmuscle fibers toward the tip of the MLP.


Phrynobatrachus have retractile type C pro-cesses, which we hoped would maximize themorphological differences between them andthe nonretractile type C processes of thedendrobatids. To gain insight into themechanism of retraction and protrusion,one of the specimens of Phrynobatrachuspetropedetoides had the MLP fully protruded,whereas the other one had it retracted belowthe surface of the tongue. We also examinedthe type A processes of Arthroleptis variabilis,Mantidactylus femoralis, Platymantis dorsa-lis, and Staurois natator (the latter twospecies were included only for comparativepurposes to delimit transformation seriesmore rigorously and were not coded forphylogenetic analysis). Due to lack of speci-mens, we did not examine any type B or Dprocesses.

The MLP of all examined taxa (i.e., typesA and C, retractile and nonretractile) isformed through the same modification of

the basal portion of the m. genioglossus,supporting the hypothesis that they arehomologous structures. As seen in sagittaland transverse section of baeobatrachus(fig. 58) the m. genioglossus basalis is ex-tended dorsally to protrude above the lingualsurface as the median lingual process. In alltaxa, muscle fibers are replaced distally byloose, presumably collagenous connectivetissue, with elastic fibers forming the wallsof the process. Although we did not stainspecifically for nervous tissue, no majornerves were detected within the MLP. Addi-tional histological findings are discussedbelow under the relevant characters.

79. MLP OCCURRENCE: absent 5 0;present 5 1.

To count the origin of the MLP as a singleevent (and not as multiple origins of each ofthe nested characters), we scored the occur-rence of the MLP as a separate character.That is, species that lack the MLP werescored as state 0 for this character andmissing (‘‘–‘‘) for the remaining MLP char-acters.

80. MLP SHAPE: short, bumplike 5 0;elongate 5 1.

We considered the MLP to be short andbumplike if it its length (height) was nogreater than its width at the base andelongate if its length was greater than itswidth at the base.

81. MLP TIP: blunt 5 0; tapering to point5 1.

Fig. 59. Character 84, median lingual process(MLP) retractility, state 1 (retractile). A: Trans-verse section of a protruded MLP in Phrynoba-trachus natalensis (AMNH 129714). B: Transversesection of a retracted MLP in Phrynobatrachuspetropedetoides (AMNH 129626).

Fig. 60. Character 86, median lingual process(MLP) epithelium. A: State 0, glandular (Phryno-batrachus petropedetoides, AMNH 129593). Thesurface of the MLP is pitted with invaginations ofthe epithelium that form alveolar glands. B: State1, nonglandular (Phrynobatrachus natalensis,AMNH 129732). The alveolar glands are absent,and the surface of the MLP consists of unmodifiedstratified epithelium.


82. MLP TEXTURE: smooth 5 0; rugose5 1.

Grant et al. (1997) found most MLPs to besmooth relative to the lingual surface (state 0)but that in some species the MLP is rugose,textured like the adjacent surfaces of thetongue.

83. MLP ORIENTATION WHEN PROTRUDED:upright 5 0; posteriorly reclined 5 1.

When protruded, Grant et al. (1997)reported upright MLPs pointing straightdorsad (state 0) and posteriorly reclinedMLPs (state 1).

84. MLP RETRACTILITY (fig. 59): nonre-tractile 5 0; retractile 5 1.

Following the reasoning of Grant et al.(1997), retractility was inferred from theposition of the MLP in preserved specimens.We were unable to detect any histologicaldifferences between retractile and nonretrac-tile processes. However, the fact that evenvery small series of some species show theMLP in various stages of retraction whereasvery large samples of others do not includea single retracted lingual process suggeststhat this is not merely an artifact ofpreservation.

Comparison of retracted and protrudedprocesses provides some clues as to themechanism involved in retractility. As seenin figure 59A of the protruded process ofPhrynobatrachus natalensis in transverseview, the connective tissue that extends tothe tip of the MLP is very loose, with largespaces between the fibers and fibroblasts. Incontrast, in a specimen of Phrynobatrachuspetropedetoides with the MLP completelyretracted below the surface of the tongue(fig. 59B), the loose connective tissue is muchdenser with no spaces between the fibers andfibroblasts, reminiscent of a squeezed sponge.This is characteristic of hydrostatic organssuch as the feet of mollusks and suggests thatprotrusion and retraction of the MLP isachieved by the displacement of some sort offluid.

85. MLP-ASSOCIATED PIT: absent 5 0;present 5 1.

Grant et al. (1997) noted the absence (state0) and presence (state 1) of a pit immediatelyposterior to the MLP into which fits theposteriorly reclined MLP of some species.Although all of the species with posteriorly

reclined MLPs sampled in the present studyalso have an associated pit, the observationsof Grant et al. (1997) establish the trans-formational independence of the two char-acters.

86. MLP EPITHELIUM (fig. 60): glandular5 0; nonglandular 5 1.

In state 0 the surface of the MLP is pittedwith invaginations of the epithelium thatform alveolar glands, as occur over the rest ofthe surface of the tongue. In state 1, theseglandular invaginations do not occur, andthe MLP is covered in unmodified, stratifiedepithelium.

87–98. LARVAE

In addition to specimens examined, larvaldata were taken from Ruthven and Gaige(1915), Fernandez (1926), Funkhouser(1956), Donoso-Barros (1965 ‘‘1964’’), Sav-age (1968), Duellman and Lynch (1969),Hoogmoed (1969), Edwards (1971, 1974b),Lynch (1971), McDiarmid (1971), Silverstone(1975a; 1976), Lescure (1976), Duellman(1978, 2004), Myers and Daly (1979), Cei(1980), Lescure and Bechter (1982), Heyer(1983), La Marca (1985), Lavilla (1987),Formas (1989), Caldwell and Myers (1990),Donnelly et al. (1990), Heyer et al. (1990),Mijares-Urrutia (1991), Myers et al. (1991),van Wijngaarden and Bolanos (1992), Ro-drıguez and Myers (1993), Haddad andMartins (1994), Junca et al. (1994), Coloma(1995), Ibanez and Smith (1995), La Marca(1996 ‘‘1994’’), Mijares-Urrutia and LaMarca (1997), Kaplan (1997), Lotters et al.(1997b), Grant and Castro-Herrera (1998),Faivovich (1998), Lindquist and Hethering-ton (1998), Lotters et al. (2000), Caldwell etal. (2002a), Caldwell and Lima (2003), Nuin(2003), and Castillo-Trenn (2004).

87. LARVAL CAUDAL COLORATION: verti-cally striped 5 0; scattered melanophoresclumped to form irregular blotches 5 1;evenly pigmented 5 2. Additive.

Caldwell et al. (2002a) figured the larvae ofcaeruleodactylus and marchesianus, the tailsof which possess conspicuous, dark, broad,vertical stripes (state 0). The larval tails of themajority of species possess variable amountsof irregular, scattered melanophores clumpedto form diffuse blotches, ranging from in-


conspicuous reticulation to large blotches(state 1). There is extensive ontogenetic andindividual variation in the amount andintensity of this diffuse spotting, as docu-mented for kingsburyi by Castillo-Trenn(2004), which prevented dividing the varia-tion observed within this character-state intoadditional states. In some species, the larvaltails are evenly pigmented brown, gray, orblack (state 2).

88. LARVAL ORAL DISC (fig. 61): ‘‘normal’’5 0; umbelliform 5 1; absent 5 2; suctorial5 3. Nonadditive.

As described by Haas (2003: 54), ‘‘[t]heoral disk is formed by the upper and lowerlips, i.e., flat, more or less expansive flaps ofskin set off from the mouth and jaws andcommonly bearing labial ridges with kerato-donts.’’ What is here referred to as the‘‘normal’’ larval oral disc (state 0) consistsof a thick, fleshy upper lip that is fullyattached medially and lacks marginal papil-lae and a lower lip that is entirely free butrelatively narrow and bears marginal papil-lae. The umbelliform (funnel-shaped) oraldisc (state 1; fig. 61) is greatly enlargedrelative to state 0. The upper lip is free andthe marginal papillae extend around theentire circumference of the disc. Amongdendrobatids, state 1 is known only inflotator, nubicola, and two unnamed speciesnot included in this study. Dendrobatids that

lack the oral disc (state 2) are endotrophic.The suctorial oral disc (state 3) is confined tothe outgroup.

89. LATERAL INDENTATION OF LARVAL

ORAL DISC: absent (not emarginate) 5 0;present (emarginate) 5 1.

90. MARGINAL PAPILLAE OF LARVAL ORAL

DISC: short 5 0; enlarged 5 1; greatlyenlarged 5 2. Additive.

The marginal papillae of most dendroba-tids (e.g., boulengeri) are numerous (.50 inlate stages) and relatively small (state 0).9 Themarginal papillae of some species (e.g.,pumilio) are fewer (,30 in late stages) anduniformly larger (state 1). For illustrationsexemplifying this state, see Silverstone(1975a) and Haddad and Martins (1994).The remarkable larvae of caeruleodactylusand marchesianus possess only 12–18 (in latestages) greatly and irregularly enlarged mar-ginal papillae (state 2). For illustrations ofthis state, see Caldwell et al. (2002a).

91. SUBMARGINAL PAPILLAE OF LARVAL

ORAL DISC (fig. 61): absent 5 0; present 5 1.

Among dendrobatids, submarginal papil-lae (see Altig and McDiarmid, 1999a) areknown to occur only in larvae with umbelli-form oral discs (e.g., nubicola). However,nondendrobatids that lack umbelliform discsalso possess submarginal papillae (e.g.,Duellmanohyla uranochroa; see Altig andMcDiarmid, 1999b), demonstrating thetransformational independence of these twocharacters.

92. MEDIAN GAP IN MARGINAL PAPILLAE

OF LOWER LABIUM: absent 5 0; present 5 1.

Among dendrobatids, the median gap inthe marginal papillae of the lower labium wasillustrated and discussed by Myers and Daly(1980; see also 1987) and was claimed bythem to be a synapomorphy for abditus,bombetes, and opisthomelas; Ruiz-Carranzaand Ramırez-Pinilla (1992) added virolinensisto the group.

Fig. 61. Character 88, larval oral disc. A, B:Ventral (A) and lateral (B) views of State 0,‘‘normal’’ (‘‘Neblina species’’, AMNH 118673). C,D: Ventral (C) and lateral (D) views of State 1,umbelliform disc (nubicola, AMNH 94849). Notealso the submarginal papillae scattered over thesurface of the oral disc (character 91).

9Castillo-Trenn (2004) documented ontogeneticvariation in the number of marginal papillae inkingsburyi, ranging from 18 in stage 25 to 62 in stage34. However, the relative size and density of papillaeremains constant, that is, as the tadpole grows thenumber of marginal papillae increases while the sizeof each papilla remains approximately the same.


93. ANTERIOR LARVAL KERATODONT

ROWS: 0 5 0; 1 5 1; 2 5 2. Additive.Haddad and Martins (1994) reported the

absence of anterior keratodont rows inhahneli, and we confirmed their observationsin tadpoles of early stages. However, tadpolesof later stages possess a single anteriorkeratodont row, and we therefore coded thisspecies as state 1. The unusual shape of themouth, illustrated by Haddad and Martins, isretained until metamorphosis.

94. POSTERIOR LARVAL KERATODONT

ROWS: 0 5 0; 1 5 1; 2 5 2; 3 5 3. Additive.Haddad and Martins (1994) reported the

absence of posterior keratodont rows inhahneli, and we confirmed their observationsin small (e.g., stages 25 or 26) tadpoles.However, tadpoles of later stages possess twoposterior keratodont rows, and we thereforecoded this species as state 2.

95. LARVAL JAW SHEATH: absent 5 0;lower jaw only, not keratinized 5 1; entire,keratinized 5 2. Additive.

96. LARVAL VENT TUBE POSITION: dextral5 0; median 5 1.

Myers (1987) claimed the position of thelarval vent tube among the differencesbetween Minyobates and Dendrobates, withthe former possessing the putatively primitivedextral position (state 0) and the latter beingmedial (state 1). Myers and Daly (1979)reported ontogenetic variation from dextralto median in silverstonei, and Donnelly et al.(1990) later reported that the vent tube ofaurotaenia, lugubris, terribilis, and vittatusmigrates from medial at Gosner (1960) stages24 and 25 to dextral by stage 37 (or earlier).Caldwell and Myers (1990: 8) reported thefrequency of dextral, sinistral, and medianvent tubes in castaneoticus and summarizedvariation in this character and noted that‘‘the rare sinistral condition [is] so far knownonly in the variation of the Dendrobatesquinquevittatus complex.’’ Similar intraspecif-ic variation has been reported in otheranurans as well (e.g., Altig and McDiarmid,1999a). Because ontogenetic series were un-available for most dendrobatids, we codedthe position of the larval vent tube asobserved in the most developed larvaeexamined. As such, for example, we codedthe vent tube of aurotaenia, lugubris, terribi-lis, and vittatus as dextral.

97. SPIRACLE: absent 5 0; present 5 1.

Among the coded dendrobatids, the un-usual condition of the absence of the spiraclehas been reported for only degranvillei(Lescure, 1984) and nidicola (Caldwell andLima, 2003), both of which possess endotro-phic larvae.

98. LATERAL LINE STITCHES: absent 5 0;present 5 1.

The lateral line system is composed ofreceptor organs (mechanoreceptive neuro-masts and electroreceptive ampullary organs)and the nerves that innervate them, both ofwhich develop from the lateral line placodes(Schlosser, 2002a). Insofar as is known, thelateral line system is entirely lacking only indirect developing anurans, but the accessoryorgans (stitches) derived from the primaryneuromasts fail to develop in some species ofmultiple families of anurans (Schlosser,2002b). That is, the apparent absence of thelateral line system in these taxa is due to theabsence of the stitches. The occurrence ofstitches varies among dendrobatids, and wecoded their absence (state 0) and presence(state 1). Only transverse stitches have beenreported for anurans (Schlosser, 2002b), anddendrobatids are no exception.

Stitches have been reported (as the lateralline system), described, and/or illustrated byseveral authors (e.g., Mijares-Urrutia, 1991;La Marca, 1996 ‘‘1994’’; Myers and Don-nelly, 1997, 2001; Castillo-Trenn, 2004), butthey are frequently overlooked, and theirabsence is usually not reported (e.g., Duell-man, 2004). Although stitches are large andconspicuous in some species, they are barelydetectable in others, owing to their small sizeand the pigmentation of the surroundingareas, so it is not safe to assume that failureto mention the lateral line stitches signifiestheir absence. As such, when coding charac-ter-states from the literature, we scored thelateral line stitches as absent when theauthors were explicit or provided adequateillustrations or unusually thorough descrip-tions that (we assume) would have notedstitches had they been visible.

In addition to the presence and absence ofstitches, we observed variation in the systemof rami they form. However, as also observedby Castillo-Trenn (2004) in kingsburyi and R.W. McDiarmid (personal commun.) in other


anurans, we found that the pattern of ramivaries extensively within species, and too littleis known about ramus ontogeny and othervariation to allow transformation series to bedelimited at present.

99–116. BEHAVIOR

The behavior of a number of dendrobatidshas been documented and, beginning withNoble (1927), interpreted phylogenetically byseveral authors (e.g., Myers and Daly, 1976b;Weygoldt, 1987; Zimmermann and Zimmer-mann, 1988; Summers et al., 1999b). How-ever, although behavior is unquestionablya valid source of evidence of phylogeneticrelationships, its interpretation as phyloge-netic characters requires special attentionbecause particular behaviors are contextdependent. Certain aspects of the behavioralrepertory of a given species may be stereo-typed and consistent across populations, butbehavioral differences among populationshave been documented (e.g., Myers andDaly, 1976b), and variation among individ-uals and over time (especially under differentconditions) in the same individual are wellknown (especially in vocalizations; e.g.,Junca, 1998; Grant and Rodrıguez, 2001),and the (external and/or internal) causes ofthis variation are, for the most part, a com-plete mystery. Even highly stereotyped re-sponses may be context dependent, whichcasts some degree of doubt on the signifi-cance of observations made on captive speci-mens and the validity of comparisons to wildindividuals. In light of these potential prob-lems, we proposed hypotheses of homologyof behaviors judiciously.

An example of overinterpretation of behav-ioral observations is Zimmermann and Zim-mermann (1988), who performed a pheneticanalysis of 62 variables (including vocaliza-tions and larval morphology) for 32 species.We disregarded some of their ‘‘characters’’because they are invariable in the ingroup andmost of the outgroup (e.g., pulsation of flanksand/or throat; upright posture; female followsmale; inflate body), defined too subjectively orarbitrarily to allow comparison (e.g., oviposi-tion near a stream; male defends largeterritory, male defends small territory), de-monstrably nonindependent (e.g., larvae car-

nivorous and/or herbivorous and larvaemostly herbivorous, microphagous), or areotherwise problematic. For example, theircharacter ‘‘larvae carried singularly or ingroup’’ is problematic because, althoughdifferences almost certainly exist among spe-cies (bombetes has only been observed to carryup to three tadpoles [T. Grant, personal obs.;A. Suarez-Mayorga, personal commun.],whereas fraterdanieli carries up to at least 12[T. Grant, personal obs.], and palmatus nursefrogs transport up to 31 tadpoles [Luddecke,2000 ‘‘1999’’]), the number of dorsal tadpolesobserved is highly variable (e.g., Myers andDaly, 1979: 326, reported a male silverstoneifound carrying a single larva and two otherscarrying nine larvae; see also Luddecke, 2000‘‘1999’’: 315, table 3) due, potentially at least,to differences in clutch size, egg survivorship,rate of development, and the fact that a singleload of tadpoles may be deposited all at onceor one or a few at a time (Ruthven and Gaige,1915). Clearly there are legitimate transfor-mation series hidden in these observations,but more information is needed beforecharacters can be delimited.

There is extensive missing data for behav-ioral characters, which necessarily limits theimpact of these characters on the presentanalysis. However, one of our motivationsfor coding it nonetheless is that standardizedcodification facilitates future work. One ofthe most difficult aspects of individuatingand scoring transformation series for phylo-genetic analysis is that the behaviors areoften complex and the ways they may bedescribed by different observers may varygreatly. By delimiting and scoring thesecharacters, we hope to draw attention tothem for use in future comparative behav-ioral studies. Especially problematic areabsences; for the present purposes, we codedconspicuous behaviors not reported in de-tailed studies as absent, but it is possiblethat they were simply not noticed. A similarproblem is that even detailed notes may failto mention expected observations, such asdiurnal activity in dendrobatids. We did notmake assumptions regarding the latter char-acters and only scored them from personalobservation or explicit statements.

In addition to personal observations andunpublished field notes, behavioral data (not


including vocalizations) were taken from thefollowing published sources: Dunn (1933,1941, 1944), Eaton (1941), Trapido (1953),Test (1954, 1956), Funkhouser (1956), Steb-bins and Hendrickson (1959), Sexton (1960),Savage (1968, 2002), Duellman and Lynch(1969, 1988), Hoogmoed (1969), Mudrack(1969), Myers (1969, 1982, 1987), Edwards(1971), Lynch (1971), Crump (1972), Silver-stone (1973, 1975a, 1975b,1976), Polder(1974), Durant and Dole (1975), Lescure(1975, 1976, 1991), Luddecke (1976, 2000‘‘1999’’), Wells (1978, 1980a, 1980b, 1980c),Myers et al. (1978, 1984), Myers and Daly(1976a, 1979, 1980, 1983), Cei (1980), Lim-erick (1980), Weygoldt (1980, 1987), Vigleand Miyata (1980), Zimmermann and Zim-mermann (1981, 1984, 1985, 1988), Kneller(1982), Hardy (1983), Heyer (1983), Dixonand Rivero-Blanco (1985), Jungfer (1985,1989), Frost (1986), Formas (1989), Summers(1989, 1990, 1992, 1999, 2000), Caldwell andMyers (1990), Praderio and Robinson (1990);Aichinger (1991), Morales (1992), van Wijn-gaarden and Bolanos (1992), Brust (1993),Duellman and Wild (1993), Giaretta et al.(1993), Rodrıguez and Myers (1993), Junca etal. (1994), Kaiser and Altig (1994), Coloma(1995), Cummins and Swan (1995), Jungfer etal. (1996a), La Marca (1996 ‘‘1994’’, 1998‘‘1996’’), Caldwell (1997), Fandino et al.(1997), Grant et al. (1997), Caldwell and deAraujo (1998, 2004, 2005), Junca (1998),Grant and Castro-Herrera (1998), Moralesand Velazco (1998), Boistel and de Massary(1999), Caldwell and de Oliveira (1999),Haddad and Giaretta (1999), Hoff et al.(1999), Summers et al. (1999b), Kohler(2000), Kok (2000), Lescure and Marty(2000), Lotters et al. (2000), Bourne et al.(2001), Downie et al. (2001), Hodl andAmezquita (2001), Lima et al. (2001, 2002),Myers and Donnelly (2001), Summers andSymula, (2001), Caldwell and Lima (2003),Giaretta and Facure (2003, 2004), Lima andKeller (2003), Grant (2004), Lehtinen et al.(2004), Toledo et al. (2004), and Summersand McKeon (2004).

99. ADVERTISEMENT CALLS: buzz 5 0;chirp 5 1; trill 5 2; retarded trill 5 3.Nonadditive.

Male advertisement calls played an impor-tant role in the systematics studies of Myers

and Daly (e.g., 1976b). For example, thehistrionicus group of Myers et al. (1984) isdelimited, in part, by a synapomorphic‘‘chirp’’ call. Data are available for numerousspecies (for partial review, see Lotters et al.,2003b), but their use in systematics has beenpredicated on their identification as a buzz(Myers and Daly, 1976b: 225), chirp (Myersand Daly, 1976b: 226), trill (Myers et al.,1978: 325), retarded trill (Myers and Daly,1979: 18), or retarded chirp (Myers andBurrowes, 1987: 16), and the diversity ofdendrobatid calls extends far beyond thesefew types. Although Lotters et al. (2003b)aimed to expand and standardize the defini-tions of these calls, they were aware thatknown calls of most species of dendrobatidsdo not correspond to any of these types, andadditional characterizations such as peeps,cricketlike chirps, croaks, whistled trills, orharsh peep train (e.g., Rodrıguez and Myers,1993; Grant and Castro-Herrera, 1998;Bourne et al., 2001) have been employed,although none of these is defined precisely.

It is clear that these call types arecomposites of temporal and spectral trans-formation series that should be decomposedinto independent characters for phylogeneticanalysis. Unfortunately, the necessary analy-sis of advertisement call variation was outsidethe scope of the present study, and we scoredadvertisement calls according to the pub-lished scheme in order to test prior hypoth-eses (e.g., the buzz call as a synapomorphy ofthe histrionicus group). This is highly sub-optimal, mainly because (1) many specieswere scored as unknown simply because theircalls did not fit within the current classifica-tion and not because data were unavailable,and (2) extensive information on spectral andtemporal modulation could not be incorpo-rated. We hope this may be rectified in futurestudies. Species were coded according toLotters et al. (2003b).

100. MALE COURTSHIP: STEREOTYPED

STRUT: absent 5 0; present 5 1.

Dole and Durant (1974), Wells (1980a),and Luddecke (2000 ‘‘1999’’) reported theoccurrence of this behavior (state 1) incollaris, panamensis (as inguinalis; see Grant,2004), and palmatus, respectively. Luddecke(2000 ‘‘1999’’: 309, see also p. 210 forillustration) described it as ‘‘a stereotyped


rigid-looking strut [the male] performs in thesilent intervals between advertisement calls’’.

101. MALE COURTSHIP: JUMPING UP AND

DOWN: Absent 5 0; present 5 1.

Wells (1980c: 195) described this characteras follows:

When a female or brown male moved near

a calling black male, the usual response of theblack male was to jump up and down on hiscalling perch. … Often the male would run fora few centimeters and jump so that his front feetrose 1–2 mm off the ground. Similar behaviorhas been reported in a closely related species (C.

collaris), although males of that species appar-ently leap higher off the substrate than do maleC. trinitatis.

102. FEMALE COURTSHIP: CROUCHING:absent 5 0; present 5 1.

According to Luddecke (2000 ‘‘1999’’:309), in this behavior the female crouches infront of, but does not slide underneath, themale.

103. FEMALE COURTSHIP: SLIDING UNDER

MALE: absent 5 0; present 5 1.

Luddecke (2000 ‘‘1999’’: 309) reported forpalmatus that the female crouches and then‘‘slides completely under the male’’ as one ofthe final stages of courtship.

104. TIMING OF SPERM DEPOSITION: afteroviposition 5 0; prior to oviposition 5 1.

In 1980 both Limerick (1980) andWeygoldt (1980) reported that sperm de-position in pumilio appears to occur prior tooviposition (state 1). This unusual occur-

rence has since been reported for additionalspecies by several authors (Jungfer, 1985;Weygoldt, 1987; Jungfer et al., 1996a, 2000;Lotters et al., 2000). Jungfer et al. (1996a)claimed this as a synapomorphy of Dendro-bates and rationale for placing Minyobatesin its synonymy, as done subsequently byJungfer et al. (2000). Nevertheless, severalspecies of Dendrobates sensu Jungfer et al.have been explicitly reported to have post-oviposition fertilization (e.g., histrionicus fideZimmermann, 1990: 69; arboreus fide Myerset al., 1984: 15), which suggests that thephylogenetic interpretation of this characteris not as straightforward as Jungfer et al.implied.

Fig. 62. Character 105, reproductive amplexus. State 2, cephalic amplexus (anthonyi, AMNH liveexhibit) shown in anterior (A) and lateral (B) aspects.

Fig. 63. Character 109, dorsal larval transport.State 1, present (fraterdanieli, specimens at UVC).This male nurse frog was transporting 12 tadpoles.


105. REPRODUCTIVE AMPLEXUS (fig. 62):absent 5 0; axillary 5 1; cephalic 5 2.Nonadditive.

Myers et al. (1978: 324–325) first describedand illustrated cephalic amplexus in tricolor.Although reproductive amplexus is absent innumerous dendrobatids (a variety of pseudo-amplectant positions–including cephalicgrasping–may be employed in aggressiveand/or courtship behavior), cephalic amplex-us was cited by numerous authors (e.g.,Duellman and Trueb, 1986; Myers and Ford,1986; Myers et al., 1991) as a dendrobatidsynapomorphy, with the absence in numer-ous dendrobatids explained as a derived loss.A similar amplectant position was reportedfor Hypsiboas faber by Martins and Haddad(1988), but the sampled outgroup speciesexhibit axillary amplexus.

106. CLOACAL APPOSITION: absent 5 0;present 5 1.

Crump (1972: 197) first reported theoccurrence of this character-state in granuli-ferus, in which the male and female faceopposite directions and bring their cloacaeinto contact.

107. EGG DEPOSITION SITE: aquatic 5 0;terrestrial: leaf litter, soil, under stones 5 1;terrestrial: phytotelmata 5 2. Additive.

This character is coded additively to reflectthe increasing or decreasing degree of asso-ciation with ground-level standing or flowingwater.

108. EGG CLUTCH ATTENDANCE: none 5 0;male 5 1; female 5 2; both 5 3. Non-additive.

See discussion of sex of nurse frog(Character 110, below) for the rationalebehind treating biparental care as a separatestate instead of polymorphism.

109. DORSAL TADPOLE TRANSPORT

(fig. 63): absent 5 0; present 5 1.Noble (1927: 103) noted that his grouping

of dendrobatids on morphological grounds‘‘receives an eloquent support from lifehistory data’’ as well, pointing out that malesof species of Dendrobates and Phyllobatestransport tadpoles on their back to pools(state 1), and, further, that ‘‘[n]o otherSalientia have breeding habits exactly likeDendrobates and Phyllobates’’ (p. 104). Forthe present purposes, we refer only to dorsaltransport by genetic parents (see below), and

we follow Ruthven and Gaige (1915: 3) inreferring to the parent that performs larvaltransport as the nurse frog.

Among outgroup taxa, males of Rhino-derma darwinii transport larvae, which Laur-ent (1942: 18) claimed as evidence of closerelationship to dendrobatids. However, maleRhinoderma transport young in their hyper-trophied vocal sacs (see Noble, 1931: 71for illustration), whereas dendrobatidstransport tadpoles on their backs. Severalother anurans transport their young ontheir backs (e.g., Hemiphractus, Stefania,Gastrotheca), but they do so beginning withthe egg clutch, whereas in dendrobatidstransport is exclusively post-hatching.Among Neotropical anurans, the only speciesreported to have terrestrial (nontransported)eggs and dorsally transported tadpoles isCycloramphus stejnegeri (Heyer and Crom-bie, 1979). Tadpole transport is not knownfor the sampled species of Cycloramphus(C. boraceiensis), but Giaretta and Facure(2003) reported male egg attendance (alsopresent in C. dubius and C. juimiria; C.F.B.Haddad, personal obs.), which leavesopen the possibility of tadpole transport.Outside of the Neotropics, apparentlyidentical parental care occurs in the dicro-glossids Limnonectes finchi and L. palavanen-sis (Inger and Voris, 1988) and the sooglossidSooglossus sechellensis (Lehtinen and Nuss-baum, 2003). In the hemisotid Hemisusmarmoratus, maternal tadpole transportmay occur as the female digs a subterraneantunnel to a water body (Lehtinen andNussbaum, 2003).

Dorsal tadpole transport by parents is herecoded as a single transformation series, buteven the extremely limited evidence availablesuggests this is much more complex andprobably involves multiple characters. Steb-bins and Hendrickson (1959: 509) reported insubpunctatus that ‘‘[t]he tadpoles are an-chored to the back of the frog by a stickymucus.’’ Myers and Daly (1980: 19) furthernoted that

[i]n some dendrobatids, this attachment isaccomplished solely by mere surface adhesionbetween the mucus and the tadpoles’ flattenedor slightly concave bellies, and the larvae areeasily moved about and dislodged. … In otherdendrobatids … the mucus attachment seems


almost gluelike and the tadpoles are veryresistant to being dislodged.

To this we add only that it is common fortadpoles to wriggle around freely on thenurse frog’s back without being prodded(especially if few tadpoles are being trans-ported by a large frog, such as bicolor), givingthe impression that they adjust themselves tothe nurse frog’s movements.

Ruiz-Carranza and Ramırez-Pinilla (1992)studied the histology of the contact surfacesof nurse frogs and transported tadpoles invirolinensis and found numerous modifica-tions in both the dorsal integument of thenurse frog and the ventral integument of thelarvae. Luddecke (2000 ‘‘1999’’) found ex-perimentally that recently hatched larvae ofpalmatus did not mount a rubber modelmoistened with water, mounted but immedi-ately abandoned a rubber model treated witheither male or female skin secretions, andwould only mount and settle on a live frog,with no sexual discrimination. In a lesscontrolled experiment with hatching anthonyiT. Grant found that gently touching the jellycapsule with a finger was sufficient tostimulate hatching, immediate mounting,settling, and attachment (i.e., the tadpolesremained attached to the finger submerged inwater for .1 min until they were forciblydislodged); however, the male nurse frog hadalready removed most of the tadpoles fromthe clutch, which may have ‘‘primed’’ theremaining embryos for hatching and trans-port. As coded in Character 110, the sex ofthe nurse frog varies among species, and littleis known about the biology of this kind of sexrole reversal. Much more research is requiredto understand the evolution of dorsal tadpoletransport in dendrobatids.

110. SEX OF NURSE FROG: male 5 0;female 5 1; both 5 2. Nonadditive.

Among species that transport larvae, therole of the nurse frog is typically assumed byone sex (Wells, 1978, 1980a, 1980b, 1980c).However, in some species, both sexes havebeen observed carrying tadpoles. Myers andDaly (1983) found experimentally that inanthonyi (as tricolor) the father was normallyresponsible for tadpole transport and wouldactively prevent the mother from approach-ing the developing clutch, but that removal of

the male shortly after breeding led to femalebrood care and larval transport. They sug-gested that parental care is competitive, thatis, the sexes compete to care for the offspring.This is at least consistent with Aichinger’s(1991) observation of 38 male nurse frogs andonly a single female nurse frog. Caldwell andde Araujo (2005) reported that in brunneusand femoralis males usually transport larvaebut females will occasionally perform thisrole, and Silverstone (1976: 38) reportednurse frogs of both sexes for petersi. It isunknown how widespread this behavior is(i.e., if both sexes are usually potentialcarriers, even though one sex predominantlyassumes this role, as in tricolor), but it is notuniversal. It is unknown how widespread thisbehavior is (i.e., if both sexes are usuallypotential carriers, even though one sex pre-dominantly assumes this role, as in tricolor),but it is not universal. Host Luddecke (in litt.,08/31/00) found experimentally that palmatusdoes not exhibit this behavior; in his experi-ments, Luddecke found that mothers atetheir eggs when the fathers were removed. Asnoted for Character 108, Luddecke (2000‘‘1999’’) also found that tadpoles mountedmales or females indiscriminately, whichsuggests that a potential for female transportmay be primitive.

Given the paucity of experimental data, itis unclear if all cases of both sexes transport-ing tadpoles are the result of the samemechanism and/or a transformation event.For the time being, we coded each speciesbased on available information. We havetherefore scored species as having males(state 0), females (state 1), or both sexes(state 2) assume the role of nurse frog. Thischaracter individuation will undoubtedly re-quire modification as more data are obtainedon this behavior.

We coded biparental transport as a sepa-rate state rather than an ambiguous poly-morphism because the behavioral modifica-tions required to achieve biparental care donot apply to male or female care alone, thatis, it involves more than just the co-occur-rence of states 1 and 2. Also, we did notspecify any particular additivity for thistransformation series, as there is no evidencethat the shift between sexes requires a co-operative (or competitive) intermediate bi-


parental stage (although this could be in-dicated by phylogenetic analysis).

Savage (2002) reported male nurse frogs intalamancae, and Summers and McKeon(2004: 62, fig. 3) scored femoralis, hahneli (as‘‘hahnei’’), talamancae, and trilineatus (as‘‘trilieatus’’) as having exclusively male pa-rental care. However, nurse frogs of bothsexes have been reported for femoralis (Silver-stone, 1976; Lescure, 1976; for a recent reportsee Caldwell and de Araujo, 2005), hahneli (aspictus; Aichinger, 1991) and trilineatus (Ai-chinger, 1991), and exclusively female nursefrogs have been reported for talamancae(Grant, 2004 and references therein). Insofaras Savage and Summers and McKeon did notdispute those reports or provide specimendocumentation for confirmation, we dismisstheir reports as erroneous.

Adding to the behavioral complexity andthe difficulty in coding this character, Wey-goldt (1980) found in pumilio that motherstransport their larvae to water, whereas malesmay perform nonparental infanticide bytransporting unelated tadpoles and not de-positing them in water, thus achieving sexualinterference. As noted above, we did notscore nonparental transport as part of thistransformation series.

111. LARVAL HABITAT: ground level poolor slow-flowing stream or other body ofwater 5 0; phytotelmata 5 1; nidicolous 5 2.Nonadditive.

Note that there is a logical dependencybetween larval habitat and dorsal tadpoletransport (Character 108) in that nidicolouslarvae are, by definition (Altig and Johnston,1989; McDiarmid and Altig, 1999), nottransported. Nevertheless, the two charactersare not coextensive and are clearly indepen-dent: Lack of transport may also be associ-ated with either state, and nurse frogs maytransport larvae to either a ground level bodyof water or phytotelm.

Although ‘‘phytotelm’’ often refers tochambers above ground (e.g., bromeliads),technically the term applies to any chambersin a plant. Moreover, whether on or abovethe ground, these phytotelmata are expectedto be biologically equivalent (e.g., bothmicrohabitats offer limited space, nutrients,and other resources, and have a potentiallyhigh risk of predation), and we therefore did

not discriminate between ground-level andhigher phytotelmata. For example, we fol-lowed Caldwell and de Araujo (1998; 2004) inscoring castaneoticus as a phytotelm breederbecause it uses Brazil nut husks.

112. LARVAL DIET: detritivorous 5 0;predaceous 5 1; oophagous 5 2; endotrophic5 3. Nonadditive.

The vast majority of anurans have detriti-vorous tadpoles (state 0). We assumed thatlarvae found in ground level pools or streamsor other large bodies of water (i.e., state 0 ofCharacter 111) are detritivorous; unless dietis actually known, larvae of other habitatswere coded as unknown (‘‘?’’) for thischaracter. Numerous species of dendrobatidsare aggressive predators that consume con-and heterospecific tadpoles and arthropodlarvae as an important component of theirdiet (Caldwell and de Araujo, 1998; state 1).Several species consume sibling oocytes(oophagous, state 2), either exclusively (his-trionicus group; Limerick, 1980) or as part ofa predaceous diet (state 1; e.g., vanzolinii;Caldwell and de Oliveira, 1999). We codedthe latter taxa as polymorphic (see alsoCharacter 113, provisioning of oocytes forlarval oophagy).

Four species with endotrophic larvae (state3) have been described: chalcopis (not in-cluded in this study), degranvillei, nidicola,and stepheni (for review and description ofnidicola see Caldwell and Lima, 2003). Someamount of larval growth prior to deposition(e.g., during transport; Wells, 1980b) isprobably widespread, but complete endotro-phy is much more limited and tends to becorrelated with a variably reduced morphol-ogy. Nevertheless, the unmodified larva ofchalcopis (Kaiser and Altig, 1994) demon-strates the transformational independence ofendotrophy and the various morphologicalreductions (see also Altig and Johnston,1989). Likewise, the occurrence of endotro-phy is independent of tadpole habitat:degranvillei is transported (Lescure andMarty, 2000; tadpole transport was alsopredicted for chalcopis by Junca et al.,1994), whereas the remaining endotrophictadpoles are nidicolous.

113. PROVISIONING OF OOCYTES FOR

LARVAL OOPHAGY: biparental 5 0; femaleonly 5 1.


Caldwell and de Oliveira (1999) reportedprovisioning of eggs for consumption bysibling tadpoles in vanzolinii, as did Bourneet al. (2001) in beebei. In these species, eggprovisioning is stimulated by male courtshipbehavior and is therefore biparental (state 0),and larval diets include a variety of foods (foradditional records, see Lehtinen et al., 2004).In other oophagous species (e.g., histrionicus)tadpole care is undertaken exclusively by thefemale. An alternative way to delimit state 1is as obligate oophagy, as it appears thattadpoles of these species feed only on eggs(obligate oophagy demonstrated experimen-tally for pumilio by Brust, 1993; Pramuk andHiler, 1999), whereas the others are pre-daceous (Caldwell and de Araujo, 1998).

Zimmermann and Zimmermann (1988)reported biparental provision of oocytes inventrimaculatus (as quinquevittatus) in captiv-ity, but Summers et al. (1999b) reportedexclusively male care in Peruvian ventrima-culatus. Caldwell and Myers (1990) hypoth-esized that ventrimaculatus is a complex ofcryptic species, which is supported by thisbehavioral variation.

114. ADULT ASSOCIATION WITH WATER:aquatic 5 0; riparian (,3 m from water) 5 1;independent of water (up to ca. 30 m fromwater) 5 2. Additive.

Myers et al. (1991) characterized nocturnusas aquatic, which they contrasted with speciessuch as panamensis (as inguinalis; see Grant,2004) and latinasus, which are riparian andindependent of streams, respectively. Post-metamorphic frogs of any species may befound in or near water, and environmentalvariation must be taken into account (i.e.,during dry seasons or at drier localities frogsthat are otherwise found well into the forestwill congregate near sources of water), butthe degree of commitment to or dependencyon an aquatic environment segregates den-drobatids into at least three groups. Amongdendrobatids, nocturnus appears to be theonly aquatic species, that is, individuals aregenerally found immersed in water (state 0).A much greater number of dendrobatids areriparian (state 1). These species are almostentirely confined to the areas immediatelyadjacent to streams, where they establish anddefend streamside territories (e.g., Wells,1980a, 1980c). When disturbed these species

seek refuge in water and not in leaf litter ordebris beside the stream. The third group ofspecies is effectively independent of water(state 3). As noted by Funkhouser (1956: 78)for espinosai, these species ‘‘scurry underdebris for safety; they do not take to watereven when it is close by’’, and territorial andcourtship behaviors occur well away fromground water. Despite their relative indepen-dence from water, the density of these frogsmay be greater nearer to streams, even inextremely wet environments such as theColombian Choco (T. Grant, personal obs.)where general moisture requirements areunlikely to be a limiting factor. This isprobably due to reproductive factors: Manyof these species are known to transportlarvae from terrestrial nests to streams orground-level pools, and it is predictable thatselection would favor preference for sitescloser to more permanent, larger bodies ofwater.

A potential fourth character-state is ar-boreality. For example, Myers et al. (1984)named arboreus in recognition of that species’arboreal habitat preference, whereas otherspecies (e.g., fraterdanieli) are active exclu-sively on the ground and only climb intovegetation (never more than 1 m) to sleep.However, between these two extremes lievariations that defy simple codification. Forexample, bombetes is a leaf-litter frog thatclimbs up to 30 m above ground to depositlarvae in bromeliads (T. Grant, personalobs.; A. Suarez-Mayorga, personal com-mun.). Similarly, histrionicus forages in leaflitter on the ground but calls from perches invegetation above ground (Silverstone, 1973;Myers and Daly, 1976b). Clearly there areevolutionary transformation events embed-ded in these behavioral variations, but theextent to which variation is obligate orfacultative is unclear, and we have chosento group putatively arboreal and terrestrialspecies as state 2. Assuming the additivity ofthis transformation series, the transformationfrom state 1 to state 2 applies to all of thesespecies (as coded), and we have failed torecognize the additional transformation(s)from state 2a (terrestrial) to state 2b (arbo-real).

115. DIEL ACTIVITY: nocturnal 5 0; di-urnal 5 1.


Myers et al. (1991) cited the transforma-tion from nocturnal to diurnal activity asevidence for the monophyly of all dendroba-tids minus nocturnus. As has been noted byseveral authors (e.g., Myers et al., 1991;Coloma, 1995; Duellman, 2004), someother species (e.g., riveroi, bocagei, nexipus)exhibit crepuscular or limited nocturnalactivity, at least facultatively (e.g., onbrightly moonlit nights). Although the con-ditions that surround this behavior areunclear, we coded these species as poly-morphic.

116. TOE TREMBLING: absent 5 0; present5 1.

We have observed several species to exhibittoe trembling or toe tapping, whereby usuallythe fourth toe (sometimes also the third)trembles or twitches rapidly up and down.Little is known about this behavior. Mostobservations derive from captive individuals,and there is no known function. It does notappear to be involved in intraspecific visualcommunication, as individuals do not altertheir behavior notably when an individualbegins toe trembling, and toe trembling maybe observed in individuals that are isolated orin groups. Toe trembling is not continuousand only occurs in active frogs. However,although quantitative data are lacking, onsetand/or vigor of toe trembling in dendrobatidsdoes not seem to correlate with any particularstimulus and does not obviously originate asan epiphenomenon (Hodl and Amezquita,2001). Toe trembling may (or may not) occurwhile foraging and during inter- and in-traspecific interactions with individuals of thesame or opposite sex. Hartmann et al. (2005)reported toe (and finger) trembling for thehylid Hyspiboas albomarginatus, which theyinterpreted to be visual signaling.

117. HYALE ANTERIOR PROCESS: absent 5

0; present 5 1.

All dendrobatids examined possess a singleanterior process on each hyale (state 1), andit is both present and absent (state 0) in thesampled outgroup species. Myers and Ford(1986) cited the occurrence of a secondanterior process on the hyalia of Atopophry-nus syntomopus as evidence that it is nota dendrobatid; we did not sample this taxonin the present study and therefore did not testtheir hypothesis.

118. SHAPE OF TERMINAL PHALANGES: T-shaped 5 0; knobbed 5 1.

Following Lynch’s (1971) terminology, thespecies sampled in this study possess T-shaped and knobbed phalanges.

119–128. EPICORACOIDS

Pectoral girdle architecture has been key inall discussions of dendrobatid relationshipssince Boulenger (1882). Character-states havegenerally been delimited in terms of theoverlap or fusion of the epicoracoids and/orthe presence or absence of epicoracoid horns(for historical usages, see Kaplan, 2004), withthe epicoracoids of dendrobatids character-ized as entirely fused and nonoverlapping andlacking epicoracoid horns, as in ‘‘firmisternal’’taxa.10 However, this is clearly an oversimpli-fication (e.g., Noble, 1926; Kaplan, 1994,1995, 2000, 2001, 2004). Recently, Kaplan(2004) divided girdle architecture variants intoseparate transformation series relating todegree of fusion (freedom) and overlap (non-overlap), and he proposed explicit character-states, which we employ here.

Of most relevance to the problem ofdendrobatid phylogeny, Noble (1926)claimed that the entirely fused epicoracoidsof subpunctatus overlap during ontogeny,a finding that was challenged by Griffiths(1959), Lynch (1971a), and Ford (1989), butultimately vindicated by Kaplan (1995).However, Kaplan (1995) interpreted differ-ences between the overlap in subpunctatusand ‘‘arciferal’’ taxa (e.g., Bufo) as evidencethat the overlap is nonhomologous andtherefore not evidence of common ancestry(contra Noble, 1926).

10We place ‘‘firmisternal’’ and ‘‘arciferal’’ in quotesand use the terms to denote the taxa they have beenassociated with rather than the pectoral girdlemorphology they purport to designate. Both firmis-terny and arcifery are clearly complexes of characters(Kaplan, 2004), the conflation of which has led tomuch unnecessary confusion in anuran systematics.Although it may be appropriate to treat them as singleunits in functional studies, the only defensibleapproach in phylogenetics is to treat each trans-formationally independent character independently,and we concur with Kaplan that the terms should beabandoned.


To date, the only dendrobatid in whichoverlap has been detected is subpunctatus.Kaplan (1995: 302) also examined abditaur-antius (adult), palmatus (adult), and viroli-nensis (Gosner, 1960, stages 42–43) and ‘‘didnot find any evidence of overlap’’, andGriffiths (1959) reported that overlap isabsent in trinitatis (not trivittatus, as reportedby Kaplan, 1995).

Although we did not perform the detailedhistology necesary to score these charactersprecisely, epicoracoid morphology has playedan important role in all previous discussionsof dendrobatid phylogeny and we believe itwould be inappropriate to exclude it alto-gether. Therefore, although we are cognizantof the potential errors that may be incorpo-rated into the analysis, we coded degree offusion and overlap in adults (or near adults)as precisely as possible through examinationof cleared and stained whole specimens.Although Kaplan (1995: 301) stated that insubpunctatus ‘‘the girdle halves overlap inadults except for a small area of ventralfusion’’, this was not visible in cleared andstained specimens, and so for consistency wecoded this species as lacking overlap. Insofaras we did not detect overlap in any otherdendrobatid, and Kaplan (1995) argued thatoverlap in subpunctatus is nonhomologouswith the overlap of ‘‘arciferal’’ taxa, codingthe occurrence of overlap in this taxon wouldresult in an autapomorphy and thereforewould not affect the results of the presentanalysis.

119. EPICORACOID FUSION: fused fromanterior tips to posterior tips 5 0; fusedfrom anterior tips of epicoracoids to levelmidway between the posterior levels of theprocoracoids and the anterior ends of thecoracoids, free posteriorly 5 1; fused fromanterior tips to a level slightly posterior tomedial ends of clavicles, free posteriorly 5 2.Additive.

States 0, 1, and 2 correspond to states E,C, and A, respectively, of Kaplan (2004: 94).State 1 is intermediate in the degree of fusion,which we considered to be evidence for thehypothesis of 0 « 1 « 2 additivity for thistransformation series.

120. EPICORACOID OVERLAP: nonoverlap-ping 5 0; overlapping from level slightlyposterior to level of procoracoids to anterior

level of sternum 5 1; overlapping from levelbetween posterior level of procoracoids andanterior ends of coracoids to posterior levelof coracoids 5 2; overlapping from levelslightly posterior to medial ends of clavicle tolevel slightly posterior to anterior level ofsternum 5 3. Nonadditive.

States 0, 1, 2, and 3 correspond to states B,E, C, and A, respectively, of Kaplan (2004:94). Because variation in overlap involvescomplex changes in epicoracoid structure, wewere unable to find evidence to select onehypothesis of additivity over another; wetherefore treated this character nonaddi-tively.

121. ANGLE OF CLAVICLES (fig. 64): di-rected laterally 5 0; directed posteriorly 5 1;directed anteriorly 5 2. Nonadditive.

In most dendrobatids each clavicle runslaterad, perpendicular to the sagittal plane(state 0). In some species, the clavicles aredirected posteriad, running approximatelyparallel to the posterior margin of thecoracoid (state 1). Clavicles directed anteriad(state 2) are confined to certain species in theoutgroup. The intermediacy of the laterallydirected clavicles is suggestive of additivity;however, pectoral girdle ontogeny does notproceed in this way, and we therefore scoredthis transformation series nonadditively.

122. ACROMION PROCESS: cartilaginous,distinct 5 0; fully calcified (or ossified),continuous with clavicle and scapula 5 1.

Fig. 64. Character 121, angle of clavicles. A:State 0, directed laterad (steyermarki, AMNH118572). B: State 1, directed anteriad, approxi-mately parallel to the posterior margin of thecoracoid (opisthomelas, AMNH 102582). C: State2, directed anteriad (Eupsophus roseus, KU 207501).


The acromion processes of some taxa arecartilaginous (state 0) in mature specimens,whereas in others they are extensively calci-fied or ossified (state 1). We did notdistinguish between extensive calcificationand ossification. As with other osteologicalcharacters that vary ontogenetically (charac-ters 127, 134–138), adult females are mostextensively ossified.

123. PREZONAL ELEMENT (OMOSTERNUM):absent 5 0; present 5 1.

124. PREZONAL ELEMENT (OMOSTERNUM)ANTERIOR EXPANSION: not expanded distally,tapering to tip 5 0; weakly expanded, to2.53 style at base of cartilage or equivalent 5

1; extensively expanded distally, 3.53 style orgreater 5 2. Additive.

125. PREZONAL ELEMENT (OMOSTERNUM)SHAPE OF ANTERIOR TERMINUS: rounded orirregularly shaped 5 0; distinctly bifid 5 1.

126. PREZONAL ELEMENT (OMOSTERNUM)SHAPE OF POSTERIOR TERMINUS: simple 5 0;notched, forming two struts 5 1; continuouswith epicoracoid cartilage 5 2. Nonadditive.

127. PREZONAL ELEMENT (OMOSTERNUN)OSSIFICATION: entirely cartilaginous 5 0;medially ossified (cartilaginous base and tip)5 1; basally ossified (cartilaginous tip) 5 2;entirely ossified 5 3. Additive.

128. SUPRASCAPULA ANTERIOR PROJEC-

TION: cartilaginous 5 0; heavily calcified 5 1.

129. STERNUM SHAPE: simple (rounded,irregular) 5 0; medially divided 5 1.

The posterior termination of the sternumis either simple (rounded or irregularlyshaped; state 0) or distinctly divided medial-ly, forming either two prongs or two broad,rounded lobes. We also observed indepen-dent variation in the lateral expansion of thesternum. For example, even though thesternum of both species is medially divided,in panamensis (UMMZ 167459) it is broadlyexpanded, whereas in juanii (ICN 5097) the

sternum is tapered. However, we also ob-served confounding intermediate and othervariation and were unable to individuatestates objectively for the current study.

130. ZYGOMATIC RAMUS OF SQUAMOSAL

(fig. 65): elongate, slender, pointed 5 0; verylong and slender 5 1; robust, truncate, andelongate 5 2; shorter and less robust but stillwell defined 5 3; well defined, moderatelength, abruptly directed ventrad 5 4; in-conspicuous, poorly differentiated 5 5; verysmall, inconspicuous, hooklike 5 6; minis-cule bump 5 7; robust, elongate, in broadcontact with the maxilla 5 8. Nonadditive.

The zygomatic ramus of the squamosalvaries extensively and forms a series ofcomplex morphological transformations. Instate 0, the zygomatic ramus is elongate(approximately half the length of the ascend-ing ramus and extending well anterior pastthe tympanic ring), slender, gently curved,and pointed. State 1 is a very long andslender process. State 2 is robust, truncate,and elongate (extending anterior to thetympanic ring, but not as long as state 2).State 3 is shorter and less robust than state 2but is still a conspicuous shaft that usuallyextends anterior to the tympanic ring. Likethe processes of states 0, 1, and 2, the axis ofstate 3 is at most only weakly inclined towardthe maxilla. The zygomatic ramus of state 4 isalso well defined, but it is distinctly andabruptly directed ventrad, its axis pointingalmost straight down at the maxilla, that is,a line from the zygomatic ramus wouldintersect the posterior extreme of maxilla,and it does not extend anterior to thetympanic ring. State 5 is an inconspicuous,poorly differentiated process. The zygomaticramus of most of the sampled species is a verysmall, inconspicuous, hooklike process (state6). McDiarmid (1971) considered the zygo-matic ramus to be absent in Melanophrynis-

R

Fig. 65. Character 130, zygomatic ramus of squamosal. A: State 0 (Eupsophus roseus, KU 207501). B:State 1 (Cycloramphus fuliginosus, KU 92789). C: State 2 (nocturnus, AMNH 130041). D: State 2 shown ina dissected whole specimen (palmatus, AMNH 20436). E: State 3 (trinitatis, AMNH 118389). F: State 4(trivittatus, AMNH 118428). G: State 5 (espinosai, AMNH 118417). H: State 6 (bocagei, UMMZ 182465).I: State 7 (Melanophryniscus stelzneri, AMNH 77710). J: State 8 (Megaelosia goeldii, AMNH 103952;squamosal colored red).


cus; however, we detected a miniscule bump(state 7) in Melanophryniscus stelzneri, whichwe considered to be homologous with thezygomatic ramus. (Regardless, we did notobserve this state in any other species in-cluded in the present analysis, so coding it as‘‘absent’’ or ‘‘a miniscule bump’’ has nobearing on the outcome of analysis.) InMegaelosia goeldii, the robust zygomaticramus extends anteroventrad to be in broadcontact with the maxilla (state 8).

131. ORIENTATION OF ALARY PROCESS OF

PREMAXILLA: directed anterolaterally 5 0;directed dorsally 5 1; directed posterodor-sally 5 2. Additive.

Myers and Ford (1986) claimed the ante-rolaterally tilted alary process as a synapo-morphy of dendrobatids, although severalother taxa also share this state (e.g., Lynch,1971). We treated this transformation seriesadditively (0 « 1 « 2) based on theargument that the rearrangement in skullarchitecture required to alter the orientationof the alary process would necessitate passingthrough the intermediate stage.

132. PALATINES: absent 5 0; present 5 1.Variation in the occurrence of the palatine

bones among dendrobatids has been docu-mented by numerous authors (e.g., Silver-stone, 1975a; Myers and Ford, 1986), andKaplan (1997) interpreted the characterphylogenetically. Trueb (1993) consideredthe neobatrachian palatine to be nonhomol-ogous with the palatine of other vertebrates,and she is almost certainly correct. Neverthe-less, this bone would unquestionably beidentified as a palatine if anurans were foundto be rooted on a neobatrachian. As such, thevalidity of Trueb’s distinction rests on thephylogenetic position of neobatrachians, thatis, it is a conclusion of phylogenetic analysis,not a premise. We therefore follow Haas(2003) in referring to this bone as thepalatine.

133. QUADRATOJUGAL–MAXILLA RELA-

TION: overlapping 5 0; separated 5 1.In dendrobatids, the quadratojugal and

maxilla are never in contact or tightly boundbut are instead loosely bound by ligamentoustissue. In some species, the two bones overlap(state 0), whereas in others the anterior tip ofthe quadratojugal does not reach the level ofthe posterior tip of the maxilla.

134. NASAL–MAXILLA RELATION (fig. 66):separated 5 0; in contact 5 1.

The nasal and maxilla may be separate(state 0) or contact each other. We did notdistinguish between overlap and fusion be-cause gross examination under a dissectingmicroscope proved inadequate to determinethe status of many specimens and thenecessary histological study was infeasiblefor the present study.

135. NASAL–SPHENETHMOID RELATION

(fig. 67): free, separate 5 0; overlapping orfused 5 1.

In state 0, the nasal and sphenethmoid donot overlap, whereas in state 1 those bonesare either overlapping or fused. We did notdistinguish between overlapping and fusionas the necessary histological analysis wasinfeasible for the present study.

136. FRONTOPARIETAL FUSION: entirelyfree 5 0; fused posteriorly 5 1; fused alongentire length 5 2. Additive.

Ontogenetic variation in frontoparietalfusion suggests that it proceeds anteriorly.We therefore treated this character additive-ly.

137. FRONTOPARIETAL–OTOCCIPITAL RE-

LATION: free 5 0; fused 5 1.Among dendrobatids, there is variation in

the relation of the frontoparietal and otoccip-ital (i.e., the fused prootic and exoccipital;Lynch, 1971: 52), being free (state 0) in sometaxa and fused (state 1) in others. Lynch(1971) documented variation in this characterin numerous outgroup taxa.

138. EXOCCIPITALS: free, separate 5 0;fused sagittally 5 1.

The exoccipital portions of the fusedotoccipital bones (see Lynch, 1971: 52) maybe separated by cartilage (i.e., chondrocranialossification may be incomplete; state 0) ormay be fused sagittally (state 1). A furtherpotential state is for them to abut but notfuse, but we did not observe this among thespecimens examined.

139. MAXILLARY TEETH: absent 5 0;present 5 1.

Variation in the occurrence of teeth hasbeen used consistently in dendrobatid sys-tematics (see Grant et al., 1997 for discus-sion). In the more recent literature, Edwards(1971: 147) stated that dendrobatids ‘‘can bedivided into two groups—those species lack-


ing maxillary teeth (Dendrobates) and thosehaving maxillary teeth (Phyllobates andColostethus).’’ Silverstone (1975a) showedthat the situation is more complicated dueto character conflict and polymorphism. (Seealso Materials and Methods for uncodedvariation in maxillary tooth size and shape.)

140. MAXILLARY TOOTH STRUCTURE: ped-icellate 5 0; nonpedicellate 5 1.

Most anurans have pedicellate teeth,whereby the tooth is divided into a pediceland crown (Parsons and Williams, 1962).Parsons and Williams (1962: 377) examinedthe teeth of bocagei (as Phyllobates bocagii)and palmatus (as Phyllobates granuliventris)and found that ‘‘the division is certainly notmarked in gross structure and is quiteprobably lacking.’’ Myers et al. (1991: 11)further pointed out that there is no ‘‘patternof physical separation of crowns frompedicels (breakage is irregular)’’, and that‘‘the loss or significant obfuscation of theusual amphibian pedicellate condition war-rants attention as a possible synapomorphyfor the Dendrobatidae.’’ We coded toothstructure from gross examination of clearedand stained specimens only, although histo-logical study is required to address thisproblem decisively.

141. VOMERINE TEETH: absent 5 0; present5 1.

142. RETROARTICULAR PROCESS OF MAN-

DIBLE (fig. 68): absent 5 0; present 5 1.

Myers and Ford (1986) noted the occur-rence of a retroarticular process on themandible as a distinguishing characteristicof dendrobatids, and Ford and Cannatella(1993) listed it as one of two uniquesynapomorphies. Although many dendroba-tids are characterized by conspicuously elon-gate retroarticular processes, Myers et al.(1991: 11) noted that in nocturnus the processis ‘‘present, but always short (compared withother dendrobatids) although somewhat vari-able in length.’’ As shown in figure 68, thereis considerable interspecific variation in thelength of the retroarticular process. However,we were unable to delimit states, in partbecause no clear choice for a standardreference point has been identified.

143. EXPANSION OF SACRAL DIAPOPHYSES

(fig. 69): unexpanded 5 0; moderately ex-panded 5 1; strongly expanded 5 2. Addi-tive.

The shape of the sacral diapophyses hasbeen used since Boulenger (1882). Ford(1989) mistakenly cited Duellman and Trueb(1986) as having placed Dendrobatidaeamong ranoids based in part on their sharinground-shaped sacral diapophyses (Duellmanand Trueb did not include that character intheir matrix), but it has, nonetheless, playedan important role in anuran systematics.

The state found in dendrobatids hasusually been referred to as round or cylindri-cal (e.g., Duellman and Trueb, 1986), but thesacral diapophyses are invariably elliptical incross section. For this reason we refer insteadto the degree of expansion of the sacraldiapophyses. Emerson (1982) quantified ex-pansion by measuring the angle formed by

Fig. 66. Character 134, nasal–maxilla relation.A: State 0, nasal and maxilla broadly separated(silverstonei, AMNH 91847). B: State 1, greatermagnification showing nasal and maxilla over-lapping or fused (bassleri, AMNH 43402).

Fig. 67. Character 135, nasal–sphenethmoidrelation. A: State 0, separate (bassleri, AMNH43402). B: State 1, overlapping or fused (nocturnus,AMNH 130014). In this case the nasals clearlyoverlap but are not fused with the sphenethmoid,but in other species the distinction between thebones is not as clear.


the expansion. Here we code this character asthe ratio of the width of the tip of thediapophysis to the width of the base of thediapophysis. Variation is not continuous; allstates are separated by gaps. Unexpandeddiapophyses are subequal at the base and tip.Moderately expanded diapophyses are 1.5–2.5 times wider at the tip than at the base;greatly expanded diapophyses are at least 2.7

times greater at the tip. Sacral diapophysesoften bear irregular flanges that we did notinclude in the measurement of width.

144–146. VERTEBRAL FUSION

Noble (1922: 15) reported fusion ofvertebrae 2 + 3 and 8 + 9 (i.e., 8 + sacrum)in two species of dendrobatids (pumilio [asDendrobates typographicus] and probablyhistrionicus or sylvaticus Funkhouser [dis-

Fig. 68. Length variation in character 142, retroarticular process of the mandible. A: nocturnus,AMNH 130041. B: riveroi, AMNH 134142. C: vittatus, AMNH 118386. D: lehmanni Myers and Daly,AMNH118442. E: pratti, AMNH118364. F: ‘‘Neblina species’’, AMNH 118667.


cussed under the tentative name Dendrobatestinctorius]). Silverstone (1975a: 5) summa-rized his observations of vertebral fusion indendrobatids as ‘‘absent in the 17 specimensof Colostethus examined, present in only twoof the 29 specimens of Phyllobates examined,and present in 28 of the 46 specimens ofDendrobates examined.’’ He also noted thatvertebral fusion varies intraspecifically, andwe also found intraspecific variation amongequivalent semaphoronts.

144. VERTEBRA 8 AND SACRUM: free 5 0;fused 5 1.

145. VERTEBRAE 1 AND 2: free 5 0; fused 5 1.

Myers et al (1991:11) reported ventralfusion of vertebrae 1 and 2 in nocturnus(AMNH 130047); however, close examina-tion showed that the vertebrae are tighlybound by ligamentous tissue but are notfused.

146. VERTEBRAE 2 AND 3: free 5 0; fused 5 1.

147–172. ALKALOID PROFILES

Dendrobatid frogs are known to possessa diverse array of over 450 alkaloids (Daly etal., 1999; J. W. Daly, in litt., 01/25/05). Use ofalkaloid profiles as transformation series iscomplicated, in part, because it appears that‘‘some, if not all … ‘dendrobatid alkaloids’may have a dietary origin’’ (Daly et al.,1994a; see also Myers and Daly, 1976b: 194–197; Myers et al., 1995), which means that theoccurrence of a given alkaloid may bedetermined not by the genotype but byavailability of the dietary source in theenvironment (making this a nonheritablecharacteristic, i.e., not a character). Saporitoet al. (2003) identified a species of siphonotidmillipede as the likely dietary source of

spiropyrrolizidine and Saporito et al. (2004)identified certain species of formicine ants asthe natural dietary source of two pumiliotox-ins found in pumilio. Dumbacher et al. (2004)identified melyrid beetles as the probabledietary source of batrachotoxins for the NewGuinean passerine birds Pitohui and Ifritaand further conjectured that this is the likelysource of the alkaloids in Phyllobates as well.There is often considerable variation in thealkaloid profiles of conspecifics from boththe same and disjunct populations (e.g.,Myers et al., 1995). Captive reared offspringof wild-caught, toxic frogs are nontoxic if fedcrickets and fruit flies, but readily accumulatealkaloids if present in the diet (either as a puresupplement to a fruit fly diet or in leaf-litterarthropods; Daly et al., 1992, 1994a, 1994c).

Nevertheless, despite the environmentaldependency, there is also clearly a heritableaspect to the alkaloid uptake system. It hasbeen found experimentally that azureiventris,panamensis, and talamancae do not accumu-late detectable amounts of alkaloids wheningested from the diet (Daly et al., 1994c;Daly, 1998). Furthermore, among sequester-ing species there is differential accumulation,as suggested indirectly by the occurrence ofdifferent alkaloid profiles among microsym-patric species (Daly et al., 1987; Myers et al.,1995) and demonstrated directly by feedingexperiments (Daly et al., 1994c, 2003; Gar-raffo et al., 2001), that is, the uptake systemsof different species either (1) are capable ofsequestering only a subset of the alkaloidsingested in the diet or (2) vary drastically inthe efficacy of accumulation of differentclasses of alkaloids. Either way, this variationis heritable. Furthermore, Daly et al. (2003)demonstrated selective alkaloid modification

Fig. 69. Character 143, expansion of sacral diapophyses. A: State 0, unexpanded (riveroi, AMNH134143l). B: State 1, moderately expanded (pumilio, AMNH 118514). C: State 2, greatly expanded(Melanophryniscus stelzneri, AMNH 77710).


by certain dendrobatid species and not others(see Character 171). As with all phenotypiccharacters, the expression of alkaloid char-acters is due to the combination of genotypeplus environment (for a detailed discussion ofthe meaning of ‘‘genetic’’ see Sarkar, 1998).Hypotheses of homology can therefore beproposed defensibly, albeit cautiously, foralkaloid profiles.

Given that it is the capability to accumu-late a class of toxin that is treated as thecharacter, we coded alkaloid profiles as ‘‘anyinstance’’ (Campbell and Frost, 1993). Thatis, we treated the demonstrated occurrence ofa given class of alkaloid in one or morepopulations of a species as evidence that theentire species is capable of accumulating thatclass of alkaloid (i.e., it is coded as present),even if that class of alkaloid was not detectedin all samples. This is not intended asa general endorsement of that method ofcodifying polymorphism (for theoretical ar-guments, see Grant and Kluge, 2003, 2004),but rather as a consequence of this particularbiological problem. Given current under-standing of the alkaloid uptake system ofthese frogs, it is most likely that the absenceof a class of alkaloid in some but not allindividuals is due to dietary deficiency andnot a character-state transformation. Thisassumption is testable, and it may be foundthat (1) this assumption is borne out (i.e., thealkaloid is sequestered when present in thediet), (2) such species are truly polymorphic(i.e., character history and species history donot track each other perfectly, either due toancestral polymorphism, a character-statetransformation event subsequent to the mostrecent cladogenetic event, or some otherphenomenon), or (3) multiple species havebeen conflated. This is exemplified by lugu-bris:

Only one of several populations of P[hyllobates]lugubris had barely detectable amounts of

batrachotoxins. Some but not all populationshad trace levels of other alkaloids. … Alkaloidsincluding a batrachotoxin, were fed to captive-raised P. lugubris and found to be readilyaccumulated into skin (J. W. Daly, unpublishedresults). Thus, the frog has a functional accu-

mulating ‘‘system’’ and the lack or near lack ofalkaloids in wild-caught frogs must reflect lowavailability or non-targeting of alkaloid- or

batrachotoxin-containing arthropods. (J. W.Daly, in litt. 02/02/00).

It is also possible that a species is capableof accumulating an alkaloid not detected inany population because the dietary source ofthe precursor is absent at all sampledlocalities (i.e., failure to detect accumulationin wild-caught specimens does not decisivelydemonstrate that the species is incapable ofsequestration). However, by coding thesetaxa as lacking the ability to accumulate thetoxin we incorporated all available evidence.The hypothesis that a taxon is incapable ofaccumulating a class of toxin is falsifiable andcan be tested both by examining morespecimens and populations and throughfeeding experiments. For example, althoughno histrionicotoxin could be detected in wildlehmanni Myers and Daly (Myers and Daly,1976b), Garraffo et al. (2001: 421) report that‘‘[f]eeding experiments indicated that D.lehmanni readily accumulated histrionico-toxin into skin when fed alkaloid-dusted fruitflies.’’

Although a dietary source is either knownor assumed for dendrobatid alkaloids, theactual arthropod(s) responsible have yet tobe discovered for the vast majority of these,that is, most of the alkaloids are unknownelsewhere in nature. Potential sources werereviewed by Daly et al. (1993: 226, 2005). Forexample, pyrrolizidines are known to occurin the ants Solenopsis xenovenenum, Mono-morium spp. from New Zealand, and Mega-lomyrmex from Venezuela. Pyrrolidines (in-cluding 2,5-pyrrolidines, known amongamphibians only in dendrobatids) occur inSolenopsis, Monomorium, and Megalomyr-mex. Decahydroquinolines were detected inextracts of virgin queens of the thief antSolenopsis (Diphorhoptrum) azteca fromPuerto Rico. 3,5-Disubstituted indolizidinesoccur in ants of the genera Monomorium andSolenopsis. Coccinellines were first discov-ered in the ladybug beetles Coccinellidae.Monocyclic piperidines occur in Solenopsis.Spiropyrrolizidine is likely sequestered froma millipede (Saporito et al., 2003), twopumiliotoxins found in pumilio are obtainedfrom formicine ants (Saporito et al., 2004),and scheloribatid mites are the probablesource of a third pumiliotoxin in pumilio


(Takada et al., 2005). The source of batra-chotoxins in dendrobatids remains unknown,although melyrid beetles are likely (Dumba-cher et al., 2004).

That the actual dietary source is unknownis an important consideration, given thediscovery by Daly et al. (2003) that somespecies convert dietary pumiliotoxin to allo-pumiliotoxin via a specific hydroxylationevent (see Character 171, below). Whereasprior to this discovery it was assumed that allof the over 450 alkaloids known in thesefrogs were incorporated without modifica-tion into the skin, one must consider thepossibility that some portion of this diversityof alkaloids may result from the modificationof precursors. Nevertheless, it is unclear howwidespread metabolic conversion may be, asthe following 12 alkaloid classes have beenadministered in feeding experiments with noevidence for any metabolism (J. W. Daly, inlitt., 01/25/05): batrachotoxin, histrionicotox-ins, allopumiliotoxins, decahydroquinolines,3,5-pyrrolizidines, 3,5-indolizidine, 5,8-indo-lizidine, 5,6,8-indolizidine, pyrrolidine, piper-idine, spiropyrrolizidine, and coccinelline-liketricyclics.

Given the dietary origin of the alkaloidsand how little is known about the alkaloiduptake system, we were conservative indelimiting alkaloid characters for phyloge-netic analysis. Instead of coding the occur-rence of each of the over 450 dendrobatidalkaloids as a separate character, we scoredthe occurrence of the major and minor classesof alkaloids, following Daly et al. (1987,1993) and incorporating more recent devel-opments (e.g., Daly et al., 1994c, 2003, 2005;Daly, 1998; Garraffo et al., 1993, 1997, 2001;Daly et al., 1999; Mortari et al., 2004; J. W.Daly, in litt., 01/25/05). We followed Myers(1987) and Myers et al. (1995) in coding 3,5-indolizidines and 5,8-methylindolizidines asdistinct characters. In only coding the occur-rence of general classes of alkaloids, weconsciously overlooked more refined, poten-tially phylogenetically informative data in anattempt to avoid introducing error due to thenature of alkaloid accumulation in thesefrogs. Furthermore, in the majority ofspecies, numerous alkaloids of the same classco-occur, which suggests that sequestrationacts at the level of the class of alkaloid, not

individual alkaloids; that is, it appears that itis the ability to sequester alkaloids withcertain chemical properties that evolves, notthe ability to sequester a particular alkaloid.

We did not distinguish between major,minor, and trace occurrences of alkaloids(i.e., we treated all as ‘‘present’’), but,following Daly’s recommendation (J. W.Daly, in litt., 02/02/00), we did not consider‘‘trace, trace’’ occurrences as evidence ofpresence of an alkaloid, as merely havingrecently consumed an alkaloid-containingprey item could give this result.11 We alsodid not discriminate based on uptake effi-ciency. For example, although uptake ofpiperidines is poor in most species (e.g.,auratus, in which they are trace alkaloids),and uptake of piperidine 241D appearshighly efficient in speciosus (in which this isa major or minor alkaloid), we codedpiperidines identically (i.e., present). Itshould be clarified that, despite the fact thatthe trivial names of the classes of alkaloidsare often derived from species that possess it(e.g., pumiliotoxin from pumilio), compoundsare assigned to a class based on molecularstructure and chemical properties, not taxo-nomic distribution.

It has been speculated that certain alka-loids could share common precursors, spe-cifically a 2,6-disubstituted(dehydro)piperi-dine as a precursor in the biosynthesis ofhistrionicotoxins, gephyrotoxins, indolizi-dines, and decahydroquinolines (Daly et al.,1987: 1065), and more generally that themonocyclic piperidines are possible precur-sors for the more complex, piperidine-ring-containing alkaloids and the monocyclicpyrrolidines for the more complex, pyrroli-dine-ring-containing dendrobatid alkaloids(Daly et al., 1993: 251). Nevertheless, withthe exception of allopumiliotoxin 267A (seeCharacter 171), there is no evidence that they

11Daly et al. (1987: 1078) reported a traceoccurrence of alkaloid 181B, a 5,8-methylindolizi-dine, from a single population of femoralis at Napo,Ecuador. However, J. W. Daly (in litt., 02/02/00)informed T. Grant that this was a trace, trace amount,and he recommended that this not be treated ‘‘asevidence for significant ability for accumulation ofalkaloids in the species’’.


share a common biosynthetic origin, andeven if they do, this would pertain to thearthropods, not the frogs’ uptake system.Historical independence is demonstrated bythe fact that no classes of alkaloids shareidentical taxonomic distributions.

We coded unambiguously only thosespecies whose alkaloid profiles have beenexamined; taxa whose profiles have not beenexamined were coded as unknown (‘‘?’’). Thediscovery that an untested species is embed-ded within a toxic clade would providea strong prediction that the species may alsosequester alkaloids and would thereforeguide chemists in their search for novel,potentially useful toxins. Negative findingsoften are not explicitly reported in theliterature; however, in cases where specieshave been examined using techniques thatwould detect a particular compound and thecompound was not reported, we coded it asabsent (e.g., epibatidine). If there was anydoubt as to the tests samples were subjectedto, we coded the character as unknown (‘‘?’’).

Data were taken from reviews (Daly et al.,1987, 1993, 1999), the primary literature(Tokuyama et al., 1992; Garraffo et al.,1993, 2001; Badio and Daly, 1994; Myers etal., 1995; Daly et al., 1997., 2003; Fitch et al.,2003; Saporito et al., 2003; Mortari et al.,2004; Darst et al., 2005), and an exhaustivesummary of published and unpublishedalkaloid profiles and corrections to previousaccounts provided by John W. Daly (in litt.,01/25/05). To facilitate coding from theliterature we list the individual nonsteroidalalkaloids for each class, following the con-vention of Daly et al. (1987). We did notinclude unclassified alkaloids, although theymay provide relevant information once theirstructures are elucidated. We did not listunpublished alkaloids in the character de-scriptions (although we did code their pres-ence in the matrix), and we only listedalkaloids that occur in the species sampledin the present study.

147. ABILITY TO SEQUESTER ALKALOIDS:absent 5 0; present 5 1.

We coded the general ability to sequesteralkaloids separately from the individualclasses of alkaloids sequestered in order tocount the gain and loss as a single trans-formation event. We scored species that are

incapable of sequestering any alkaloid asstate 0 for this character and missing (‘‘–‘‘)for all other lipophilic alkaloid characters; wecoded species that are able to sequester anyalkaloid as state 1 for this character andpresent and absent for each of the particularalkaloid classes. That is, although the originof the ability to sequester alkaloids necessar-ily entails the ability to sequester someparticular class(es) of alkaloid(s) (i.e., thereis a logical relation of nested dependencybetween these characters), the fact that notaxon possesses only a single class of alkaloidwould mean that the alternative approach oftreating each origin and loss as entirelyunrelated events would count the origin ofsequestration as multiple events. The biolog-ical assumption underlying this coding is thatthere exists a single genetic basis for thesequestration of all classes of lipophilicalkaloids and that modifications to it accountfor the differential ability to sequester distinctclasses. This assumption is consistent withthe limited understanding of the uptakemechanism but has not been subjected tocritical test (i.e., no attempt has been made toisolate the genetic basis of sequestration).

Darst et al. (2005) reported the occurrenceof alkaloids based on thin layer chromatog-raphy, which allowed us to score severaladditional species. Unfortunately, that meth-od does not discriminate between or lead tothe identification of different alkaloids, sothis is the only character that can be codedfrom their results.

148. BATRACHOTOXINS (BTX): absent 5 0;present 5 1.

The steroidal batrachotoxins are known tooccur in only five species of frogs (aurotaenia,bicolor, lugubris, terribilis, and vittatus), andtheir shared occurrence was treated asevidence of the monophyly of those speciesin a restricted Phyllobates (Myers et al.,1978). Given the extreme toxicity of BTXrelative to other dendrobatid alkaloids, theability of these frogs (and the inability ofother dendrobatids) to sequester BTX islikely related to their modified sodiumchannel that is insensitive to BTX (asdemonstrated for aurotaenia and terribilis;Daly et al., 1980).

149. HISTRIONICOTOXINS (HTX): absent 5

0; present 5 1.


235A, 237F, 239H, 259A, 261A, 263C,265E, 283A, 285A, 285B, 285C, 285E, 287A,287B, 287D, 291a.

Alkaloid 283A9 (found in sylvaticus Funk-houser) is closely related to and was treatedas an HTX by Daly et al. (1987), but it wasnot included by Daly et al. (1993) or Daly etal. (2005).

150. PUMILIOTOXIN (PTX): absent 5 0;present 5 1.

207B, 209F, 225F, 237A, 251D, 253F,265D, 265G 267C, 267D, 277B, 281A, 293E,297B, 305B, 307A, 307B, 307D, 307F 307G,307H, 309A, 309C, 321A, 323A, 325B, 353A.

151. ALLOPUMILIOTOXINS (aPTX): absent5 0; present 5 1.

225E, 237B, 241H, 251I, 253A, 267A,297A, 305A, 307C, 309D, 321C, 323B,325A, 339A, 339B, 341A, 341B, 357.

152. HOMOPUMILIOTOXINS (hPTX): absent5 0; present 5 1.

223G, 249F, 251L, 256R, 265N, 317, 319A,319B, 321B.

153. DECAHYDROQUINOLINE (DHQ): ab-sent 5 0; present 5 1.

193D, 195A, 209A, 209J, 211A, 211K,219A, 219C, 219D, 221C, 221D, 223F,223Q, 223S, 231E, 243A, 245E, 249D,249E, 251A, 253D, 267L, 269AB, 269A,269B, 271D, 275B.

154. 3,5-DISUBSTITUTED PYRROLIZIDINES:absent 5 0; present 5 1.

167F, 195F, 209K, 223B, 223H, 237G,251K, 253I, 265H, 265J, 267H.

167F and 209K were formerly classified asthe 3,5-disubstituted indolizidines 167B and209D.

155. 3,5-DISUBSTITUTED INDOLIZIDINES:absent 5 0; present 5 1.

195B, 211E, 223AB, 223R, 237E, 239AB,239CD, 239E, 249A, 271F, 275C, 275F.

156. 5,8-DISUBSTITUTED INDOLIZIDINES:absent 5 0; present 5 1.

181B, 193E, 197C, 203A, 205A, 207A,209B, 209I, 217B, 219F, 221A, 221K, 223D,223J, 225D, 231C, 233D, 235B, 237D, 237H,239A, 239B, 239C, 239D, 239F, 239G, 241C,241F, 243B, 243C, 243D, 245B, 245C, 245D,251B, 251U, 253B, 263F, 257C, 259B, 261D,271A, 273B, 279D, 295A, 295B.

157. DEHYDRO-5,8-INDOLIZIDINES: absent5 0; present 5 1.

245F, 245H.

158. 5,6,8-INDOLIZIDINES: absent 5 0;present 5 1.

195G, 207Q, 223A, 231B, 233G, 237L,249H, 251M, 253H, 259C, 263A, 263D, 265I,265L, 267J, 273A, 275E, 277C, 277E, 279F,293C.

159: 4,6-QUINOLIZIDINES: absent 50; pres-ent 5 1.

195C, 237I.160. 1,4-QUINOLIZIDINES: absent 5 0;

present 5 1.207I, 217A, 231A, 233A, 235E’, 247D,

257D.161. LEHMIZIDINES: absent 5 0; present 5 1.275A.162. EPIQUINAMIDE: absent 5 0; present 5 1.196.163. 2,5-PYRROLIDINE (PYR): present 5 0;

absent 5 1.183B, 197B, 223N, 225C, 225H, 277D,

279G.164. 2,6-PIPERIDINES (PIP): absent 5 0;

present 5 1.197E, 211I, 211J, 213, 221L, 223K, 225B,

225I, 237J, 239I, 239L, 239O, 241D, 241G,253J, 255A, 267K, 267C.

165. GEPHYROTOXIN (GTX): absent 5 0;present 5 1.

287C, 289B.166. COCCINELLINE-LIKE TRICYCLICS: ab-

sent 5 0; present 5 1.191B, 193A, 193C, 201B, 205B, 205E,

207J, 207P, 207R, 219I, 221G, 221M,235M, 235P.

167. SPIROPYRROLIZIDINES: absent 5 0;present 5 1.

Referred to as pyrrolizidine oximes byDaly et al. (1993).

222, 236, 252A, 254.168. INDOLIC ALKALOIDS (CHIMONAN-

THINE/CALYCANTHINE): absent 5 0; present5 1.

346B, 346C.169. EPIBATIDINES: absent 5 0; present 5 1.208/210, 308/310.170. NORANABASAMINE (5PYRIDYL-PIPER-

IDINES): absent 5 0; present 5 1.This pyridine alkaloid is known in nature

only from aurotaenia, bicolor, and terribilis(Daly et al., 1993, 1999).

239J.171. PUMILIOTOXIN 7-HYDROXYLASE: ab-

sent 5 0; present 5 1.


Feeding experiments by Daly et al. (2003)demonstrated the existence in several speciesof dendrobatids of an enantioselective mech-anism that converts PTX (+)-251D to themore highly toxic allopumiliotoxin (aPTX)(+)-267A. That is, contrary to other alkaloidcharacters, which code the ability to seques-ter a class of alkaloid, this character appliesto the occurrence of the 7-hydroxylase, asevidenced by the occurrence of the hydrox-ylated compound.

Coding this character is somewhat morecomplicated than coding the other alkaloidcharacters, because in this case occurrence ofaPTX 267A may be due to either (1) thehydroxylation of PTX 251D or (2) thesequestration of aPTX 267A from a dietarysource (aPTX is known to occur in somearthropods). This creates the potential forboth false negatives and false positives. Directevidence for the occurrence of 7-hydroxylasemay only be obtained through feeding experi-ments. Further evidence on the distribution ofthe pumiliotoxin 7-hydroxylase obtained in-directly from the alkaloid profiles of wild-caught specimens (see Daly et al., 2003:11095, table 1) requires the assumption thatall aPTX 267A occurs through metabolism ofingested PTX 251D, which, at least in the caseof anthonyi (reported as tricolor; for taxono-my of these species, see Graham et al., 2004),is false (assuming multiple species have notbeen conflated). Daly et al. (2003) reportedwild-caught specimens as possessing traceamounts of aPTX 267A, but feeding experi-ments revealed that the species is incapable ofhydroxylating PTX 251D and the occurrenceof aPTX 267A represents a false positive forthe presence of 7-hydroxylase. Nevertheless,in the absence of direct evidence from feedingexperiments, such as is available for anthonyi,we coded all trace, minor, and major occur-rences of aPTX 267A as the presence of the 7-hydroxylase, which allows the results ofphylogenetic analysis to serve as a tool fordesigning future feeding experiments to testhypothesized occurrence of 7-hydroxylase(e.g., finding that a species that possessesaPTX 267A is embedded in a clade of speciesincapable of 7-hydroxylation would suggestthe occurrence may be due to sequestrationfrom a dietary source and not biosyntheticconversion).

Conversely, the absence of 7-hydroxylasecan only be assured in the presence of PTX251D. We coded the failure to detect aPTX267A as ‘‘absent’’ (state 0) only when PTX251D was detected. If PTX 251D was notdetected (but other PTXs were), we codedthis character as unknown (‘‘?’’) (e.g., trunca-tus). If available evidence indicates thata species is incapable of sequestering pumi-liotoxins, we coded this character as missing(‘‘–’’) (e.g., trivittatus).

Direct evidence for the presence of thepumiliotoxin 7-hydroxylase through feedingexperiments was found for auratus, galacto-notus, and castaneoticus, and direct evidencefor the absence of pumiliotoxin 7-hydroxy-lase through feeding experiments was foundin tricolor and bicolor (Daly et al., 2003).Other species are coded on the basis of wild-caught specimens, with data derived fromDaly et al. (1987, 1993, 2003).

172. TETRODOTOXIN (TTX): absent 5 0;present 5 1.

Daly et al. (1994b) reported the occurrenceof TTX in panamensis (as Colostethus in-guinalis; see Grant, 2004). They also exam-ined aqueous extracts of eight additionalspecies referred to Colostethus (the ‘‘Colos-tethus species’’ reported as being ‘‘common,nr Villa Marıa, Caldas, Colombia’’ is frater-danieli), and nocturnus, pumilio, and bicolor.Daly et al. (1994b: 283) cautioned that thenegative results for pumilio and bicolor werebased on methanol extracts,

which would have extracted only minimalamounts of tetrodotoxin. … Thus, very lowlevels of tetrodotoxin-like compounds … mighthave escaped detection because of the lowefficiency of methonol in extracting suchcompounds. But levels approaching those re-ported for C. inguinalis [5 panamensis] …would have been detected even in methanolextracts.

173. CHROMOSOME NUMBER: 18 5 0; 20 5

1; 22 5 2; 24 5 3; 26 5 4; 28 5 5; 30 5 5.Additive.

Karyological data have been reported for35 of the dendrobatids included in the presentstudy: panamensis and pumilio (Duellman,1967), auratus and pumilio (Leon, 1970),trivittatus (Bogart, 1970, 1973, 1991), trinita-tis (Rada de Martınez, 1976), auratus, gran-


uliferus, histrionicus, lugubris, pumilio, andsylvaticus Funkhouser (as histrionicus fromNW Ecuador) (Rasotto et al., 1987), con-spicuus [as brunneus], femoralis, fraterdanieli,olfersioides, palmatus, pictus, subpunctatus,talamancae, trivittatus, truncatus, vanzolinii[as quinquevittatus], vertebralis, and an un-described species referred to Colostethus(Bogart, 1991), caeruleodactylus, marchesia-nus (sensu stricto; see Caldwell et al., 2002b)and two undescribed species referred toColostethus (Veiga-Menoncello et al.,2003a), nidicola and stepheni (Veiga-Menon-cello et al., 2003b), chalcopis, leopardalis,herminae, neblina, olmonae, and trinitatis(Kaiser et al., 2003), flavopictus, femoralis,hahneli, and trivittatus (Aguiar et al., 2002).Thirty of those species are included in thepresent study.

For outgroup taxa, data were taken fromKuramoto’s (1990) review. Data publishedsubsequently were taken from Silva et al.(2001) for Cycloramphus boraceiensis, Rosaet al. (2003) for Megaelosia, Ramos et al.(2002) for Atelopus zeteki, and Aguiar et al.(2004) for Crossodactylus and Hylodes phyl-lodes.

Coding chromosome variation as trans-formation series is complicated by impreci-sion in determining chromosome homology.For the most part, chromosomes are simplyarranged according to size and named(numbered) consecutively. That all variationin chromosome morphology is reported inrelation to chromosome identity (which isa function of relative chromosome size) isa serious problem. Rarely, more detailedconsiderations are brought to bear (e.g., seeBogart, 1991, regarding the homology ofchromosome 4 in pictus and chromosome 5in trivittatus), but this is done so infrequentlyas to be of little use in the present study. Afurther limitation of available karyologicaldata relates to the variation in techniques andkinds of data reported. For example, nucle-olar organizing regions (NORs) are reportedfor only 11 of the dendrobatids included inthis study, and in just those few species atleast six NOR states are apparent. Likewise,in light of the confounding variation heobserved, Bogart (1991: 245) cautioned that‘‘[i]t is evident that analysis of chromosomearms would be of little value for understand-

ing karyotype evolution in the family Den-drobatidae. It is also evident that dendroba-tid chromosomes have undergone extensiverestructuring via translocations and inver-sions.’’

There are undoubtedly many additionaltransformation series in chromosome mor-phology, but we coded only chromosomenumber because (1) it is reported in allkaryological studies, (2) it is less dependenton individual chromosome identity (seereferences above), and (3) it has beenemployed previously in studies of dendroba-tid systematics. Nevertheless, inferring trans-formation series solely from chromosomenumber necessarily assumes that the samechromosome(s) are gained or lost in eachchange in total number of chromosomes,which future research will undoubtedly de-termine to have been an oversimplification.

RESULTS

GENERAL RESULTS

Direct optimization parsimony analysisfollowed by the parsimony ratchet analysisof the implied alignment resulted in 37equally parsimonious solutions of 46,520steps. Forty nodes collapse in the strictconsensus of these optimal topologies, all ofwhich involve conspecific terminals only.Further swapping those 37 trees using theimplied alignment recovered a total of 25,872trees, with no additional nodes collapsed inthe strict consensus. The CI (Kluge andFarris, 1969) and RI (Farris, 1989) for thephenotypic characters12 on the total evidencesolutions are 0.14 and 0.76, respectively. Webegin by summarizing higher-level relation-ships, shown in figure 70, and proceed to therelationships among dendrobatids in fig-ures 71–76. Rather than describe the clado-gram and associated support values exhaus-

12Although one may calculate the CI and RI forDNA sequences based on the implied alignment, thesignificance of these homoplasy-based indices in thecontext of dynamic homology is questionable. Twoequally parsimonious explanations may have differentCIs and RIs, and explanations with the same CI andRI may have different costs. For details, see Kluge andGrant (2006).


tively, we emphasize information not de-picted on the cladogram, especially theunambiguous transformations that delimitclades and the bearing of the current resultson species-level problems. Only unambiguoussynapomorphies shared by all most parsimo-nious trees are reported.

DENDROBATID MONOPHYLY AND

OUTGROUP RELATIONSHIPS

Dendrobatid monophyly was supportedstrongly in the present analysis (fig. 70).Unambiguous phenotypic transformationsinclude the loss of supernumerary tubercleson the hand (Character 2, 1 R 0), gain of thetarsal keel (Character 28, 0 R 1), the ‘‘ranid’’

type insertion of the distal tendon of the m.semitendinosus (Character 69, 0 R 1), gain ofthe m. semitendinosus binding tendon (Char-acter 70, 0 R 1), occurrence of the dorsal flapof the m. depressor mandibulae (Character72, 0 R 1), relation of the tympanum and m.depressor mandibulae (Character 75, 0 R 1),orientation of the m. intermandibularis sup-plementary element (Character 78, 0 R 1),maxillary tooth structure (Character 140, 0 R1), the occurrence of the retroarticular pro-cess of the mandible (Character 142, 0 R 1),and the reduction in chromosome numberfrom 26 to 24 (Character 175, 4 R 3).Behavioral synapomorphies include the lossof reproductive amplexus (Character 104, 1R 0), the gain of dorsal tadpole transport

Fig. 70. Strict consensus of 25,872 most parsimonious trees of 46,520 steps: outgroup relationships andmonophyly and placement of dendrobatids. Numbers above branches are Bremer support values. Familygroup names applied as in Frost et al. (2006). Cycloramphinae and Hylodinae were nested withinCycloramphidae in Frost et al.’s study. Numbers following terminal names are unique sample identifiers.Terminals without numbers or with alphanumeric identifiers (GenBank numbers) were not sequenced forthe present study or Frost et al. (2006) and were taken from GenBank. Upper right inset shows entirecladogram and corresponding figure numbers, with present view in black.


(108, 0 R 1), and the origin of toe trembling(Character 115, 0 R 1).

Our results generally resemble those ofFrost et al. (2006) regarding the phylogeneticposition of dendrobatids and the relation-ships among outgroup taxa, but with a fewsignificant exceptions. Of greatest relevanceto the problem of dendrobatid relationships,the current study refuted Frost et al.’s (2006)placement of Thoropa and dendrobatids assister groups and instead placed Thoropainside Cycloramphidae, with Hylodinae re-covered as the sister group of dendrobatids(as first suggested by Noble, 1926). Inaddition to the genotypic transformationsthat optimize unambiguously to this node,phenotypic transformations include the ori-

gin of digital scutes (Character 1, 0 R 1) andthe formation of digital discs (Character 6,0 R 1), the origin of T-shaped terminalphalanges (Character 118, 1 R 0), and theoccurrence of an oblique lateral stripe (Char-acter 55, 0 R 1). Except for the removal ofhylodines and insertion of Thoropa, therelationships among cycloramphines areidentical to those of Frost et al. (2006). Aswas found by Frost et al., the next moreinclusive clade includes Bufonidae, and thenCycloramphinae.

The greatest difference between Frost etal.’s (2006) results and the present hypothesisinvolves the placement of leptodactylids. Theclades here labeled Leptodactylidae 1 andLeptodactylidae 2 were a monophyletic

Fig. 71. Strict consensus of 25,872 most parsimonious trees of 46,520 steps: relationships amongdendrobatids. Numbers above branches are Bremer support values. Numbers following terminal namesare unique sample identifiers. Terminals without numbers or with alphanumeric identifiers (GenBanknumbers) were not sequenced for the present study or Frost et al. (2006) and were taken from GenBank.Unidentified species taken from GenBank are labeled as originally published. Upper right inset showsentire cladogram and corresponding figure numbers, with present view in black.


group in Frost et al.’s study, and that cladewas sister to Centrolenidae, together formingthe Amnibatrachia. Here, centrolenids arethe sister of all included taxa except hylids,Leptodactylidae 1 is sister to all but thecentrolenids and hylids, and Leptodactylidae2 is sister to Bufonidae + Hylodinae +Dendrobatidae, that is, it is separated fromLeptodactylidae 1 by ceratophryids andcycloramphines.

RELATIONSHIPS AMONG DENDROBATIDS

The eastern Colombian species palmatus issister to a clade that includes all species thatpossess the median lingual process (MLP)(fig. 71). Eight unambiguous phenotypictransformations unite palmatus with theMLP clade, including the origins of fringeson the preaxial edges of fingers II and III(Characters 13 and 15, 0 R 1).

Within the MLP clade, tepuyensis and theundescribed species ‘‘Thomasing’’ are rela-tively robust frogs with extensive webbing.Their monophyly is strongly supported (Bre-mer support 5 41), although only a singlephenotypic synapomorphy optimizes unam-biguously to this node (expansion of toe discI, Character 31, 1 R 2). Percent pairwisedistances are shown in table 4.

The identification of sample 606 astepuyensis will likely require revision. Thatspecies was described by La Marca (1998‘‘1996’’) from Auyantepui, whereas sample606 was taken over 200 km away on Mt.Ayanganna (ca. 60 km WNW of Kaieteur,Guyana). Given the high degree of endemismof many tepui species, it is doubtful that thesesamples are conspecific. Nevertheless, wecompared the voucher specimen of the tissuesample (ROM 39637, the only specimen of

this species collected at this locality) to a seriesof 33 specimens of tepuyensis from the typelocality and failed to detect diagnosticdifferences. Our prediction is that additionalspecimens and/or molecular data will revealthat these are different species, but for thepresent we apply the name tepuyensis tospecimens from both localities. ‘‘Thomasing’’is an undescribed species from Mt. Thomas-ing, Mazaruni-Potaro, Guyana.

The monophyly of the remaining species ofthe MLP clade is supported by three un-ambiguous phenotypic transformations: thegain of the pale paracloacal marks (Charac-ter 49, 0 R 1), completion of the obliquelateral stripe (Character 56, 0 R 1), and theorigin of an endotrophic larval diet (Charac-ter 112, 0 R 3).

Among the species of this clade are severalthat resemble, superficially at least, degran-villei. Species delimitation is hindered byextensive morphological variation withinsyntopic series, making this a prime exampleof the relevance of DNA sequence data indiscovering cryptic diversity. Samples 278,279, and 1336 were all collected in Guyana(details below). Although we did not detectmorphological differences, our results indi-cate that they are not conspecific withdegranvillei sensu stricto. The degranvilleidata obtained from GenBank were generatedby Vences et al. (2003), who stated that theirsample of degranvillei was from Saul, FrenchGuiana, which is quite close to the typelocality and, therefore, likely to representdegranvillei sensu stricto. We therefore referto the Guyanan species as ‘‘degranvillei.’’Cytochrome b sequences for the Vences et al.specimen were not available, but the pairwisedistance between ‘‘Tafelberg’’ and the ‘‘de-granvillei’’ is 17.5%.

The two samples of praderioi were collect-ed at 1,310 m on Roraima, Guyana. Sample1336 of the Guyanan ‘‘degranvillei’’ was alsocollected on Roraima but was taken at1,075 m. The two remaining Guyanan ‘‘de-granvillei’’samples were taken in the Merumemountains, and ‘‘Ayanganna’’ was collectedon Mt. Ayanganna, ca. 50 km WNW ofKaieteur, Guyana.

Despite the morphological similarity andgeographic proximity of praderioi and ‘‘de-granvillei’’ on Roraima, and only ,300 m

TABLE 4Percent Uncorrected Pairwise Distances Between

Cytochrome b Sequences of tepuyensisand ‘‘Thomasing’’a

Sample ID 1 2 3

1 tepuyensis 606 —

2 ‘‘Thomasing’’ 1332 4.4 —

3 ‘‘Thomasing’’ 1333 4.4 0.0 —

a Gray lines separate species.


difference in elevation between the localities,the pairwise distance is 10.1–10.4% (seetable 5). The pairwise distance between thethree ‘‘degranvillei’’ samples is only 1.6–1.8%,despite the much greater geographic distance.

According to the topology alone, it ispossible that ‘‘Ayanganna’’ and praderioimay be conspecific. Nevertheless, they differmorphologically (e.g., webbing) and at 8.3%of their cytochrome b sites, leaving littledoubt that they are different species.

The sister species beebei and roraima arediminutive, geographically proximate speciesthat both possess the median lingual processand breed in phytotelmata (for breedingbehavior in beebei, see Bourne et al., 2001).Pairwise distances are shown for beebei androraima in table 6. There is no confusionsurrounding the identity of beebei, with theexception that the French Guianan speciesdiscussed under that name (e.g., Kok, 2000;Lescure and Marty, 2000) is not conspecificwith beebei sensu stricto from Guyana(among other differences, the French Guia-nan species lacks the median lingual process).

La Marca (1998 ‘‘1996’’) described roraimabased on a single immature specimen from

2,700 m near the peak of Mt. Roraima.Although there are several inconsistencies inLa Marca’s description and illustrations, andthe immaturity of the holotype impedesidentification, the material included in thepresent study was collected at the typelocality and agrees with the descriptionsufficiently to conclude that it is roraima.Samples 1337 and 1338 were taken fromadults CPI 10216 and CPI 10217. Sample1339 is from an untagged tadpole collected ina bromeliad, which establishes conclusivelyadult and larval conspecificity.

The clade composed of baeobatrachus, de-granvillei, stepheni, ‘‘Tafelberg’’, and ‘‘Brown-sberg’’ has a Bremer value of 24. Five unam-biguous phenotypic changes occur at this node,including the loss of the distal subarticulartubercle on finger IV (Character 3, 1 R 0).

The nomenclatural history of baeobatra-chus and stepheni is convoluted. The name‘‘baeobatrachus’’ first appeared in Edwards’swidely distributed but formally unpublishedPh.D. dissertation (Edwards, 1974a). Thetype locality Edwards intended to designatewas Ducke Reserve in Amazonas State, justoutside Manaus (Brazil). The two samplesincluded here (514, 515) are from thatlocality. On the 15th anniversary of thecompletion of Edwards’s dissertation, Mar-tins (1989) described stepheni with the explicitintent of providing a name for Edwards’s‘‘baeobatrachus’’. The type locality of ste-pheni is Presidente Figueiredo, also in Ama-zonas State and approximately 100 km fromManaus. Apparently unaware of this de-velopment or the fact that Edwards’s ‘‘baeo-batrachus’’ was not an available name, anddespite having cited a paper that deals withthe reproductive biology of stepheni (viz.,

TABLE 5Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of

‘‘degranvillei ’’ from Guyana, praderioi, and the Undescribed Species ‘‘Ayanganna’’a

Sample ID 1 2 3 4 5 6

1 ‘‘degranvillei’’ 278 Mereme —

2 ‘‘degranvillei’’ 279 Mereme 0.3 —

3 ‘‘degranvillei’’ 1336 Roraima 1.8 1.6 —

4 ‘‘Ayanganna’’ 607 9.6 9.4 9.6 —

5 praderioi 1334 Roraima 10.4 10.1 10.4 8.3 —

6 praderioi 1335 Roraima 10.4 10.1 10.4 8.3 0.0 —

a Gray lines separate localities and species.

TABLE 6Percent Uncorrected Pairwise Distances BetweenCytochrome b Sequences of beebei and roraimaa

Sample ID 1 2 3 4 5

1 beebei 605 —

2 beebei 608 0.3 —

3 roraima 1337 5.7 5.5 —

4 roraima 1338 5.5 5.2 0.3 —

5 roraima 1339 5.7 5.5 0.5 0.3 —



Junca et al., 1994), in a popular article Boisteland Massary (1999) presented a color pho-tograph and brief but validating diagnosisunder the name Colostethus baeobatrachus.Boistel and Massary did not specify a typelocality or voucher specimen, but Kok (2000)provided a complete redescription based onmaterial from Montagne Belvedere in FrenchGuiana and deposited at IRSNB. The foursamples included here (14, 42, 43, 44) weretaken from that series. Immediately there-after, Kok (2001) determined that baeobatra-chus and stepheni were indistinguishable andplaced them in synonymy. Published sono-grams of stepheni at Reserva Ducke (Junca,1998) and baeobatrachus in French Guiana(Lescure and Marty, 2000) are very similar,the sole potentially relevant difference occur-ring in the dominant frequencies: in stepheni itis given as 4.6–4.8 kHz and in baeobatrachus5.12–5.83 kHz. Sample sizes are very smallthough, and such minor differences arecommonly observed within species.

Nevertheless, the ca. 17% pairwise distancebetween the Reserva Ducke and MontagneBelvedere samples strongly suggests they arenot conspecific (see table 7). Furthermore,tadpoles of stepheni are nidicolous withreduced mouth parts and a median vent tube(Junca et al., 1994; Junca, 1998), whereasa male nurse frog was collected at Serra doNavio, Amapa, Brazil transporting threetadpoles with fully developed mouth partsand dextral vent tube.13 Assuming that theMontagne Belvedere and Serra do Naviosamples are conspecific, there is strongevidence that these baeobatrachus and ste-

pheni are not conspecific, despite the appar-ent lack of diagnostic characters in adultmorphology. That baeobatrachus and ste-pheni are valid species is further supported byphylogenetic analysis, which places the un-described species ‘‘Brownsberg’’ (1328), fromGuyana, as sister to baeobatrachus to theexclusion of stepheni.

The remaining species, ‘‘Tafelberg’’ (sam-ple 1326) is another undescribed species fromGuyana that is closely related to (andpotentially conspecific with) the GenBankdegranvillei (see above for comments on theidentity of this sample). Cytochrome b se-quences were unavailable for the GenBankdegranvillei sample and morphological com-parisons were not made, but the number ofunambiguous transformations in 12S and


stepheni, baeobatrachus, and the Undescribed Species ‘‘Tafelberg’’and ‘‘Brownsberg’’a

Sample ID 1 2 3 4 5 6 7 8

1 stepheni 514 —

2 stepheni 515 0.3 —

3 baeobatrachus 14 17.4 17.1 —

4 baeobatrachus 42 17.7 17.4 0.3 —

5 baeobatrachus 43 17.4 17.1 0.0 0.3 —

6 baeobatrachus 44 17.4 17.1 0.0 0.3 0.0 —

7 ‘‘Tafelberg’’ 1326 19.0 19.2 16.9 17.1 16.9 16.9 —

8 ‘‘Brownsberg’’ 1328 17.1 16.9 11.9 12.2 11.9 11.9 15.6 —


13Lescure and Marty (2000: 320) also claimeddifferences in larval morphology between stepheni(described by Junca et al., 1994) and baeobatrachus(described, according to Lescure and Marty, byEdwards, 1974a, in his dissertation). However, theyfailed to note that Edwards’s description was based onfree-swimming larvae from Reserva Ducke, and yetJunca’s nidicolous larvae were also from ReservaDucke. This suggests that either (1) stepheni has bothfree-swimming, exotrophic and nidicolous, endotro-phic larvae, (2) stepheni and baeobatrachus occur insympatry at Reserva Ducke, or (3) Edwards’s free-swimming tadpoles were not stepheni. Given that atleast one additional dendrobatid (Colostethus march-esianus fide Junca, 1998) occurs at Reserva Ducke andEdwards never explained his rationale for associatingthese tadpoles and adults, (3) is the most plausibleexplanation.


16S sequences that occur on the terminalbranches (13 for ‘‘Tafelberg,’’ 14 for degran-villei) suggests they are not conspecific.

The other clade shown in figure 71 iscomposed mainly of species currently re-ferred to Aromobates, Mannophryne, andNephelobates. The monophyly of this cladeis strongly supported (Bremer support 5 38)and is delimited by 54 unambiguous geno-typic changes, although there are no un-ambiguous phenotypic transformations atthis node. Following the current taxonomy,Aromobates nocturnus and Colostethus salt-uensis are nested within Nephelobates. The

latter species was included in the alboguttatusgroup of Rivero (1990 ‘‘1988’’), but wasexcluded without comment when La Marca(1992) named his own version of the albo-guttatus group formally as Nephelobates.Likewise, the affinities of nocturnus and thespecies of both Nephelobates and Manno-phryne were noted when Aromobates wasdescribed (referring to those as yet unnamedgenera as the alboguttatus and collarisgroups, respectively; Myers et al., 1991),and Kaiser et al. (1994), Meinhardt andParmelee (1996), and Grant et al. (1997)questioned the monophyly of those genera



relative to Aromobates. More specifically,prior to naming Nephelobates, La Marca(1993) considered nocturnus to be mostclosely related to his alboguttatus group,which is corroborated in this study.

Although Nephelobates is paraphyletic, themonophyly of the equally controversialMannophryne is solidly corroborated in thisanalysis. This clade has a Bremer value of40 and may be diagnosed morphologicallyby the synapomorphic dermal collar, whichoptimizes unambiguously to this node (Char-acter 59, 0 R 1). The conclusions, basedon morphological criteria, that thecollarlike gular–chest markings of severalEcuadorian species (e.g., elachyhistus) arenot homologous with the dermal collar ofthese Venezuelan species and that thediffuse collar of nocturnus is due to non-homologous subdermal pigmentation (seeCharacters 58 and 59) are supported by thedistant relationships of these taxa in theoptimal cladograms.

Among the nominal species included in thecladogram, the herminae samples are notconspecific; at this time we are unable tospeculate as to which of these (if either)corresponds to herminae sensu stricto. Thecytochrome b sequences for the two samplesof nocturnus are identical.

The clade shown in figure 72 is a large,primarily cis-Andean (east of the Andes)group. Ten unambiguous phenotypic trans-formations delmit this clade, including theorigin of swelling of finger III in adult males(Character 20, 0 R 1) and the loss ofpalatines (Character 132, 1 R 0).

The sister of the remainder of this clade isalagoanus, from the Atlantic forest of Brazil,followed by the undescribed ‘‘Neblina spe-cies’’ and undulatus. ‘‘Neblina species’’ wascollected at the base of the tepui Neblina, inVenezuela; the cytochrome b sequences of the

two specimens are identical. Myers andDonnelly (2001) described undulatus fromthe Yutaje massif, also in Venezuela. Thethree samples were all collected in the samevicinity (see table 8 for pairwise distances).

Among the species included in the presentanalysis, the only trans-Andean (west of theAndes) species in this clade are the sisterspecies talamancae and the undescribed‘‘Magdalena’’. The affinities of talamancaehave never been clear (e.g., Rivero, 1990‘‘1988’’ was unable to assign it to any of hisgroups), probably because it differs consid-erably from the trans-Andean species withwhich it was compared. However, the place-ment of these species among these cis-Andean species is strongly supported andhighlights the overall resemblance of thesespecies (e.g., for photographs of talamancaeand kingsburyi, see Coloma, 1995). More-over, the discovery of the undescribed‘‘Magdalena’’ species fills in the gap in thedistribution between talamancae and theremaining species.

‘‘Magdalena’’ and talamancae share theunambiguous transformation from an evenlystippled to solid dark throat in adult males(Character 61, 2 R 4). These two species areallopatric. ‘‘Magdalena’’ is known only fromsites on the floor of the middle Magdalenariver valley in Colombia, and talamancae iswidespread from the Pacific lowlands ofSouth America in Ecuador and Colombianorth to Nicaragua. Pairwise distances forcytochrome b sequences of these species areshown in table 9. The talamancae samples arefrom two localities in Panama (Bocas delToro: 325, 326; Cocle: 1147) and one inNicaragua (361, 362, 408).

Although the most parsimonious clado-gram recovers monophyletic Panamanianand Nicaraguan samples of talamancae, thegenetic distances between the samples areconsistent with the hypothesis of continuousgene flow. The greatest pairwise distance isbetween the samples from Nicaragua andCocle, with the intermediate sample fromBocas del Toro also intermediate genetically.

Much of the diversity of small, brown,relatively nondescript cis-Andean dendro-batids has been associated with the old (seeappendix 1) names brunneus, marchesianus,and trilineatus. Progress in documenting the

TABLE 8Percent Uncorrected Pairwise Distances BetweenCytochrome b Sequences of Syntopic Specimens

of undulatus

Sample ID 1 2 3

1 undulatus 331 —

2 undulatus 332 0.8 —

3 undulatus 333 0.3 1.0 —


diversity of Amazonian dendrobatids hasbeen hindered by confusion surroundingthese nominal species. Grant and Rodrıguez(2001) clarified the identity of the westernAmazonian trilineatus, and Caldwell et al.(2002b) redescribed marchesianus based onextensive new material and vocalizationsfrom the type locality (in the vicinity of theRio Uaupes in Amazonian Brazil) andclarified that all populations referred to thatspecies from elsewhere (e.g., Santa Cecilia,Ecuador) were heterospecific (T. Grant hasalso examined material of this species fromthe adjacent region of Colombia). In the sameyear, Morales (2002 ‘‘2000’’) provided anaccount for marchesianus based on examina-tion of a syntype and specimens from otherlocalities, but his redescription is incomplete(e.g., it does not address intraspecific varia-tion or make comparisons with other species)and disagrees in several key points with thatof Caldwell et al. (2002b), as well as theColombian material examined by T. Grant,and the account is therefore rejected. Weincluded DNA sequences for numerous speci-mens referred to trilineatus by Grant andRodrıguez (2001), as well as material fromthe same or nearby localities, but we wereunable to include sequences for marchesianussensu stricto.

The remaining taxonomic problem involv-ing an old name is brunneus. Grant andRodrıguez (2001) provided data for topotyp-ic and other material, but they did notattempt to address decisively the problem ofbrunneus identity. La Marca et al. (2004)improved matters considerably by clarifyingthat the ‘‘brunneus’’ from northern Venezuela

was in fact a new species (which they namedColostethus pittieri) most closely related tohumilis. In what appears on the surface to bethe most thorough study of the systematics ofthese frogs, Morales (2002 ‘‘2000’’) providedan account for brunneus. Like the remainderof the accounts in that paper—includingthose for the 11 new species named there-in—the account of brunneus does not addressvariation within brunneus or compare thatspecies to others and is therefore highlyunsatisfactory. Nevertheless, Morales’s ac-count of brunneus is the most recent attemptto clarify its identity, and we therefore applythe name in his sense. We included DNAsequences from several of the specimensexamined by Morales and referred by himto several species, including brunneus and hisnew species conspicuus and gasconi.

Although we apply the name brunneus inthe sense of Morales (2002 ‘‘2000’’), andsamples 352 and 1278 were both referred tothat species by him, Morales also referredsample 354 of a distantly related species fromCurua-Una (shown in fig. 73) to brunneus.The minimum pairwise distance between thatsample and either of the others he referred tobrunneus is 16.6%. We therefore exclude thatsample from the pairwise comparisons intable 10 and instead include it with the othersamples from Curua-Una (see below). Thepairwise cytochrome b distances betweenbrunneus and its sister species from ParqueEstadual Guajara-Mirim (‘‘PEG-M2’’) are14.3–15.3%.

Terminals identified as ‘‘PEG-M2’’ repre-sent one of three undescribed species ofdendrobatids collected at Parque Estadual


talamancae and ‘‘Magdalena’’a

Sample ID 1 2 3 4 5 6 7

1 talamancae 325 Bocas del Toro —

2 talamancae 326 Bocas del Toro 0.5 —

3 talamancae 1147 Cocle 2.6 2.6 —

4 talamancae 361 Nicaragua 5.2 5.2 5.7 —

5 talamancae 362 Nicaragua 5.2 5.2 5.7 0.0 —

6 talamancae 408 Nicaragua 5.2 5.2 5.7 0.0 0.0 —

7 ‘‘Magdalena’’ 1358 16.1 16.6 16.1 15.6 15.6 15.6 —



Guajara-Mirim, Rondonia, Brazil (see ta-ble 11) The pairwise distances between thesamples of this species and brunneus are 14.3–15.3%.

The next clade includes juanii, fromVillavicencio, Colombia, zaparo, from east-ern Ecuador, and the widespread femoralis.The monophyly of zaparo and femoralis isstrongly supported (BS 5 147), and they areunited by 161 unambiguous transformations.

The occurrence of zaparo and femoralis inthis clade of otherwise cryptically colored,nontoxic frogs conflicts strongly with thetraditional placement of these species withbright, toxic species such as petersi and pictus(e.g., Silverstone, 1976). Nevertheless, thedistant placement of these species found byprevious studies (e.g, Santos et al., 2003;Vences et al., 2003a) could not be refuted bythe inclusion of phenotypic and additionalDNA evidence. Furthermore, both femoralisand zaparo appear to be incapable ofaccumulating alkaloids (Darst et al., 2005;J. W. Daly in litt., 02/02/00), which suggeststhat the remarkable resemblance of thesenontoxic and toxic species may be due toBatesian mimicry.

The type locality of femoralis is Yurim-aguas, Peru, but the taxon is distributedthroughout much of the Amazon basin(Silverstone, 1976). Morphologically, speci-mens referred to femoralis exhibit minorvariations in coloration (e.g., thickness oflateral stripes, size and extent of bright thighflash marks; see Silverstone, 1976). Wegenerated cytochrome b sequences for 17samples of femoralis collected at the follow-ing eight localities, covering much of thenominal species’ range (pairwise distancesshown in table 12): Porto Walter, Acre,Brazil (397, 398, 1309); Cusco Amazonico,Madre de Dios, Peru (78, 128, 129); Cuya-beno, Sucumbıos, Ecuador (399, 400, 1306);Parque Estadual Guajara-Mirim, Rondonia,Brazil, (393, 394); Reserva Ducke, Amazo-nas, Brazil (520); Panguana, Huanuco,Peru (nearest to the type locality andtherefore tentatively treated as femoralissensu stricto) (526); Sipaliwini, Suriname(1325); Rio Curua-Una, Para, Brazil (395,396, 1298).

The taxonomy of zaparo is less problem-atic, but we include it in table 12 as a point ofreference for femoralis. Vences et al. (2003a)

TABLE 11Uncorrected Pairwise Distances Between Cytochrome b Sequences of ‘‘PEG-M2’’

Sample ID 1 2 3 4 5 6

1 ‘‘PEG-M2’’ 617 —

2 ‘‘PEG-M2’’ 618 1.3 —

3 ‘‘PEG-M2’’ 1237 1.6 0.3 —

4 ‘‘PEG-M2’’ 1276 1.3 0.5 0.8 —

5 ‘‘PEG-M2’’ 1282 1.6 0.3 0.5 0.8 —

6 ‘‘PEG-M2’’ 1289 1.6 0.3 0.0 0.8 0.5 —

TABLE 10Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of brunneus

Sample ID 1 2 3 4 5 6 7 8 9 10

1 brunneus 352 —

2 brunneus 612 0.8 —

3 brunneus 613 0.3 0.5 —

4 brunneus 1264 0.0 0.8 0.3 —

5 brunneus 1271 0.0 0.8 0.3 0.0 —

6 brunneus 1278 1.0 0.3 0.8 1.0 1.0 —

7 brunneus 1281 0.0 0.8 0.3 0.0 0.0 1.0 —

8 brunneus 1286 1.3 0.5 1.0 1.3 1.3 0.3 1.3 —

9 brunneus 1294 1.3 0.5 1.0 1.3 1.3 0.3 1.3 0.5 —

10 brunneus 1316 0.8 0.0 0.5 0.8 0.8 0.3 0.8 0.5 0.5 —


united these two species formally in Allo-bates. The species Duellman and Mendelson(1995) referred to as zaparo is a distantlyrelated, probably toxic species (details dis-cussed below).

As shown in table 12, the pairwise dis-tances between zaparo and femoralis samplesare 12.2–15.3%. Forty-three unambiguoustransformations unite the zaparo samples,and 38 unite those of femoralis. Although the

cladogram is consistent with the recognitionof a single species for material currentlyreferred to femoralis (i.e., the taxon is mono-phyletic), the extensive patristic and pairwisedistances are suggestive of multiple species.Cytochrome b distance is low within localities(0.0–0.8%) and much higher between locali-ties (3.9–14.6%). This is strongly suggestivethat a different species occurs at each of theselocalities (i.e., eight species), which would


femoralis and zaparoa

Sample ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

1 femoralis 397

PW

—

2 femoralis 398

PW

0.0 —

3 femoralis 1309

PW

0.0 0.0 —

4 femoralis 78 CA 5.2 5.2 5.2 —

5 femoralis 128

CA

5.2 5.2 5.2 0.0 —

6 femoralis 129

CA

5.2 5.2 5.2 0.0 0.0 —

7 femoralis 399

CU

11.7 11.7 11.7 14.3 14.3 14.3 —

8 femoralis 400

CU

11.4 11.4 11.4 14.0 14.0 14.0 0.8 —

9 femoralis 1306

CU

11.7 11.7 11.7 14.3 14.3 14.3 0.5 0.3 —

10 femoralis 393

PEG-M

14.6 14.6 14.6 14.0 14.0 14.0 9.9 10.7 10.4 —

11 femoralis 394

PEG-M

14.6 14.6 14.6 14.0 14.0 14.0 9.9 10.7 10.4 0.0 —

12 femoralis 520

RD

13.8 13.8 13.8 14.0 14.0 14.0 10.4 10.7 10.4 7.3 7.3 —

13 femoralis 526

PAN

12.7 12.7 12.7 13.3 13.3 13.3 9.9 10.7 10.4 7.3 7.3 3.9 —

14 femoralis 1325

SIP

12.0 12.0 12.0 12.2 12.2 12.2 8.1 8.8 8.6 6.2 6.2 4.7 4.2 —

15 femoralis 395

RCU

13.3 13.3 13.3 13.8 13.8 13.8 8.6 9.4 9.1 6.2 6.2 5.7 5.7 3.6 —

16 femoralis 396

RCU

13.5 13.5 13.5 14.0 14.0 14.0 8.8 9.6 9.4 6.5 6.5 6.0 6.0 3.9 0.3 —

17 femoralis 1298

RCU

13.3 13.3 13.3 13.8 13.8 13.8 8.6 9.4 9.1 6.2 6.2 5.8 5.7 3.6 0.0 0.3 —

18 zaparo 321 13.3 13.3 13.3 13.0 13.0 13.0 12.2 12.0 12.2 14.3 14.3 14.6 15.6 13.3 14.3 14.6 14.3 —

19 zaparo 328 13.0 13.0 13.0 12.7 12.7 12.7 12.5 12.2 12.5 14.6 14.6 14.8 15.3 13.5 14.6 14.8 14.5 0.3 —

a Gray lines separate localities and species. Abbreviations are: PW (Port Walter), CA (Cusco Amazonico), CU

(Cuyabeno), PEG-M (Parque Estadual Guajara-Mirim), RD (Reserva Ducke), SIP (Sipaliwini), RCU (Rio Curua-Una).


greatly increase the known diversity of thisclade.

Thirty-four unambiguous transformationsestablish the monophyly of the clade shownin figure 73, with a Bremer value of 10. Thecaeruleodactylus–’’PEG-M3’’ clade is unitedby 36 unambiguous transformations. Limaand Caldwell (2001) named caeruleodactylus,and Caldwell et al. (2002a) described itsdistinctive tadpole. Although we did notinclude marchesianus in the present analysis,the conspicuous dark vertical stripes on thetail and greatly enlarged marginal papillaeprovide evidence that it is the sister species ofcaeruleodactylus (see Caldwell et al., 2002a).

Pairwise distances between specimens ofcaeruleodactylus are shown in table 13. Thesamples were collected at the type locality.

Sample 1277 from Rio Ituxi was referredto gasconi by Morales (2002 ‘‘2000’’), as wasthe distantly related sample 356 from PortoWalter (the pairwise distance between cyto-chrome b sequences of these two specimens is15.6%). The type locality given for thisspecies is ‘‘Jainu on the left side of the RıoJurua, Amazonas, Brazil’’ (Morales, 2002‘‘2000’’: 30, translated from the Spanish).Although Porto Walter is located on the RioJurua, Rio Ituxi is slightly closer, and on thatbasis we refer these terminals to gasconi.

Fig. 73. Strict consensus of 25,872 most parsimonious trees of 46,520 steps: relationships amongdendrobatids. Numbers above branches are Bremer support values. Numbers following terminal namesare unique sample identifiers. Terminals without sample numbers were taken from GenBank. Note thatMorales (Morales, 2002 ‘‘2000’’) identified ‘‘PortoWalter2’’ 356 as gasconi, and ‘‘Curua-Una’’ 354 asbrunneus (see fig. 72). Upper right inset shows entire cladogram and corresponding figure numbers, withpresent view in black.


Comparison with topotypes will be requiredto confirm the identity of these samples. Thepairwise cytochrome b distance between thisspecies and caeruleodactylus is 14.0–14.5%,between this species and the undescribed‘‘Curua-Una’’ (see below) 13.5–14.2%, andbetween this species and ‘‘PEG-M3’’ 15.8–16.4%. Pairwise cytochrome b distances with-in gasconi are given in table 14.

Samples from near Rio Curua-Una repre-sent an undescribed species (‘‘Curua-Una’’;see table 15). The pairwise distances betweenthe samples of this species and those of‘‘PEG-M3’’ (see below) are 9.9–10.9%. Mor-ales (2002 ‘‘2000’’) referred sample 354 to thedistantly related brunneus; as mentionedabove, the minimum pairwise distance be-tween this specimen and either of the othersMorales referred to brunneus is 16.6%.

‘‘PEG-M3’’ is one of three undescribedspecies of dendrobatids collected at ParqueEstadual Guajara-Mirim, Rondonia, Brazil.The pairwise distance between the samples ofthis species and ‘‘Curua-Una’’ is 9.9–10.9%(for distances within ‘‘PEG-M3’’ see ta-ble 16). In the current analysis, humilis isnested within the samples of ‘‘PEG-M3’’.However, it is highly unlikely that thepopulations are conspecific: The sample of

humilis was collected at 2,100 m in theVenezuelan Andes (La Marca et al., 2002),whereas ‘‘PEG-M3’’ is from the Amazonianlowlands of western Brazil. Sequence data forhumilis are limited to approximately 500 bpof 16S, so additional molecular data is likelyto overturn this result. La Marca et al. (2004)considered humilis to be most closely relatedto their new species pittieri, which we did nottest here.

The clade composed of conspicuus, insper-atus, the unidentified Ecuadorian speciesreported by Santos et al. (2003; no localitygiven), and an undescribed species fromCuyabeno, Ecuador is well supported (Bre-mer support 5 56) and united by 70 un-ambiguous transformations. The samples ofthe Brazilian species conspicuus were collect-ed at Porto Walter and sample 614 wasreferred to conspicuus by Morales (2002‘‘2000’’). Bremer support for the monophylyof these specimens is 157, and 157 synapo-morphies optimize to it unambiguously. Theremaining three species in this clade (branchlength 5 62, Bremer support 5 56) are allfrom Ecuador. Cytochrome b data are notavailable for insperatus and the unnamed‘‘Colostethus sp.’’, but pairwise distances are

TABLE 13Percent Uncorrected Pairwise Distances BetweenCytochrome b Sequences of Syntopic Specimens

of caeruleoactylus

Sample ID 1 2 3 4

1 caeruleoactylus 406 —

2 caeruleoactylus 621 0.3 —

3 caeruleoactylus 1261 0.0 0.3 —

4 caeruleoactylus 1287 0.3 0.5 0.3 —


Cytochrome b Sequences of gasconia

Sample ID 1 2 3 4

1 gasconi 357 —

2 gasconi 358 0.0 —

3 gasconi 1277 0.3 0.3 —

4 gasconi 1284 0.3 0.3 0.0 —

a Sample 1277 was referred to gasconi by Morales

(2002 ‘‘2000’’).

TABLE 15Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of an

Undescribed Species from Near the Rio Curua-Una in Brazila

Sample ID 1 2 3 4 5 6

1 ‘‘Curua-Una’’ 351 —

2 ‘‘Curua-Una’’ 353 0.3 —

3 ‘‘Curua-Una’’ 354 0.3 0.0 —

4 ‘‘Curua-Una’’ 1260 0.8 0.5 0.5 —

5 ‘‘Curua-Una’’ 1268 0.0 0.3 0.3 0.8 —

6 ‘‘Curua-Una’’ 517 0.8 0.5 0.5 0.0 0.8 —

a Morales (2002 ‘‘2000’’) referred sample 354 to the distantly related brunneus.


shown in table 17 for conspicuus and ‘‘Cuya-beno’’.

The remaining terminals in figure 73 areallied to trilineatus. In their redescription oftrilineatus based on extensive material fromPeru, Grant and Rodrıguez (2001) notedvariation within and between localities thatcould be representative of greater speciesdiversity. The present study included DNAsequences of putative trilineatus samplesfrom Cusco Amazonico, Madre de Dios,Peru (74 and 112; specimens not examined byGrant and Rodrıguez, 2001, but referredexplicitly to trilineatus by Morales, 2002‘‘2000’’) and Panguana, Huanuco, Peru(527; Grant and Rodrıguez, 2001, referredmaterial from this locality to trilineatus;Morales, 2002 ‘‘2000’’, referred specimensfrom this locality to marchesianus, but thatspecies is endemic to the Rio Uaupes of

Brazil and adjacent Rio Vaupes of Colombia;Caldwell et al., 2002a, 2002b; T. Grant,personal obs.). The type locality of Yuri-maguas is closest to Panguana. Also includedhere are samples of one of two dendrobatidspecies collected at Porto Walter, referred toas ‘‘PortoWalter2’’. As mentioned above, oneof these specimens (356) was referred togasconi by Morales (2002 ‘‘2000’’). A singlesample each is available from Sao Francisco(516), Reserva Ducke (524), and Rio Ituxi(404), and several samples each from ParqueEstadual Guajara-Mirim (359, 616, 1288),Manaus (620, 405, 1318), Cusco Amazonico(74, 112), and Porto Walter (355, 356, 1265,1269, 1273).

The trilineatus clade has a Bremer supportvalue of 18, with 25 unambiguous transfor-mations at this node. As shown in figure 73and table 18, the pattern of diversification is


conspicuus and an Undescribed Species from Cuyabeno, Ecuadora

Sample ID 1 2 3 4 5 6 7 8 9 10 11 12

1 conspicuus 614 —

2 conspicuus 615 0.0 —

3 ‘‘Cuyabeno’’ 346 13.5 13.5 —

4 ‘‘Cuyabeno’’ 347 12.5 12.5 1.6 —

5 ‘‘Cuyabeno’’ 348 13.2 13.2 0.3 1.6 —

6 ‘‘Cuyabeno’’ 349 14.0 14.0 0.5 1.8 2.3 —

7 ‘‘Cuyabeno’’ 350 13.5 13.5 0.3 2.1 2.1 0.8 —

8 ‘‘Cuyabeno’’ 402 12.5 12.5 1.6 0.0 1.8 2.1 1.8 —

9 ‘‘Cuyabeno’’ 403 13.5 13.5 0.0 1.5 1.8 0.5 0.3 1.6 —

10 ‘‘Cuyabeno’’ 1262 13.0 13.0 0.5 1.0 1.8 1.0 0.8 1.0 0.5 —

11 ‘‘Cuyabeno’’ 1283 12.5 12.5 1.6 0.0 1.8 2.1 1.8 0.0 1.6 1.0 —

12 ‘‘Cuyabeno’’ 1317 12.5 12.5 1.6 0.0 1.8 2.1 1.8 0.0 1.6 1.0 0.0 —


TABLE 16Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of ‘‘PEG-M3’’

Sample ID 1 2 3 4 5 6 7 8 9

1 ‘‘PEG-M3’’ 360 —

2 ‘‘PEG-M3’’ 619 0.0 —

3 ‘‘PEG-M3’’ 1263 0.3 0.3 —

4 ‘‘PEG-M3’’ 1272 0.0 0.0 0.3 —

5 ‘‘PEG-M3’’ 1274 0.3 0.3 0.0 0.3 —

6 ‘‘PEG-M3’’ 1279 0.3 0.3 0.0 0.3 0.0 —

7 ‘‘PEG-M3’’ 1295 0.3 0.3 0.0 0.3 0.0 0.0 —

8 ‘‘PEG-M3’’ 1296 0.0 0.0 0.3 0.0 0.3 0.3 0.3 —

9 ‘‘PEG-M3’’ 1319 0.0 0.0 0.3 0.0 0.3 0.3 0.3 0.0 —


suggestive of eight species—one at eachlocality. The minimum pairwise cytochromeb distances between localities is 5.7% betweenthe trilineatus from Cusco Amazonico andthe samples from Porto Walter. Despite thegeographic and cladistic proximity of theReserva Ducke and Sao Francisco samples,these samples (which are united by 25unambiguous transformations) are sepa-rated by a patristic distance of 244 stepsand their cytochrome b sequences are 15.3%different.

The clade shown in figure 74 is united by89 unambiguous transformations, includingreduction in the length of finger IV (Charac-ter 4, 0 R 1), lengthening of finger I(Character 5, 2 R 3), and swelling of fingerIII in adult males (Character 20, 0 R 1). Thenext clade, shown at the top of figure 74,includes the nubicola group and Silverstone’s(1976) tricolor group + machalilla. Thisinclusive clade is delimited by 84 unambigu-ous transformations in DNA sequences.

The nubicola group, represented by flota-tor, nubicola, and the undescribed species‘‘nubicola-spC’’ to be named by T. Grant andC. W. Myers (in preparation) is delimited by46 unambiguous transformations, includingthe gain of a straight pale ventrolateral stripe(Character 54, 0 R 2), pale male abdomencolor (Character 63, 3 R 0), anteriorpigmentation of the large intestine (Character66, 0 R 1), and several synapomorphiesrelating to the larval oral disc (Character 88,0 R 1; Character 89, 1 R 0; Character 91,0 R 1; and Character 94, 3 R 0). This cladeincludes data downloaded from GenBankthat were attributed to pratti from westernColombia by Vences et al. (2003a). However,one of the authors of that study informed usthat they did not examine a voucher speci-men (S. Lotters, in litt. 2/23/2005), andnubicola and pratti occur in sympatry andare often confused by collectors. The threesampled species resemble each other greatly;the Central American species flotator was


trilineatus and Related Undescribed Speciesa

Sample ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

1 SF 516 —

2 RD 524 15.3 —

3 PEG-M1

359

16.1 16.4 —

4 PEG-M1

616

16.1 16.4 0.0 —

5 PEG-M

1288

16.1 16.4 0.0 0.0 —

6 RI 404 15.1 16.4 8.6 8.6 8.6 —

7 MAN1 620 17.4 14.8 9.1 9.1 9.1 7.3 —

8 MAN1 405 17.4 14.8 9.1 9.1 9.1 7.3 0.0 —

9 MAN1 1318 17.7 15.1 9.4 9.4 9.4 7.3 0.3 0.3 —

10 triPAN 517 14.3 16.9 15.6 15.6 15.6 16.4 16.1 16.1 16.4 —

11 triCA 74 13.5 15.6 15.3 15.3 15.3 15.8 15.6 15.6 15.8 17.4 —

12 triCA 112 13.8 15.3 15.6 15.6 15.6 16.1 15.3 15.3 15.6 17.7 0.3 —

13 PW2 355 14.0 14.8 13.0 13.0 13.0 13.5 13.8 13.8 14.0 14.8 5.5 5.7 —

14 PW2 356 13.5 14.3 13.0 13.0 13.0 13.5 13.8 13.8 14.0 14.3 6.0 6.2 0.5 —

15 PW2 1265 13.8 14.6 12.7 12.7 12.7 13.2 13.5 13.5 13.8 14.5 5.7 6.0 0.3 0.3 —

16 PW2 1269 14.0 14.8 13.0 13.0 13.0 13.5 13.8 13.8 14.0 14.8 6.0 6.2 0.5 0.5 0.3 —

17 PW2 1273 14.0 14.8 12.7 12.7 12.7 13.2 13.5 13.5 13.8 14.5 6.0 6.2 0.5 0.5 0.3 0.5 —

a Abbreviations are as follows: MAN1 (‘‘Manaus1’’), PW2 (‘‘PortoWalter2’’), RD (‘‘ReservaDucke’’), RI

(‘‘RioItuxi’’), ‘‘PEG-M1’’, SF (‘‘SaoFrancisco’’), triCA (trilineatus Cusco Amazonico), triPAN (trilineatus, Panguana).

Gray lines separate localities.


considered a synonym of nubicola until 1995(Ibanez and Smith, 1995), but these twospecies are not sisters and differ in 18.4% oftheir cytochrome b sites (table 19).

The sister of the nubicola clade includesseveral taxonomically problematic taxa. Lot-ters et al. (2003b) noted differences in thevocalizations of boulengeri and concludedthat more than one species was probablyinvolved. Cytochrome b sequences are un-available for the GenBank specimen forcomparison, but only eight unambiguoustransformations group boulengeri 280with the other species of this clade. Cyto-chrome b sequences are also unavailablefor GenBank specimen Epipedobates sp.QCAZ16589, but the occurrence of only fourunambiguous transformations to distinguishit from espinosai 1139 suggests that it may beconspecific with espinosai. Like Santos et al.(2003), we found that machalilla is nestedwithin this clade of otherwise toxic species.However, the specimens we sequenced falltogether, whereas the Santos et al. sequenceobtained from GenBank is sister to tricolor.Graham et al. (2004) reported this sample ofmachalilla to be most closely related toanthonyi instead of tricolor, although thatconclusion was not supported in their anal-ysis (the critical node had a Bremer value of0). In our analysis, only a single unambigu-ous synapomorphy unites tricolor and ma-chalilla, and the clade has a Bremer value ofonly 1. There is little unambiguous evidenceto group these samples with anthonyi to theexclusion of machalilla samples 73 and 132(only five transformations), but it is worthnoting that samples 73 and 132 are united by24 unambiguous transformations and differin only 9.

Grant and Castro-Herrera (1998) notedthe extensive within- and among-populationvariation in fraterdanieli and left open thepossibility that this may be a complex ofsimilar species. The samples of fraterdanieliwere collected in Colombia near Popayan,Cauca, at approximately 1,800 m in theCordillera Occidental (1226, 1227, 1228).An additional sample was collected at2,800 m in the Departamento de Caldas inthe Cordillera Central (1230). All localitiesface the Cauca valley. As seen in table 20, thethree Cauca fraterdanieli cytochrome b se-quences are identical. The pairwise distancebetween those samples and the Caldasspecimen is 6.5%. Likewise, the three Caucaspecimens are united by 42 unambiguoustransformations, and the Caldas samplediffers by an additional 63 unambiguoustransformations. This pattern of diversity isstrongly suggestive that these are two differ-ent species. The type locality of frateranieli isin the Cordillera Central in Antioquia, atapproximately 1,900 m (Silverstone, 1971).Minimally, comparison with samples fromlower elevations in the Cordillera Central isrequired to determine which of these popula-tions (i.e., that from approximately the sameelevation but much further south in theCordillera Occidental or that from theCordillera Central in roughly the same regionbut much higher elevation) is conspecific withfraterdanieli sensu stricto or if additionalspecies have been conflated under this name.

The terminals labeled pratti 1144 and and‘‘pratti-like’’ 1224 are morphologically in-distinguishable but are almost certainly notconspecific. Sample 1144 was collected at ElCope, Cocle, central Panama, and 1224 isfrom Jungurudo, Darien, near the boarder


flotator, nubicola, and ‘‘nubicola-spC’’a

Sample ID 1 2 3 4 5 6

1 flotator 1143 —

2 flotator 1145 0.0 —

3 nubicola 1142 18.4 18.4 —

4 nubicola 1146 18.4 18.4 0.0 —

5 ‘‘nubicola-spC’’ 496 15.3 15.3 22.1 22.1 —

6 ‘‘nubicola-spC’’ 497 15.3 15.3 22.1 22.1 0.0 —



with Colombia. Roberto Ibanez noted differ-ences in their vocalizations (in litt., 12/20/2003). Further, only female nurse frogs areknown to occur in pratti (Grant, 2004),whereas a male nurse frog was collected atJungurudo. The behavioral sample size at theDarien locality is inadequate to eliminate thepossibility that this is simply intraspecificvariation, but these behavioral differences arefurther reinforced by the observation that,although these two samples are united by 104unambiguous transformations, the patristicdistance between them is 163. Finally, thepairwise distance between their cytochromeb sequences is 10.6%. As such, despite thelack of morphological differences betweenthese frogs, there is considerable evidencethat they represent different species. Resolu-tion of this problem, though evidentiallystraightforward, is nomenclaturally compli-cated. The type locality of pratti is in westernColombia, and the relationship of topotypicpratti to either of these samples has not yetbeen assessed. The proximity of the Darienspecies to the type locality suggests it may bepratti sensu stricto, but direct evidence isrequired. As noted above, the specimenreported as pratti from western Colombiaby Vences et al. (2003a) is probably a mis-identified specimen of nubicola, but that toorequires confirmation.

Grant (2004) removed panamensis fromthe synonymy of inguinalis on morphologicalgrounds, and, although there are severalpoints of resemblance, imbricolus differsextensively from both species (e.g., ventralcoloration, color and definition of flashmarks, degree of webbing, sexual dimor-phism, and occurrence of tetrodotoxin).Although the identities of inguinalis andimbricolus are not problematic, panamensis

is widespread and highly variable. Dunn(1933) and Savage (1968) drew attention todifferences between western and easternsamples. Grant (2004) found that variationbetween localities was no greater than isobserved in samples from each locality andtherefore concluded that the samples ofpanamensis constituted a single species. Thetwo panamensis samples included for DNAsequence data are from distant localities:1150 is from El Cope in central Panama,whereas 1223 is from extreme eastern Pana-ma at Cana, Darien at the eastern extreme ofthe distribution, near the Colombian border.

Both the cladistic results and the pairwisecytochrome b comparisons (table 21) supportGrant’s (2004) conclusion that inguinalis isnot conspecific with the Panamanian speciespreviously assigned to its synonymy. How-ever, the present results suggest that the twosamples of panamensis represent differentspecies. The pairwise distance between thecytochrome b sequences for these two sam-ples is 11.4%. Furthermore, the cladogramshows the western sample to be more closelyrelated to imbricolus, from which its cyto-chrome b sequence differed by only 3.9%.Denser sampling at intervening localities, aswell as additional data (e.g., vocalizations,behavior), are required to address thisproblem decisively.

The next large clade includes the majorityof the species referred to Phyllobates bySilverstone (1976), Ameerega by Bauer(1986), and Epipedobates by Myers (1987).More specifically, it is equivalent to thecombination of Silverstone’s (1976) pictusand trivittatus groups, with the addition ofspecies described subsequently. The clade isdelimited by 131 unambiguous transforma-tions, including the almost unique gain of


Cytochrome b Sequences of fraterdanielia

Sample ID 1 2 3 4

1 fraterdanieli Cauca 1226 —

2 fraterdanieli Cauca 1227 0.0 —

3 fraterdanieli Cauca 1228 0.0 0.0 —

4 fraterdanieli Caldas 1230 6.5 6.5 6.5 —

a Gray lines separate localities.

TABLE 21Percent Uncorrected Pairwise Distances BetweenCytochrome b Sequences of imbricolus, inguinalis,

and Two Distant Localities of panamensisa

Sample ID 1 2 3 4

1 imbricolus 1229 —

2 inguinalis 1348 15.6 —

3 panamensis 1150 El Cope 11.7 16.1 —

4 panamensis 1223 Cana 3.9 14.8 11.4 —



conspicuously granular dorsal skin (Charac-ter 0, 1 R 2) and ability to sequesterlipophilic alkaloids (Character 146, 0 R 1).

Unlike other widespread Amazonian spe-cies, such as femoralis (discussed above), anddespite the known degree of external colorand color pattern variation (Silverstone,1976) of nominal trivittatus, the pattern andextent of diversity are not suggestive of morethan a single species (see table 22). Weincluded samples of trivittatus from sevenlocalities covering (albeit sparsely) most ofthe known range of the species, as follows(listed approximately from southwest tonortheast): Tambopata Reserve, Madre deDios, Peru (319, 320, 322); Porto Walter,Acre, Brazil (385, 387, 1297, 1305); Pan-guana, Huanuco, Peru (518); Leticia, Ama-zonas, Colombia (1350); Balbina, north ofManaus, Amazonas, Brazil (519); south ofManaus, Amazonas, Brazil (627, 628); andPara, Suriname (1329).

The monophyly of trivittatus is establishedby 96 unambiguous transformations, and thepairwise cytochrome b distances betweenthese samples and Guyanan pictus are 13.2–14.3% and southeastern Brazilian flavopictusare 10.9–12.5%. Conversely, the variationwithin trivittatus is low, despite the greatdistances between localities. Pairwise cyto-chrome b distances between localities are 0.5–

3.4%. Although the higher values are as greatas or greater than those between some closelyrelated species (e.g., auratus and truncatus;see below), there are no major gaps (i.e.,pairwise distances appear to vary continu-ously) or geographic trends, and cladisticrelationships do not suggest historicallyisolated populations.

Duellman and Mendelson (1995) referredsample 127 from northern Peru to zaparo,but they also noted that theirs was the firstrecord of that taxon outside the Rıo Pastazadrainage. The present results demonstrateconclusively that this species is not conspe-cific with zaparo, despite their morphologicalresemblance, and we therefore place thename in quotes. Sufficient data (e.g., locality)are unavailable to determine if ‘‘zaparo’’ andSantos et al.’s (2003) parvulus are conspecific.

One of the more unexpected species-levelresults is the grouping of the GenBanksample of pictus from Bolivia, near the typelocality, with pictus 1331 from Guyana.Despite the great geographic distance be-tween these localities, the samples appear tobe conspecific.

‘‘PortoWalter1’’ is another undescribedspecies from Porto Walter, Brazil. The sisterof this species is rubriventris. Although only12 unambiguous transformations diagnose‘‘PortoWalter1’’ from rubriventris (three

TABLE 22Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of trivittatusa

Sample ID 1 2 3 4 5 6 7 8 9 10 11 12 13

1 trivittatus 319 TAM —

2 trivittatus 320 TAM 0.0 —

3 trivittatus 322 TAM 0.0 0.0 —

4 trivittatus 385 PW 1.8 1.8 1.8 —

5 trivittatus 387 PW 2.1 2.1 2.1 3.4 —

6 trivittatus 1297 PW 2.1 2.1 2.1 3.4 0.52 —

7 trivittatus 1305PW 2.1 2.1 2.1 3.4 0.52 0.52 —

8 trivittatus 518 PAN 1.2 1.2 1.2 2.6 1.3 1.3 1.3 —

9 trivittatus 1350 LET 2.1 2.1 2.1 3.4 1.6 1.6 1.6 1.3 —

10 trivittatus 519 BAL 1.6 1.6 1.6 1.3 3.1 3.1 3.1 2.3 3.1 —

11 trivittatus 627 MAN 1.3 1.3 1.3 0.5 2.9 2.9 2.9 2.1 2.9 0.8 —

12 trivittatus 628 MAN 1.3 1.3 1.3 0.5 2.9 2.9 2.9 2.1 2.9 0.8 0.0 —

13 trivittatus 1329 PAR 1.6 1.6 1.6 1.3 3.1 3.1 3.1 2.3 3.1 0.0 0.8 0.8 —

a Gray lines separate localities. Abbreviations are: TAM (Tambopata Reserve), PW (Porto Walter), PAN (Panguana),

LET (Leticia), BAL (Balbina), MAN (Manaus), and PAR (Para).


‘‘PortoWalter1’’ synapomorphies and ninerubriventris autapomorphies) only 566 bp of16S data were available for rubriventris.Cytochrome b sequences are identical insamples of ‘‘PortoWalter1’’, except for sam-ple 626, which differs from the others ina single nucleotide.

Like trivittatus and femoralis, hahneli isanother widespread Amazonian species. Thetype locality for hahneli is Yurimaguas, Peru.We included 12 samples of hahneli from fourlocalities, as follows: Cusco Amazonico, Peru(79, 109, 110), Leticia, Colombia (1354);south of Manaus, Brazil (386, 391, 392,1304), and Porto Walter, Brazil (382, 388,389, 390). Pairwise distances between thecytochrome b sequences of these samples aregiven in table 23, and these sequences differfrom those of pulchripectus in 10.4–11.7% oftheir sites. The hahneli samples are united by51 unambiguous transformations. The sam-ple from Leticia differs from the others in6.5–7.5% of its cytochrome b sequence—much more than occurs between othersamples, despite the greater geographic dis-tance between other samples (e.g., CuscoAmazonico and Manaus). Likewise, the cladecontaining the remaining hahneli samples isunited by 31 unambiguous transformations.This suggests that samples from Leticia andother localities are not conspecific. It isunclear which of these species is conspecificwith hahneli sensu stricto. The cytochrome

b distances between the remaining hahnelilocalities are 2.1–3.1%. Although Santos etal. (2003) did not provide specimen data, thetopology suggests their specimen (QCAZ-13325) is conspecific with the Leticia hahneli.

In figure 75, the Colombian species sub-punctatus is sister to a clade diagnosed by 76unambiguous transformations, including sev-eral changes in hand and foot morphology(Characters 13, 15, 36–44), the appearance ofposteriorly angled clavicles (Character 121,0 R 1), gain of palatine bones (Character132, 0 R 1), and the shift to riparian habitat(Character 113, 2 R 1).

Santos et al. (2003) resurrected maculosusfrom synonymy with bocagei, where it hadbeen placed by Coloma (1995). Key to thatinterpretation is the identity of the specimenthey identified as bocagei sensu stricto,because that species falls out with sauli bothhere and in Santos et al.’s (2003) analysis.However, no locality or other data wereprovided for that specimen, and an alterna-tive possibility is that the remaining samples(including those identified here as bocageifrom Cuyabeno) are conspecific with topo-typic bocagei and the sister of sauli is anundescribed species. Additional data arerequired to assess the alternative hypotheses.

Santos et al. (2003) also omitted localitydata for the unidentified specimens Colos-tethus sp. QCAZ16511, Colostethus sp.QCAZ16504, and Colostethus sp. QCAZ-

TABLE 23Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of hahnelia

Sample ID 1 2 3 4 5 6 7 8 9 10 11 12

1 hahneli 79 CA —

2 hahneli 109 CA 0.3 —

3 hahneli 110 CA 0.3 0.0 —

4 hahneli 382 PW 2.6 2.6 2.6 —

5 hahneli 388 PW 2.9 2.9 2.9 0.3 —

6 hahneli 389 PW 2.6 2.6 2.6 1.0 1.3 —

7 hahneli 390 PW 3.1 3.1 3.1 1.6 1.8 1.0 —

8 hahneli 1354 LET 7.5 7.3 7.3 6.8 7.0 6.0 6.5 —

9 hahneli 386 MAN 2.1 2.1 2.1 2.1 2.3 2.1 2.6 7.3 —

10 hahneli 391 MAN 2.1 2.1 2.1 2.1 2.3 2.1 2.6 7.3 0.0 —

11 hahneli 392 MAN 2.1 2.1 2.1 2.1 2.3 2.1 2.6 7.3 0.0 0.0 —

12 hahneli 1304 MAN 2.1 2.1 2.1 2.1 2.3 2.1 2.6 7.3 0.0 0.0 0.0 —

a Gray lines separate localities. Abbreviations are: CA (Cusco Amazonico), PW (Porto Walter), LET (Leticia), and

MAN (Manaus).


16503, which complicates understanding thediversification of these dendrobatids. Asnoted above, one possibility is that theseand related terminals are conspecific. Never-theless, in light of the patristic distancesbetween these terminals (128 steps betweenQCAZ16511 and QCAZ16504; 87 betweenQCAZ16504 and QCAZ16503), our pro-visional interpretation is that Colostethus sp.QCAZ16511 and Colostethus sp. QCAZ-16504 are different, possibly undescribedspecies, and that Colostethus sp. QCAZ16503is conspecific with the terminals from Cuya-beno (e.g., patristic distance between

QCAZ16503 bocagei 1267 5 8 steps). Al-though the topology is consistent withColostethus sp. QCAZ16511 being conspecif-ic with maculosus sensu Santos et al. (2003),50 and 47 unambiguous transformations inmtDNA subunit H1 occur at these terminalnodes, respectively, suggesting they representdifferent species.

The clade composed of delatorreae, pul-cherrimus, and sylvaticus Barbour and Nobleis delimited by 38 unambiguous transforma-tions. These species are all from mid- to highelevations in the Andes of northern Ecuador(delatorreae) and northern Peru (pulcherrimus



and sylvaticus Barbour and Noble). Duell-man (2004) recently named pulcherrimus andcompared it to the similar sylvaticus Barbourand Noble. The two samples of pulcherrimus(118 and 119) are topoparatypes (Cajamarca,Peru), and both samples of sylvaticus Bar-bour and Noble were collected at 2,820 m inAyacaba, Peru. The species are closely re-lated, but the pairwise distances betweentheir cytochrome b sequences are 13.0–13.3% (see table 24) and each is diagnosedby approximately 100 unambiguous trans-formations.

The sister group to that clade includesnexipus, azureiventris, chlorocraspedus, andan unidentified species sequenced by Santoset al. (2003; no locality data given). Theknown species form a distinctive group ofrelatively brightly colored frogs with dorso-lateral stripes, the latter being an unambig-uous synapomorphy of the clade (Character52, 0 R 3). In total, the clade is delimited by51 unambiguous transformations. Lotters etal. (2000) proposed the genus Cryptophyllo-bates for the putatively aposematic azurei-ventris. However, Daly (1998:171) reportedthat in a feeding experiment this species didnot accumulate dietary alkaloids. The re-cently described species chlorocraspedus is asbrightly colored as azureiventris, and wild-caught samples lacked detectable levels ofalkaloids also (J. W. Daly, in litt., 01/28/05).Although they were not included in thepresent study, the two recently named speciespatitae (Lotters et al., 2003a) and eleuther-odactylus (Duellman, 2004) are also likelypart of this clade. The samples of chlorocras-pedus have identical cytochrome b sequences,with the exception of sample 385, whichdiffers in two nucleotides (0.5%). Samples ofnexipus were included from two localities at

different elevations (Cataratas Ahuashiyacu,14 km NE of Tarapoto, 730 m: 75, 130, 131;and San Martın, 6 km ESE of Shapaja,300 m: 123). The specimen from the lowerelevation is identical to two of the threespecimens from the higher elevation; thosespecimens differ from one of the 730 mspecimens in two nucleotides. Santos et al.(2003) omitted locality data for the samplethey identified as nexipus, but 45 unambigu-ous transformations optimize to the terminalnode (all from mtDNA subunit H1) and 42unambiguous changes delimit the remainingspecimens as a clade (all from Peru), suggest-ing that they may not be conspecific.

The terminals referred to as ‘‘Ibague’’ arean undescribed species from the slopes of theMagdalena valley in Colombia (see table 25for pairwise cytochrome b distances). Thespecies possesses the black arm band in adultmales and is thus the sole exemplar of theramosi group included in present study(Grant and Castro, 1998; Grant and Ardila-Robayo, 2002). Other species that possessthis structure (and are included in the ramosigroup) are anthracinus, cevallosi, fascianiger,exasperatus, lehmanni Silverstone, ramosi,and saltuarius. ‘‘Ibague’’ is nested in a cladewith vertebralis and pulchellus, all of whichare small, identically striped, and similarlycolored Andean frogs. Forty-two changesoptimize unambiguously to the node includ-ing vertebralis and 34 unambiguous transfor-mations unite ‘‘Ibague’’ with pulchellus.

Rivero (1991a) described idiomelus basedon a single specimen (from Venceremos,Peru), but Duellman’s (2004) account wasbased on extensive material, including adultsand larvae from several localities; all of thesamples sequenced in the present study werereferred to idiomelus by Duellman (2004).Three specimens (120–122) are from 2,180 mat Abra Pardo de Miguel, San Martın, andthe other two (77 and 126) are from 2,150 mat Pomachochas, Amazonas. The five speci-mens form a clade, but the Abra Pardolocality is paraphyletic with respect toPomachochas. Pairwise cytochrome b dis-tances are given in table 26.

Originally described from Loja, Ecuador,elachyhistus is a widespread, highly variableAndean species. Duellman (2004) recentlyredescribed elachyhistus from several locali-


Cytochrome b Sequences of pulcherrimus andsylvaticus Barbour and Noblea

Sample ID 1 2 3 4

1 pulcherrimus 118 —

2 pulcherrimus 119 0.3 —

3 sylvaticus 76 13.0 13.3 —

4 sylvaticus 113 13.0 13.3 0.0 —



ties in northern Peru, including those in-cluded in the present study. Based on thecurrent results, it is clear two species havebeen conflated, a southern species fromCajamarca, Peru (105, 106, and 107), whichis sister to insulatus, and a northern speciesfrom Piura, Peru (108, 114, 115, 116, 117); seetable 27 for pairwise cytochrome b distances.Locality data were not given by Santos et al.(2003) for the GenBank elachyhistus includ-ed, but the data provided in GenBank reportit as being from Ecuador, which suggests thatthe northern species is elachyhistus and thesouthern species is undescribed.

Although toxic species also occur else-where in the cladogram (e.g., anthonyi,petersi), the remaining clade, shown infigure 76, consists of exclusively brightlycolored and (insofar as is known) toxicspecies. Evidence for the monophyly of thisclade is given by 63 unambiguous changes,including origin of smooth dorsal skin(Character 0, 1 R 0), loss of the obliquelateral stripe (Character 55, 1 R 0), the lossof metallic pigmentation of the iris (Charac-ter 65, 1 R 0), use of phytotelmata as larvalhabitat (Character 110, 0 R 1), and origin ofthe ability to sequester lipophilic alkaloids(Character 147, 0 R 1).

The clade composed of aurotaenia, bicolor,lugubris, terribilis, and vittatus constitutes

Phyllobates sensu Myers et al. (1978), andits monophyly is established by 146 un-ambiguous transformations, including thelengthening of finger I (Character 5, 2 R 3)and the ability to accumulate batrachotoxin(Character 148, 0 R 1). Species identities areuncontroversial, the sole potential exceptionbeing the possibility that terribilis representsthe southern extreme of clinal variation ofbicolor (Myers et al., 1978; Lotters et al.,1997a). That hypothesis is rejected in thecurrent phylogenetic analysis, which placesaurotaenia and terribilis as sister species tothe exclusion of bicolor. This result is alsoconsistent with cytochrome b pairwise dis-tances (table 28). The pairwise distancebetween terribilis and bicolor is 7.0%, where-as the pairwise distance between terribilis andaurotaenia is only 5.7%. The two lugubrissamples are from Panama (329) and Nicar-agua (366), representing opposite extremes inthe species’ distribution. These specimensform a clade, and there is no indication inmorphology or otherwise that lugubris mayrefer to more than a single species. Neverthe-less, the distance between cytochrome b se-quences of the two samples is 6.0%. Theterribilis samples are from the type locality inwestern Colombia (1135) and bred in captiv-ity (1232). The aurotaenia, bicolor, and one ofthe vittatus samples (839) were bred incaptivity; the second vittatus sample isGenBank sequence AF128582.

Maxson and Myers (1985) proposed thatthe South American species bicolor andterribilis were sisters and were, in turn, sisterto a clade composed of lugubris, aurotaenia,and vittatus, the latter two being sisters. Inaddition to a plausible biogeographic argu-ment, bicolor and terribilis were grouped onthe basis of the shared ontogenetic ‘‘loss’’


Cytochrome b Sequences of ‘‘Ibague’’a

Sample ID 1 2 3

1 ‘‘Ibague’’ 1225 —

2 ‘‘Ibague’’ 1347 0.3 —

3 ‘‘Ibague’’ 1345 La Mesa 0.5 0.3 —


TABLE 26Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of idiomelusa

Sample ID 1 2 3 4 5

1 idiomelus 120 Abra Pardo —

2 idiomelus 121 Abra Pardo 0.0 —

3 idiomelus 122 Abra Pardo 0.0 0.0 —

4 idiomelus 77 Pomachochas 0.5 0.5 0.5 —

5 idiomelus 126 Pomachochas 0.5 0.5 0.5 0.0 —



(through hypertrophy; see Character 52,above) of dorsolateral stripes (DLS). Widmeret al. (2000) tested that hypothesis with a 520-bp cytochrome b dataset and concurred thatbicolor and terribilis were sister-species.However, they found that aurotaenia wasplaced in a clade with the other SouthAmerican species, and that the two CentralAmerican species, lugubris and vittatus, weresisters. Our greatly enlarged dataset corrobo-rates Widmer et al.’s (2000) hypothesis ofCentral and South American monophyly, butwe found that bicolor is the sister ofaurotaenia + terribilis. Maxson and Myers(1985; see also Myers et al., 1978) hypothe-sized that the persistent DLS was ‘‘a primi-tive pattern that is retained by the adults ofaurotaenia, lugubris, and vittatus.’’ Accordingto our results, it is equally parsimonious forthe persistent DLS to have evolved in theancestor of Phyllobates, with either two‘‘losses’’ of the DLS in adults in bicolor andterribilis or a single ‘‘loss’’ of the DLS inadults in the ancestor of the South Americanclade and reversion in aurotaenia, or for theDLS in juveniles only to be ancestral forPhyllobates and the persistent DLS to haveevolved independently in aurotaenia and theancestor of lugubris and vittatus.

Myers (1987) designated steyermarki as thetype species of Minyobates, which he pro-posed for several species previously includedin Silvestone’s (1975a) minutus group ofdiminutive Dendrobates. Further, Myers hy-pothesized that steyermarki and its relativeswere placed outside of a Phyllobates +Dendrobates clade that included the remain-der of Silverstone’s minutus group. In our

results, steyermarki is placed as the sisterspecies of all species traditionally associatedwith Dendrobates, including a clade thatcorresponds to the bulk of Silverstone’sminutus group. The placement of thesespecies in a clade exclusive of Phyllobatesrefutes Myers’s hypothesis (but see discussionof castaneoticus and quinquevittatus, below),but is identical to the findings of Vences et al.(2003a). These clades are not strongly sup-ported, owing to the lack of evidence forsteyermarki, for which only phenotypic char-acters and 547 bp of 16S (the latter se-quenced by Vences et al., 2003a) could beincluded.

Silverstone (1976) named the distinctivespecies fulguritus from the Choco region ofwestern Colombia, and the included sampleis from near Bahıa Solano. The sister-speciesclaudiae and minutus strongly resemble eachother morphologically. Nevertheless, theircytochrome b sequences are 8.1–8.3% dis-similar (see table 29). The monophyly of thisgroup of three species is established by 69unambiguous transformations, including theoccurrence of dorsolateral and oblique lateralstripes (Characters 52 and 55) and the fusionof vertebrae 2 + 3 (Character 146, 0 R 1).

The sister group of the fulguritus cladecontains most of the Amazonian species ofSilverstone’s minutus group. Evidence for themonophyly of this group is given by 67unambiguous transformations, including theexpansion of finger discs II–IV (Characters8–10). Caldwell and Myers (1990) removedventrimaculatus from the synonymy of quin-quevittatus (see below), but they noted thatthe nominal taxon, which occurs throughout

TABLE 27Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of Two

Species Currently Referred to elachyhistusa

Sample ID 1 2 3 4 5 6 7 8

1 elachyhistus Cajamarca 105 —

2 elachyhistus Cajamarca 106 0.3 —

3 elachyhistus Cajamarca 107 0.0 0.3 —

4 elachyhistus Piura 108 13.0 13.2 13.0 —

5 elachyhistus Piura 114 11.4 11.7 11.4 2.1 —

6 elachyhistus Piura 115 11.7 11.9 11.7 2.3 0.3 —

7 elachyhistus Piura 116 11.7 11.9 11.7 2.3 0.3 0.0 —

8 elachyhistus Piura 117 11.4 11.7 11.4 2.1 0.0 0.3 0.3 —



the Amazon region from Peru to FrenchGuiana, probably consists of a complex ofsimilar species. Symula et al. (2003) presentedmolecular evidence that at least two distantlyrelated species are included in Peruvian‘‘ventrimaculatus’’. Assuming the validity of

amazonicus and variabilis (but see Lotters andVences, 2000, and Caldwell and Myers, 1990,respectively) the current results suggest fivespecies of ‘‘ventrimaculatus’’, one at RioItuxi, Brazil (376, 377), a second at Manaus,Brazil (1310), a third at Leticia, Colombia



(1349) and French Guiana (GenBankAY263248), a fourth at Pompeya, Ecuador(374), and a fifth at Porto Walter, Brazil (375,1311). Although the cladogram does notfalsify the hypothesis that the samples fromRio Ituxi and Manaus are a single species, 72unambiguous synapomorphies unite the twoRio Ituxi specimens, 58 autapomorphiesoptimize unambiguously to the Manausterminal node, and cytochrome b sequencesdiffer in 8.1% of their sites (table 30). Despitethe considerable geographic distance betweenthe Leticia and French Guiana samples, theyform a clade, and the patristic differencebetween them is minimal (14 steps; cyto-chrome b data are unavailable for the FrenchGuiana sample); however, this is based onanalysis of only 560 bp of 16S for the FrenchGuiana specimen. The Leticia and Pompeyalocalities are closest to the type locality ofSarayacu, Ecuador, suggesting that one ofthem may be conspecific with ventrimaculatussensu stricto.

The sister clade to the minutus groupconsists of the remaining species traditionallyreferred to Dendrobates. Fifty-four unambig-uous transformations delimit this node, in-cluding the origin of even caudal pigmenta-tion in larvae (Character 87, 1 R 2).

Caldwell and Myers (1990) clarified theidentity of quinquevittatus (removing the un-related ventrimaculatus from its synonymy inthe process; see above). They proposed a closerelationship between quinquevittatus and theclearly heterospecific castaneoticus. They didnot discuss the relationships of galactonotus,but its placement in the tinctorius group bySilverstone (1975a) was uncontroversial. Themonophyly of galactonotus, castaneoticus,and quinquevittatus was first proposed byVences et al. (2003a), although these threetaxa were unresolved in their topology. In thepresent study, 105 unambiguous synapomor-phies optimize to this node, and Bremersupport is 37, leaving little doubt as to thereality of this clade. Nevertheless, the occur-rence of galactonotus in this clade is un-expected, as its morphology shares little withthe diminutive castaneoticus and quinquevit-tatus. Pairwise cytochrome b distances forthese species are shown in table 31.

The next clade is delimited by 37 un-ambiguous transformations. The first cladeincluded in this group consists of thehistrionicus group of Myers et al. (1984).The evidence for the monophyly of thisgroup is overwhelming, consisting of 132unambigously optimized synapomorphies.These include several larval modifications

TABLE 29Percent Uncorrected Pairwise Distances BetweenCytochrome b Sequences of claudiae, fulguritus,

and minutusa

Sample ID 1 2 3 4 5

1 claudiae 323 —

2 claudiae 324 0.0 —

3 claudiae 330 0.3 0.3 —

4 fulguritus 499 14.0 14.0 13.8 —

5 minutus 1149 8.3 8.3 8.1 15.1 —



aurotaenia, bicolor, lugubris, terribilis, and vittatusa

Sample ID 1 2 3 4 5 6 7 8

1 aurotaenia 840 (CR) —

2 bicolor 1233 (CR) 6.0 —

3 lugubris 329 Panama 16.6 17.9 —

4 lugubris 366 Nicaragua 17.1 17.9 6.0 —

5 terribilis 1135 5.7 7.0 17.7 18.7 —

6 terribilis 1232 (CR) 5.7 7.0 17.7 18.7 0.0 —

7 vittatus 839 (CR) 16.4 16.9 6.5 5.7 17.9 17.9 —

8 vittatus (GB) 16.1 14.1 5.2 6.7 17.3 17.3 2.1 —

a CR 5 captive reared. GB 5 GenBank. Gray lines separate species.


(Characters 90, 93, and 94), origin of tadpoletransport by female nurse frogs (Character109, 0 R 1) and larval oophagy (Character111, 1 R 2), and fusion of the sacrum andvertebra 8 (Character 143, 0 R 1) andvertebrae 2 and 3 (Character 145, 0 R 1).

Myers and Daly (1976b) illustrated anddiscussed the extensive variation within whatthey considered to be the single specieshistrionicus, distributed throughout the Pa-cific lowlands of western Colombia andnorthwestern Ecuador. In the same paper,they named lehmanni Myers and Daly, basedprimarily on differences in vocalizations,coloration and color pattern, and, especially,the absence of histrionicotoxins from skinalkaloid profiles. Nevertheless, E. Zimmer-mann (1986:135) claimed that histrionicusand lehmanni Myers and Daly crosses pro-duced fertile offspring, and Garraffo et al.(2001) showed experimentally that lehmanniMyers and Daly efficiently sequesters histrio-

nicotoxins administered in the diet. Based ondifferences in vocalizations and colorationand color pattern, Lotters et al. (1999)resurrected sylvaticus Funkhouser from thesynonymy of histrioncus for the southernmostpopulations in southern Colombia andnorthern Ecuador.

The histrionicus samples included here wereboth collected in Choco department, Colom-bia, but are from distant localities and involvedifferent color morphs. Sample 336 was takenalong Quebrada Vicordo (locality D of Myersand Daly, 1976b; see their Plate 1C for colormorph), whereas sample 498 is from SierraMecana (approximately 6u159N, 77u219W),north of Bahıa Solano; the two localities areseparated by .100 km. The lehmanni Myersand Daly sample is from the region of the typelocality. The sample of sylvaticus Funkhouseris from Ecuador. Cytochrome b sequenceswere not available for the GenBank speci-mens shown in the cladogram.

TABLE 30Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of Nominal ventrimaculatusa

Sample ID 1 2 3 4 5 6 7

1 Rio Ituxi 376 —

2 Rio Ituxi 377 0.0 —

3 Manaus 1310 8.1 8.1 —

4 Pompeya 374 15.6 15.6 16.4 —

5 Leticia 1349 14.0 14.0 13.2 11.4 —

6 Porto Walter 375 17.4 17.4 16.7 16.4 13.2 —

7 Porto Walter 1311 16.9 16.9 16.1 16.1 12.7 1.0 —



castaneoticus, galactonotus, and quinquevittatusa

Sample ID 1 2 3 4 5 6 7 8

1 castaneoticus 363 —

2 galactonotus 533 (CR) 18.6 —

3 galactonotus 647 18.6 0.5 —

4 quinquevittatus 368 Rio Ituxi 17.4 15.8 16.4 —

5 quinquevittatus 369 Rio Ituxi 17.4 15.8 16.4 0.0 —

6 quinquevittatus 370 Rio Formoso 17.1 16.1 16.6 0.3 0.3 —

7 quinquevittatus 371 Rio Formoso 17.4 15.8 16.4 0.0 0.0 0.3 —

8 quinquevittatus 1312 Rio Formoso 17.4 15.8 16.4 0.0 0.0 0.3 0.0 —

a CR 5 captive reared. Gray lines separate localities and species.


The cladogram is consistent with the validityof these three species. The two samples ofhistrionicus were recovered as monophyletic;their cytochrome b sequences differ from eachother in only a single base (0.3%) and areapproximately 5% different from both leh-manni Myers and Daly and sylvaticus Funk-houser (see table 32). Similarly, sylvaticusFunkhouser, which had been in the synonymyof histrionicus until recently, is more closelyrelated to lehmanni Myers and Daly. Al-though it has never been postulated that thesetwo nominal species may be conspecific to theexclusion of histrionicus, that hypothesis is notruled out by the current results. Theircytochrome b sequences are only 2.9% dis-similar, which is less than the distance betweenthe closely related species bicolor and terribilis(7.0%) and minutus and claudiae (8.1–8.3%),for example, but is greater than is observedbetween some specimens of the clearly hetero-specific auratus and truncatus (2.3–3.1%; seebelow). Regardless of their low degree ofpairwise dissimilarity, lehmanni Myers andDaly and sylvaticus Funkhouser are stilldiagnosable on the basis of phenotypicevidence (Myers and Daly, 1976b; Lotters etal., 1999) and therefore are valid species.

Also included in this clade are a number ofsmall species allied phenetically to pumilio.The systematics of these species has beenconfounded by the astonishing intra- andinterpopulational variation in coloration(e.g., Myers and Daly, 1983). Only pumiliois not represented by singletons in the clado-gram, and, as such, the monophyly of thosespecies was not tested. Nevertheless, consid-eration of patristic and pairwise (table 33)distance supports the historical reality ofthese species.

The sister of the histrionicus group isequivalent to Silverstone’s (1975a) tinctoriusgroup, with the exclusion of galactonotus (seeabove). This clade is individuated by 92unambiguously optmized synapomorphies.Hoogmoed (1969) described azureus fromVier Gebroeders Mountain in southernSipaliwini, near the Brazilian border. Itsresemblance to tinctorius was noted in theoriginal description, and Silverstone (1975a)considered it to be closely related to andpotentially derived from that species. Theextensive variation in tinctorius discoveredsubsequently has only strengthened the sus-picion that these two nominal taxa areconspecific. The two samples of azureus wereobtained from the region of the type localityin Suriname (1330) and in adjacent Brazil(534). One of the tinctorius samples is alsofrom near the Tafelberg airstrip, Sipaliwini,Suriname (1327), and the other is fromBrazil.

The cladogram indicates that tinctorius isparaphyletic with respect to azureus. Fur-thermore, as shown in the pairwise compar-isons (table 34), the two azureus samples areidentical and differ from the Braziliantinctorius sample in only a single nucleotide(0.3%). The pairwise distance between theBrazil and Suriname tinctorius is greater thanthat between it and azureus. All of this isconsistent with the hypothesis that thesesamples are conspecific.

Despite the considerable variation in colorand color pattern in auratus, there are noknown problems surrounding the identitiesof auratus and truncatus (see table 35).Silverstone (1975a) hypothesized that thesetwo species are closely related, and the

TABLE 32Percent Uncorrected Pairwise Distances BetweenCytochrome b Sequences of histrionicus, lehmanni

Myers and Daly, and sylvaticus Funkhousera

Sample ID 1 2 3 4

1 histrionicus 336 Vicordo —

2 histrionicus 498 Mecana 0.3 —

3 lehmanni 338 5.2 4.9 —

4 sylvaticus 364 5.2 4.9 2.9 —



Cytochrome b Sequences of arboreus, pumilio,speciosus, and vicenteia

Sample ID 1 3 4 5 6

1 arboreus 340 —

3 pumilio 367 5.7 —

4 pumilio 1313 4.2 5.5 —

5 speciosus 341 5.5 3.6 4.4 —

6 vicentei 1148 4.9 4.7 3.9 3.6 —



available evidence corroborates that claimwith a total of 84 unambiguously optimizedsynapomorphies. Samples of auratus arefrom two localities in Bocas del Toro,Panama (327, 334, 335) and one in Nicaragua(365). One truncatus sample was captiveraised (1151); the other two were taken inwestern Colombia.

SUMMARY OF RELATIONSHIPS

AMONG DENDROBATIDS

This study resolved the relationshipsamong most of the included terminals.Results are generally consistent with priorfindings, especially the species groups pro-posed by Silverstone (1975a, 1976). At thelevel of genera, Allobates (whether restrictedto the femoralis group or applied moregenerally; see below), Ameerega, Dendrobates(including or excluding Minyobates, Oo-phaga, and Ranitomeya), Epipedobates, Man-nophryne, Oophaga, Phyllobates, and Ranito-meya were all found to be monophyletic.Nephelobates was found to be paraphyleticwith respect to Aromobates nocturnus andColostethus saltuensis. As expected, the great-

est incongruence between generic groupingand phylogeny involves Colostethus, whichwas shown to be vastly nonmonophyletic.Nevertheless, the density of taxon samplingallowed coherent clades to be delimited,which will permit a monophyletic taxonomyto be developed below.

In addition to resolving the relationshipsamong species, this study sheds light on theidentities of numerous problematic species.Comparison with topotypic material is re-quired to determine which species are new,and in several cases the lack of locality datafor the sequences reported by Santos et al.(2003) makes it difficult to assess speciesidentity (especially in relation to bocagei).Nevertheless, consideration of cladistic andpatristic distances suggests the 367 dendro-batid terminals included in this analysisrepresent 156 species. Available evidenceclearly delimits two distantly related speciesconflated under the name elachyhistus, cor-roborating Duellman’s (2004) suspicions.The widespread Amazonian taxon ventrima-culatus is paraphyletic with respect to ama-zonicus, duellmani, fantasticus, reticulatus,variabilis, and an unidentified species re-ported by Santos et al. (2003), which, incombination with patterns of patristic dis-tances for all data and pairwise distances forcytochrome b sequences, suggests this taxonincludes five species. Although femoralis,fraterdanieli, hahneli, and nexipus were allrecovered as monophyletic, pairwise cyto-chrome b distances and patristic distances forall data suggest they may include multiplespecies (femoralis: 8 spp.; fraterdanieli: 2 spp.;hahneli: 2 spp.; nexipus: 2 spp.). Grant andRodrıguez (2001) speculated that trilineatus


Cytochrome b Sequences of azureus and tinctoriusa

Sample ID 1 2 3 4

1 azureus 1330 —

2 azureus 534 0.0 —

3 tinctorius 1327 2.6 2.6 —

4 tinctorius 535 0.3 0.3 2.3 —


TABLE 35Percent Uncorrected Pairwise Distances Between Cytochrome b Sequences of auratus and truncatusa

Sample ID 1 2 3 4 5 6 7

1 auratus 327 Panama —

2 auratus 334 Panama 0.0 —

3 auratus 335 Panama 0.0 0.0 —

4 auratus 365 Nicaragua 0.0 0.0 0.0 —

5 truncatus 1151 (CR) 2.3 2.3 2.3 2.3 —

6 truncatus 1351 3.1 3.1 3.1 3.1 1.3 —

7 truncatus 1352 3.1 3.1 3.1 3.1 1.3 0.0 —

a CR 5 captive reared. Gray lines separate species and localities.


may be a complex of species, which is borneout by the current analysis; we identifyprovisionally eight species in the trilineatusclade. Similarly, although the diversity ofsmall, cryptically colored Amazonian frogs isgreater than the current taxonomy indicates,and the names proposed by Morales (2002‘‘2000’’) are available to associate with severalof these species, Morales’s taxonomy isdifficult to apply because it treated somespecimens of distantly related species asconspecific and some conspecific specimensas different taxa. Although we have referredpopulations to Morales’s names, this is pro-visional and topotypic material must beexamined to clarify identities. Specimens fromGuyana that are mophologically indistin-guishable from degranvillei are not conspecificwith a sample from near the type locality inFrench Guiana. Among controversial species,our results confirm the independence ofbaeobatrachus and stepheni. Our evidence alsosuggests that azureus is nested within, andconspecific with, tinctorius.

As a quick heuristic to help identifyspecies, pairwise comparisons of cytochromeb sequences proved useful, but they are nota panacea. Focusing on well-delimited, un-controversial species, intraspecific cyto-chrome b sequence distances ranged from0.0 to 6.0%. The greatest intraspecific dis-tances were between Nicaraguan and Pana-manian samples of lugubris (6.0%) andtalamancae (5.7%). The localities for thesepairs of samples are also separated by largegeographic distance, but the cytochromeb sequences of auratus samples from Nicar-agua and Panama are identical. Similarly, weexpected evidence to indicate that the wide-spread and phenotypically variable Amazo-nian species trivittatus is composed of multi-ple species, as was found in femoralis;however, trivittatus DNA sequences arerelatively homogeneous across its distribu-tion, suggesting the existence of a singlespecies. Minimally, these results highlight thepitfalls of generalizing across taxa andsuggest caution in interpeting pairwise com-parisons alone.

Among closely related species of unprob-lematic identity, the least interspecific cyto-chrome b distance is 2.3% and 3.9% in theauratus–truncatus and vicentei–pumilio pairs,

respectively. Among putative sister-speciespairs, the greatest cytochrome b distance is18.6% between castaneoticus and galactono-tus. Given the degree of morphological di-vergence between these species, this is un-surprising. However, it is only slightly greaterthan that observed between the morpholog-ically more similar (but more distantly re-lated) castaneoticus and quinquivittatus(17.4%). As mentioned above, the CentralAmerican species flotator and nubicola wereconsidered conspecific until recently (Ibanezand Smith, 1995), yet they are not each other’sclosest relatives and their pairwise cyto-chrome b distance is 18.4%. Whether thesedifferences in pairwise distances betweenclosely related species are due to incompletetaxon sampling (i.e., they are not as closelyrelated as they were presumed to be) orvariation in evolutionary rates is unknown.

A MONOPHYLETIC TAXONOMY

PRELIMINARY CONSIDERATIONS

Evolutionary relationships are the explan-atory framework that unifies all areas ofbiology, and the results of the present studyprovide a coherent foundation to understandthe many fascinating and useful characteris-tics of dendrobatid frogs. To facilitate un-derstanding and application of the phyloge-netic results, we propose a revised taxonomyfor dendrobatid frogs that reflects as closelyas is presently feasible (see below) currentknowledge of phylogeny (figs. 77, 78).

Our adherence to Linnaean nomenclatureand the strictures of the Code (ICZN, 1999)is pragmatic and not intended as a completeendorsement. The imposition of Linnaeanranks is arbitrary and artificial, skewing boththought and analysis as they continue to betreated as identifying objectively equivalententities, despite pleas to the contrary. Ifscientific language is to accurately reflectunderstanding of evolutionary relationships,then it is clear that sooner or later Linnaeannomenclature will have to be abandoned ortransformed significantly.

The best known alternative is the Phylo-Code (e.g., de Queiroz and Gauthier, 1990),which eliminates ranks. However, the Phylo-Code also institutes a number of conventions


that would, should they be adopted, surelyimpede scientific progress, for example, byincreasing the frequency with which minorchanges in topology would lead to extremechanges in taxonomy (i.e., nomenclaturalinstability). Consider, for example, that thefinding of Darst and Cannatella (2004; seealso Faivovich et al., 2005; Frost et al., 2006)that hemiphractines are distant relatives ofother hylids makes Ford and Cannatella’s(1993) node-based definition of Hylidae applyto all hyloids except brachycephalids (parsi-mony) or all hyloids (maximum likelihood).

Kluge (2005) recently proposed a novelsystem to represent phylogeny exactly andeliminate the drawbacks of the Linnaeansystem without abandoning its strengths(e.g., designation of ‘‘types’’ for bookkeepingpurposes, the principle of priority to encour-age progress), all or much of which is likelyto be implemented (if not endorsed explicitly)simply because it is designed expressly toencourage scientific progress. Indeed, some

aspects of his proposal, such as the naming ofall clades, may be inevitable by-products ofthe growth of scientific knowledge, whetherthe Code is overhauled or not (e.g., by simplyshifting ranked names toward the tips, thuspushing the bulk of cladistic structure abovethe family level where the Code does notapply). Nevertheless, Kluge’s proposal hasnot yet been vetted by the scientific commu-nity, and for the immediate need to translatethe phylogeny of dendrobatids into a mono-phyletic taxonomy we continue to apply theexisting Code.

Over the past four decades the number ofrecognized dendrobatid species has explodedfrom 70 to 247, and there is no indicationthat discovery of new species in this clade willwane in the foreseeable future. Compared toother vertebrate groups, anuran families arelarge and cumbersome. Consider, for exam-ple, that Frost et al.’s (2006) new taxonomyrecognizes only 42 families for approximately5,000 species of anurans—prior to the Frost

Fig. 77. Graphic summary of the proposed higher-level taxonomy of Athesphatanura.


Fig. 78. Graphic summary of the proposed taxonomy of Dendrobatoidea.


et al. update there were only 33 recognizedfamilies of anurans. In comparison, currenttaxonomy recognizes approximately 220families of birds to accommodate roughly10,000 species, 500 families of fishes for28,000 species, and 130 families for roughlythe same number of mammals as there areanurans. Indeed, in all of these groups theorders are approximately equivalent to thefamilies in anuran nomenclature (e.g., thereare 26 recognized orders of mammals).

This recognition of few families for frogs isnot due to an active decision by theherpetological community but rather tradi-tion and the fact that, as exemplified bydendrobatids, much of the diversity of frogshas been discovered so recently and rapidly(over 45% of recognized amphibian specieshave been named since 1985; D. R. Frost,unpublished data) without any major re-vamping of the higher-level taxonomy. Thisis understandable, given that monophyly ismore important than the arbitrary ranking ofclades, and by that argument there is no needto elevate the rank of the dendrobatid clade.However, the retention of the old family unitsalso results in an underappreciation of di-versity and actually obscures patterns ofdiversification. Insofar as the purpose ofnaming clades is to facilitate further research,Linnaean ranks, artificial as they are, area useful means of carving off chunks ofdiversity for scientific discussion (if this werenot the case, then the optimal Linnaeansolution to nonmonophyly would always beaccretion, with recognized taxa growing everlarger and obscuring more phylogeneticstructure as knowledge increases), and in thissense anuran taxonomy is much less refinedthan in other vertebrate groups. We thereforeelevate the rank of the dendrobatid clade tosuperfamily (Dendrobatoidea) and proposea new arrangement of families, subfamilies,and genera to better reflect the diversity andphylogeny of this clade.

For taxonomic purposes, we have exam-ined specimens of all but a few species ofdendrobatids, but available material of manyspecies was not adequate to permit theirinclusion in the phylogenetic analysis (generaand species included are listed below inboldface). We therefore refer them to generaprovisionally as both an efficient means of

summarizing what is known about thosespecies and as explicit phylogenetic hypothe-ses to be tested in future studies. To permitprovisional reference of species that were notincluded, we fit names to the cladogramsomewhat loosely, that is, names refer todemonstrably monophyletic groups, butmuch of the finer cladistic structure remainsunnamed. This was done as a working com-promise between two extreme alternatives.

The two alternatives are (1) to maintainthe status quo until knowledge is ‘‘sufficientlycomplete’’ to merit taxonomic revision byallowing all species to be placed withcertainty, or (2) to propose a new taxonomyfor the species included in this analysis andtreat all others as incertae sedis. Alternative(1) is tantamount to a plea for ignorance andis antiscientific. There is no objective basisfor determining when any system of scientificknowledge is ‘‘sufficiently complete’’ for anypurpose. It is a fundamental characteristic ofscience that future evidence (or discoveryoperations) may overturn any prior hypoth-esis, and rejecting current knowledge simplybecause it may ultimately be wrong wouldprevent all progress. Alternative (2) is equallyunsatisfactory because it effectively hides theevidence that already exists regarding therelationships of those species. New taxo-nomies build upon prior ones, and thoseprior ones had some empirical basis, howeverlimited. Finally, provisional placement facil-itates content increasing progressive problemshifts (sensu Lakatos, 1978) by increasing thetestability of phylogenetic hypotheses (logi-cally, the more species included in thehypotheses, the greater the potential to falsifyit) and, further, by facilitating alpha taxon-omy and the discovery of new species. Forexample, in the current system, a new speciesof Colostethus should, in principle, be com-pared to ca. 120 species ranging fromNicaragua to southeastern Brazil and Boli-via. Most taxonomists are regional specialistsand lack the resources to undertake suchcomparisons, which frequently leads to ex-tensive errors by either referring to differentspecies under the same name or namingspecies that are not diagnosable in a broadercontext. A taxonomy that reflects currentknowledge of phylogeny will point to appro-


priate comparisons and thereby greatly facil-itate species-level work.

As noted above (Materials and Methods),we included genotypic and phenotypic datafor 13 type species (genus name in parenthe-ses): azureiventris (Cryptophyllobates), bicolor(Phyllobates), femoralis (Allobates), inguinalis(Prostherapis), nocturnus (Aromobates), pul-chellus (Phyllodromus), pumilio (Oophaga),reticulatus (Ranitomeya), silverstonei (Phobo-bates), steyermarki (Minyobates), tinctorius(Dendrobates), tricolor (Epipedobates), andtrivittatus (Ameerega). We did not include thetype species alboguttatus (Nephelobates), fuli-ginosus (Hyloxalus), latinasus (Colostethus),or yustizi (Mannophryne) because adequatedata were not available. Nevertheless, weincluded numerous putatively closely relatedspecies and made taxonomic changes accord-ingly. That is, we treated the sampled speciesas proxies for the type species in the sameway that the sampled species were treated asrepresentative of the complete diversity ofdendrobatids. In both cases, further samplingmay prove these assumptions to be false, butin the meantime it is better to presenta taxonomy derived from a hypothesis ofrelationships supported by evidence that canform the basis for future testing than toretain the current taxonomy that misrepre-sents current understanding of phylogeny. Inthe following accounts, species included inthe phylogenetic analysis are in bold.

Unless otherwise noted, only named spe-cies are listed in the following accounts.Additional undescribed species of severaltaxa were included in the phylogeneticanalysis, but these are discussed only in theComment sections. For each named clade wereport a summary of the unambiguoustransformations (including phenotypic syna-pomorphies and branch length) and Bremersupport. Additionally, generic acounts in-clude a standardized diagnosis designed toallow species to be referred to taxa efficientlyfollowing the examination of few, conspicu-ous, and, insofar as is possible, easilyaccessible characters, as well as generalitiesthat are taxonomically useful but difficult toindividuate as hypotheses of homology. Thepurpose of these general characterizations isto facilitate rapid identification, and, as such,descriptions are much less precise than in the

delimitation and analysis of transformationseries. Unambiguous molecular transforma-tions for each named group are given inappendix 8.

THE HIGHER-LEVEL TAXONOMY OF

ATHESPHATANURA FROST ET AL., 2006

As noted above, the higher-level relation-ships within Athesphatanura differ from theFrost et al. (2006) hypothesis in two majorways: (1) Leptodactylidae is divided into twodistantly related groups, neither of which isthe sister of Centrolenidae (i.e., Diphyaba-trachia is refuted); one is placed as the sisterto all Leptodactyliformes except Centroleni-dae, and the other is placed inside Hestico-batrachia as the sister to bufonids, dendro-batids, and hylodines. (2) Thoropa (andtherefore Thoropidae) move from being thesister of dendrobatids to be nested amongcycloramphids, while the hylodines movefrom being placed with cycloramphids tobeing the sister of dendrobatoids. Whereregulated taxa were concerned, we appliedthe Code. However, there are different waysin which changes to unregulated taxa can beincorporated, with the opposing strategiesfocusing on content (i.e., any change incontent entails a different phylogenetic hy-pothesis and, therefore, demands a differentname) and nomenclatural stability (i.e.,changes in content entail only a reformulationof the existing name and do not necessitatea new name). There are merits to bothapproaches, and we devised a taxonomy thatintroduces the fewest new names required toaccomodate the new phylogenetic structureand reformulates the remaining groups.Specifically, we propose new names for theadditional nodes created by the movement ofthe leptodactylids, and we reformulateChthonobatrachia and Hesticobatrachia(which are otherswise unchanged) to includeLeiuperidae, and Agastorophrynia to includeHylodidae and exclude Thoropa.

The higher-level taxonomy of Athesphata-nura is summarized in table 36. The para-phyly of Hylidae in our analysis owes tohaving rooted on the hylid species Hypsiboasboans. Insofar as our results do not otherwisediffer from Frost et al. (2006) with regard to


Hylidae or Leptodactyliformes, we omitthose taxa from the following accounts.

FAMILY: CENTROLENIDAE TAYLOR, 1951

Centrolenidae Taylor, 1951. Type genus: Centro-lene Jimenez de la Espada, 1872.

Allophrynidae Goin et al., 1978: 240. Type genus:Allophryne Gaige, 1926.

IMMEDIATELY MORE INCLUSIVE TAX-

ON: Leptodactyliformes Frost et al., 2006.SISTER GROUP: Cruciabatrachia new tax-

on.CONTENT (4 GENERA): Allophryne Gaige,

1926; ‘‘Centrolene’’ Jimenez de la Espada,1872; Cochranella Taylor, 1951; Hyalino-batrachium Ruiz-Carranza and Lynch, 1991.

CHARACTERIZATION, DIAGNOSIS, AND

SUPPORT: Branch length 5 39. Bremersupport 5 15. All unambiguously optimizedsynapomorphies for this clade are from DNAsequences (see appendix 8). See Frost et al.(2006) for discussion of synapomorphies.

DISTRIBUTION: Tropical southern Mex-ico to Bolivia, northeastern Argentina, andsoutheastern Brazil.

COMMENT: Our optimal hypothesis ofcentrolenid relationships is identical to thatof Frost (2006), and we direct the reader tothat paper for comments. Our results differ inthe placement of Centrolenidae relative to

other hyloid lineages. Frost et al. (2006)found Centrolenidae to be grouped withLeptodactylidae, together forming Diphya-batrachia. We did not find support forDiphyabatrachia in the present analysis,and instead found Centrolenidae to be thesister group of a large, predominantly Neo-tropical radiation, named below.

UNRANKED TAXON: CRUCIABATRACHIA

NEW TAXON


ON: Leptodactyliformes Frost et al., 2006.

SISTER GROUP: Centrolenidae Taylor,1951.

CONTENT: Leptodactylidae Werner, 1896(1838); Chthonobatrachia Frost et al., 2006.


SUPPORT: Branch length 5 21. Bremersupport 5 14. All unambiguously optimizedsynapomorphies for this clade are from DNAsequences (see appendix 8).

DISTRIBUTION: Cosmopolitan in temper-ate and tropical areas (but with most speciesconcentrated in the southern hemisphere)except for the Australo-Papuan region, Ma-dagascar, Seychelles, and New Zealand.

ETYMOLOGY: Cruciabatrachia, frogs ofthe Southern Cross (Crux), referencing thepredominantly southern range of this clade.

COMMENT: Following Frost et al.(2006), we include Somuncuria in this groupprovisionally on the basis of the evidencesuggested by Lynch (1978), who placedSomuncuria as the sister taxon of Pleuro-dema.

FAMILY: LEPTODACTYLIDAE WERNER,

1896 (1838)

Cystignathi Tschudi, 1838. Type genus: Cy-

stignathus Wagler, 1830.

Plectromantidae Mivart, 1869, Proc. Zool. Soc.

London, 1869: 291. Type genus: Plectromantis

Peters, 1862.

Adenomeridae Hoffmann, 1878, In Bronn (ed.).

Type genus: Adenomera Steindachner, 1867.

Leptodactylidae Werner, 1896. Type genus: Lep-

todactylus Fitzinger, 1826.


ON: Cruciabatrachia new taxon.

SISTER GROUP: Agastorophrynia Frostet al., 2006.

TABLE 36The Higher-Level Taxonomy of

Athesphatanura Frost et al., 2006

Hylidae Rafinesque, 1815

Leptodactyliformes Frost et al., 2006

Centrolenidae Taylor, 1951

Cruciabatrachia new taxon

Leptodactylidae Werner, 1896 (1838)

Chthonobatrachia Frost et al., 2006

Ceratophryidae Tschudi, 1838

Batrachylinae Gallardgo, 1965

Ceratophryninae Tschudi, 1838

Telmatobiinae Fitzinger, 1843

Hesticobatrachia Frost et al., 2006

Cycloramphidae Bonaparte, 1850

Calamitophrynia new taxon

Leiuperidae Bonaparte, 1850

Agastorophrynia Frost et al., 2006

Bufonidae Gray, 1825

Nobleobatia new taxon

Hylodidae Gunther, 1858

Dendrobatoidea Cope, 1865


CONTENT (4 GENERA): HydrolaetareGallardo, 1963; Leptodactylus Fitzinger,1826; Paratelmatobius Lutz and Carvalho,1958; Scythrophrys Lynch, 1971



DISTRIBUTION: Tropical Mexico through-out Central and South America.

COMMENT: Following Frost et al. (2006),we place Hydrolaetare in this group becauseof its presumed close relationship to Lepto-dactylus (Heyer, 1970), although we suggestthat this proposition merits further study.

UNRANKED TAXON: CHTHONOBATRACHIA

FROST ET AL., 2006


ON: Cruciabatrachia new taxon.SISTER GROUP: Leptodactylidae Werner,

1896 (1838).CONTENT: Ceratophryidae Tschudi,

1838; Hesticobatrachia Frost et al., 2006.CHARACTERIZATION, DIAGNOSIS, AND


DISTRIBUTION: Cosmopolitan in temper-ate and tropical areas except for the Aus-tralo-Papuan region, Madagascar, Sey-chelles, and New Zealand.

COMMENT: Our Chthonobatrachia isequivalent to the group proposed by Frostet al. (2006) with the inclusion of LeiuperidaeBonaparte, 1850, which was previously in thesynonymy of Leptodactylidae. See Frost etal. (2006) for further comments.

FAMILY: CERATOPHRYIDAE TSCHUDI, 1838

Ceratophrydes Tschudi, 1838: 26. Type genus:Ceratophrys Wied-Neuwied, 1824.


ON: Chthonobatrachia Frost et al., 2006.SISTER GROUP: Hesticobatrachia Frost

et al., 2006.CONTENT (3 SUBFAMILIES): Batrachyli-

nae Gallardo, 1965; CeratophryninaeTschudi, 1838; Telmatobiinae Fitzinger, 1843.


SUPPORT: Branch length 5 30. Bremersupport 5 19. No phenotypic character-states optimize unambiguously to this node(see appendix 8 for diagostic DNA sequencetransformations). See Frost et al. (2006) fordiscussion of additional synapomorphies.

DISTRIBUTION: Andean and tropicallowland South America from Colombia andVenezuela south to extreme southern Argen-tina and Chile.

COMMENT: Frost et al. (2006) recognizedtwo subfamilies within Ceratophryidae: Tel-matobiinae (for Telmatobius) and Cerato-phryinae (for the tribes Batrachylini andCeratophryini). Our results show Cerato-phryinae to be paraphyletic with respect toTelmatobiinae. Rather than dissolve all ofthese groups in the synonymy of Cerato-phryidae, below we elevate Batrachylini andCeratophryini of Frost et al. (2006) tosubfamilies.

SUBFAMILY: BATRACHYLINAE

GALLARDO, 1965

Batrachylinae Gallardo, 1965: 83. Type genus:Batrachylus Bell, 1843.


ON: Ceratophryidae Tschudi, 1838.

SISTER GROUP: Unnamed group com-posed of Ceratophryinae Tschudi, 1838;Telmatobiinae Fitzinger, 1843.

CONTENT (2 GENERA): AtelognathusLynch, 1978 and Batrachyla Bell, 1843.



DISTRIBUTION: Central to extremesouthern Argentina and Chile (Patagonia).

COMMENT: This taxon is equal to Ba-trachylini of Frost et al. (2006).

SUBFAMILY: CERATOPHRYINAE TSCHUDI, 1838

Ceratophrydes Tschudi, 1838: 26. Type genus:Ceratophrys Wied-Neuwied, 1824; StombinaeGallardo, 1965: 82. Type genus: Stombus Graven-horst, 1825.




SISTER GROUP: Telmatobiinae Fitzinger,1843.

CONTENT (3 GENERA): CeratophrysWied-Neuwied, 1824; Chacophrys Reig andLimeses, 1963; Lepidobatrachus Budgett,1899.



DISTRIBUTION: Andean and tropicallowlands of South America from Colombiaand Venezuela south to extreme southernArgentina and Chile.

COMMENT: This taxon is equal to Cer-atophryini of Frost et al. (2006).

SUBFAMILY: TELMATOBIINAE FITZINGER, 1843

Telmatobii Fitzinger, 1843: 31. Type genus:Telmatobius Wiegmann, 1834.



SISTER GROUP: Ceratophryinae Tschudi,1838.

CONTENT (1 GENUS): Telmatobius Wieg-mann, 1834.



DISTRIBUTION: Andean South America,Ecuador to Chile and Argentina.

COMMENT: This taxon is equal to Tel-matobiinae of Frost et al. (2006).

UNRANKED TAXON: HESTICOBATRACHIA

FROST ET AL., 2006


ON: Chthonobatrachia Frost et al., 2006.

SISTER GROUP: Ceratophryidae Tschudi,1838.

CONTENT: Cycloramphidae Bonaparte,1850; Calamitophrynia new taxon.

CHARACTERIZATION, DIAGNOSIS, AND SUP-

PORT: Branch length 5 26. Bremer support5 14. All unambiguously optimized syna-pomorphies for this clade are from DNAsequences (see appendix 8). See Frost et al.(2006) for discussion of synapomorphies.


COMMENT: Our Hesticobatrachia isequivalent to the group proposed by Frostet al. (2006) with the inclusion of LeiuperidaeBonaparte, 1850, which was previously in thesynonymy of Leptodactylidae. See Frost etal. (2006) for further comments.

FAMILY: CYCLORAMPHIDAE BONAPARTE, 1850

Cyclorhamphina Bonaparte, 1850. Type genus:Cycloramphus Tschudi, 1838.

Rhinodermina Bonaparte, 1850. Type genus:Rhinoderma Dumeril and Bibron, 1841.

Alsodina Mivart, 1869. Type genus: Alsodes Bell,1843.

Grypiscina Mivart, 1869. Type genus: Grypiscus

Cope, 1867 ‘‘1866’’.

Odontophrynini Lynch, 1969. Type genus: Odon-

tophrynus Reinhardt and Lutken, 1862 ‘‘1861’’.(Odontophrynini subsequently named moreformally by Lynch, 1971: 142.)

Thoropidae Frost et al., 2006. Type genus:Thoropa Cope, 1865.


ON: Hesticobatrachia Frost et al., 2006.

SISTER GROUP: Calamitophrynia newtaxon.

DISTRIBUTION: Southern tropical andtemperate South America.

CONTENT (12 GENERA): Alsodes Bell,1843; Crossodactylodes Cochran, 1938; Cy-cloramphus Tschudi, 1838; Eupsophus Fitzin-ger, 1843; Hylorina Bell, 1843; LimnomedusaFitzinger, 1843; Macrogenioglottus Carvalho,1946; Odontophrynus Reinhardt and Lutken,1862 ‘‘1861’’; Proceratophrys Miranda-Ri-beiro, 1920; Rhinoderma Dumeril and Bi-bron, 1841; Thoropa Cope, 1865; ZachaenusCope, 1866.




COMMENT: Cycloramphidae, as definedhere, differs from Cycloramphidae sensuFrost et al. (2006) in that it excludes


Hylodinae (see below) and includes Thoropa(and therefore Thoropidae). We do notrecognize subfamilial divisions within Cyclo-ramphidae. See Frost et al. (2006) foradditional discussion.

UNRANKED TAXON: CALAMITOPHRYNIA

NEW TAXON


ON: Hesticobatrachia Frost et al., 2006.SISTER GROUP: Cycloramphidae Bona-

parte, 1850.CONTENT: Leiuperidae Bonaparte, 1850;

Agastorophrynia Frost et al., 2006.CHARACTERIZATION, DIAGNOSIS, AND



ETYMOLOGY: Calamitophrynia, fromthe Greek calamitas (misfortune), phryne(toad), and suffix ia (having the nature of),referencing the loud calls of so many of thespecies in this clade.

COMMENT: The placement of Leiuperi-dae at such cladistic distance from Lepto-dactylidae was unexpected and contradictsthe findings of Frost et al. (2006).

FAMILY: LEIUPERIDAE BONAPARTE, 1850

Leiuperina Bonaparte, 1850. Type genus: Leiu-perus Dumeril and Bibron, 1841.

Paludicolina Mivart, 1869. Type genus: PaludicolaWagler, 1830.

Pseudopaludicolinae Gallardo, 1965. Type genus:Pseudopaludicola Miranda-Ribeiro, 1926.


ON: Calamitophrynia new taxon.SISTER GROUP: Agastorophrynia Frost

et al., 2006.CONTENT (7 GENERA): Edalorhina Jime-

nez de la Espada, 1871 ‘‘1870’’; EngystomopsJimenezde la Espada, 1872; EupemphixSteindachner, 1863; Physalaemus Fitzinger,1826; Pleurodema Tschudi, 1838; Pseudopa-ludicola Miranda-Ribeiro, 1926; SomuncuriaLynch, 1978.


SUPPORT: Branch length 5 36. Bremer

support 5 23. All unambiguously optimizedsynapomorphies for this clade are from DNAsequences (see appendix 8).

DISTRIBUTION: Mexico throughout Cen-tral and South America.

COMMENT: Following Frost et al.(2006), we include Somuncuria in this groupprovisionally on the basis of the evidencesuggested by Lynch (1978), who placedSomuncuria as the sister taxon of Pleuro-dema.

UNRANKED TAXON: AGASTOROPHRYNIA

FROST ET AL., 2006


ON: Hesticobatrachia Frost et al., 2006.

SISTER GROUP: Leiuperidae Bonaparte,1850.

CONTENT: Bufonidae Gray, 1825; No-bleobatia new taxon.


SUPPORT: Branch length 5 22. Bremersupport 5 5. All unambiguously optimizedsynapomorphies for this clade are fromDNA sequences (see appendix 8). Althoughthe origin of diurnal activity (Character 115)is ambiguous due to incomplete coding ofphenotypic characters in the presentanalysis, it is a likely synapomorphy forAgastorophrynia. Melanophryniscus, Dendro-phryniscus, Atelopus, and almost all speciesof Nobleobatia new taxon are diurnal,whereas leiuperids, cycloramphids, cerato-phryids, leptodactylids, centrolenids, andhylids are entirely or predominantly noctur-nal.


COMMENT: Agastorophrynia here isequivalent to that of Frost et al. (2006) withthe inclusion of their Hylodinae, which isremoved from their Cycloramphidae, andthe exclusion of Thoropidae, which isfound to be a junior synonym of Cycloram-phidae.

FAMILY: BUFONIDAE GRAY, 1825

Bufonina Gray, 1825. Type genus: Bufo Laurenti,1768.

Atelopoda Fitzinger, 1843. Type genus: AtelopusDumeril and Bibron, 1841.


Phryniscidae Gunther, 1858. Type genus: Phrynis-

cus Wiegmann, 1834.

Adenomidae Cope, 1861 ‘‘1860’’. Type genus:Adenomus Cope, 1861.

Dendrophryniscina Jimenezde la Espada, 1871‘‘1870’’. Type genus: Dendrophryniscus Jimenezde la Espada, 1871 ‘‘1870’’.

Platosphinae Fejervary, 1917. Type genus: Plato-

sphus d’Isle, 1877 (fossil taxon considered to bein this synonymy because Platosophus 5 Bufo

sensu lato).

Bufavidae Fejervary, 1920. Type genus: Bufavus

Portis, 1885 (fossil taxon considered to be inthis synonymy because Bufavus 5 Bufo sensulato).

Tornierobatidae Miranda-Ribeiro, 1926. Typegenus: Tornierobates Miranda-Ribeiro, 1926.

Nectophrynidae Laurent, 1942. Type genus: Nec-tophryne Buchholz and Peters, 1875.

Stephopaedini Dubois, 1987 ‘‘1985’’. Type genus:Stephopaedes Channing, 1978.


ON: Agastorophrynia Frost et al., 2006.

SISTER GROUP: Nobleobatia new taxon.

CONTENT (47 GENERA): AdenomusCope, 1861 ‘‘1860’’; Altiphrynoides Dubois,1987 ‘‘1986’’; Amietophrynus Frost et al.,2006; Andinophryne Hoogmoed, 1985; Anax-yrus Tschudi, 1845; Ansonia Stoliczka, 1870;Atelophryniscus McCranie, Wilson, and Wil-liams, 1989; Atelopus Dumeril and Bibron,1841; Bufo Laurenti, 1768; Bufoides Pillai andYazdani, 1973; Capensibufo Grandison, 1980;Chaunus Wagler, 1828; Churamiti Channingand Stanley, 2002; Cranopsis Cope, 187‘‘1876’’; Crepidophryne Cope, 1889; Dendro-phryniscus Jimenez de la Espada, 1871‘‘1870’’; Didynamipus Andersson, 1903; Dut-taphrynus Frost et al., 2006; Epidalea Cope,1865; Frostius Cannatella, 1986; Ingerophry-nus Frost et al., 2006; Laurentophryne Tihen,1960; Leptophryne Fitzinger, 1843; Melano-phryniscus Gallardo, 1961; MertensophryneTihen, 1960; Metaphryniscus Senaris, Ayar-zaguena, and Gorzula, 1994; NannophryneGunther, 1870; Nectophryne Buchholz andPeters, 1875; ‘‘Nectophrynoides’’ Noble, 1926;Nimbaphrynoides Dubois, 1987 ‘‘1986’’; Or-eophrynella Boulenger, 1895; OsornophryneRuiz-Carranza and Hernandez-Camacho,1976; Parapelophryne Fei, Ye, and Jiang,2003; Pedostibes Gunther, 1876 ‘‘1875’’;Pelophryne Barbour, 1938; Peltophryne Fit-zinger, 1843; Phrynoidis Fitzinger, 1843;

Pseudobufo Tschudi, 1838; PseudepidaleaFrost et al., 2006; Rhaebo Cope, 1862;Rhamphophryne Trueb, 1971; Rhinella Fitzin-ger, 1826; Schismaderma Smith, 1849; True-bella Graybeal and Cannatella, 1995; Vandij-kophrynus Frost et al., 2006; Werneria Poche,1903; ‘‘Wolterstorffina’’ Mertens, 1939.


SUPPORT: Branch length 5 78. Bremersupport 5 37.

Unambiguously optimized phenotypic sy-napomorphies of this clade are (1) origin ofspiculate skin texture (Character 0, 1 R 3),(2) m. depressor mandibulae origin notextending posterior to the squamosal (Char-acter 73, 1 R 0), (3) loss of m. intermandi-bularis supplementary element (Character 77,1 R 0), (4) origin of median gap in marginalpapillae of lower labium of tadpoles (Char-acter 92, 0 R 1), (5) epicoracoids overlappingfrom level between posterior level of procor-acoids and anterior ends of coracoids toposterior level of coracoids (Character 120, 1R 2), (6) prezonal element of pectoral girdle(omosternum) absent (Character 123, 1 R 0),(7) maxillary teeth absent (Character 139, 1R 0), (8) reduction of chromosome numberto 22 (Character 173, 4 R 2). See Frost et al.(2006) for discussion of additional synapo-morphies.


COMMENT: The internal structure ofBufonidae is identical to that of Frost et al.(2006) for the terminals sampled, as is theplacement of Bufonidae relative to othergroups, the exception being the content ofthe sister group, named below. See Frost etal. (2006) for further comments and structurewithin Bufonidae.

UNRANKED TAXON: NOBLEOBATIA

NEW TAXON


ON: Calamitophrynia new taxon.SISTER GROUP: Bufonidae Gray, 1825.CONTENT: Hylodidae Gunther, 1858;

Dendrobatoidea Cope, 1865.CHARACTERIZATION, DIAGNOSIS, AND



Unambiguously optimized phenotypic sy-napomorphies of this clade are (1) origin ofpaired dorsal digital scutes (Character 1, 0 R1), (2) differentiation of digital discs (Char-acter 6, 0 R 3), (3) reduction of postaxialwebbing of toe II (Character 39, 3 R 1/2), (4)reduction of postaxial webbing of toe III(Character 41, 3 R 2), (5) reduction ofpreaxial webbing of toe IV (Character 42, 3R 2), (6) reduction of postaxial webbing oftoe IV (Character 43, 4 R 1), (7) reduction ofpretaxial webbing of toe V (Character 44, 4 R1), (8) origin of the pale oblique lateral stripe(Character 55, 0 R 1), (9) origin of T-shapedterminal phalanges (Character 118, 1 R 0).

DISTRIBUTION: Most of tropical Centraland South America and Atlantic forest ofBrazil.

ETYMOLOGY: Nobleobatia, formed fromNoble (a surname) and batia (from the Greekbates, a walker). We name this taxon inhonor of G. K. Noble, who, in 1926, was thefirst to propose the immediate relationshipbetween the genera now referred to Hylodi-dae and Dendrobatoidea.

COMMENT: The strongly supportedplacement of dendrobatids and hylodids assister taxa corroborates one of the mostcontroversial hypotheses of anuran relation-ships.

FAMILY: HYLODIDAE GUNTHER, 1858

Hylodinae Gunther, 1858. Type genus: Hylodes

Fitzinger, 1826.

Elosiidae Miranda-Ribeiro, 1923. Type genus:Elosia Tschudi, 1838.


ON: Nobleobatia new taxon.

SISTER GROUP: Dendrobatoidea Cope,1865.


CONTENT (3 GENERA): CrossodactylusDumeril and Bibron, 1841; Hylodes Fitzin-ger, 1826; Megaelosia Miranda-Ribeiro,1923.



Unambiguously optimized phenotypic sy-napomorphies of this clade are (1) origin ofpreaxial fringe on finger II (Character 13, 0 R

1), (2) origin of preaxial fringe on finger III(Character 15, 0 R 1), (3) origin of tarsalfringe (Character 30, 0 R 1), (4) origin ofpreaxial fringe on toe I (Character 36, 0 R 1),(5) origin of fringe on postaxial fringe on toeV (Character 45, 0 R 1), (6) loss of oocytepigmentation (Character 68, 1 R 0), (7) lossof fibers of m. depressor mandibulae origi-nating from the annulus tympanicus (Char-acter 74, 1 R 0), (8) origin of paired lateralvocal sacs (Character 76, 1 R 2), (9) gain oflateral line stitches (Character 98, 0 R 1).

DISTRIBUTION: The Atlantic forest ofBrazil.

COMMENT: Frost et al. (2006) foundthese genera to form a clade nested withinCycloramphidae, which they referred to asHylodinae.

SUPERFAMILY: DENDROBATOIDEA COPE, 1865

Phyllobatae Fitzinger, 1843. Type genus: Phyllo-bates Dumeril and Bibron, 1841.

Eubaphidae Bonaparte, 1850. Type genus: Euba-phus Bonaparte, 1831.

Hysaplesidae Gunther, 1858. Type genus: Hysa-plesia Boie in Schlegel, 1826. (Note that thistaxon was named as Hylaplesidae, derived fromHylaplesia, an incorrent subsequent spelling ofHysaplesia.)

Dendrobatidae Cope, 1865. Type genus: Dendro-bates Wagler, 1830.


ON: Agastorophrynia Frost et al., 2006.

SISTER GROUP: Hylodidae Gunther,1858.

CONTENT (2 FAMILIES): DendrobatidaeCope, 1865 and Aromobatidae new family.



Unambiguous phenotypic transformationsare (1) (Character 2, 0 R 1), (2) gain of thetarsal keel (Character 28, 0 R 1), (3) origin ofthe metatarsal fold (Character 46, 0 R 1) (4)the ‘‘ranid’’ type insertion of the distaltendon of insertion of the m. semitendinosus(Character 69, 0 R 1), (5) gain of the m.semitendinosus binding tendon (Character70, 0 R 1), (6) occurrence of the dorsal flapof the m. depressor mandibulae (Character72, 0 R 1), (7) m. depressor mandibulaeoverlapping posterodorsal portion of tympa-num (Character 75, 0 R 1), (8) orientation of


the m. intermandibularis supplementary ele-ment (Character 78, 0 R 1), (9) loss ofreproductive amplexus (Character 105, 1 R0), (10) dorsal tadpole transport (Character109, 0 R 1), (11) origin of toe trembling(Character 116, 0 R 1), (12) complete fusionof epicoracoids (Character 119, 1/2 R 0), (13)nonoverlapping epicoracoids (Character 120,1 R 0), (14) medial ossification of prezonalelement (omosternum) of pectoral girdle(Character 127, 0 R 1), (15) maxillary teethnonpedicellate (Character 140, 0 R 1), (16)occurrence of the retroarticular process of themandible (Character 142, 0 R 1), and (17)reduction in chromosome number from 26 to24 (Character 173, 4 R 3).

Additional characteristics useful in diag-nosing dendrobatoids are the occurrence ofdorsal scutes on the digital tip, shared onlywith the sister clade Hylodidae (amongNeotropical frogs).

DISTRIBUTION: Dendrobatoid frogs oc-cur throughout large parts of Nicaragua,Costa Rica, Panama, Colombia, Ecuador,Peru, Bolivia, Venezuela, Guyana, Suriname,

French Guiana, Brazil, and the LesserAntilles.

COMMENT: As discussed above, the ele-vation of the dendrobatid clade to superfam-ily is proposed to allow more information onthe phylogeny and biology of the group to beconveyed in the working taxonomy. Tomaintain rank equivalency, Dubois (1992)recognized Dendrobatoidae as an epifamily(redundant with Dendrobatidae) within thesuperfamily Ranoidea. Frost et al. (2006)applied Dendrobatoidea to the clade ofdendrobatids + Thoropa (i.e., Dendrobatidaesensu lato + Thoropidae). In the presentanalysis Thoropa is nested among cycloram-phids, and the sister group of dendrobatids isthe cycloramphid subfamily Hylodinae(Crossodactylus, Hylodes, and Megaelosia).In recognition of the placement of thehylodine genera outside of Cycloramphidae,we recognize them as a family, HylodidaeGunther, 1858 (see above).

Gross examination reveals fused, nonover-lapping epicoracoid cartilages (i.e., firmis-terny in the traditional sense) in dendroba-toids, although histological study hasshown this to differ in one species(Noble, 1926; Kaplan, 1995; see also Kaplan,2004).

THE TAXONOMY OF DENDROBATOIDEA

COPE, 1865

Below we propose a formal taxonomy fordendrobatoid frogs. The structure of thistaxonomy is outlined in table 37, and theplacement of all species is detailed inappendix 1.

FAMILY: AROMOBATIDAE NEW FAMILY

Aromobatidae new family. Type genus: Aromo-

bates Myers, Daly, and Paolillo, 1991.


ON: Dendrobatoidea.

SISTER GROUP: Dendrobatidae.

CONTENT (2 SUBFAMILIES): Anomalo-glossinae new subfamily; Aromobatinae newsubfamily.


SUPPORT: Branch length 5 71. Bremersupport 5 41. No phenotypic character-

TABLE 37The Taxonomy of Dendrobatoidea Cope, 1865

Aromobatidae new family

Anomaloglossinae new subfamily

Anomaloglossus new genus

Rheobates new genus

Aromobatinae new subfamily

Aromobates Myers, Daly, and Paolillo, 1991

Mannophryne La Marca, 1992

Allobatinae new subfamily

Allobates Zimmermann and Zimmermann, 1988

Dendrobatidae Cope, 1865

Colostethinae Cope, 1867

Ameerega Bauer, 1986

Colostethus Cope, 1866

Epipedobates Myers, 1987

Silverstoneia new genus

Hyloxalinae new subfamily

Hyloxalus Jimenez de la Espada, 1871 ‘‘1870’’

Dendrobatinae Cope, 1865

Adelphobates new genus

Dendrobates Wagler, 1830

Minyobates Myers, 1987

Oophaga Bauer, 1988

Phyllobates Dumeril and Bibron, 1841

Ranitomeya Bauer, 1988


states optimize unambiguously to this node(see appendix 8 for diagostic DNA sequencetransformations).

DISTRIBUTION: Central and SouthAmerica and the Lesser Antilles, with mostspecies occurring on the eastern slopes of theAndes, throughout the Amazon region, andin the Atlantic forest of Brazil.

COMMENT: Insofar as is known, allspecies of Aromobatidae lack the ability tosequester alkaloids.

SUBFAMILY: ANOMALOGLOSSINAE

NEW SUBFAMILY

Anomaloglossinae. Type genus: Anomaloglossusnew genus.


ON: Aromobatidae.

SISTER GROUP: Aromobatinae new sub-family.

CONTENT: Anomaloglossus new genusand Rheobates new genus.



Unambiguously optimized phenotypic syn-apomorphies are (1) fringe present on pre-axial surface of finger II (Character 13, 0 R1), (2) fringe present on preaxial surface offinger III (Character 15, 0 R 1), (3) toe disc IImoderately expanded (Character 32, 1 R 2),(4) fringe on preaxial side of toe I present(Character 36, 0 R 1), (5) distal 1.5 phalangesof postaxial side of toe I free of webbing(Character 37, 2 R 3/4), (6) fringe present onpostaxial side of toe V (Character 45, 0 R 1),(7) strong metatarsal fold (Character 46,1 R 2), (8) male abdomen with irregular(clumped) stippling or faint, diffuse spotting(Character 63, 3 R 4).

DISTRIBUTION: Almost exclusively cis-Andean, with most species in eastern Ama-zonia, the Orinoco drainage, and tepuiregions. Three species also occur on thePacific slopes of Colombia and Ecuador.

GENUS: ANOMALOGLOSSUS NEW GENUS

Anomaloglossus new genus. Type species: Colos-tethus beebei Noble, 1923.


ON: Anomaloglossinae new subfamily.SISTER GROUP: Rheobates new genus.

CONTENT (19 SPECIES): Anomaloglossusatopoglossus (Grant, Humphrey, and Myers,1997) new combination; A. ayarzaguenai (LaMarca, 1998 ‘‘1996’’) new combination; A.baeobatrachus (Boistel and Massary, 1999new combination; A. beebei (Noble, 1923) newcombination; A. ‘‘chocoensis’’ auctorum (notof Boulenger, 1912; see Grant et al., 1997); A.degranvillei (Lescure, 1975) new combination;A. guanayensis (La Marca, 1998 ‘‘1996’’) newcombination; A. lacrimosus (Myers, 1991) newcombination; A. murisipanensis (La Marca,1998 ‘‘1996’’) new combination; A. parimae(La Marca, 1998 ‘‘1996’’) new combination;A. parkerae (Meinhardt and Parmelee, 1996)new combination; A. praderioi (La Marca,1998 ‘‘1996’’) new combination; A. roraima(La Marca, 1998 ‘‘1996’’) new combination;A. shrevei (Rivero, 1961) new combination; A.stepheni (Martins, 1989) new combination; A.tamacuarensis (Myers and Donnelly, 1997)new combination; A. tepuyensis (La Marca,1998 ‘‘1996’’) new combination; A. triunfo(Barrio-Amoros, Fuentes, and Rivas, 2004)new combination; A. wothuja (Barrio-Amoros,Fuentes, and Rivas, 2004) new combination.



The only unambiguously optimized phe-notypic synapomorphies of this clade is (1)the unique and unreversed origin of themedian lingual process (Character 79, 0 R 1).

Other characteristics include: (1) Dorsalcoloration cryptic, brown or gray; (2) paleoblique lateral stripe present or absent; (3)pale dorsolateral stripe present or absent; (4)pale ventrolateral stripe absent; (5) dorsalskin texture posteriorly granular; (6) toewebbing basal to extensive; (7) third fingerof adult males swollen or not; (8) finger Ishorter than finger II; (9) finger discs weaklyexpanded; (10) median lingual process pres-ent; (11) larval vent tube usually dextral; (12)larval oral disc shape usually ‘‘normal’’ (notumbelliform), variably reduced in endotro-phic species; (13) larval oral disc emarginate(variably reduced in endotrophic species);(14) lipophilic alkaloids absent; (15) chromo-some number 2n 5 24 (known in Anomalo-glossus stepheni); (16) testes unpigmented ormedially pigmented; (17) dark throat collarabsent.


DISTRIBUTION: Most species are cis-An-dean; among these, none occurs west orsouth of the region of Manaus, and there isa large number of tepui species. Three speciesalso occur on the Pacific slopes of Colombiaand Ecuador.

ETYMOLOGY: Anomaloglossus, formedfrom the Greek anomalos (irregular, unusual)and glossa (tongue), in reference to theunusual tongue bearing the median lingualprocess. Gender masculine. (This name shouldnot be mistaken for Anomaloglossa Percival,1978, which is a genus of brachiopod.)

COMMENT: Anomaloglossus is most sim-ply diagnosed by the synapomorphic occur-rence of the median lingual process (Grantet al., 1997). Owing to the shared occurrenceof the median lingual process (MLP) inthe potential sister taxa specified by theOld World ranoid hypothesis of dendrobatidorigins (e.g., Ford and Cannatella, 1993) andits absence in all hyloids, Grant et al.interpreted the MLP as symplesiomorphicin dendrobatids. However, Frost et al. (2006)showed decisively that dendrobatoids are notclosely related to Old World ranoids, andtheir MLP is independently derived.

La Marca (1998 ‘‘1996’’) did not note thepresence or absence of the MLP in A.ayarzaguenai, A. guanayensis, A. murisipa-nensis, or A. parimae, and we have notexamined these species; as such, their in-clusion in this genus is a prediction based ondistribution and their resemblance to MLP-possessing species, and it must be confirmed.The presence of the MLP is confirmed for allother species referred to this genus.

Within Anomaloglossus there are basicallytwo ‘‘flavors’’ of frogs: small, slender frogswith minimal toe webbing (e.g., A. stepheni),and larger, more robust frogs with moderateto extensive toe webbing (e.g., A. tepuyensis).The former group is strictly cis-Andean,whereas the latter occurs east of the Andesand on the Pacific slopes of Colombia andEcuador. In the present analysis these twogroups are reciprocally monophyletic, butgreater taxon sampling is required to thor-oughly test this result. Similarly, the trans-Andean MLP-possessing species must beincluded explicitly in phylogenetic analysisto corroborate their placement in Anomalo-glossus.

GENUS: RHEOBATES NEW GENUS

Rheobates new genus. Type species: Phyllobates

palmatus Werner, 1899.


ON: Anomaloglossinae new subfamily.

SISTER GROUP: Anomaloglossus new ge-nus.

CONTENT (2 SPECIES): Rheobates palma-tus (Werner, 1899) new combination; R.pseudopalmatus (Rivero and Serna, 2000‘‘1995’’) new combination.


SUPPORT: Branch length 5 28. Bremersupport 5 28.14 Insofar as this genus isrepresented by a single species in thisanalysis, we cannot distinguish betweenautapomorphies and synapomorphies andtherefore do not report the apomorphicstates.

Other characteristics include: (1) Dorsalcoloration cryptic, brown or gray; (2) paleoblique lateral stripe present or absent, oftenmore conspicuous in juveniles; (3) paledorsolateral stripe absent; (4) pale ventrolat-eral stripe absent; (5) dorsal skin textureposteriorly granular; (6) toe webbing exten-sive; (7) third finger of adult males notswollen; (8) finger I shorter than finger II;(9) finger discs weakly expanded; (10) medianlingual process absent; (11) larval vent tubedextral; (12) larval oral disc shape ‘‘normal’’(not umbelliform); (13) larval oral discemarginate; (14) lipophilic alkaloids un-known (presumed absent); (15) chromosomenumber 2n 5 24 (known in Rheobatespalmatus); (16) testes unpigmented; (17) darkthroat collar absent.

DISTRIBUTION: Colombia, eastern amdwesterm slopes of the Cordillera Oriental andacross the Magdalena valley on the easternslope of the Cordillera Central. The eleva-tional distribution extends from ca. 400 m toover 2,000 m.

ETYMOLOGY: Rheobates, from theGreek rheo (stream, current) and bates (awalker) in reference to the riparian habitat ofthe type species R. palmatus.

14The relevance of these values is minimal, giventhat this clade consists only of two specimens of thesame species, but we report them for consistency.


COMMENT: Phylogenetic analysis showedRheobates palmatus to be the sister taxon ofAnomaloglossus, from which it differs moststrikingly in lacking the median lingualprocess. Otherwise, this taxon most resem-bles several extensively webbed species ofHyloxalus, from which it differs in having anelongate, robust zygomatic ramus of thesquamosal and more extensive toe webbing,and Colostethus, from which it differs inlacking the swollen third finger in adultmales. It differs from species of Aromobatesin lacking a pale dorsolateral stripe, and fromspecies of Mannophryne in lacking a darkthroat collar.

We refer Rheobates pseudopalmatus to thisgenus provisionally based on Rivero and

Serna’s (2000 ‘‘1995’’) assertion that R.

palmatus and R. pseudopalmatus are sisterspecies. Nevertheless, we caution that the

diagnostic characters provided by Rivero andSerna are inadequate to validate their claim

and exclude R. pseudopalmatus from Hylox-

alus or Aromobates. That said, the diagnostic

differences given by Rivero and Serna arealso inadequate to distinguish this species

from R. palmatus, and given that the typelocality of Amalfi lies within the known

distribution of R. palmatus on the easternslope of the Cordillera Central, it is likelythat two taxa are conspecific.

Bernal et al. (2005) recently examinedRheobates palmatus on the Amazonian and

western flanks of the Cordillera Oriental andreported bioacoustic and genetic evidence of

lineage differentiation between populationson either side of the Andes. Should sampling

of additional localities corroborate this find-ing, names are available: Phyllobates (Hypo-

dictyon) palmatus Werner, 1899 was named

from Fusagasuga and would therefore applyto the western slope populations, whereas

Hyloxalus granuliventris Boulenger, 1919was described from ‘‘Bogota’’ and, although

that is a vague locality, a first reviser couldapply the name to the eastern slope popula-

tions.

SUBFAMILY: AROMOBATINAE

NEW SUBFAMILY

Aromobatinae new subfamily. Type genus: Aromo-

bates Myers, Paolillo, and Daly, 1991.


ON: Aromobatidae new family.SISTER GROUP: Anomaloglossinae new

subfamily.CONTENT (2 GENERA): Aromobates and

Mannophryne.CHARACTERIZATION, DIAGNOSIS, AND


DISTRIBUTION: Merida Andes of Vene-zuela and adjacent Cordillera Oriental ofColombia, Cordillera de la Costa, andPeninsula de Parıa, Trinidad and Tobago.

COMMENT: Aromobates and Manno-phryne form a morphologically and geo-graphically compact clade.

GENUS: AROMOBATES MYERS, PAOLILLO, AND

DALY, 1991

Aromobates. Type species Aromobates nocturnusMyers, Paolillo, and Daly, 1991 by originaldesignation.

Nephelobates La Marca, 1994. Type species:Phyllobates alboguttatus Boulenger, 1903 byoriginal designation.


ON: Aromobatinae new subfamily.SISTER GROUP: Mannophryne La Marca,

1991.CONTENT (12 SPECIES): Aromobates al-

boguttatus (Boulenger, 1903) new combina-tion; A. capurinensis (Pefaur, 1993) newcombination; A. duranti (Pefaur, 1985) newcombination; A. haydeeae (Rivero, 1978‘‘1976’’) new combination; A. leopardalis(Rivero, 1980 ‘‘1978’’) new combination; A.mayorgai (River, 1980 ‘‘1978’’); A. meridensis(Dole and Durant, 1972) new combination; A.molinarii (La Marca, 1985) new combination;A. nocturnus Myers, Paolillo, and Daly, 1991;A. orostoma (Rivero, 1978 ‘‘1976’’) newcombination; A. saltuensis (Rivero 1980‘‘1978’’) new combination; A. serranus (Pe-faur, 1985) new combination.


SUPPORT: Branch length 5 61. Bremersupport 5 25. All unambiguously optimizedsynapomorphies for this clade are from DNAsequences.

Other characteristics include: (1) Dorsalcoloration cryptic, brown or gray; (2) pale


oblique lateral stripe present or absent; (3)pale dorsolateral stripe present; (4) paleventrolateral stripe absent; (5) dorsal skintexture posteriorly granular; (6) toe webbingbasal to extensive; (7) third finger of adultmales not swollen; (8) finger I shorter than,equal to, or longer than finger II; (9) fingerdiscs weakly to moderately expanded; (10)median lingual process absent; (11) larval venttube dextral; (12) larval oral disc shape‘‘normal’’ (not umbelliform); (13) larval oraldisc emarginate; (14) lipophilic alkaloidsabsent; (15) chromosome number 2n 5 24(known in Aromobates leopardalis); (16) testesunpigmented; (17) dark throat collar absent.

DISTRIBUTION: Merida Andes of Vene-zuela and adjacent Cordillera Oriental ofColombia.

COMMENT: The inclusion of A. capuri-nensis in this genus is provisional because wehave not examined specimens and osteolog-ical data (e.g., length of zygomatic ramus)have not been published. Nevertheless, Pe-faur’s (1993) description called attention tothe resemblance of this species to the otherspecies here included in Aromobates, and itsdistribution at approximately 2,400 m in theMerida Andes lends indirect support to thisrelationship.

GENUS: MANNOPHRYNE LA MARCA, 1991

Mannophryne La Marca, 1992. Type species:Colostethus yustizi La Marca, 1989 by originaldesignation.


ON: Aromobatinae new subfamily.

SISTER GROUP: Aromobates Myers, Pao-lillo, and Daly, 1991.

CONTENT (12 SPECIES): Mannophryne ca-quetio Mijares-Urrutia and Arends R., 1999;M. collaris (Boulenger, 1912); M. cordillerianaLa Marca, ‘‘1994’’ 1995; M. herminae (Boett-ger, 1893); M. lamarcai Mijares-Urrutia andArends R., 1999; M. larandina (Yustiz, 1991);M. neblina (Test, 1956); M. oblitterata (Riv-ero, 1986 ‘‘1984’’); M. olmonae (Hardy, 1983);M. riveroi (Donoso-Barros, 1965 ‘‘1964’’); M.trinitatis (Garman, 1887); M. yustizi (LaMarca, 1989).



Phenotypic synapomorphies that optimizeunambiguously to this node are (1) presenceof a dermal collar (Character 59, 0 R 1) and(2) male abdomen color evenly stippled(Character 63, 3 R 2).

Other characteristics include: (1) Dorsalcoloration cryptic, brown; (2) pale obliquelateral stripe present; (3) pale dorsolateralstripe present; (4) pale ventrolateral stripeabsent; (5) dorsal skin texture posteriorlygranular; (6) toe webbing moderate to exten-sive; (7) third finger of adult males notswollen; (8) finger I shorter than finger II;(9) finger discs narrow to moderately expand-ed; (10) median lingual process absent; (11)larval vent tube dextral; (12) larval oral discshape ‘‘normal’’ (not umbelliform); (13) larvaloral disc emarginate; (14) lipophilic alkaloidsabsent; (15) chromosome number 2n 5 24(known in Mannophryne herminae, M. olmo-nae, M. neblina, M. trinitatis); (16) testesunpigmented; (17) dark throat collar present.

DISTRIBUTION: Andes, Cordillera de laCosta, and Peninsula de Parıa in Venezuela;Trinidad and Tobago.

COMMENT: The content of Mannophrynedoes not change with this study.

SUBFAMILY: ALLOBATINAE NEW SUBFAMILY


ON: Aromobatidae new family.SISTER GROUP: Aromobatinae new sub-

family.CONTENT: Allobates Zimmermann and

Zimmermann, 1988.CHARACTERIZATION, DIAGNOSIS, AND


Unambiguously optimized phenotypic sy-napomorphies of this clade are (1) finger IVreaching distal half of distal subarticulartubercle of finger III (character 4, 0 R 1/2),(2) finger III swollen in adult males (Charac-ter 20, 0 R 1), (3) webbing absent on postaxialside of toe I (Character 37, 2 R 0), (4)webbing absent on preaxial side of toe II(Character 38, 1 R 0), (5) webbing absent onpostaxial side of toe II (Character 39, 1 R 0),(6) webbing absent on preaxial side of toe III(Character 40, 2 R 0), (7) pale paracloacalmark present (Character 49, 0 R 1), (8)oblique lateral line diffuse (Character 57, 0 R2), (9) male abdomen pale, free or almost free


of melanophores (Character 63, 3 R 0), (10)palatines absent (Character 132, 1 R 0).

DISTRIBUTION: South America and theLesser Antilles. South American species arecis-Andean with two exceptions: (1) Allobatestalamancae occurs in the Pacific lowlands ofColombia and Ecuador and north throughCentral America to Nicaragua, (2) its unde-scribed sister species A. ‘‘Magdalena’’ fromthis study, which occurs in the MagdalenaValley of Colombia. A single species (A.chalcopis) occurs in Martinique.

COMMENT: This name is currently re-dundant with Allobates. However, Allobatesis a large, broadly distributed, and heteroge-neous clade. Current knowledge is inade-quate to name additional genera and assignspecies not included explicitly in the presentanalysis, and the need for a functionaltaxonomy outweighs the need to nameadditional clades. Given the rapid accumula-tion of data over the last few years, weanticipate that the paucity of knowledge willbe remedied quickly and this large genus willbe partitioned as knowledge of its phylogenyaccumulates, making Allobatinae an infor-mative name (for recognized species groups,see Comments for Allobates, below). Recog-nition of Allobatinae is necessitated by therecognition of Aromobatinae for the clade ofAromobates and Mannophryne.

GENUS: ALLOBATES ZIMMERMANN AND

ZIMMERMANN, 1988

Allobates Zimmermann and Zimmermann, 1988.

Type species Prostherapis femoralis Boulenger,

1884, by original designation.

IMMEDIATELY MORE INCLUSIVE TAX

ON: Aromobatidae new family.

SISTER GROUP: Aromobatinae new sub-family.

CONTENT (42 SPECIES): Allobates ala-goanus (Bokermann, 1967); A. alessandroi(Grant and Rodrıguez, 2001) new combina-tion; A. bromelicola (Test, 1956) new combi-nation; A. brunneus (Cope, 1887) new combi-nation; A. caeruleodactylus (Lima andCaldwell, 2001) new combination; A. capixaba(Bokermann, 1967) new combination; A.carioca (Bokermann, 1967) new combination;A. cepedai (Morales, 2002 ‘‘2000’’) newcombination; A. chalcopis (Kaiser, Coloma,

and Gray, 1994) new combination; A. con-spicuus (Morales, 2002 ‘‘2000’’) new combi-nation; A. craspedoceps (Duellman, 2004) newcombination; A. crombei (Morales, 2002‘‘2000’’) new combination; A. femoralis (Bou-lenger, 1883); A. fratinescus (Morales, 2002‘‘2000’’) new combination; A. fuscellus (Mor-ales, 2002 ‘‘2000’’) new combination; A.gasconi (Morales, 2002 ‘‘2000’’) new combi-nation; A. goianus (Bokermann, 1975) newcombination; A. humilis (Rivero, 1980‘‘1978’’) new combination; A. insperatus(Morales, 2002 ‘‘2000’’) new combination; A.juanii (Morales, 1994) new combination; A.kingsburyi (Boulenger, 1918) new combina-tion; A. mandelorum (Schmidt, 1932) newcombination; A. marchesianus (Melin, 1941)new combination; A. masniger (Morales, 2002‘‘2000’’) new combination; A. mcdiarmidi(Reynolds and Foster, 1992) new combina-tion; A. melanolaemus (Grant and Rodrıguez,2001) new combination; A. myersi (Pyburn,1981) new combination; A. nidicola (Caldwelland Lima, 2003) new combination; A. olfer-sioides (Lutz, 1925) new combination; A.ornatus (Morales, 2002 ‘‘2000’’) new combi-nation; A. picachos (Ardila-Robayo, Acosta-Galvis, and Coloma, 2000 ‘‘1999’’); A. pittieri(La Marca, Manzanilla, and Mijares-Urru-tia, 2004) new combination; A. ranoides(Boulenger, 1918) new combination; A. rufulus(Gorzula, 1990 ‘‘1988’’); A. sanmartini (Riv-ero, Langone, and Prigioni, 1986) newcombination; A. sumtuosus (Morales, 2002‘‘2000’’) new combination; A. talamancae(Cope, 1875) new combination; A. trilineatus(Boulenger, 1884 ‘‘1883’’) new combination;A. undulatus (Myers and Donnelly, 2001) newcombination; A. vanzolinius (Morales, 2002‘‘2000’’) new combination; A. wayuu (Acosta,Cuentas, and Coloma, 2000 ‘‘1999’’) newcombination; A. zaparo (Silverstone, 1976).


SUPPORT: As for Allobatinae, above.

Other characteristics include: (1) Dorsalcoloration cryptic in most species (brighter inA. femoralis group); (2) pale oblique lateralstripe present in most (but not all) species; (3)pale dorsolateral stripe present or absent; (4)pale ventrolateral stripe present or absent; (5)dorsal skin texture posteriorly granularexcept in A. femoralis group, which isstrongly granular; (6) toe webbing absent to


moderate (basal in most species); (7) thirdfinger of adult males swollen or not swollen;(8) finger I longer than finger II in mostspecies (equal or shorter in some); (9) fingerdiscs weakly expanded; (10) median lingualprocess absent; (11) larval vent tube dextral;(12) larval oral disc shape ‘‘normal’’ (notumbelliform); (13) larval oral disc emargin-ate; (14) lipophilic alkaloids absent; (15)chromosome number 2n 5 24 (known in A.femoralis, A. olfersioides, A. talamancae) and2n 5 22 (known in A. nidicola, A. caeruleo-dactylus, A. chalcopis); (16) testes unpigmen-ted; (17) dark throat collar absent.

DISTRIBUTION: As for Allobatinae,above.

COMMENT: With 42 nominal species,Allobates includes nearly half of the speciespreviously referred to the polyphyletic genusColostethus. Although the monophyly ofAllobates is strongly supported, given thenumber of species and their morphological,genetic (e.g., chromosome numbers), andbehavioral diversity, additional partitioningwill likely be required. Although formalrecognition at this time is premature becauseit would leave the remaining species ina paraphyletic group and there are inade-quate data to refer all species to particularclades, a restricted Allobates may be appliedto the A. femoralis group. This group ispresently composed of only four nominalspecies (A. femoralis, A. myersi, A. zaparo,and A. rufulus—the latter based on minimalevidence), but numerous additional speciesawait description.

FAMILY: DENDROBATIDAE COPE, 1865


ON: Dendrobatoidea Cope, 1865.

SISTER GROUP: Aromobatidae new fam-ily.

CONTENT (3 SUBFAMILIES): Colostethi-nae Cope, 1867; Dendrobatinae Cope, 1865;and Hyloxalinae new subfamily.



Unambiguously optimized phenotypic sy-napomorphies for this clade are (1) webbingon the postaxial side of toe I absent(Character 37, 2 R 0), (2) webbing on the

preaxial side of toe II absent (Character 38, 1/2 R 0), (3) webbing on the postaxial side oftoe II absent (Character 39, 1/2 R 0), (4)webbing on the preaxial side of toe III absent(Character 40, 2/3/4 R 0), and (5) palatinesabsent (Character 132, 1 R 0).

DISTRIBUTION: As for Dendrobatoidea.

COMMENT: For synonymy see Dendro-batoidea, above.

SUBFAMILY: COLOSTETHINAE COPE, 1867

Colostethidae Cope, 1867. Type genus: Colostethus

Cope, 1866 by monotypy.


ON: Dendrobatidae Cope, 1865.

SISTER GROUP: Unnamed clade com-posed of Dendrobatinae Cope, 1865 andHyloxalinae new subfamily.

CONTENT (4 GENERA): AmeeregaBauer, 1986; Colostethus Cope, 1866; Epipe-dobates Myers, 1987; Silverstoneia new genus.



Unambiguously optimized phenotypic sy-napomorphies for this clade are (1) finger IVreaching the distal half of the subarticulartubercle of finger III (Character 4, 0 R 1), (2)finger I longer than finger II (Character 5, 2R 3), (3) finger III swollen in adult males(Character 20, 0 R 1), (4) female crouching incourtship (Character 102, 0 R 1), and (5) gainof cephalic amplexus (Character 105, 0 R 2).

DISTRIBUTION: As for Dendrobatidae.Silverstoneia and Epipedobates are exclusivelytrans-Andean, Colostethus is almost exclu-sively trans-Andean (see below), and Ameer-ega is almost exclusively cis-Andean.

COMMENT: Mivart’s (1869) Calostethinais derived from the subsequent misspelling ofColostethus Cope, 1866 and ColostethidaeCope 1867 as Calostethus and Calostethidae,respectively, and is therefore not an availablename.

GENUS: AMEEREGA BAUER, 1986

Ameerega Bauer, 1986. Type species: Hyla trivit-

tata Spix, 1824 by original designation.

Phobobates Zimmermann and Zimmermann, 1988.

Type species: Dendrobates silverstonei Myers

and Daly, 1979 by original designation.


Paruwrobates Bauer, 1994. Type species: Dendro-bates andinus Myers and Burrowes, 1987 byoriginal designation.

Pseudendrobates Bauer, 1988. Type species: Den-drobates silverstonei by original designation.


ON: Colostethinae Cope, 1867.SISTER GROUP: Colostethus Cope, 1866.CONTENT (25 SPECIES): Ameerega andina

(Myers and Burrowes, 1987) new combina-tion; A. bassleri (Melin, 1941); A. bilinguis(Jungfer, 198915) new combination; A. bolivi-ana (Boulenger, 1902); A. braccata (Stein-dachner, 1864) new combination; A. cainar-achi (Schulte, 1989) new combination; A.erythromos (Vigle and Miyata, 1980) newcombination; A. flavopicta (Lutz, 1925); A.hahneli (Boulenger, 1883) new combination;A. ingeri (Cochran and Goin, 1970) newcombination; A. labialis (Cope, 1874) newcombination; A. macero (Rodrıguez andMyers, 1993) new combination; A. maculataW. Peters, 1873; A. parvula (Boulenger, 1882)new combination; A. peruviridis (Bauer, 1986)new combination; A. petersi (Silverstone,1976) new combination; A. picta (Tschudi,1838); A. planipaleae (Morales and Velazco,1998); A. pongoensis (Schulte, 1999) newcombination; A. pulchripecta (Silverstone,1976); A. rubriventris (Lotters, Debold,Henle, Glaw, and Kneller, 1997) new combi-nation; A. silverstonei (Myers and Daly, 1979)new combination; A. simulans (Myers, Rodri-guez, and Icochea, 2000) new combination; A.smaragdina (Silverstone, 1976); A. trivittata(Spix, 1824).



Unambiguously optimized phenotypic sy-napomorphies for this clade include (1)granular dorsal skin (Character 0, 1 R 2;unreversed, this being the most conspicuoussynapomorphy of this genus), (2) femaleabdomen dark with pale (usually blue)

spotting/reticulation/marbling (Character 64,0 R 3), and (3) the ability to sequesterlipophilic alkaloids (Character 147, 0 R 1).

Other characteristics include: (1) Dorsalcoloration variable (brown, red, bright or-ange, bright metallic green); (2) pale obliquelateral stripe usually present (often incom-plete), absent in A. silverstonei; (3) paledorsolateral stripe absent; (4) pale ventrolat-eral stripe absent or wavy series of elongatespots; (5) dorsal skin texture strongly gran-ular; (6) toe webbing lacking in most species,at most basal; (7) third finger of adult malesswollen in most (but not all) species; (8)finger I equal to finger II in almost all species;(9) finger discs narrow to moderately ex-panded; (10) median lingual process absent;(11) larval vent tube dextral; (12) larval oraldisc shape normal (not umbelliform); (13)larval oral disc emarginate; (14) lipophilicalkaloids present; (15) chromosome number2n 5 24 (known in Ameerega flavopicta, A.hahneli, A. picta, and A. trivittata); (16) testespigmented in most species (unpigmented inA. flavopicta and A. petersi); (17) dark throatcollar absent.

DISTRIBUTION: This clade is almost en-tirely cis-Andean, the sole exceptions beingthe presumed sister species Ameerega andinaand A. erythromos, which occur at low tomoderate elevations of the Pacific Andeanslopes, and A. maculata, known only fromthe Panamanian holotype. Most speciesoccur in lowlands, but some reach as highas ca. 1,400 m.

COMMENT: Ameerega is most easilyidentified by the conspicuously granulardorsal skin texture, consisting of rounded orflattened granules distributed densely andevenly, as was underscored by Jungfer (1989)in his study of the ‘‘red-backed granulated’’species. In most dendrobatoids, includingEpipedobates, granules or tubercles are scat-tered irregularly over the dorsal surfaces,being more distinct and prevalent posteriorly,especially in the sacral region and on thethigh and/or shank, and absent or weakerand sparser anteriorly, and often distinctlyelevated and conical. (For detailed discussionand illustrations see Character 0, above.)Other species that possess strongly granulardorsal skin are Allobates femoralis, A. zaparo,and D. granuliferus.

15In a recent book on amphibian conservation,Amezquita et al. (2004) explicitly placed Epipedobatesbilinguis in the synonymy of Dendrobates ingeri.However, they offered no evidence for this taxonomicchange and did not dispute the differences cited byJungfer (1989) to distinguish the two species. As such,we continue to recognize both taxa as valid species.


In content, Ameerega is equivalent to thecombination of Silverstone’s (1976) pictusand trivittatus groups. Most species pre-viously referred to Epipedobates (sensuMyers, 1987) pertain to this group, that is,it is equivalent to Phyllobates sensu Silver-stone (1975a) following the removal of thebicolor and femoralis groups.

Vigle and Miyata (1980) described Ameer-ega erythromos as part of Silvertone’s (1976)pictus group, and Myers and Burrowes (1987)considered A. andina to be its sister species.There would be little reason to question theinclusion of these species in Ameerega if itwere not for their biogeographically anoma-lous placement west of the Andes, whereasthe remainder of the clade is cis-Andean (butsee also below). The name ParuwrobatesBauer, 1994 is available for these species,should they be found not to be nested withinAmeerega. Ameerega erythromos possessesseveral skin toxins, which suggests it is notclosely related to Hyloxalus azureiventris (seebelow). Ameerega andina egg clutches aredeposited in bromeliads, and presumablytadpoles are transported to phytotelmata,which suggests these species could be part ofDendrobatinae (see below).

Finally, our placement of Ameerega macu-lata in this genus is provisional and deservesfurther investigation. The species was rede-scribed by Myers (1982), who removed itfrom the synonymy of Dendrobates auratuswhere it had been placed by Dunn (1931).Myers (1987) subsequently transferred it tohis Epipedobates. The species remains knownonly from the western Panamanian holotype,which has teeth on the maxillary arch, basalwebbing between toes II–IV, and a long firstfinger, like species of Ameerega. However, italso possesses a spotted dorsum and smoothskin, thus differing from all known species ofAmeerega. As noted by Myers (1982; see alsoPhenotypic Characters, above), skin granu-lation can be lost in preserved specimens, andinsofar as most of the species Myers (1987)placed in Epipedobates are here consideredAmeerega, we transfer this species to thatgenus as well.

GENUS: COLOSTETHUS COPE, 1867

Colostethus Cope, 1867. Type species: Phyllobateslatinasus by original designation.

Prostherapis Cope, 1868. Type species: Prosthera-

pis inguinalis by original designation.


ON: Colostethinae Cope, 1867.

SISTER GROUP: Ameerega Bauer, 1986.

CONTENT (18 SPECIES): Colostethus agi-lis Lynch and Ruiz-Carranza, 1985; C. alacrisRivero and Granados-Diaz, 1990 ‘‘1989’’;C. brachistriatus Rivero and Serna, 1986;C. dysprosium Rivero and Serna, 2000‘‘1995’’; C. fraterdanieli Silverstone, 1971;C. fugax Morales and Schulte, 1993;C. furviventris Rivero and Serna, 1991;Colostethus imbricolus Silverstone, 1975;C. inguinalis Cope, 1868; C. jacobuspetersiRivero, 1991; C. mertensi (Cochran andGoin, 1964); C. latinasus (Cope, 1863); C.lynchi Grant, 1998; C. panamensis (Dunn,1933); C. pratti (Boulenger, 1899); C. ruthveniKaplan, 1997; C. thorntoni (Cochran andGoin, 1970); C. yaguara Rivero and Serna,1991.



Unambiguously optimized phenotypic sy-

napomorphies for this clade are (1) toe disc II

moderately expanded (Character 32, 1 R 2)

and (2) male abdomen color pale, free or

almost free of melanophores (Character 63, 3

R 0).

Other characteristics include: (1) Dorsal

coloration cryptic, brown; (2) pale oblique

lateral stripe present (may be broken or

incomplete); (3) pale dorsolateral stripe

usually absent (present in C. pratti); (4) pale

ventrolateral stripe present or absent; (5)

dorsal skin texture posteriorly granular; (6)

toe webbing absent or basal to extensive; (7)

third finger of adult males swollen; (8) finger

I equal to or longer than finger II; (9) finger

discs moderately expanded; (10) median

lingual process absent; (11) larval vent tube

dextral; (12) larval oral disc shape normal

(not umbelliform); (13) larval oral disc

emarginate; (14) lipophilic alkaloids absent;

(15) chromosome number 2n 5 24 (known in

Colostethus fraterdanieli and C. panamensis);

(16) testes entirely pigmented in most species,

partially or unpigmented in others; (17) dark

throat collar absent.


DISTRIBUTION: Colostethus is a primarilytrans-Andean clade, extending from easternCentral America to northwestern Ecuador,with most species occurring at cloud forestlocalities in the western Andes. The only cis-Andean species is C. fugax, which is knownfrom the eastern slope of the CordilleraOriental in southern Ecuador, 600–700 m(see Comment).

COMMENT: Colostethus, as applied inthis revised taxonomy, refers to a morpholog-ically compact group of species. Nevertheless,the type species, Colostethus latinasus, wasnot included in the phylogenetic analysis dueto inadequate material, and the name isapplied to this clade based on its assumedclose relationship to C. inguinalis and C.panamensis (for comparisons, see Grant,2004). Among dendrobatids, Colostethusdiffers from all species of Hyloxalus inpossessing a swollen third finger, and fromall species of Silverstoneia in larger size(maximum of 22 mm SVL in Silverstoneia,greater than 24 mm SVL in Colostethus) andpossessing a ‘‘normal’’ larval mouth (umbelli-form in Silverstoneia). Among aromobatids,Allobates talamancae is sympatric with sev-eral species of Colostethus in Pacific Colom-bia and Ecuador and in Central America.Allobates talamancae differs from all speciesof Colostethus in lacking a pale obliquelateral stripe and swelling of finger III inadult males.

The moderately to extensively webbedspecies Colostethus agilis, C. mertensi, andC. thorntoni are referred to this genus becausethey (1) have a short zygomatic ramus of thesquamosal (thus differing from Rheobates),(2) possess a swollen third finger in adultmales (thus differing from Hyloxalus), (3)lack dorsolateral stripes (thus differing fromAllobates), and (4) lack a median lingualprocess (thus differing from Anomaloglossus);other genera lack moderate to extensivewebbing.

DNA sequence data for Colostethus fugaxwere deposited on GenBank by Santos et al.(2003), who did not provide locality data.Additional samples of this species froma known locality are required to further testthe placement of this species from theAmazon slopes in this otherwise trans-An-dean clade. Nevertheless, it resembles other

species of Colostethus in possessing a swollenthird finger in adult males (unlike Hyloxalus)and lacking a dorsolateral stripe (unlikealmost all species of Allobates).

GENUS: EPIPEDOBATES MYERS, 1987

Epipedobates Myers, 1987. Type species: Prosthe-

rapis tricolor Boulenger, 1899 by originaldesignation.



SISTER GROUP: Silverstoneia new genus.

CONTENT (5 SPECIES): Epipedobates an-thonyi (Noble, 1921); E. boulengeri (Barbour,1909); E. espinosai (Funkhouser, 1956); E.machalilla (Coloma, 1995) new combination;E. tricolor (Boulenger, 1899).



Due to the lack of phenotypic data for theGenBank sample of E. boulengeri, all pheno-typic transformations that occur at this nodeare optimization ambiguous. Assuming fastoptimization, phenotypic transformations forEpipedobates are (1) loss of metatarsal fold(Character 46, 1 R 0), (2) female throat andchest color dark with pale median longitudi-nal stripe (Character 62, 0 R 5), and (3)female abdomen color dark with discrete palespotting/reticulation/marbling (Character 64,0 R 3).

Other characteristics include: (1) Dorsalcoloration cryptic, brown; (2) pale obliquelateral stripe present; (3) pale dorsolateralstripe present or absent; (4) pale ventrolateralstripe present or absent; (5) dorsal skintexture smooth or with granules or tuberclesscattered irregularly over dorsal surfaces,most distinct and prevalent posteriorly; (6)toe webbing basal; (7) third finger of adultmales swollen; (8) finger I longer than fingerII; (9) finger discs narrow to moderatelyexpanded; (10) median lingual process ab-sent; (11) larval vent tube dextral; (12) larvaloral disc shape ‘‘normal’’ (not umbelliform);(13) larval oral disc emarginate; (14) lipo-philic alkaloids present; (15) chromosomenumber unknown; (16) testes entirely pig-mented; (17) dark throat collar absent.

DISTRIBUTION: All species of Epipedo-bates are trans-Andean. Epipedobates boulen-


geri, E. espinosai, and E. machalilla occur inthe Pacific lowlands of northern SouthAmerica. Epipedobates anthonyi and E. tri-color are montane species, occurring up to1,800 m on the western versant of the Andes.Following Graham et al. (2004), E. anthonyiis applied to populations in central Ecuador,while E. tricolor is applied to populations insouthern Ecuador and northern Peru (seeComment).

COMMENT: Epipedobates, as appliedhere, is equivalent to the femoralis group ofSilverstone (1976), with the exclusion ofPhyllobates femoralis and Phyllobates zaparo(both of which are placed in the aromobatidgenus Allobates; see above).

Silverstone (1976: 29) expressed doubtregarding the identity of some Ecuadorianspecimens he referred to E. boulengeri, andLotters et al. (2003b) considered the possi-bility that a complex of species may beconcealed within this nominal taxon. Theseviews seem to be validated by the currentstudy, which found E. boulengeri to benonmonophyletic. However, insofar as San-tos et al. (2003) provided no specimen datafor the DNA sequence data they depositedon GenBank, it is impossible to address thisproblem.

For the same reason, it is impossible toaddress the identity of Santos et al.’s (2003)Epipedobates sp. QCAZ16589, although itsplacement with, and few differences from, E.espinosai suggest they may be conspecific.

Graham et al. (2004) generated DNAsequence data for a specimen from the typelocality of E. tricolor and found that it didnot form a clade with samples from furthersouth (although that result was contradictedby alternative, equally parsimonious clado-grams). As such, they restricted E. tricolor tothe northern populations and applied E.anthonyi to the southern ones. We followtheir usage here, although morphologicalcharacters to consistently diagnose the twotaxa have yet to be identified.

GENUS: SILVERSTONEIA NEW GENUS

Silverstoneia new genus. Type species: Phyllobates

nubicola Dunn, 1924.



SISTER GROUP: Epipedobates Myers,1987.

CONTENT (3 SPECIES): Silverstoneia flo-tator (Dunn, 1931) new combination; S.nubicola (Dunn, 1924) new combination; S.erasmios (Rivero and Serna, ‘‘1995’’ 2000)new combination.



Unambiguously optimized phenotypic sy-napomorphies are (1) occurrence of a com-plete ventrolateral stripe (Character 54, 0 R2), (2) male abdomen color (Character 63, 3R 0), (3) anteriorly pigmented large intestine(Character 66, 0 R 1), umbelliform larvalmouth (Character 88, 0 R 1), (4) loss ofemargination of the oral disc (89, 1 R 0), (5)origin of submarginal larval papillae (Char-acter 91, 0 R 1), and (6) the loss of posteriorkeratodont rows in larvae (Character 94, 3 R0).

Other characteristics include: (1) Dorsalcoloration cryptic, brown; (2) pale obliquelateral stripe present; (3) pale dorsolateralstripe usually absent (present in some popula-tions of S. flotator in Costa Rica); (4) paleventrolateral stripe present; (5) dorsal skintexture posteriorly granular; (6) toe webbingbasal between toes III and IV; (7) thirdfinger of adult males swollen in named species(not swollen in two undescribed species; seebelow); (8) finger I longer than finger II;(9) finger discs moderately expanded; (10) me-dian lingual process absent; (11) larval venttube dextral; (12) larval oral disc shapeumbelliform; (13) larval oral disc not emar-ginate; (14) lipophilic alkaloids absent;(15) chromosome number unknown; (16) tes-tes entirely pigmented; (17) dark throat collarabsent.

DISTRIBUTION: Costa Rica to southwest-ern Colombia (southern Valle del Cauca).Nominal and undescribed species (see below)all occur below 1,600 m.

ETYMOLOGY: Silverstoneia (gender femi-nine) is named in honor of Phillip A.Silverstone for his outstanding contributionto knowledge of dendrobatoid frogs. Silver-stone named 11 species of dendrobatoids (allof which are still considered valid), and after30 years, his superb monographs (Silver-stone, 1975a, 1976) remain an essential


starting point for students of dendrobatoidfrogs. Furthermore, Silverstone carried outextensive field studies in South America inthe late 1960s and early 1970s, particularly inthe Pacific lowlands of Colombia, which havebeen key to understanding dendrobatoiddiversity (e.g., Grant, 2004) and are especiallycentral to discovering the diversity of thisgenus (see Comment).

COMMENT: These species have longbeen considered to form a distinct group,the first to recognize it (for Silverstoneiaflotator and S. nubicola) being Dunn (1931)based on the swollen third finger of adultmales and the unique larval mouth. Atpresent, Silverstoneia contains only threespecies. However, the descriptions of fiveadditional species (including ‘‘nubicola-spC’’from the present analysis) are currently inmanuscript form (T. Grant and C. W. Myers,in progress). All known larvae in this cladehave an umbelliform oral disc with sub-marginal papillae and reduced keratodontrows.

SUBFAMILY: HYLOXALINAE NEW SUBFAMILY


ON: Dendrobatidae Cope, 1865.SISTER GROUP: Dendrobatinae Wagler,

1865.CONTENT (1 GENUS): Hyloxalus Jimenez

de la Espada, 1871 ‘‘1870’’.CHARACTERIZATION, DIAGNOSIS, AND


DISTRIBUTION: Andes and adjacent Am-azonian lowlands of South America.

COMMENT: Although this name is cur-rently redundant with Hyloxalus, we anticipatethat the available names Cryptophyllobatesand Phyllodromus will be resurrected in thenear future, making Hyloxalinae an informa-tive name (for species groups, see Commentsfor Hyloxalus, below). Moreover, recognitionof Hyloxalinae is necessitated by the recog-nition of Dendrobatinae for the five generaof brightly colored and highly toxic species.

GENUS: HYLOXALUS JIMENEZ DE LA ESPADA,

1871 ‘‘1870’’

Hyloxalus Jimenez de la Espada, 1871 ‘‘1870’’.Type species: Hyloxalus fuliginosus Jimenez de

la Espada, 1871 ‘‘1870’’ by subsequent designa-tion by Savage (1968).

Phyllodromus Jimenez de la Espada, 1871 ‘‘1870’’.Type species: Phyllodromus pulchellum Jimenezde la Espada, 1871 ‘‘1870, by monotypy.

Cryptophyllobates Lotters, Jungfer, and Widmer,2000. Type species: Phyllobates azureiventris byoriginal designation.


ON: Hyloxalinae new subfamily.

SISTER GROUP: Dendrobatinae Wagler,1865.

CONTENT (57 SPECIES): Hyloxalus abdi-taurantius (Silverstone, 1975) new combina-tion; H. aeruginosus (Duellman, 2004) newcombination; H. anthracinus (Edwards, 1971)new combination; H. argyrogaster (Moralesand Schulte, 1993) new combination; H. awa(Coloma, 1995) new combination; H. azurei-ventris (Kneller and Henle, 1985) new combi-nation; H. betancuri (Rivero and Serna, 1991)new combination; H. bocagei Jimenez de laEspada, 1871; H. borjai (Rivero and Serna,2000 ‘‘1995’’) new combination; H. breviquar-tus (Rivero and Serna, 1986) new combina-tion; H. cevallosi (Rivero, 1991) new combi-nation; H. chlorocraspedus (Caldwell, 2005)new combnation; H. chocoensis Boulenger,1912; H. delatorreae (Coloma, 1995) newcombination; H. edwardsi (Lynch, 1982) newcombination; H. elachyhistus (Edwards, 1971)new combination; H. eleutherodactylus (Duell-man, 2004) new combination; H. exasperatus(Duellman and Lynch, 1988) new combina-tion; H. excisus (Rivero and Serna 2000‘‘1995’’) new combination; H. faciopuntulatus(Rivero, 1991) new combination; H. fallax(Rivero, 1991) new combination; H. fasciani-ger (Grant and Castro-H., 1998) new combi-nation; H. fuliginosus Jimenez de la Espada,1871; H. idiomelus (Rivero, 1991) new com-bination; H. infraguttatus (Boulenger, 1898)new combination; H. insulatus (Duellman,2004) new combination; H. lehmanni (Silver-stone, 1971) new combination; H. leucophaeus(Duellman, 2004) new combination; H. littor-alis (Pefaur, 1984) new combination; H.maculosus (Rivero, 1991) new combination;H. maquipucuna (Coloma, 1995) new combi-nation; H. marmoreoventris (Rivero, 1991)new combination; H. mittermeieri (Rivero,1991) new combination; H. mystax (Duellmanand Simmons, 1988) new combination; H.


nexipus (Frost, 1985) new combination; H.parcus (Rivero, 1991) new combination; H.patitae (Lotters, Morales, and Proy, 2003)new combination; H. peculiaris (Rivero, 1991)new combination; H. peruvianus (Melin, 1941)new combination; H. pinguis (Rivero andGranados-Diaz, 1990 ‘‘1989’’) new combina-tion; H. pulchellus (Jimenez de la Espada,1871) new combination; H. pulcherrimus(Duellman, 2004) new combination; H. pumi-lus (Rivero, 1991) new combination; H. ramosi(Silverstone, 1971) new combination; H. ruizi(Lynch, 1982) new combination; H. saltuarius(Grant and Ardila-Robayo, 2002) new com-bination; H. sauli (Edwards, 1974) newcombination; H. shuar (Duellman and Sim-mons, 1988) new combination; H. sordidatus(Duellman, 2004); H. spilotogaster (Duell-man, 2004) new combination; H. subpunctatus(Cope, 1899) new combination; H. sylvaticus(Barbour and Noble, 1920) new combination;H. toachi (Coloma, 1995) new combination;H. utcubambensis (Morales, 1994) new com-bination; H. vergeli Hellmich, 1940; H.vertebralis (Boulenger, 1899) new combina-tion; H. whymperi (Boulenger, 1882) newcombination.


SUPPORT: As for Hyloxalinae, above.

Other characteristics include: (1) Dorsalcoloration usually cryptic, brown, gray, orblack (conspicuous and bright in H. azur-eiventris); (2) pale oblique lateral stripepresent; (3) pale dorsolateral stripe absentin most (but not all) species; (4) paleventrolateral stripe usually absent; (5) dorsalskin texture posteriorly granular; (6) toewebbing varies from absent in most speciesto basal or extensive in some species; (7) thirdfinger of adult males not swollen; (8) finger Ishorter than finger II; (9) finger discs narrowto moderately expanded; (10) median lingualprocess absent; (11) larval vent tube dextral;(12) larval oral disc shape ‘‘normal’’ (notumbelliform); (13) larval oral disc emargin-ate; (14) lipophilic alkaloids absent; (15)chromosome number 2n 5 24 (known inHyloxalus subpunctatus and H. vertebralis);(16) testes unpigmented in most species(reported as pigmented in H. toachi byColoma, 1995); (17) dark throat collarabsent.

DISTRIBUTION: Andean South America.

COMMENT: Hyloxalus contains approxi-mately half of the species previously referredto the large, polyphyletic genus Colostethus(the other half being referred to the aromo-batid genus Allobates). Hyloxalus is anexclusively Andean radiation, although somespecies occur in the adjacent foothills.

Unfortunately, available material of thetype species, Hyloxalus fuliginosus, was in-adequate to allow its inclusion in the presentanalysis, and the name is applied based onthe presumed close relationship of thatspecies and H. bocagei, that is, H. bocagei istreated herein as a proxy for H. fuliginosus.In the event that H. fuliginosus is found notto be part of this clade, the oldest availablename would be Phyllodromus, for which thetype species is H. pulchellus.

Given the number and diversity of speciesreferred to Hyloxalus, additional partitioningwill be warranted as knowledge of the groupincreases. We include two previously recog-nized groups in Hyloxalus: (1) The Hyloxalusramosi group is delimited by the uniqueoccurrence of black, apparently glandulartissue on the inner surface of the arm. Inaddition to the undescribed H. ‘‘Ibague’’,included in the present analysis, we haveobserved this character-state in H. anthraci-nus, H. cevallosi, H. exasperatus, H. fasciani-ger, H. lehmanni, H. ramosi, and H. salt-uarius. No genus-group name exists for thisclade. (2) A group we refer to herein as the H.azureiventris group is strongly supported asmonophyletic in our analysis, and the exter-nal resemblance of the species is undeniable.An unambiguously optimized morphologicalsynapomorphy for the clade is the occurrenceof a pale dorsolateral stripe. In addition to H.azureiventris, H. nexipus, and H. chlorocras-pedus (all included in the present analysis),this group includes H. eleutherodactylus andH. patitae. A species sequenced by Santos etal. (2003) is also part of this clade, but itsidentity remains to be clarified. The genus-group name Cryptophyllobates is availablefor this clade. Formal taxonomic recognitionof these clades would render Hyloxalusparaphyletic.

SUBFAMILY: DENDROBATINAE COPE, 1865


ON: Dendrobatidae Cope, 1865.


SISTER GROUP: Hyloxalinae new subfam-ily.

CONTENT: Adelphobates new genus; Den-drobates Wagler, 1830; Minyobates Myers,1987; Oophaga Bauer, 1988; PhyllobatesDumeril and Bibron, 1941; RanitomeyaBauer, 1988.



Unambiguously optimized phenotypic syn-

apomorphies for this clade are (1) dorsal skin

texture smooth (Character 0, 1 R 0), (2) pale

oblique lateral stripe absent (Character 55, 1

R 0), (3) iris coloration lacking metallic

pigmentation and pupil ring (Character 65,

1 R 0), (4) larvae deposited in phytotelmata

(Character 111, 0 R 1), and (5) the ability to

sequester lipophilic alkaloids (Character 147,

0 R 1).

DISTRIBUTION: As for Dendrobatoidea,excluding the Atlantic forest of Brazil andhigher elevations of the Andes.

COMMENT: For synonymy see Dendro-batoidea, above.

GENUS: PHYLLOBATES DUMERIL AND

BIBRON, 1841

Phyllobates Dumeril and Bibron, 1841. Type

species: Phyllobates bicolor Dumeril and Bibron,

1841 by monotypy.


ON: Dendrobatinae Cope, 1865.

SISTER GROUP: Unnamed clade com-posed of Adelphobates new genus; Dendro-bates Wagler, 1830; Minyobates Myers, 1987;Oophaga Bauer, 1988; Ranitomeya Bauer,1988.

CONTENT (5 SPECIES): Phyllobates auro-taenia (Boulenger, 1913); P. bicolor Dumeriland Bibron, 1841; P. lugubris (Schmidt,1857); P. terribilis Myers, Daly, and Malkin,1978; and P. vittatus (Cope, 1893).



Unambiguously optimized phenotypic sy-napomorphies of this clade are (1) finger Ilonger than finger II (Character 5, 2 R 3) and(2) the uniquely derived ability to sequesterbatrachotoxin (Character 148, 0 R 1).

Other characteristics include: (1) Dorsalcoloration bright, composed of either shinyblack with bright yellow, orange, or greendorsolateral stripes or solid bright yellow,orange or green; (2) pale oblique lateralstripe absent; (3) pale dorsolateral stripepresent in all juveniles, lost ontogeneticallyin P. bicolor and P. terribilis; (4) paleventrolateral stripe absent in most, a wavyseries of elongate spots in P. vittatus;(5) dorsal skin texture smooth; (6) toewebbing absent; (7) third finger of adultmales not swollen; (8) finger I longer thanfinger II; (9) finger discs narrow to moder-ately expanded; (10) median lingual processabsent; (11) larval vent tube dextral; (12)larval oral disc shape ‘‘normal’’ (not umbelli-form); (13) lipophilic alkaloids present; (14)larval oral disc emarginate; (15) chromosomenumber 2n 5 24 (known in Phyllobateslugubris); (16) testes unpigmented; (17) darkthroat collar absent.

DISTRIBUTION: Phyllobates is an exclu-sively trans-Andean group with species occur-ring from Costa Rica through the Chocoregion of southwestern Colombia up to amaximum elevation of approximately 1500 m.

COMMENT: Phyllobates in the presenttaxonomy is unchanged from that proposedby Myers et al. (1978).

GENUS: MINYOBATES MYERS, 1987

Minyobates Myers, 1987. Type species: Dendro-

bates steyermarki Rivero, 1971 by originaldesignation.



SISTER GROUP: Unnamed clade com-posed of Adelphobates new genus; Dendro-bates Wagler, 1830; Oophaga Bauer, 1994;Ranitomeya Bauer, 1988.

CONTENT (1 SPECIES): Minyobates steyer-marki (Rivero, 1971).


SUPPORT: Branch length 5 39.

Other characteristics include: (1) Dorsalcoloration conspicuous, bright; (2) pale ob-lique lateral stripe absent; (3) pale dorsolat-eral stripe absent; (4) pale ventrolateral stripeabsent; (5) dorsal skin texture smooth; (6) toewebbing absent; (7) third finger of adultmales not swollen; (8) finger I longer than


finger II; (9) finger discs II–IV weaklyexpanded; (10) median lingual process ab-sent; (11) larval vent tube dextral or medial;(12) larval oral disc shape ‘‘normal’’ (notumbelliform); (13) larval oral disc emargin-ate; (14) lipophilic alkaloids present; (15)chromosome number unknown; (16) testescolor polymorphic; (17) dark throat collarabsent.

DISTRIBUTION: Minyobates steyermarkiis known only from Cerro Yapacana, Vene-zuela.

COMMENT: The placement of Minyo-bates steyermarki is poorly supported, asindicated by the low Bremer values ofassociated nodes. During the course of theanalysis its placement alternated betweenbeing sister to all dendrobatines exceptPhyllobates—as in our optimal solutionsand the findings reported by Vences et al.(2003a)—and sister to all species referred toRanitomeya (see below) in near-optimalsolutions. Recognizing Minyobates asa monotypic genus is consistent with bothof those solutions and expedient in thatit does not require additional taxa to benamed.

GENUS: RANITOMEYA BAUER, 1988

Ranitomeya Bauer, 1988. Type species: Dendro-

bates reticulatus Boulenger, 1884 ‘‘1883’’ byoriginal designation.



SISTER GROUP: Unnamed clade com-posed of Adelphobates new genus; Dendro-bates Wagler, 1830; Oophaga Bauer, 1994.

CONTENT (24 SPECIES): Ranitomeya ab-dita (Myers and Daly, 1976) new combination;R. altobueyensis (Silverstone, 1975) newcombination; R. amazonica (Schulte, 1999)new combination; R. biolat (Morales, 1992)new combination; R. bombetes (Myers andDaly, 1980) new combination; R. claudiae(Jungfer, Lotters, and Jorgens, 2000) newcombination; R. duellmani (Schulte, 1999) newcombination; R. fantastica (Boulenger, 1884‘‘1883’’); R. flavovittata (Schulte, 1999) newcombination; R. fulgurita (Silverstone, 1975)new combination; R. ignea (Melin, 1941) newcombination; R. imitator (Schulte, 1986) newcombination; R. intermedia (Schulte, 1999)

new combination; R. lamasi (Morales, 1992)new combination; R. minuta (Shreve, 1935)new combination; R. opisthomelas (Boulenger,1899) new combination; R. reticulata (Bou-lenger, 1884 ‘‘1883’’); R. rubrocephala(Schulte, 1999); R. sirensis (Aichinger, 1991)new combination; R. vanzolinii (Myers, 1982)new combination; R. variabilis (Zimmermannand Zimmermann, 1988); R. ventrimaculata(Shreve, 1935) new combination; R. viridis(Myers and Daly, 1976) new combination; R.virolinensis (Ruiz-Carranza and Ramırez-Pinilla, 1992) new combination.



The sole unambiguously optimized pheno-typic synapomorphy for this clade is (1) thegreatly reduced length of finger I (Character5, 1 R 0).

Other characteristics include: (1) Dorsalcoloration conspicuous, bright; (2) pale ob-lique lateral stripe absent; (3) pale dorsolat-eral stripe absent in most species; (4) paleventrolateral stripe absent; (5) dorsal skintexture smooth; (6) toe webbing absent; (7)third finger of adult males not swollen; (8)finger I shorter than finger II; (9) finger discsII–IV greatly expanded in most species; (10)median lingual process absent; (11) larvalvent tube dextral or medial; (12) larval oraldisc shape ‘‘normal’’ (not umbelliform); (13)larval oral disc emarginate; (14) lipophilicalkaloids present; (15) chromosome number2n 5 20 (known in Ranitomeya vanzolinii);(16) testes pigmented in most species (poly-morphic in R. imitator); (17) dark throatcollar absent.

DISTRIBUTION: As for Dendrobatinae,above.

COMMENT: Ranitomeya is equivalent toSilverstone’s (1975a) minutus group with theremoval of steyermarki and quinquevittatussensu stricto. Within Ranitomeya we recov-ered a monophyletic radiation equivalent toMinyobates sensu Myers (1987) minus steyer-marki (the type species of Minyobates). Thisclade is found in Central America and theColombian Choco and is absent from theAmazon basin and eastern slope of theCordillera Oriental. The sister clade to thatradiation is an exclusively Amazonian group.We recommend referring to these clades as


the minutus and ventrimaculatus groups,respectively, pending further study of theplacement of steyermarki and the Andeanspecies.

GENUS: ADELPHOBATES NEW GENUS

Adelphobates new genus. Type species: Dendrobates

castaneoticus Caldwell and Myers, 1990.



SISTER GROUP: Unnamed clade com-posed of Dendrobates Wagler, 1830; OophagaBauer, 1994

CONTENT (4 SPECIES): Adelphobates cap-tivus (Myers, 1982) new combination; A.castaneoticus (Caldwell and Myers, 1990)new combination; A. galactonotus (Steindach-ner, 1864) new combination; A. quinquevitta-tus (Steindachner, 1864) new combination.



Other characteristics include: (1) Dorsalcoloration conspicuous, bright; (2) pale ob-lique lateral stripe absent or present; (3) paledorsolateral stripe absent or present; (4) paleventrolateral stripe absent or present; (5)dorsal skin texture smooth; (6) toe webbingabsent; (7) third finger of adult males notswollen; (8) finger I shorter than finger II; (9)finger discs of fingers II–IV greatly expanded;(10) median lingual process absent; (11)larval vent tube medial; (12) larval oral discshape ‘‘normal’’ (not umbelliform); (13)larval oral disc emarginate; (14) lipophilicalkaloids present; (15) chromosome numberunknown; (16) testes unpigmented in Adel-phobates castaneoticus and A. galactonotus,pigmented in A. quinquevittatus; (17) darkthroat collar absent.

DISTRIBUTION: Amazonia.

ETYMOLOGY: Adelphobates, from theGreek adelphos (twin, brother) and bates (awalker). Gender masculine. We take greatpleasure in proposing this name in honor ofCharles W. Myers and John W. Daly.Through an ambitious field and laboratoryresearch program that began nearly fourdecades ago and remains active, the Myersand Daly collaboration has led to an un-

precedented increase in scientific knowledgeof dendrobatoid frogs. Although they pur-sued independent investigations, their nameshave become synonymous with this group,and it is only fitting that the contribution ofthese scientific ‘‘brothers’’ be commemoratedby this generic name.

COMMENT: Adelphobates galactonotuswas previously considered to be a species ofthe tinctorius species group (e.g., Silverstone,1975a), and the remaining species wereplaced in what is herein called Ranitomeya.Caldwell and Myers (1990) considered A.castaneoticus and A. quinquevittatus to besister species; however, in addition to theextensive support from DNA sequence evi-dence for the proposed relationships, A.castaneoticus and A. galactonotus share theloss of testis pigmentation. We includeAdelphobates captivus in this genus insteadof Ranitomeya on the basis of its distinctivedorsal pattern of elongate spots, found alsoin A. castaneoticus but not known in anyspecies of Ranitomeya.

GENUS: OOPHAGA BAUER, 1994

Oophaga Bauer, 1988. Type species: Dendrobates

pumilio Schmidt, 1857 by original designation.



SISTER GROUP: Dendrobates Wagler,1830.

CONTENT (9 SPECIES): Oophaga arborea(Myers, Daly, and Martınez, 1984); O.granulifera (Taylor, 1958); O. histrionica(Berthold, 1845); O. lehmanni (Myers andDaly, 1976); O. occultator (Myers and Daly,1976); O. pumilio (Schmidt, 1857); O. speciosa(Schmidt, 1857); O. sylvatica (Funkhouser,1956); O. vicentei (Jungfer, Weygoldt, andJuraske, 1996) new combination.



Unambiguously optimized phenotypicsynapomorphies of this clade are (1) larvalmarginal papillae enlarged (Character 90, 0 R1), (2) occurrence of a single anterior larvaltooth keratodont (Character 93, 2 R 1),(3) single posterior larval tooth keratodont(Character 94, 3 R 1), (4) the chirp call(Character 99, 0 R 1); (5) cloacal touching


during courtship/oviposition (Character106, 0 R 1), (6) female nurse frog (Character110, 0 R 1), (7) omosternum entirely carti-laginous (Character 127, 1 R 0), (8) anteriorprojection of suprascapula heavily calcified(Character 128, 0 R 1), (9) sacrum andvertebra 8 fused (Character 144, 0 R 1),(10) vertebrae 2 and 3 fused (Character 146,0 R 1).

Other characteristics include: (1) Dorsalcoloration conspicuous, bright; (2) pale ob-lique lateral stripe absent; (3) pale dorsolat-eral stripe absent; (4) pale ventrolateral stripeabsent; (5) dorsal skin texture smooth in allbut O. granulifera, in which it is stronglygranular; (6) toe webbing absent; (7) thirdfinger of adult males not swollen; (8) finger Ishorter than finger II; (9) finger discsmoderately expanded; (10) median lingualprocess absent; (11) larval vent tube medial;(12) larval oral disc shape ‘‘normal’’ (notumbelliform); (13) larval oral disc not emar-ginate; (14) lipophilic alkaloids present; (15)chromosome number 2n 5 20 (known inOophaga granulifera, O. pumilio, and O.sylvatica); (16) testes pigmented (entirely inmost; medially in O. sylvatica); (17) darkthroat collar absent.

DISTRIBUTION: Nicaragua through theColombian Choco to northern Ecuador atelevations below 1,200 m.

COMMENT: Oophaga is identical to thehistrionicus group of Myers et al. (1984), withthe addition of newly discovered taxa. This isone of the most conspicuous and well-knownclades within Dendrobatoidea, thanks to itschirp call, tadpole morphology, and repro-ductive behavior.

GENUS: DENDROBATES WAGLER, 1830

Hysaplesia Boie in Schlegel, 1826. Type species:Calamata punctatus Schneider, 1799 by sub-sequent designation by Stejneger, 1937.

Dendrobates Wagler, 1830. Type species: Rana

tinctoria Cuvier, 1797 by subsequent designationby Dimeril and Bibron, 1841.

Eubaphus Bonaparte, 1832. Type species: Rana

tinctoria Shaw 1802, by monotypy.

Dendromedusa Gistel, 1848. Replacement name forHylaplesia Boie, 1827 (an incorrect subsequentspelling of Hysaplesia).



SISTER GROUP: Oophaga Bauer, 1988.CONTENT (6 SPECIES): Dendrobates aur-

atus Girard, 1855; Dendrobates azureusHoogmoed, 1969; Dendrobates leucomelasSteindachner, 1864; Dendrobates nubeculosusJungfer and Bohme, 2004; Dendrobates tinc-torius (Cuvier, 1797); Dendrobates truncatus(Cope, 1861).



Other characteristics include: (1) Dorsalcoloration conspicuous, bright; (2) pale ob-lique lateral stripe absent; (3) pale dorsolat-eral stripe absent in most species (present inD. truncatus); (4) pale ventrolateral stripeabsent; (5) dorsal skin texture smooth; (6) toewebbing absent; (7) third finger of adultmales not swollen; (8) finger I shorter thanfinger II; (9) finger discs moderately togreatly expanded; (10) median lingual processabsent; (11) larval vent tube medial; (12) lar-val oral disc shape ‘‘normal’’ (not umbelli-form); (13) larval oral disc emarginate;(14) lipophilic alkaloids present; (15) chro-mosome number 2n 5 18 (known in Den-drobates auratus and D. truncatus); (16)testes pigmented; (17) dark throat collarabsent.

DISTRIBUTION: As for Dendrobatinae,above.

COMMENT: This greatly restricted Den-drobates clade is equivalent to the combina-tion of Silverstone’s (1975a) Dendrobatestinctorius group (minus galactonotus) andDendrobates auratus group. See Savage etal. (in press) for nomenclatural notes onDendrobates.

We include Dendrobates azureus as a validspecies. However, Wollenberg et al. (2006)argue for its synonymy with D. tinctorius,which is also supported by our results (seeabove).

INCERTAE SEDIS AND NOMINA DUBIA

Of the 304 taxa that were named as orsubsequently transferred into the currentDendrobatoidea, five are of dubious status.The identity of Phyllobates peruensis Stein-dachner, 1867 has been uncertain for well


over a century. Boulenger (1882: 194) sus-pected it may be a species of Hylodes, andSilverstone (1976: 6) questioned whether itwas a dendrobatid. The holotype does notexist at the Vienna Natural History Museum(F. Tiedemann, in litt. 10/10/02), and thedescription is inadequate to relate any knownpopulation to this name. We thereforeconsider this taxon to be nomen dubium.

The second species is Prostherapis dunniRivero, 1961. La Marca (2004) redescribedthe species, and we refer the reader to thatpaper for a complete account. La Marcaclarified that, contrary to previous accounts,the species does not possess a collar (and istherefore not referable to Mannophryne) andappears to be confined to the central part ofthe Venezuelan Coastal Range. The feet areextensively webbed, the zygomatic ramus iselongate and robust, and the tongue lacks themedian lingual process (T. Grant, personalobs.). Given these character-states and geo-graphic distribution, it seems most likely thatProstherapis dunni is a species of Aromobates.However, we are unable to rule out thepossibility that it is a species of Rheobates,also an aromobatid, given that R. palmatusalso shares these character-states. Insofar aswe have no further insights to offer into thesystematics of Prostherapis dunni, we followLa Marca (2004: 24) in concluding that ‘‘[i]tsphylogenetic relationships remain enigmatic’’and refer it only to Aromobatidae, incertasedis.

The third species is Dendrobates myster-iosus Myers, 1982. Myers (1982) placed thisspecies in the captivus group with Adelpho-bates captivus, stating that ‘‘[t]he large palespots on either the dorsal or especially theventral surface of the thigh may providea necessary synapomorphy in support of sucha relationship.’’ Schulte (1990) concludedthat it is not related to A. captivus (asDendrobates) or any part of the quinquevitta-tus group (sensu Silverstone, 1975a, pre-sumably) and is instead related to Oophaga(as the Dendrobates histrionicus group). Hebased this conclusion on shared size, absenceof omosternum, occurrence of round spotson a dark background, similar reproductivebehavior, an elevated number of small ova,and an apparently (no audiospectrographicdata were presented) similar fundamental

frequency of the call. None of these char-acters is unique to the Oophaga, and severalother reported character-states conflict withthis relationship (e.g., larval mouth parts). Assuch, given the current evidence, we considerDendrobates mystersiosus to be Dendrobati-nae, incerta sedis.

The fourth species is Colostethus ramireziRivero and Serna, 2000 ‘‘1995’’. Rivero andSerna (2000 ‘‘1995’’: 50) described this speciesfrom the northwestern Colombian Andes as‘‘medium sized, surely referable to eithergroup IV or IX’’ (translated freely from theSpanish), species of which are placed in thedendrobatid genera Colostethus, Hyloxalus,and Silverstoneia in the current taxonomy.The data reported by Rivero and Serna (2000‘‘1995’’) do not permit the species to beallocated defensibly to any of these genera,and we therefore consider it to be Dendro-batidae, incerta sedis.

Finally, Colostethus poecilonotus Rivero,1991 is known only from the type locality at500 m in the Peruvian department of Ama-zonas. Rivero (1991a) described this speciesas ‘‘probably belonging to Group IX’’,species of which are placed in Anomaloglos-sus, Colostethus, and (mostly) Hyloxalus inthe current taxonomy. This species lacks themedian lingual process (which excludes itfrom Anomaloglossus), but evidence is lack-ing to place it in either Colostethus orHyloxalus. Likewise, there is no evidence toindicate that this is not a species of Allobates.As such, we consider this species to beDendrobatoidea, incerta sedis.

DISCUSSION:CHARACTER EVOLUTION

The novel knowledge claims that emergefrom phylogenetic analysis have implicationsbeyond the immediate problems of systemat-ics. By providing a causally relevant frame-work of reference, knowledge of phylogenyimposes meaningful structure on otherwisedisparate biological data from unrelatedfields of biology, which often leads tounanticipated insights and identifies novelproblems for further investigation. It is thispotential for cross-discipline unification thatmakes phylogenetic systematics a fundamen-


tal part of an ampliative, progressive researchprogram.

In this section, we analyze the implicationsof the phylogeny of Dendrobatoidea for theevolution of several characters and charactersystems. Although this does not entailphylogenetic analysis in the strict sense ofcladogram searching, our approach remainsdecidedly phylogenetic. Rather than searchfor statistical correlations to explain biolog-ical variation in terms of its adaptive value,functional significance, or selective pressures(e.g., Summers and Earn, 1999; Caldwell andde Araujo, 2004), we explain it in terms of itsevolutionary origins.

The following analysis of character evolu-tion should be interpreted in light of twocaveats: First, the analysis necessarily as-sumes the veracity and completeness ofreported observations. For the most part thisis not likely to be problematic. Data weretaken either from personal observations, fieldnotes and photographs, or published sourcesthat were vetted by peer review. However,increased sampling may lead to alternativescorings. For example, nurse frog sex isusually known from one or a few observa-tions, but detection of biparental transportrequires multiple observations, by definition.Second, there are extensive missing data forseveral of the characters we analyze below,and, although the most parsimonious opti-mization often allows unambiguous retrodic-tion of unknown states, it is possible thatfuture discoveries will overturn some retro-dictions and favor alternative evolutionaryexplanations. Unless otherwise stated, onlyunambiguous optimizations are considered.There is no defensible basis for choosingbetween fast (accelerated) and slow (delayed)optimizations, making any evolutionary in-ference drawn from such optimizations un-tenable.

As noted above, many aspects of thenatural history of dendrobatoids have beenstudied. Here we focus on adult habitatselection and reproductive biology, includingparental care, larval habitat, and larval diet.Parental care in dendrobatoids involves atleast three distinct components, each of whichmay be undertaken by one or both parents:clutch attendance, tadpole transport, andoocyte provision for larval consumption.

Few data on clutch attendance are available(but were coded nonetheless as Character109), and we therefore focus only on tadpoletransport and provision of oocytes for larvalconsumption, the latter in the context of larvaldiet.

In terms of species diversity, the mostthorough comparative study of dendrobatoidreproductive biology to date is that ofSummers and McKeon (2004). However,the phylogeny used in that study wasa composite ‘‘derived from several of therecent molecular phylogenetic analyses’’(p. 56). The means of resolving conflictamong those studies was not specified.Furthermore, species not included in any ofthose analyses were placed in the cladogrambased on their assumed position (e.g., Ar-omobates nocturnus, Dendrobates mysterio-sus). Additionally, as mentioned above (Phe-notypic Characters), some character-stateswere misattributed by Summers and McKeon(2004), which has implications for the evolu-tionary scenarios they proposed.

ADULT HABITAT SELECTION

The traditional view inherited from Noble(1926) is that dendrobatoids evolved pro-gressively from more aquatic to more terres-trial species. This was also manifest in thephylogeny proposed by Myers et al. (1991),in which the fully aquatic Aromobates noc-turnus was sister to all other dendrobatids,which, in turn, were divided into the riparian‘‘Hyloxalus sensu stricto’’ and more terres-trial ‘‘Colostethus sensu stricto’’ and apose-matic taxa. Adult association with water wascoded here as Character 114.

The ancestral state for Dendrobatoidea isambiguous in our analysis. However, theancestral state for Dendrobatidae optimizesunambiguously as terrestrial (i.e., indepen-dent of bodies of water, adults reaching 30 mor more into the forest), with no fewer thansix independent origins of riparian habitatpreference (i.e., adults occurring alongstreams or pools, extending no further than3 m from the water’s edge) and one sub-sequent origin of terrestriality (in Hyloxalustoachi; see Coloma, 1995: 54).

Among aromobatids the situation is lessclear. Under slow optimization the ancestral


state is riparian, with five independent originsof terrestriality and one subsequent return toa riparian lifestyle. Under fast optimizationthe ancestral state is terrestrial, with fiveindependent origins of riparian habitat pref-erence and one reversal to terrestriality.Nevertheless, it is clear that rather than beingthe primitive state for dendrobatoids (asproposed by Myers et al., 1991), the fullyaquatic behavior of Aromobates nocturnus isunambiguously derived.

It is also clear that neither clade evolvedthrough a simple progression from a moreaquatic lifestyle to a more terrestrial one, andthe pattern that emerges is complex. Never-theless, adult association with water isconserved phylogenetically, with a retentionindex of 0.71. In some cases, the transition isaccompanied by morphological transforma-tions that are presumably associated with thedegree of association with water, such as thegain or loss of webbing (e.g., Hyloxalusbocagei, H. nexipus, Rheobates palmatus,and Anomaloglossus tepuyensis all possessextensive toe webbing). However, althoughthere are no extensively webbed species codedas independent of water, species with in-termediate webbing may be terrestrial (e.g.,Colostethus fraterdanieli, with basal webbingbetween toes II and III) or riparian (e.g.,Hyloxalus insulatus, with the same degree ofwebbing between toes II and III).

REPRODUCTIVE AMPLEXUS

Cephalic reproductive amplexus has longbeen considered a synapomorphy of Den-drobatoidea, with the absence in numerousdendrobatids explained as a derived losswithin the clade (e.g., Duellman and Trueb,1986; Myers and Ford, 1986; Myers et al.,1991; Haas, 2003). Although data are lackingfor many species, our results indicate un-ambiguously that axillary amplexus was lost(Character 105, 1 R 0) in the most recentcommon ancestor of Dendrobatoidea, withcephalic amplexus derived independently(Character 105, 0 R 2) in Anomaloglossus(either in the common ancestor of the genusor within the genus; data are only availablefor A. beebei), the most recent commonancestor of Colostethinae, and Minyobatessteyermarki. This reversal of the polarity of

this character in dendrobatoids necessitatesa fundamental rethinking of the evolution ofamplexus in Dendrobatoidea. For example,rather than the cephalic grasping that occursduring wrestling and courtship being a vesti-gial remnant of the ancestral cephalic re-productive amplexus (e.g., Myers et al.,1991), our results suggest this behavior maybe a first, intermediate step toward cephalicreproductive amplexus.

SEX OF NURSE FROGS

Previous studies have claimed dorsal tad-pole transport as a synapomorphy of Den-drobatoidea (e.g., Myers, 1987; Weygoldt,1987), which is corroborated unambiguouslyin the present study. Moreover, the twoincluded dendrobatoids known to lack dorsaltransport (Allobates nidicola and Anomalo-glossus stepheni; both aromobatids) lost itindependently, as discussed in greater detailbelow in the context of larval endotrophy.

Tadpole transport by male nurse frogs isalso the unambiguously primitive state fordendrobatoids, with transport by femalenurse frogs and biparental transport havingevolved repeatedly. Among aromobatids,transport exclusively by female nurse frogsevolved only in Allobates talamancae. Tad-pole transport remains unknown in theundescribed sister species of A. talamancae(A. ‘‘Magdalena’’), but no other aromobatidis known to have exclusively female nursefrogs.

Biparental transport evolved independent-ly in the ancestor of the Allobates femoraliscomplex and A. trilineatus, although theparticulars of each case are unclear. First,tadpole transport is unknown in A. zaparo.Second, we coded all specimens presumed tobe ‘‘Allobates femoralis’’ on morphologicalgrounds as having biparental transport.However, this is based on reports by Silver-stone (1976: 31) of female nurse frogs fromPeru and Suriname, Lescure (1976a: 487,1976b) of male nurse frogs from FrenchGuiana, and Aichinger (1991) of male nursefrogs from Peru (explicit reports of both sexesare by Weygoldt, 1987 and Caldwell and deAraujo, 2005). In light of the evidence that A.femoralis is a complex of species it is possiblethat at least some of these species may have


nurse frogs of a single sex. Nevertheless,Caldwell and de Araujo (2005) reportednurse frogs of both sexes for this species ata single locality (Rio Curua-Una, Brazil), andthere is no evidence to suggest more than onespecies is involved. Observations of biparen-tal transport in A. trilineatus also occurred ata single locality (Panguana, Peru; Aichinger,1991), and, whether or not this is viewed asa complex of species, there is no evidence thatmore than one trilineatus complex speciesoccurs there. Larval transport is unknown inthe closest relatives of A. trilineatus, but A.insperatus has exclusively male transport.

Among dendrobatids, transport by exclu-sively female nurse frogs evolved two or threetimes: once or twice in Colostethus and oncein the ancestor of Oophaga. Among species ofColostethus, C. panamensis and C. prattipossess female nurse frogs, whereas C.fraterdanieli and the undescribed species C.‘‘pratti-like’’ are known only to have malenurse frogs. The ambiguity is due to theunknown states of C. imbricolus and, inparticular, C. inguinalis (note that priorreports of C. inguinalis transport apply toC. panamensis; Grant, 2004). Finding that C.inguinalis has male nurse frogs would entailindependent origins of female nurse frogs inC. panamensis and C. pratti; finding that C.inguinalis has female nurse frogs would implya single origin of female nurse frogs, witha reversal to male nurse frogs in C. ‘‘pratti-like’’.

Female larval transport optimizes unam-biguously as homologous in all species ofOophaga. In this clade, the shift to femaletransport was accompanied by the produc-tion of maternal oocytes for larval consump-tion (see Character 112). The adaptivesignificance, if any, of this correlation isunknown, but the independent evolution offemale nurse frogs in lineages that lack larvaloophagy demonstrates that the relation is notnecessary biologically. It should also benoted that larval use of phytotelmata (Char-acter 111) arose in the common ancestor ofDendrobatinae and is therefore not coupledwith female transport (or oophagy; seebelow).

As in Aromobatidae, biparental transportappears to have evolved multiple times inDendrobatidae. Coloma (1995:20) reported

a male nurse frog for Hyloxalus awa, butMudrack’s (1969) detailed observations ofthe breeding behavior of H. awa (as Phyllo-bates sp.) in captivity showed that either sexmay transport tadpoles.16 Ameerega hahneliand A. petersi are closely related species, butbiparental care (reported for A. hahneli [asEpipedobates pictus] by Aichinger, 1991 [seealso Haddad and Martins, 1994:291; Kok,2000:13] and A. petersi by Silverstone,1976:38) optimizes unambiguously as inde-pendently evolved.

LARVAL HABITAT AND DIET

Three habitats are exploited by larvaldendrobatoids (Character 111). The primitivestate for dendrobatoids is for larvae tooccupy ground level pools or streams, as istypical of most anurans. Larval use ofphytotelmata (i.e., phytotelm breeding)evolved three times: twice in aromobatidsand once in dendrobatids. Among aromoba-tids, phytotelm breeding was reported forAnomaloglossus beebei by Bourne et al.(2001). In the present study, we also foundthat its sister species A. roraima is a phytotelmbreeder. Adults and tadpoles of A. roraimawere collected from tank bromeliads near thetype locality, and tadpole identification wasaccomplished by analysis of DNA sequences.The cytochrome b sequences of the threespecimens sampled (two adults, one tadpole)differ in only 1–3 bp (0.3–0.8% uncorrectedpairwise distance). Larval habitat is unknownfor all close relatives of A. beebei and A.roraima. As such, it is unclear if phytotelmbreeding is homologous in just these twospecies or a more inclusive clade.

The second origin of larval use of phyto-telmata in aromobatids occurred in Allobatesfemoralis, as reported by Caldwell and deAraujo (2004). Nevertheless, in this species,

16Weygoldt (1987: 55) disputed Mudrack’s (1969)claim of biparental care in Hyloxalus awa (asColostethus sp.), stating that it ‘‘may be a captivityartifact because under crowded conditions many frogsoccasionally attempt to sit on or close to eggs’’.However, that does not address Mudrack’s observa-tion that both males and females actually transporttadpoles. We therefore accept Mudrack’s report atface value.


phytotelm breeding is most likely opportu-nistic, that is, ground-level phytotelmata areprobably exploited like any other ground-level body of water and not targeted prefer-entially. This species is not known to exploitaboveground phytotelmata. (Caldwell and deAraujo also mentioned finding ‘‘Colostethus’’[probably Allobates] larvae in ground-levelphytotelmata, but they did not identify thespecies.)

Among dendrobatids, available evidenceindicates that phytotelm breeding evolvedonly once, in the most recent commonancestor of Dendrobatinae (Phyllobates +Minyobates + Ranitomeya + Adelphobates +Oophaga + Dendrobates). Within that clade,Dendrobates leucomelas reevolved the larvaluse of ground-level streams and pools, andD. auratus and D. truncatus evolved a gener-alist strategy whereby they transport larvaeto aboveground phytotelmata or ground-level water bodies.

Larval oophagy (i.e., larval consumptionof nutritive eggs provided by the mother;Character 112) evolved independently inphytotelm breeders of Dendrobatidae andAromobatidae. In Oophaga, females performall parental care and deposit nutritive oocytesfor larval consumption without any involve-ment of the male. Brust (1993) and Pramukand Hiler (1999) demonstrated the obligateoophagy of O. pumilio larvae, and it is likelythat this is the case for the remainder of theclade as well. Insofar as is known, this hasnot evolved in Aromobatidae.

Nevertheless, in both Aromobatidae andDendrobatidae a form of biparental care hasevolved in which courtship culminates in thefemale depositing oocytes directly into thewater for larval consumption, that is, maleinvolvement in courtship is required tostimulate the female to release oocytes(Character 113). This cooperative behaviorwas first reported for the dendrobatinesRanitomeya reticulatus (Kneller, 1982; Zim-mermann and Zimmermann, 1984), R. van-zolinii (Caldwell, 1997; Caldwell and deOliveira, 1999) and R. ventrimaculatus (Zim-mermann and Zimmermann, 1988, as quin-quevittatus; note that exclusively male carewas observed in Peruvian R. ventrimaculatusby Summers et al., 1999b, further supportingCaldwell and Myers’s, 1990, conjecture that

this is a complex of cryptic species) and morerecently for the aromobatid Anomaloglossusbeebei (Bourne et al., 2001). Even in thesecases of biparental care, oocytes are notfertilized and are deposited directly into thewater (and not on the dry surfaces abovewater), which indicates that they are de-posited solely for larval consumption and notmerely as a biproduct of repeated mating.This reproductive mode therefore differsfrom larval oophagy in Osteocephalus (Hyli-dae), in which parents mate repeatedly at thesame sites and freshly laid eggs are eitherconsumed by older siblings or survivethrough competition to metamorphosis(Jungfer and Weygoldt, 1999; see also Had-dad et al., 2005, in regard to Aplastodiscusperviridis). However, the oophagy resemblesthat of the foam-nest breeder Leptodactylusfallax, in which maternal provisioning ofnutritive oocytes is unaccompanied by themale (Gibson and Buley, 2004).

According to available data, nidicolouslarvae evolved at least twice in Aromobatidaeand never in Dendrobatidae. Anomaloglossusstepheni (Junca et al., 1994; Junca, 1996,1998) and Allobates nidicola are not closelyrelated. Anomaloglossus degranvillei is alsoendotrophic, but this species is exoviviparous(Altig and Johnston, 1989), that is, tadpolesdevelop while being transported by the malenurse frog (see review by Caldwell and Lima,2003). Allobates chalcopis is also endotrophic(Kaiser and Altig, 1994) and is predicted tobe exoviviparous (Junca et al., 1994). Thephylogenetic placement of A. chalcopis issomewhat unclear in that it was not includedexplicitly in the present study. Nevertheless, itlacks the median lingual process, whichsuggests it is not closely related to Anom-aloglossus stepheni, and the fact that it isendotrophic and has 2n 5 22 chromosomessuggests it may be closely related to A.nidicola (see A Monophyeletic Taxonomy,above, for further discussion of this species).

As coded for the present analysis, endo-trophy optimizes unambiguously as theprimitive state for the nonwebbed clade ofAnomaloglossus. Nevertheless, this must beinterpreted in light of (1) the extensivemissing data and (2) the fact that we codedobserved specimens of A. ‘‘degranvillei’’ fromGuyana according to reproductive observa-


tions made on A. degranvillei sensu strictofrom French Guiana. As discussed above,these are likely different species. In that case,and assuming that A. ‘‘degranvillei’’ is notendotrophic, endotrophy would optimize ashomologous in the less inclusive clade thatincludes A. stepheni. In either case, currentevidence indicates at least two independentorigins of endotrophy in aromobatid frogs(depending on the exact placement of Allo-bates chalcopis). Further investigation will berequired to determine if the independentorigins of endotrophy are accompanied bydifferent developmental modifications aswell. Detailed developmental data exist foronly a few anurans (reviewed by Thibaudeauand Altig, 1999; Callery et al., 2001; Desnits-kiy, 2004) and are entirely lacking foraromobatids.

As noted by Junca et al. (1994) andCaldwell and Lima, (2003), the timingmodifications that produced endotrophiclarvae differ. The exoviviparous larvae ofAnomaloglossus degranvillei lack the kerati-nized jaw sheath, oral disc, and spiracle,whereas nidicolous larvae of the closelyrelated A. stepheni lack the keratinized jawsheath and oral disc but possess a spiracle.The inverse occurs in Allobates, in which thepresumably exoviviparous larvae of Allobateschalcopis have a complete larval morpholgyand the nidicolous larvae of Allobates nidi-cola possess an unkeratinized lower jaw andlack the oral disc and spiracle.

The close phylogenetic relationship be-tween Anomaloglossus degranvillei and A.stepheni to A. beebei draws attention toa previously unappreciated relationship be-tween endotrophy and oophagy. Conceptu-ally, endotrophy and oophagy are differentphysiological and behavioral means to thesame end: the female’s reproductive biologyis altered to provide additional nutrients forlarval development, either through pre-ovi-position enrichment of the oocyte or post-oviposition provision of nutritive oocytes.This observation that oophagy and endotro-phy are in this sense adaptationally equiva-lent raises more questions than answers. Asmentioned above, the unambiguous optimi-zation of endotrophy as the primitive statefor this clade may be an artifact of taxonomy.Nevertheless, assuming that relationship to

be true implies that oophagous speciesevolved from an endotrophic ancestor. Dataare unavailable on the relative metaboliccosts of production of normal-sized oocytesfor larval consumption versus expansion ofthe nutritive endoderm, but they will beessential to understanding the trade-offsinvolved in these transformations.

Further, in terms of reproductive success,under what conditions would natural selec-tion favor one or the other strategy? Sum-mers and Earn (1999) analyzed the condi-tions under which entirely female care(including provision of nutritive oocytes)would be favored, but the relative costs andbenefits of endotrophy have not been con-sidered in this context. Summers and Earnsuggested that the transition from all male toall female care may have been driven in partby males suffering a cost of lost matingopportunities due to investment in parentalcare. Male investment in parental care is notappreciably less, and may actually be greater,in nidicolous species (Junca, 1996) than otherdendrobatoids, the difference being thatmales guard clutches throughout develop-ment in nidicolous species, which lengthensthe duration of male investment, but musttransport tadpoles to water in nonnidicolousspecies, which is also costly and may increasethe risk of predation (potentially throughdecreased locomotor performance [but seeDownie et al., 2005] or increased exposureand lack of known escape routes) and loss ofterritory (Cummins and Swan, 1995). Thepotential exists for tadpole transport to bea cost to male foraging, but in Mannophrynetrinitatis male nurse frogs do not appear toforage less than calling males (Downie et al.,2005). The fact that the male remains in (andtherefore does not risk losing) his territoryand continues to vocalize and mate success-fully (Junca, 1996) lends support to Summersand Earn’s model, with the subtle clarifica-tion that it is not the paternal investment thatmatters per se, but the cost it entails in termsof lost mating opportunities.

SUMMARY AND CONCLUSIONS

DNA sequences totaling approximately6,100 bp (x 5 3,740 bp per terminal; totaldataset <1.55 million bp) were generated for


five mitochondrial and six nuclear loci, and174 phenotypic characters were individuatedfrom adult and larval morphology, alkaloidprofiles, and behavior. The complete datasetincluded 414 terminals: 367 terminals of 156ingroup species, and 47 outgroup terminals.Direct optimization parsimony analysis re-sulted in a 25,872 most parsimonious solu-tions of 46,520 transformations, with allconflict restricted to conspecific terminals.Dendrobatids were recovered as a monophy-letic group, identified in the new taxonomy asDendrobatoidea Cope, 1865, and the sistergroup was found to consist of CrossodactylusDumeril and Bibron, 1841, Hylodes Fitzin-ger, 1826, and Megaelosia Miranda-Ribeiro,1923, recognized herein as Hylodidae Gun-ther, 1858. The latter finding disagrees withthe results of Frost et al. (2006) but is basedon greatly increased character sampling forimmediately relevant terminals and includeda large sample of taxa from the Frost et al.study. Additional changes to outgroup tax-onomy are: proposal of Cruciabatrachia newtaxon for the sister group of CentrolenidaeTaylor, 1951; elevation of the tribes Batra-chylini and Ceratophrynini to subfamilies(Batrachylinae Gallardo, 1965 and Cerato-phryinae Tschudi, 1838); proposal of Cala-mitophrynia new taxon for the sister group ofCycloramphidae (including ThoropidaeFrost et al., 2006), resurrection of Leiuper-idae Bonaparte, 1850 for several generareferred previously to Leptodactylidae Wer-ner, 1896 (1938); and proposal of Nobleoba-tia new taxon for the clade composed ofHylodidae Gunther, 1858 and Dendrobatoi-dea Cope, 1865.

As expected, Colostethus Cope, 1866 wasfound to be violently polyphyletic, and weproposed a new monophyletic taxonomy.The sampled dendrobatoids were distributedapproximately symmetrically in two clades:Aromobatidae new family and Dendrobati-dae Cope, 1865. Insofar as is known, allaromobatids are nontoxic. Within Aromoba-tidae, a diverse clade of species that possessthe median lingual process was discoveredand named Anomaloglossus n.gen. All in-cluded species of Anomaloglossus occur eastof the Andes, but three species (A. atopoglos-sus Grant, Humphrey, and Myers, 1997; A.‘‘chocoensis’’ auctorum [not Hyloxalus cho-

coensis Boulenger, 1912; see Grant et al.,1997]), and A. lacrimosus Myers, 1991), aredistributed in the Pacific slopes and lowlandsof Colombia and Ecuador. Several species ofAnomaloglossus possess highly derived re-productive biology, including nidicolous andexoviviparous endotrophic larvae, phytotelmbreeding, and the biparental production ofnutritive oocytes for larval consumption. Thesister of that genus is Rheobates n. gen. fromthe eastern and central Andes of Colombia.The inclusive Anomaloglossus + Rheobatesclade was named Anomaloglossinae newsubfamily.

Colostethus saltuensis Rivero 1980 ‘‘1978’’and Aromobates nocturnus Myers, Daly, andPaolillo, 1991, the latter being the typespecies of Aromobates Myers, Daly, andPaolillo, 1991, were found to be nested withina clade of species referred previously toNephelobates La Marca, 1994. Consequently,Nephelobates is a junior synonym of Aromo-bates. Mannophryne La Marca, 1992, whosestatus has been controversial, was found tobe monophyletic. The Aromobates + Manno-phryne clade was recognized as Aromobati-nae new subfamily. Aromobatinae is distrib-uted primarily in the Andes of Venezuela,with minor incursions into adjacent Colom-bia and a few lowland species that alsoextend to Trinidad.

The remaining species of aromobatidsform a large, predominantly cis-Andeanradiation referred to the existing nameAllobates Zimmermann and Zimmermann,1988. Within this clade is a complex ofsuperficially similar species traditionallyplaced in Silverstone’s (1976) femoralis group(or directly in A. femoralis Boulenger, 1883),including A. femoralis, A. zaparo Silverstone,1976, A. myersi Pyburn, 1981, and A. rufulusGorzula, 1990 ‘‘1988’’. Also in this clade areroughly half of the species formerly referredto Colostethus, including Allobates nidicolaCaldwell and Lima, 2003 and A. chalcopisKaiser, Coloma, and Gray, 1994, whichpossess nidicolous and exoviviparous endo-trophic larvae, respectively. Given the di-versity of species in this clade (in terms of thenumber of species and their morphological,behavioral, and reproductive variation), it islikely that further progress will allow addi-tional clades in this group to be recognized


formally and for Allobates to be restricted tothe femoralis group. We therefore proposethe name Allobatinae new subfamily for thisclade.

All poisonous species of dendrobatoids arerestricted to Dendrobatidae Cope, 1865.Numerous conspicuous clades occur in thisclade, many of which can be referred toavailable names. Colostethinae Cope, 1868includes four genera. Silverstoneia n.gen. isnamed for the nubicola group of species,a clade of three nominal and at least fiveas yet undescribed species (one of whichwas included in our analysis) with highlymodified larvae. The sister group of Silver-stoneia is Epipedobates Myers, 1987, whichis here applied to the species related to E.tricolor. This clade includes the toxicspecies E. tricolor and E. anthonyi. Darst etal. (2005) reported E. boulengeri as lackingalkaloids, and we coded this species accord-ingly, but we suggest that result deservesfurther invesigation, either through addi-tional sampling of wild populations orcontrolled feeding experiments. Likewise,the toxicity of E. machalilla has yet to beinvestigated.

The sister group of those genera includesColostethus Cope, 1866 and AmeeregaBauer, 1986. Colostethus in our restrictedsense is a moderate-sized clade (18 recognizedspecies) from the Andes of Colombiaand Ecuador, the inter-Andean valleys ofColombia, and a single species (C. fugaxMorales and Schulte, 1993) known fromthe Amazon slope of the Ecuadorean Andes.No species of Colostethus is known tosequester lipophilic alkaloids, but C. pana-mensis (Dunn, 1933) possesses tetrodotoxin(Daly et al., 1994b). Parental care variesamong species of Colostethus; C. pratti(Boulenger, 1899) and C. panamensis havefemale nurse frogs, whereas C. fraterdanieliSilverstone, 1971 and the undescribed speciesreferred to as C. ‘‘pratti-like’’ have malenurse frogs.

Ameerega consists of most of the speciesreferred to Epipedobates recently and Phyllo-bates Dumeril and Bibron, 1841 sensuSilverstone (1976) before that. The bulk ofthis radiation is cis-Andean, with only A.andina (Myers and Burrowes, 1987) and A.erythromos (Vigle and Miyata, 1980) known

on the Pacific slopes of Colombia andEcuador, respectively, and A. maculata (W.Peters, 1873) from Panama. Although we didnot include these western species in thephylogenetic analysis, we allied them withthis clade on the basis of their previousplacement in the pictus group and Epipedo-bates, respectively. Insofar as is known, allspecies of Ameerega sequester lipophilicalkaloids.

Hyloxalus Jimenez de la Espada, 1871‘‘1870’’ is applied to a large clade of nontoxic,primarily (but not exclusively) Andean spe-cies, including approximately half of thespecies referred previously to Colostethus(with most others referred to Allobates,above). Available names included in thesynonymy of Hyloxalus are Cryptophyllo-bates Lotters, Jungfer, and Widmer, 2000and Phyllodromus Jimenez de la Espada,1871 ‘‘1870’’. The type species of both ofthese genera were included in the analysis,and both fall out in strongly supportedclades. The clade that would be referred toCryptophyllobates is morphologically con-spicuous, and referring species not analyzedexplicitly (such as H. eleutherodactylus[Duellman, 2004]) is unproblematic. Howev-er, owing to its placement as sister to theclade that includes the type species ofPhyllodromus (Phyllodromus pulchellum Jime-nez de la Espada, 1871 ‘‘1870’’), recognitionof Cryptophyllobates would require thatspecies not analyzed explicitly be allocatedto either Hyloxalus or Phyllodromus, which isnot feasible at present.

The remaining clade consists of the fivegenera most widely recognized as dart-poisonfrogs. All of these species are toxic and mostplace larvae in phytotelmata. Given thedistinctiveness of this clade and its impor-tance in many areas of biology, we recognizeit as Dendrobatinae Cope, 1865. As such, weproposed Hyoxalinae new subfamily for thesister group, that is, Hyloxalus. This solutionis not entirely satisfactory because it pro-duces a redundant name. Nevertheless, asdiscussed above, available genus-groupnames exist within Hyloxalus and at leasttwo conspicuous clades are known, and weexpect the formal nomenclatural recognitionof additional clades as knowledge of den-drobatid phylogeny increases.


Phyllobates Dumeril and Bibron, 1841 isidentical to the group proposed by Myers etal. (1978) and is here recovered as the sistergroup of the remaining dendrobatines. Werecognize the sister of the next less-inclusiveclade as the monotypic genus MinyobatesMyers, 1987 for M. steyermarki (Rivero,1971). Support for the placement of thisspecies is weak, owing primarily to thesmall amount of available DNA sequencedata (see Vences et al., 2003a), but its recog-nition as a monotypic genus is expedient inthat it does not require that more names beproposed.

Ranitomeya Bauer, 1986 includes most ofthe diminutive species included in Silver-stone’s (1975a) minutus group prior to theplacement by Myers (1987) of several ofthose species in Minyobates Myers, 1987(additional species otherwise referable toSilverstone’s minutus group are not relatedto these species; see below). Ranitomeyaincludes two conspicuous clades: a trans-Andean/Central American clade and a strictlyAmazonian cis-Andean clade; we refer tothese clades as the minutus and ventrimacula-tus groups, respectively.

Oophaga Bauer, 1988 is applied to thehistrionicus group of Myers et al. (1984).These species have unique vocalizations andexhibit all-female parental care, includingfemale tadpole transport and the productionof nutritive oocytes solely for the purpose offeeding larvae.

Adelphobates n.gen. was proposed for theclade containing A. castaneoticus (Caldwelland Myers, 1990), A. quinquevittatus (Stein-dachner, 1864), A. galactonotus (Steindach-ner, 1864), and (provisionally) A. captivus(Myers, 1982). The close relationships be-tween A. castaneoticus and A. quinquevittatuswere expected, but the placement of A.galactonotus here is somewhat heterodoxin that it alone was previously referred tothe tinctorius group of Silverstone (1975a).Nevertheless, this result was also obtainedby Vences et al. (2003a), and, insofar asmorphology for this species was included inthe present analysis, there is no empiricalbasis to challenge its placement. Finally,Dendrobates Wagler, 1830 was applied tothe remainder of the tinctorius group ofSilverstone (1975a).

ACKNOWLEDGMENTS

We are grateful to Maureen Donnelly,Alexander Haas, and Roberto Ibanez fortheir thorough and insightful reviews of themanuscript. For access to collections andinstitutional specimen and tissue loans, will-ingness to share field notes and collectiondata, unpublished data, photographs, andfield collaboration, we thank Andres Acosta,Adolfo Amezquita, Gonzalo Andrade, MarıaCristina Ardila, Santiago Ayerbe, MarcusBreece, Kent Beaman, Wilmar Bolıvar, God-frey Bourne, Jonathan Campbell, FernandoCastro, Darıo Correa, John Daly, Rafael deSa, William Duellman, Linda Ford, GlennFox, Carl Franklin and Donna Dittman,Steve Gotte, Ron Heyer, Roberto Ibanez,Gustavo Kattan, David Kizirian, ArnoldKluge, Pablo Lehmann, Karen Lips, StefanLotters, Horst Luddecke, Jhon Lynch, RossMacCulloch, Roy McDiarmid, Paula Mik-kelsen, Jhon Jairo Mueses-C., Paulo Nuin,Vivian Paez, Juan Manuel Rengifo, GilsonRivas, Lily Rodrıguez, Ralph Saporito, JaySavage, Gerhard Schlosser, Greg Schneider,Philip Silverstone-Sopkin, John Simmons,Angela Suarez, Franz Tiedemann, LindaTrueb, David Wake, Addison Wynn, andRichard Zweifel.

For going out of their way to obtaincritical tissues and/or sequences, we thankAndres Acosta, Godfrey Bourne, MarcusBreece, Roberto Ibanez, Karen Lips, JhonLynch, Paulo Nuin, Marco Rada, GilsonRivas, and Vanessa Verdade.

The research reported herein was presentedby TG to the Department of Ecology,Evolution, and Environmental Biology, Co-lumbia University, in partial fulfillment ofthe degree of Doctor of Philosophy.Chris Raxworthy, Don Melnick, andEleanor Sterling’s critical comments onTG’s dissertation led to significant improve-ments in the final paper. Julian Faivovich,Arnold Kluge, Diego Pol, and Leo Smithprovided encouragement and criticismthroughout the study. That the presentstudy was completed is due in no small partto the willingness of Charles Myers and JohnDaly to share their seemingly endless knowl-edge of dendrobatid frogs at all stages of thisstudy.


Colombian institutional support to TG forfield and laboratory study was provided bythe Corporacion Autonoma Regional delValle del Cauca, EcoAndina, Instituto deCiencias Naturales and Departamento deBiologıa of the Universidad Nacional (Bo-gota), Instituto de Investigaciones de Recur-sos Biologicos Alexander von Humboldt,Serraniaguas, Universidad del Cauca, andUniversidad del Valle. Luis Fernando Garcıagranted TG access to molecular laboratoryfacilities at the Universidad Nacional deColombia (Bogota).

For assistance in all aspects of funding, TGis grateful to Diane Bynum and BarbaraGreen of the AMNH Office of Grants andFellowships. TG was supported by anAMNH Graduate Student Fellowship anda Columbia University Center for Environ-mental Research and Conservation FacultyFellowship. Funding to TG for field workwas provided by the AMNH KalbfleischFund and the Declining Amphibian Popula-tions Task Force (DAPTF). TG’s disserta-tion research was supported by an NSFDoctoral Dissertation Improvement Grant(DEB 0309226) awarded to TG and DRF.This publication is based upon work sup-ported by NASA under award NAG5-13028to DRF and WCW.

JPC thanks Teresa Cristina Avila-Pires forarranging logistics and collaborating onfieldwork at various sites in AmazonianBrazil under a research convenio betweenthe Oklahoma Museum of Natural Historyand Museu Paraense E. Goeldi, and Alber-tina P. Lima for arranging fieldwork at a sitenear Manaus. For help in the field, JPC isgrateful to Laurie J. Vitt, Maria Carmozinade Araujo, Pete Zani, Shawn S. Sartorius,and Veronica R. L. de Oliveira. Permits toconduct research and collect specimens inBrazil were issued by Conselho Nacional deDesenvolvimento Cientıfico e Tecnologico(CNPq, Portaria MCT no. 170, de 28/09/94)and the Instituto Brasileiro do Meio Am-biente e dos Recursos Naturais Renovaveis(IBAMA, no. 073/94-DIFAS), respectively.Amazonian research was funded by NSFgrants DEB-9200779 and DEB-9505518 toLaurie J. Vitt and JPC.

RG acknowledges the generous supportof Mrs. J.B. Fuqua and Ronald Determann.

Conservation International, The Consortiumfor Conservation Medicine, and DAPTFprovided additional funding for amphibianconservation projects. The guidance andgenerosity of Joseph R. Mendelson III,George Rabb, and Chris Shaw werecrucial in gathering the financial and in-tellectual resources necessary to carry outamphibian work in Atlanta. Paul Huggett,Brian Kubicki, and many volunteers pro-vided critical assistance to the amphibianprogram.

CFBH acknowledges Biota-FAPESP (01/13341-3) and CNPq for financial support,and Alice M. Haddad, Marılia T. A.Hartmann, Mark Moffett, and Ricardo J.Sawaya for assistance in field work. Expor-tation permits of Brazilian samples wereissued by CITES Lic. 081968 BR; Autoriza-coes de Acesso e de Remessa de Amostras deComponentes do Patrimonio Genetico nos02001002851/2004; 02001.002669/2004; per-mits for collection were issued by IBAMA/RAN, licencas 057/03 and 054/05.

PJRK appreciates the field assistance ofDavid Rignon, Jose Tarin, and the Couf-fignal family, and thanks the MinistereFrancais de l’Amenagement du Territoire etde l’Environnement and the Section Envir-onnement de la Prefecture de la RegionGuyane for permits (respectively 98/198/AUT and arrete 626 1D/1B/ENV du 07Mai 1999). Fieldwork in French Guianawas partly funded by the Systematics andBiochemical Taxonomy Section of IRSNB.

DBM acknowledges the USGS Herpetol-ogy Project, National Geographic Society,and Conservation International for financialsupport for field work in Guyana.

Tissues obtained by BPN were collectedduring fieldwork supported by grants fromNSF (DEB-0206562) to BPN and P. T.Chippindale and National Geographic(#7509-03) to BPN.

Use of Venezuelan tissues was authorizedby the Ministerio del Medio Ambiente y deRecursos Naturales Renovables (MARNR)in Oficio #41-0180, issued for the project‘‘Evaluacion taxonomica de algunas pobla-ciones de Mannophryne spp. (Anura: Den-drobatidae) en el norte de Venezuela’’under the auspices of the Museo de HistoriaNatural La Salle, Caracas (MHNLS). We


are grateful to Cesar Molina (MARNR)and Celsa Senaris and Gilson Rivas(MHNLS) for their assistance in processingthis permit.

REFERENCES

Aguiar, O., Jr., K.A. Carvalho, A.A. Giaretta, andS.M. Recco-Pimentel. 2004. Cytogenetics ofHylodes and Crossodactylus species (Anura,Leptodactylidae) with comments on Hylodi-nae/Dendrobatidae relationships. Genetica, 121:43–53.

Aguiar, O., Jr., A.A. Garda, A.P. Lima, G.R.Colli, S.N. Bao, and S.M. Recco-Pimentel.2003. Biflagellate spermatozoon of the poison-dart frogs Epipedobates femoralis and Colos-

tethus sp. (Anura, Dendrobatidae). Journal ofMorphology, 255: 114–121.

Aguiar, O., Jr., A.P. Lima, A.A. Giaretta, andS.M. Recco-Pimentel. 2002. Cytogenetic analy-sis of four poison frogs of the Epipedobates

genus (Anura: Dendrobatidae). Herpetologica,58: 293–303.

Aichinger, M. 1987. Annual activity patterns ofanurans in a seasonal neotropical environment.Oecologia, 71: 583–592.

Aichinger, M. 1991. Tadpole transport in relationto rainfall, fecundity and body size in fivespecies of poison-dart frogs from AmazonianPeru. Amphibia-Reptilia, 12: 49–55.

Altig, R., and G.F. Johnston. 1989. Guilds ofanuran larvae: relationships among develop-mental modes, morphologies, and habitats.Herpetological Monographs, 3: 81–109.

Altig, R., and R.W. McDiarmid. 1999a. Bodyplan: development and morphology. In R.W.McDiarmid, and R. Altig (editors), Tadpoles:the biology of anuran larvae: 24–51. Chicago:University of Chicago Press.

Altig, R., and R.W. McDiarmid. 1999b.Diversity: familial and generic characterizations.In R.W. McDiarmid, and R. Altig (editors),Tadpoles: the biology of anuran larvae:295–337. Chicago: University of ChicagoPress.

Amezquita, A., J.V. Rueda-Almonacid, and J.N.Rueda-Martınez. 2004. Rana venenosa de Inger,Epipedobates ingeri, familia Dendrobatidae. In

J.V. Rueda-Almonacid, J.D. Lynch, and A.Amezquita (editors), Libro Rojo de Anfibios deColombia: 346–349. Bogota: Conservacion In-ternacional Colombia, Instituto de CienciasNaturales–Universidad Nacional de Colombia,Mnisterio del Medio Ambiente.

Anonymous. 1985. Remarks on poison frogs. HetPaludarium April, 1985: 1–3.

Ardila-Robayo, M.C. 1979. Status sistematico delgenero Geobatrachus Ruthven, 1915 (Amphibia:Anura). Caldasia, 12: 383–495.

Austin, J.D., S.C. Lougheed, K. Tanner, A.A.Chek, J.P. Bogart, and P.T. Boag. 2002. Amolecular perspective on the evolutionary affin-ities of an enigmatic Neotropical frog, Allo-phryne ruthveni. Zoological Journal of theLinnean Society. London, 134: 335– 346.

Ba-Omar, T.A., J.R. Downie, and W.J.P. Barnes.2000. Development of adhesive toe-pads in thetree-frog (Phyllomedusa trinitatis). Journal ofZoology (London), 250: 267–282.

Badio, B., and J.W. Daly. 1994. Epibatidine,a potent analgetic and nicotonic agonist. Mo-lecular Pharmocology, 45: 563–569.

Baker, A. 2003. Quantitative parsimony andexplanatory power. The British Journal for thePhilosophy of Science, 54: 245–259.

Barbour, T., and G.K. Noble. 1920. Someamphibians from northwestern Peru, with a re-vision of the genera Phyllobates and Telmato-bius. Bulletin of the Museum of ComparativeZoology, 63: 395–427.

Barnes, E.C. 2000. Ockham’s razor and the anti-superfluity principle. Erkenntnis, 53: 353–374.

Bauer, L. 1986. A new genus and a new specificname in the dart poison frog family (Dendro-batidae, Anura, Amphibia). RIPA November,1986: 1–12.

Bauer, L. 1988. Pijlgifkikkers en verwanten: defamilie Dendrobatidae. Het Paludarium 1 No-vember, 1988: 1–6.

Bauer, L. 1994. New names in the familyDendrobatidae (Anura, Amphibia). RIPA Fall,1994: 1–6.

Bernal, X.E., C. Guarnizo, and H. Luddecke.2005. Geographic variation in advertisementcall and genetic structure of Colostethus palma-tus (Anura, Dendrobatidae) from the Colom-bian Andes. Herpetologica, 61: 395–408.

Bhaduri, J.L. 1953. A study of the urinogeneticalsystem of Salientia. Proceedings of the Zoolog-ical Society of Bengal, 6: 1–111.

Biju, S.D., and F. Bossuyt. 2003. New frog familyfrom India reveals an ancient biogeograpic linkwith the Seychelles. Nature, 425: 711–713.

Blommers-Schlosser, R.M.A. 1993. Systematicrelationships of the Mantellinae Laurent 1946(Anura Ranoidea). Ethology Ecology & Evolu-tion, 5: 199–218.

Bogart, J.P. 1970. Systematic problems in theamphibian family Leptodactylidae (Anura) asindicated by karyotypic analysis. Cytogenetics,9: 369–383.

Bogart, J.P. 1973. Evolution of anuran karyotypes.In J.L. Vial (editor), Evolutionary biology of theanurans: contemporary research on major


problems: 337–349. Columbia: University ofMissouri Press.

Bogart, J.P. 1991. The influence of life history onkaryotypic evolution in frogs. In D.M. Greenand S.K. Sessions (editors), Amphibian cytoge-netics and evolution: 233–258. San Diego:Academic Press.

Bogert, C.M. 1960. The influence of sound on thebehavior of amphibians and reptiles. In W.E.Lanyon and W.N. Tavolga (editors), AnimalSounds and Communication. American Insti-tute of Biological Sciences, 7: 137–320.

Boistel, R., and J.-C. de Massary. 1999. Lesamphibiens veneneux de la famille des dendro-batides. Le Courrier de la Nature, 176: 34–39.

Bonacum, J., R. DeSalle, P. O’Grady, D. Olivera,J. Wintermute, and M. Zilversmit. 2001. Newnuclear and mitochondrial primers for system-atics and comparative genomics in Drosophili-dae. Drosophila Information Service, 84:201–204.

Bonaparte, C.L.J.L. 1850. Conspectus system-atum. Herpetologiae et amphibiologiae. Editioaltera reformata. Leiden: Brill.

Bossuyt, F., and M.C. Milinkovitch. 2000. Con-vergent adaptive radiations in Madagascan andAsian ranid frogs reveal covariation betweenlarval and adult traits. Proceedings of the NationalAcademy of Science USA, 97: 6585–9590.

Boulenger, G.A. 1882. Catalogue of the BatrachiaSalientia s. Ecaudata in the Collection of theBritish Museum. London: Taylor and Francis.

Boulenger, G.A. 1888. On the ‘‘nursing’’ habits ofDendrobates. Annals and Magazine of NaturalHistory, Series 6, 2: 122–123.

Boulenger, G.A. 1910. Les batraciens et principla-mement ceux d’Europe. Paris: Octave Doin etFils.

Boulenger, G.A. 1912. Descriptions of new ba-trachians from the Andes of South America,preserved in the British Museum. Annals andMagazine of Natural History, Series 8, 10:185–191.

Bourne, G.R., A.C. Collins, A.M. Holder, andC.L. McCarthy. 2001. Vocal communicationand reproductive behavior of the frog Colos-tethus beebei in Guyana. Journal of Herpetolo-gy, 35: 272–281.

Breder, C.M., Jr. 1946. Amphibians and reptilesof the Rio Chucunaque drainage, Darien,Panama, with notes on their life histories andhabits. Bulletin of the American Museum ofNatural History, 86: 375–436.

Bremer, K. 1994. Branch support and treestability. Cladistics, 10: 295–304.

Brust, D.G. 1993. Maternal brood care byDendrobates pumilio: a frog that feeds its young.Journal of Herpetology, 27: 96–98.

Bunnell, P. 1973. Vocalizations in the territorialbehavior of the frog Dendrobates pumilio.Copeia, 1973: 277–284.

Burton, T.C. 1998a. Pointing the way: thedistribution and evolution of some charactersof the finger muscles of frogs. AmericanMuseum Novitates, 3229: 1–13.

Burton, T.C. 1998b. Variation in the hand andsuperficial throat musculature of Neotropicalleptodactylid frogs. Herpetologica, 54: 53–72.

Caldwell, J.P. 1993. Brazil nut fruit capsules asphytotelmata: interactions among anuran andinsect larvae. Canadian Journal of Zoology, 71:1193–1201.

Caldwell, J.P. 1996. The evolution of myrmeco-phagy and its correlates in poison frogs (familyDendrobatidae). Journal of Zoology (London),240: 75–101.

Caldwell, J.P. 1997. Pair bonding in spotted poisonfrogs. Nature, 385: 211.

Caldwell, J.P. 2005. A new Amazonian species ofCryptophyllobates (Anura: Dendrobatidae).Herpetologica, 61: 449–461.

Caldwell, J.P., and M.C. de Araujo. 1998.Cannibalistic interactions resulting from indis-criminate predatory behavior in tadpoles ofpoison frogs (Anura: Dendrobatidae). Biotro-pica, 30: 92–103.

Caldwell, J.P., and M.C. de Araujo. 2004.Historical and ecological factors influence sur-vivorship in two clades of phytotelm-breedingfrogs (Anura: Bufonidae, Dendrobatidae). Mis-cellaneous Publications, Museum of Zoology,University of Michigan, 193: 11–21.

Caldwell, J.P., and M.C. de Araujo. 2005.Amphibian faunas of two eastern Amazonianrainforest sites in Para, Brazil. OccasionalPapers of the Sam Noble Oklahoma Museumof Natural History, 16: 1–41.

Caldwell, J.P., and V.R.L. de Oliveira. 1999.Determinants of biparental care in the spottedpoison frog, Dendrobates vanzolinii (Anura:Dendrobatidae). Copeia, 1999: 565–575.

Caldwell, J.P., and A.P. Lima. 2003. A newAmazonian species of Colostethus (Anura:Dendrobatidae) with a nidicolous tadpole.Herpetologica, 59: 219–234.

Caldwell, J.P., A.P. Lima, and G.M. Biavatia.2002a. Descriptions of tadpoles of Colostethusmarchesianus and Colostethus caeruleodactylus(Anura: Dendrobatidae) from their type local-ities. Copeia, 2002: 166–172.

Caldwell, J.P., A.P. Lima, and C. Keller. 2002b.Redescription of Colostethus marchesianus (Me-lin, 1941) from its type locality. Copeia, 2002:157–165.

Caldwell, J.P., and C.W. Myers. 1990. A newpoison frog from Amazonian Brazil, with


further revision of the quinquevittatus group ofDendrobates. American Museum Novitates,2988: 1–21.

Callery, E.M., H. Fang, and R.P. Elinson. 2001.Frogs without polliwogs: evolution of anurandirect development. BioEssays, 23: 223–232.

Campbell, J.A., and D.R. Frost. 1993. Anguidlizards of the genus Abronia: revisionary notes,descriptions of new species, a phylogeneticanalysis, and key. Bulletin of the AmericanMuseum of Natural History, 216: 1–121.

Castillo-Trenn, P. 2004. Description of the tadpoleof Colostethus kingsburyi (Anura: Dendrobati-dae) from Ecuador. Journal of Herpetology, 38:600–606.

Cei, J.M. 1980. Amphibians of Argentina. Mon-itore Zoologico Italiano. Nuova Serie, Mono-graphia, 2: 1–609.

Clough, M., and K. Summers. 2000. Phylogeneticsystematics and biogeography of the poisonfrogs: evidence from mitochondrial DNA se-quences. Biological Journal of the LinneanSociety, 70: 515–540.

Cochran, D.M. 1966. Taxonomy and distributionof arrow-poison frogs in Colombia. Memoriasdo Instituto Butantan, 33: 61–65.

Cochran, D.M., and C.J. Goin. 1964. Descriptionof a new frog of the genus Phyllobates fromColombia (Amphibia, Ranidae, Dendrobati-nae). Senckenbergiana Biologica, 45: 255–257.

Cochran, D.M., and C.J. Goin. 1970. Frogs ofColombia. Bulletin of the U.S. National Muse-um, 288: 1–655.

Colgan, D.J., A. McLauchlan, G.D.F. Wilson,S.P. Livingston, G.D. Edgecombe, J. Macar-anas, and G. Cassis. 1999. Histone H3 and U2snRNA DNA sequences and arthropod molec-ular evolution. Australian Journal of Zoology,46: 419–437.

Coloma, L.A. 1995. Ecuadorian frogs of the genusColostethus (Anura: Dendrobatidae). The Uni-versity of Kansas Natural History MuseumMiscellaneous Publication, 87: 1–72.

Cope, E.D. 1865. Sketch of the primary groups ofBatrachia Salientia. Natural History Review,New Series, 5: 97–120.

Cope, E.D. 1867. On the families of the raniformAnura. Journal of the Academy of NaturalSciences of Philadelphia ser. 3, 6: 189–206.

Cope, E.D. 1868. An examination of the Reptiliaand Batrachia obtained by the Orton Expedi-tion to Ecuador and the upper Amazon, withnotes on other species. Proceedings of theAcademy of Natural Sciences of Philadelphia,20: 96–140.

Cope, E.D. 1871. On the system of the BatrachiaAnura of the British Museum Catalogue.American Journal of Science and Arts, 1: 1–6.

Cope, E.D. 1875. Check-list of North American

Batrachia and Reptilia; with a systematic list of

the higher groups, and an essay on geographical

distribution based on specimens contained in

the U.S. National Museum. Bulletin of the U.S.

National Museum, 1: 1–104.

Cope, E.D. 1887. Synopsis of the Batrachia and

Reptilia obtained by H. H. Smith, in the

Province of Mato Grosso, Brazil. Proceedings

of the American Philosophical Society, 24:

44–60.

Cormen, T.H., C.E. Leiserson, R.L. Rivest, and C.

Stein. 2001. Introduction to algorithm, 2nd ed.

Cambridge, MA: The MIT Press.

Crump, M.L. 1971. Quantitative analysis of the

ecological distribution of a tropical herpeto-

fauna. Occasional Papers of the Museum of

Natural History, The University of Kansas, 3:

1–62.

Crump, M.L. 1972. Territoriality and mating

behavior in Dendrobates granuliferus (Anura:

Dendrobatidae). Herpetologica, 22: 195–198.

Cummins, C.P., and M.J.S. Swan. 1995. Variation

in reproductive characteristics of the stream frog

Colostethus trinitatis on the island of Trinidad.

Journal of Tropical Ecology, 11: 603–618.

Cuvier, G.L.C.F.D. 1797. Tableau elementaire de

l’histoire naturelle des animaux. Paris: Bau-

douin.

D’Haese, C.A. 2003. Sensitivity analysis and tree-

fusing: Faster, better. Cladistics, 19: 150–151.

Daly, J.W. 1998. Thirty years of discovering

arthropod alkaloids in amphibian skin. Journal

of Natural Products, 61: 162–172.

Daly, J.W., N.R. Andriamaharavo, M. Andriant-

siferana, and C.W. Myers. 1996. Madagascan

poison frogs (Mantella) and their skin alkaloids.

American Museum Novitates, 3177: 1–34.

Daly, J.W., H.M. Garraffo, and C.W. Myers.

1997. The origin of frog skin alkaloids: an

enigma. Pharmaceutical News, 4: 9–14.

Daly, J.W., H.M. Garraffo, and T.F. Spande.

1993. Amphibian alkaloids. In G.A. Cordell

(editor), The Alkaloids: 185–288. San Diego:

Academic Press.

Daly, J.W., H.M. Garraffo, and T.F. Spande.

1999. Alkaloids from amphibian skins. In S.W.

Pelletier (editor), Alkaloids: chemical and bi-

ological perspectives: 1–161. New York: Perga-

mon.

Daly, J.W., H.M. Garraffo, T.F. Spande, V.C.

Clark, J. Ma, H. Ziffer, and J.F. Cover, Jr.

2003. Evidence for an enantioselective pumlio-

toxin 7-hydroxylase in dendrobatid poison frogs

of the genus Dendrobates. Proceedings of the

National Academy of Science USA, 100:

11092–11097.


Daly, J.W., H.M. Garraffo, T.F. Spande, M.W.Decker, J.P. Sullivan, and M. Williams. 2000.Alkaloids from frog skins: the discovery ofepibatidine and the potential for developingnovel non-opioid analgesics. Natural ProductsReport, 17: 131–135.

Daly, J.W., H.M. Garraffo, T.F. Spande, C.Jaramillo, and A.S. Rand. 1994a. Dietarysource for skin alkaloids of poison frogs(Dendrobatidae)? Journal of Chemical Ecology,20: 943–955.

Daly, J.W., F. Gusovsky, C.W. Myers, M. Yotsu-Yamashita, and T. Yasumoto. 1994b. Firstoccurrence of tetrodotoxin in a dendrobatidfrog (Colostethus inguinalis), with further re-ports for the bufonid genus Atelopus. Toxicon,32: 279–285.

Daly, J.W., and C.W. Myers. 1967. Toxicity ofPanamanian poison frogs (Dendrobates): somebiological and chemical aspects. Science, 156:970–973.

Daly, J.W., C.W. Myers, J.E. Warnick, and E.X.Albuquerque. 1980. Levels of batrachotoxinand lack of sensitivity to its action in poison-dart Frogs (Phyllobates). Science, 208: 1383–85.

Daly, J.W., C.W. Myers, and N. Whittaker. 1987.Further classification of skin alkaloids fromneotropical poison frogs (Dendrobatidae), witha general survey of toxic/noxious substances inthe Amphibia. Toxicon, 25: 1023–1095.

Daly, J.W., S.I. Secunda, H.M. Garraffo, T.F.Spande, A. Wisnieski, and J.F. Cover, Jr. 1994c.An uptake system for dietary alkaloids inpoison frogs (Dendrobatidae). Toxicon, 32:657–663.

Daly, J.W., S.I. Secunda, H.M. Garraffo, T.F.Spande, A. Wisnieski, C. Nishihira, and J.F.Cover, Jr. 1992. Variability in alkaloid profilesin neotropical poison frogs (Dendrobatidae):genetic versus environmental determinants.Toxicon, 30: 887–898.

Daly, J.W., T.F. Spande, and H.M. Garraffo.2005. Alkaloids from amphibian skin: a tabula-tion of over eight-hundred compounds. Journalof Natural Products, 68: 1556–1575.

Darst, C.R., and D.C. Cannatella. 2004. Novelrelationships among hyloid frogs inferred from12S and 16S mitochondrial DNA sequences.Molecular Phylogenetics and Evolution, 31:462–475.

Darst, C.R., P.A. Menendez-Guerrero, L.A. Co-loma, and D.C. Cannatella. 2005. Evolution ofdietary specialization and chemical defense inpoison frogs (Dendrobatidae): a comparativeanalysis. The American Naturalist, 165: 56–69.

Davis, D.D. 1935. A new generic and familyposition for Bufo borbonica. Field Museum of

Natural History Publications Zoologal Series,20: 87–92.

Davis, J.I., and K.C. Nixon. 1992. Populations,genetic variation, and the delimitation ofphylogenetic species. Systematic Biology, 41:421–435.

de Queiroz, K., and J.A. Gauthier. 1990. Phylog-eny as a central principle in taxonomy: phylo-genetic definitions of taxon names. SystematicZoology, 39: 307–322.

de Sa, R.O. 1998. Chondrocranial anatomy andskeletogenesis in Dendrobates auratus. Journalof Herpetology, 32: 205–210.

Desnitskiy, A.G. 2004. Evolutionay transforma-tions of ontogenesis in anuran amphibians.Russian Journal of Developmental Biology,35: 125–130.

Dixon, J.R., and C. Rivero-Blanco. 1985. A newdendrobatid frog (Colostethus) from Venezuelawith notes on its natural history and that ofrelated species. Journal of Herpetology, 19:177–184.

Dole, J.W., and P. Durant. 1974. Courtshipbehavior in Colostethus collaris (Dendrobati-dae). Copeia, 1974: 988–990.

Donnelly, M.A. 1989a. Demographic effects ofreproductive resource supplementation in a ter-ritorial frog, Dendrobates pumilio. EcologicalMonographs, 59: 207–221.

Donnelly, M.A. 1989b. Effects of reproductiveresource supplementation on space–use patternsin Dendrobates pumilio. Oecologia, 81: 212–218.

Donnelly, M.A. 1989c. Reproductive phenologyand age structure of Dendrobates pumilio innortheastern Costa Rica. Journal of Herpetol-ogy, 23: 362–367.

Donnelly, M.A. 1991. Feeding patterns of thestrawberry poison frog, Dendrobates pumilio(Anura: Dendrobatidae). Copeia, 1991:723–730.

Donnelly, M.A., C. Guyer, and R.O. de Sa. 1990.The tadpole of a dart-poison frog Phyllobateslugubris (Anura: Dendrobatidae). Proceedingsof the Biological Society of Washington, 103:427–431.

Donoso-Barros, R. 1965 ‘‘1964’’. Anew Dendro-batidae frog, Prostherapis riveroi from Vene-zuela. Caribbean Journal of Science, 4: 485–489.

Downie, J.R., S.R. Livingstone, and J.R. Cor-mack. 2001. Selection of tadpole deposition sitesby male Trinidadian stream frogs, Mannophrynetrinitatis (Dendrobatidae): an exampe of anti–predator behaviour. Herpetological Journal, 11:91–100.

Downie, J.R., E. Robinson, R.J. Linklater-McLennan, E. Somerville, and N. Kamenos.2005. Are there costs to extended larval trans-port in the Trinidadian stream frog, Manno-


phryne trinitatis (Dendrobatidae)? Journal ofNatural History, 39: 2023–2034.

Dubois, A. 1982. Dendrobates Wagler, 1830 andDendrobatidae Cope, 1865 (Amphibia, Anura):proposed conservation. The Bulletin of Zoolog-ical Nomenclature, 39: 267–278.

Dubois, A. 1984. La nomenclature supragemer-ique des amphibiens anoures. Memoires duMuseum national d’Histoire naturelle, 131:1–64.

Dubois, A. 1992. Notes sur la classification desRanidae (Amphibiens anoures). Bulletin Men-suel de la Societe Linneenne de Lyon, 61:305–352.

Duellman, W.E. 1966. Aggressive behavior indendrobatid frogs. Herpetologica, 22: 217–221.

Duellman, W.E. 1967. Additional studies ofchromosomes of anuran amphibians. Systemat-ic Zoology, 16: 38–43.

Duellman, W.E. 1975. On the classification offrogs. Occasional Papers of the Museum ofNatural History, The University of Kansas, 42:1–14.

Duellman, W.E. 1978. The biology of an equato-rial herpetofauna in Amazonian Ecuador. Uni-versity of Kansas Museum of Natural HistoryMiscellaneous Publication, 65: 1–352.

Duellman, W.E. 1995. Temporal fluctuations inabundances of anuran amphibians in a seasonalAmazonian rainforest. Journal of Herpetology,29: 13–21.

Duellman, W.E. 2004. Frogs of the genus Colos-tethus (Anura; Dendrobatidae) in the Andes ofnorthern Peru. Scientific Papers Natural HistoryMuseum, The University of Kansas, 35: 1–49.

Duellman, W.E., and J.D. Lynch. 1969. Descrip-tions of Atelopus tadpoles and their relevance toatelopodid classification. Herpetologica, 25:231–240.

Duellman, W.E., and J.D. Lynch. 1988. Anuranamphibians from the Cordillera de Cutucu,Ecuador. Proceedings of the Academy ofNatural Sciences of Philadelphia, 140: 125–142.

Duellman, W.E., and J.R. Mendelson III. 1995.Amphibians and reptiles from northern Depar-tamento Loreto, Peru: taxonomy and biogeog-raphy. University of Kansas Museum of Natu-ral History Science Bulletin, 55: 329–376.

Duellman, W.E., and J.E. Simmons. 1988. Twonew species of dendrobatid frogs, genus Colos-tethus, from the Cordillera del Condorm Ecua-dor. Proceedings of the Academy of NaturalSciences of Philadelphia, 140: 115–124.

Duellman, W.E., and L. Trueb. 1986. Biology ofamphibians. New York: McGraw-Hill.

Duellman, W.E., and E.R. Wild. 1993. Anuranamphibians from the Cordillera de Huanca-bamba, northern Peru: systematics, ecology,

and biogeography. Occasional Papers of theMuseum of Natural History University ofKansas, 157: 1–53.

Dumbacher, J.P., A. Wako, S.R. Derrickson, A.Samuelson, T.F. Spande, and J.W. Daly. 2004.Melyrid beetles (Choresine): a putative sourcefor the batrachotoxin alkaloids found in poison-dart frogs and toxic passerine birds. Proceedingsof the National Academy of Science USA, 101:15857–15860.

Dumeril, A-.M.-C., and G. Bibron. 1841. Erpeto-logie general ou histoire naturelle complete desreptiles. Paris: Librairie Encyclopedique deRoret.

Dunlap, D.G. 1960. The comparative morphologyof the pelvic appendage in the Salientia. Journalof Morphology, 106: 1–76.

Dunn, E.R. 1924. Some Panamanian frogs.Occasional Papers of the Museum of Zoology,University of Michigan, 151: 1–17.

Dunn, E.R. 1931. New frogs from Panama andCosta Rica. Occasional Papers of the BostonSociety of Natural History, 5: 385–401.

Dunn, E.R. 1933. Amphibians and reptiles from ElValle de Anton, Panama. Occasional Papers ofthe Boston Society of Natural History, 8: 65–79.

Dunn, E.R. 1940. New and noteworthy herpeto-logical material from Panama. Proceedings ofthe Academy of Natural Sciences of Philadel-phia, 92: 105–122.

Dunn, E.R. 1941. Notes on Dendrobates auratus.Copeia, 1941: 88–93.

Dunn, E.R. 1944. Notes on the breeding habits ofthe tadpole-carrying frog Hyloxalus granuliven-tris. Caldasia, 2: 397–398.

Dunn, E.R. 1957. Neotropical frog genera: Pros-therapis versus Hyloxalus, with remarks onPhyllobates. Copeia, 1957: 77–78.

Durant, P., and J.W. Dole. 1975. Aggressivebehavior in Colostethus (5Prostherapis) collaris(Anura: Dendrobatidae). Herpetologica, 31:23–26.

Eaton, T. 1941. Notes on the life history ofDendrobates auratus. Copeia, 1941: 93–95.

Edwards, S.R. 1971. Taxonomic notes on SouthAmerican Colostethus with descriptions of twonew species (Amphibia, Dendrobatidae). Pro-ceedings of the Biological Society of Washing-ton, 84: 147–162.

Edwards, S.R. 1974a. A phenetic analysis of thegenus Colostethus (Anura: Dendrobatidae).Ph.D. dissertation. University of Kansas,Lawrence, 419 pp.

Edwards, S.R. 1974b. Taxonomic notes on SouthAmerican dendrobatid frogs of the genusColostethus. Occasional Papers of the Museumof Natural History, The University of Kansas,30: 1–14.


Emerson, S.B. 1982. Frog postcranial morphology:

identification of a functional complex. Copeia,

1982: 603–613.

Emerson, S.B., and D. Diehl. 1980. Toe pad

morphology and adhesive mechanisms in frogs.

Biological Journal of the Linnean Society, 13:

199–216.

Emerson, S.B., C.M. Richards, R.C. Drewes, and

K.M. Kjer. 2000. On the relationships among

ranoid frogs: a review of the evidence. Herpe-

tologica, 56: 209–230.

Faivovich, J. 1998. Comments on the larvae of the

Argentine species of the genus Crossodactylus

(Leptodactylidae, Hylodinae). Alytes, 16: 61–67.

Faivovich, J., C.F.B. Haddad, P.C.A. Garcia,

D.R. Frost, J.A. Campbell, and W.C. Wheeler.

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.

Fandino, M.C., H. Luddecke, and A. Amezquita.

1997. Vocalisation and larval transportation of

male Colostethus subpunctatus (Anura: Dendro-

batidae). Amphibia-Reptilia, 18: 39–48.

Farris, J.S. 1967. The meaning of relationship and

taxonomic procedure. Systematic Zoology, 16:

44–51.

Farris, J.S. 1973. On comparing the shapes of

taxonomic trees. Systematic Zoology, 22: 50–54.

Farris, J.S. 1979. The information content of the

phylogenetic system. Systematic Zoology, 28:

483–519.

Farris, J.S. 1983. The logical basis of phylogenetic

analysis. In N.I. Platnick, and V.A. Funk

(editors), Advances in Cladistics: 7–36. New

York: Columbia University Press.

Farris, J.S. 1989. The retention index and the

rescaled consistency index. Cladistics, 5:

417–419.

Feller, A.E., and S.B. Hedges. 1998. Molecular

evidence for the early history of living amphi-

bians. Molecular Phylogenetics and Evolution,

9: 509–516.

Fernandez, K. 1926. Sobre la biologıa y reproduc-

cion de algunos batracios argentinos (segunda

parte). Boletın de la Academia Nacional de

Ciencias, 29: 271–320.

Fitch, R.W., H.M. Garraffo, T.F. Spande, H.J.C.

Yeh, and J.W. Daly. 2003. Bioassay-guided

isolation of epiquinamide, a novel quinolizidine

alkaloid and nicotinic agonist from an Ecua-

doran poison frog, Epipedobates tricolor. Jour-

nal of Natural Products, 66: 1345–1350.

Fitzinger, L.I. 1843. Systema reptilium. Fasciculus

primus. Amblyglossae. Braumuller et Seidel,

Wien.

Fitzinger, L.I. 1860. Die Ausbeute der osterrei-

chischen Naturforscher an Saugethieren und

Reptilien wahrend der Weltumsegelung Sr.

Majestat Fregatte Novara. Sitzungsberichte

Oesterreichische Akademie der Wissenschaften,

42: 383–416.

Flier, J., M.W. Edwards, J.W. Daly, and C.W.

Myers. 1980. Widespread occurrence in frogs

and toads of skin compounds interacting with

the Ouabain site of Na+, K+-ATPase. Science,

208: 503–505.

Folmer, O., M. Black, W. Hoeh, R. Lutz, and R.

Vrijenhoek. 1994. DNA primers for amplifica-

tion of mitochondrial cytochrome oxidase c

subunit I from diverse metazoan invertebrates.

Molecular Marine Biology and Biotechnology,

3: 294–299.

Ford, L.S. 1989. The phylogenetic position of

poison-dart frogs (Dendrobatidae): reassess-

ment of the neobatrachian phylogeny with

comments on complex character systems.

Ph.D. dissertation, University of Kansas, Law-

rence, 307 pp.

Ford, L.S. 1993. The phylogenetic position of the

dart-poison frogs (Dendrobatidae) among anur-

ans: an examination of the competing hypoth-

eses and their characters. Ethology Ecology &

Evolution, 5: 219–231.

Ford, L.S., and D.C. Cannatella. 1993. The major

clades of frogs. Herpetological Monographs, 7:

94–117.

Formas, J.R. 1989. A new species of Eupsophus

(Amphibia: Anura: Leptodactylidae) from

southern Chile. Proceedings of the Biological

Society of Washington, 102: 568–576.

Frost, D.R. 1986. A new species of Colostethus

(Anura: Dendrobatidae) from Ecuador. Pro-

ceedings of the Biological Society of Washing-

ton, 99: 214–217.

Frost, D.R. 2000. Species, descriptive efficiency,

and progress in systematics. In R.C. Bruce, R.G.

Jaeger, and L.D. Houck (editors), The biology

of plethodontid salamanders: 7–29. New York:

Kluwer Academic / Plenum Publishers.

Frost, D.R., H.M. Crafts, L.A. Fitzgerald, and

T.A. Titus. 1998. Geographic variation, species

recognition, and molecular evolution of cyto-

chrome oxidase I in the Tropidurus spinulosus

complex (Iguania: Tropiduridae). Copeia, 1998:

839–851.

Frost, D.R., T. Grant, J. Faivovich, R. Bain, A.

Haas, C.F.B. Haddad, R.O. de Sa, S.C.

Donnellan, C.J. Raxworthy, M. Wilkinson, A.

Channing, J.A. Campbell, B.L. Blotto, P.

Moler, R.C. Drewes, R.A. Nussbaum, J.D.

Lynch, D. Green, and W.C. Wheeler. 2006.

The amphibian tree of life. Bulletin of the


American Museum of Natural History, 297:1–370.

Frost, D.R., and A.G. Kluge. 1994. A consider-ation of epistemology in systematic biology,with special reference to species. Cladistics, 10:259–294.

Frost, D.R., M.T. Rodrigues, T. Grant, and T.A.Titus. 2001. Phylogenetics of the lizard genusTropidurus (Squamata: Tropiduridae: Tropidur-inae): direct optimization, descriptive efficiency,and sensitivity analysis of congruencebetween molecular data and morphology. Mo-lecular Phylogenetics and Evolution, 21: 352–371.

Funkhouser, J.W. 1956. New frogs from Ecuadorand southwestern Colombia. Zoologica, 41:73–79.

Gadow, H. 1901. Amphibia and Reptiles [sic].London: MacMillan.

Garda, A.A., G.R. Colli, O. Aguiar-Junior, S.M.Recco-Pimentel, and S.N. Bao. 2002. Theultrastructure of the spermatozoa of Epipedo-bates flavopictus (Amphibia, Anura, Dendroba-tidae), with comments on its evolutionarysignificance. Tissue and Cell, 34: 356–364.

Garey, M.R., R.L. Graham, and D.S. Johnson.1977. The complexity of computing Steinerminimal trees. SIAM Journal on AppliedMathematics, 32: 835–859.

Garey, M.R., and D.S. Johnson. 1977. Therectilinear Steiner tree problem is NP–complete.SIAM Journal on Applied Mathematics, 32:826–834.

Garraffo, H.M., P. Jain, T.F. Spande, and J.W.Daly. 1997. Alkaloid 223A: the first trisubsti-tuted indolizidine from dendrobatid frogs.Journal of Natural Products, 60: 2–5.

Garraffo, H.M., P. Jain, T.F. Spande, J.W. Daly,H.B. Jones, L.J. Smith, and V.E. Zottig. 2001.Structure of alkaloid 275A, a novel 1-azabicy-clo[5.3.0]decane from a dendrobatid frog, Den-drobates lehmanni: synthesis of the tetrahydro-diastereomers. Journal of Natural Products, 64:421–427.

Garraffo, H.M., T.F. Spande, J.W. Daly, A.Baldessari, and E.G. Gros. 1993. Alkaloidsfrom bufonid toads (Melanophryniscus): deca-hydroquinolines, pumiliotoxins and homopumi-liotoxins, indolizidines, pyrrolizodines, and qui-nolizidines. Journal of Natural Products, 56:357–373.

Gerhardt, H.C., and J. Rheinlaender. 1980.Accuracy of sound localization in a miniaturedendrobatid frog. Naturwissenschaften, 67:362–363.

Giaretta, A.A., W.C.A. Bokermann, and C.F.B.Haddad. 1993. A review of the genus Megae-losia (Anura: Leptodactylidae) with a descrip-

tion of a new species. Journal of Herpetology,27: 276–285.

Giaretta, A.A., and K.G. Facure. 2003. Cyclor-amphus boraceiensis (flattened waterfall frog).Clutch attendance. Herpetological Review, 34:50.

Giaretta, A.A., and K.G. Facure. 2004. Repro-ductive ecology and behavior of Thoropamiliaris (Spix, 1824) (Anura, Leptodactylidae,Telmatobiinae). Biota Neotropical, 4: 1–10.

Gibson, R.C., and K.R. Buley. 2004. Maternalcare and obligatory oophagy in Leptodactylusfallax: A new reproductive mode in frogs.Copeia, 2004: 128–135.

Goebel, A.M., J.M. Donnelly, and M.E. Atz.1999. PCR primers and amplification methodsfor 12S ribosomal DNA, the control region,cytochrome oxidase I, and cytochrome b inbufonids and other frogs, and an overview ofPCR primers which have amplified DNA inamphibians successfully. Molecular Phyloge-netics and Evolution, 11: 163–199.

Goin, C.J., O.B. Goin, and G.R. Zug. 1978.Introduction to herpetology, 3rd ed. SanFrancisco: W. H. Freeman and Co.

Goloboff, P.A. 1999. Analyzing large data sets inreasonable times: solutions for composite opti-ma. Cladistics, 15: 415–428.

Goloboff, P.A., and J.S. Farris. 2001. Methods ofquick consensus estimation. Cladistics, 17:S26–S34.

Goloboff, P.A., J.S. Farris, and K.C. Nixon. 2003.TNT: tree analysis using new technology, Pro-gram and documentation available at www.zmuc.dk/public/phylogeny.

Goodman, D.E. 1971. Territorial behavior ina neotropical frog, Dendrobates granuliferus.Copeia, 1971: 365–370.

Gosner, K.L. 1960. A simplified table for staginganuran embryos and larvae with notes onidentification. Herpetologica, 16: 183–190.

Graham, C.H., S.R. Ron, J.C. Santos, C.J.Schneider, and C. Moritz. 2004. Integratingphylogenetics and environmental niche modelsto explore speciation mechanisms in dendroba-tid frogs. Evolution, 58: 1781–1793.

Grant, T. 1998. Una nueva especie de Colostethusdel grupo edwardsi de Colombia. Revista de laAcademia Colombiana de Ciencias Exactas,Fisicas y Naturales, 22: 423–428.

Grant, T. 2002. Testing methods: The evaluationof discovery operations in evolutionary biology.Cladistics, 18: 94–111.

Grant, T. 2004. On the identities of Colostethusinguinalis (Cope, 1868) and C. panamensis(Dunn, 1933), with comments on C. latinasus(Cope, 1863) (Anura: Dendrobatidae). Ameri-can Museum Novitates, 3444: 1–24.


Grant, T., and M.C. Ardila-Robayo. 2002. A newspecies of Colostethus (Anura: Dendrobatidae)from the eastern slopes of the CordilleraOriental of Colombia. Herpetologica, 58:252–260.

Grant, T., and F. Castro-Herrera. 1998. The cloudforest Colostethus (Anura, Dendrobatidae) ofa region of the Cordillera Occidental ofColombia. Journal of Herpetology, 32: 378–392.

Grant, T., D.R. Frost, R. Ibanez, D., C.W. Myers,and J.M. Savage. 2006. Hyloxalus panamensisDunn, 1933: proposed emendation of spellingfrom Hyloxalus panamansis (currently Colos-tethus panamansis; Amphibia, Anura). Bulletinof Zoological Nomenclature, 63: 39–41.

Grant, T., E.C. Humphrey, and C.W. Myers.1997. The median lingual process of frogs:a bizarre character of Old World ranoidsdiscovered in South American dendrobatids.American Museum Novitates, 3212: 1–40.

Grant, T., and A.G. Kluge. 2003. Data explora-tion in phylogenetic inference: Scientific, heu-ristic, or neither. Cladistics, 19: 379–418.

Grant, T., and A.G. Kluge. 2004. Transformationseries as an ideographic character concept.Cladistics, 20: 23–31.

Grant, T., and L.O. Rodrıguez. 2001. Two newspecies of frogs of the genus Colostethus(Dendrobatidae) from Peru and a redescriptionof C. trilineatus (Boulenger, 1883). AmericanMuseum Novitates, 3355: 1–24.

Graybeal, A. 1997. Phylogenetic relationships ofbufonid frogs and tests of alternate macroevo-lutionary hypotheses characterizing their radia-tion. Zoological Journal of the Linnean Society,119: 297–338.

Green, D.M. 1979. Tree frog toe pads: compara-tive surface morphology using scanning electronmicroscopy. Canadian Journal of Zoology, 57:2033–2046.

Griffiths, I. 1954. On the ‘‘otic element’’ inAmphibia Salientia. Proceedings of the Zoolog-ical Society of London, 124: 3–50.

Griffiths, I. 1959. The phylogeny of Sminthilluslimbatus and the status of the Brachycephalidae(Amphibia Salientia). Proceedings of the Zoo-logical Society of London, 132: 457–487.

Griffiths, I. 1963. The phylogeny of the Salientia.Biological Review, 38: 241–292.

Gunther, A. 1858. Catalogue of the BatrachiaSalientia in the Collection of the BritishMuseum. London: Taylor and Francis.

Haas, A. 1995. Cranial features of dendrobatidlarvae (Amphibia: Anura: Dendrobatidae).Journal of Morphology, 224: 241–264.

Haas, A. 2001. Mandibular arch musculature ofanuran tadpoles, with comments on homologies

of amphibian jaw muscles. Journal of Morphol-ogy, 247: 1–33.

Haas, A. 2003. Phylogeny of frogs as inferred fromprimarily larval characters (Amphibia: Anura).Cladistics, 19: 23–89.

Haddad, C.F.B., J. Faivovich, and P.C.A. Garcia.2005. The specialized reproductive mode of thetreefrog Aplastodiscus perviridis (Anura: Hyli-dae). Amphibia Reptilia, 26: 87–92.

Haddad, C.F.B., and A.A. Giaretta. 1999. Visualand acoustic communication in the Braziliantorrent frog, Hylodes asper (Anura: Leptodac-tylidae). Herpetologica, 55: 324–333.

Haddad, C.F.B., and M. Martins. 1994. Fourspecies of Brazilian poison frogs related toEpipedobates pictus (Dendrobatidae): yaxonomyand natural history. Herpetologica, 50: 282–295.

Hall, T.A. 1999. BioEdit: a user-friendly biologicalsequence alignment editor and analysis programfor Windows 95/98/NT. Nucleic Acids Sympo-sium Series, 41: 95–98.

Hardy, J.D.J. 1983. A new frog of the genusColostethus from the island of Tobago, WestIndies (Anura: Dendrobatidae). Bulletin of theMaryland Herpetological Society, 19: 47–57.

Hartmann, M.T., L.O.M. Giasson, P.A. Hart-mann, and C.F.B. Haddad. 2005. Visual com-munication in Brazilian species of anurans fromthe Atlantic forest. Journal of Natural History,39: 1675–1685.

Hay, J.M., I. Ruvinsky, S.B. Hedges, and L.R.Maxson. 1995. Phylogenetic relationships ofamphibian families inferred from DNA se-quences of mitochondrial 12S and 16S ribosom-al RNA genes. Molecular Biology and Evolu-tion, 12: 928–937.

Hecht, M.K. 1963. A reevaluation of the earlyhistory of the frogs. Part II. Systematic Zoolo-gy, 12: 20–35.

Hedges, S.B. 1994. Molcular evidence for theorigin of birds. Proceedings of the NationalAcademy of Science USA, 91: 2621–2624.

Hedges, S.B., and L.R. Maxson. 1993. A molec-ular perspective on lissamphibian phylogeny.Herpetological Monographs, 7: 27–42.

Hennig, W. 1966. Phylogenetic systematics. Chi-cago: University of Illinois Press.

Heyer, W.R. 1970. Studies on the frogs of thegenus Leptodactylus (Amphibia: Leptodactyli-dae). VI. Biosystematics of the melanonotusgroup. Natural History Museum of Los AngelesCounty Contributions in Science, 191: 1–48.

Heyer, W.R. 1983. Variation and systematics offrogs of the genus Cycloramphus (Amphibia,Leptodactylidae). Arquivos de Zoologia Museuda Universidade de Sao Paulo, 30: 235–339.

Heyer, W.R., and R.I. Crombie. 1979. Naturalhistory notes on Craspedoglossa stejnegeri and


Thoropa petropolitana (Amphibia: Salientia:Leptodactylidae). Journal of the WashingtonAcademy of Science, 69: 17–20.

Heyer, W.R., A.S. Rand, C.A.G. Cruz, O.L.Peixoto, and C.E. Nelson. 1990. Frogs ofBoraceia. Arquivos de Zoologia Museu daUniversidade de Sao Paulo, 31: 231–410.

Hillis, D.M., L.K. Ammerman, M.T. Dixon, andR.O. de Sa. 1993. Ribosomal DNA and thephylogeny of frogs. Herpetological Mono-graphs, 7: 118–131.

Hillis, D.M., and M.T. Dixon. 1991. RibosomalDNA: molecular evolution and and phyloge-netic inference. The Quarterly Review of Bi-ology, 66: 411–453.

Hodl, W., and A. Amezquita. 2001. Visualsignaling in anuran amphibians. In M.J. Ryan(editor), Anuran Communication: 121–141.Washington: Smithsonian Institution Press.

Hoff, K.v., A.R. Blaustein, R.W. McDiarmid, andR. Altig. 1999. Behavior: interactions and theirconsequences. In R.W. McDiarmid, and R.Altig (editors), Tadpoles: the biology of anuranlarvae: 215–239. Chicago: University of ChicagoPress.

Holthius, L.B., and A. Dubois. 1983. Commentson the proposed conservation of DendrobatesWagler, 1830 and Dendrobatidae Cope, 1965.Z.N.(S.)1930. Bulletin of Zoological Nomencla-ture, 40: 197–199.

Hoogmoed, M.S. 1969. Notes on the herpetofaunaof Surinam: 3. A new species of Dendrobates(Amphibia Salientia, Dendrobatidae) from Sur-inam. Zoologische Mededelingen Leiden, 44:133–141.

Ibanez, R., and E.M. Smith. 1995. Systematicstatus of Colostethus flotator and C. nubicola(Anura: Dendrobatidae) in Panama. Copeia,1995: 446–456.

ICZN. 1999. International Code of ZoologicalNomenclature. London: International Trust forZoological Nomenclature.

Inger, R.F. 1967. The development of a phylogenyof frogs. Evolution, 21: 369–384.

Inger, R.F., and H.K. Voris. 1988. Taxonomicstatus and reproductive biology of Borneantadpole caryying frogs. Copeia, 1988:1060–1061.

Izecksohn, E., and E. Gouvea. 1987 ‘‘1985’’. Novaespecie de Megaelosia de Itatiaia, Estado do Riode Janeiro. Arquivos de Universidade FederalRural do Rio de Janeiro, 8: 17–22.

Junca, F.A. 1996. Parental care and egg mortalityin Colostethus stepheni. Journal of Herpetology,30: 292–294.

Junca, F.A. 1998. Reproductive biology of Colos-tethus stepheni and Colostethus marchesianus(Dendrobatidae), with the description of a new

anuran mating behavior. Herpetologica, 54:

377–387.

Junca, F.A., R. Altig, and C. Gascon. 1994.

Breeding biology of Colostethus stepheni, a den-

drobatid with a nontrasnported nidicolous

tadpole. Copeia, 1994: 747–750.

Jungfer, K.-H. 1985. Beitrag zur Kenntnis von

Dendrobates speciosus O. Schmidt, 1857 (Sal-

ienti: Dendrobatidae). Salamandra, 21:

263–280.

Jungfer, K.-H. 1989. Pfeilgiftfrosche der Gattung

Epipedobates mit rot granuliertem Rucken aus

dem Oriente von Ecuador und Peru. Salaman-

dra, 25: 81–98.

Jungfer, K.-H., H. Birkhahn, V. Kulpmann, and

K. Wassmann. 1996a. Haltung und Fortpflan-

zung von Dendrobates fulguritus Silverstone,

1975, mit Anmerkungen zur Gattung Minyo-

bates Myers, 1987. Herpetofauna, 15: 19–27.

Jungfer, K.-H., and W. Bohme. 2004. A new

poison-dart frog (Dendrobates) from northern

central Guyana (Amphibia: Anura: Dendroba-

tidae). Salamandra, 99–103.

Jungfer, K.-H., S. Lotters, and D. Jorgens. 2000.

Der kleinste Pfeilgiftfrosch-eine neue Dendro-

bates-Art aus West-Panama. Herpetofauna, 22:

11–18.

Jungfer, K.-H., and P. Weygoldt. 1999. Biparental

care in the tadpole-feeding Amazonian treefrog

Osteocephalus oophagus. Amphibia–Reptilia,

20: 235–249.

Jungfer, K.-H., P. Weygoldt, and N. Juraske.

1996b. Dendrobates vicentei, ein neuer Pfeilgift-

frosch aus Zentral–Panama. Herpetofauna, 18:

17–26.

Kaiser, H., and R. Altig. 1994. The atypical

tadpole of the dendrobatid frog, Colostethus

chalcopis, from Martinique, French Antilles.

Copeia, 1994: 374–378.

Kaiser, H., L.A. Coloma, and H.M. Gray. 1994. A

new species of Colostethus (Anura: Dendroba-

tidae) from Martinique, French Antilles. Her-

petologica, 50: 23–32.

Kaiser, H., C. Steinlein, W. Feichtinger, and M.

Schmid. 2003. Chromosome banding of six

dendrobatid frogs (Colostethus, Mannophryne).

Herpetologica, 59: 203–218.

Kaplan, M. 1994. Analysis of some long-standing

controversies concerning the pectoral girdle of

Atelopus (Bufonidae) using ontogenetic studies.

Journal of Herpetology, 28: 128–131.

Kaplan, M. 1995. On the presence of overlap

during the development of the pectoral girdle of

Colostethus subpunctatus (Amphibia: Anura)

and its relevance in the classification of Den-

drobatidae. Journal of Herpetology, 29: 300–

304.


Kaplan, M. 1997. A new species of Colostethus

from the Sierra Nevada de Santa Marta

(Colombia) with comments on intergeneric

relationships within the Dendrobatidae. Journal

of Herpetology, 31: 369–375.

Kaplan, M. 2000. The pectoral girdles of Rana

rugulosa (Ranidae) and Nesomantis thomasseti

(Sooglossidae). Herpetologica, 56: 188–195.

Kaplan, M. 2001. On the relevance of the

character ‘‘absence of epicoracoid horns’’ in

the systematics of anurans. Alytes, 19: 196–204.

Kaplan, M. 2004. Evaluation and redefinition of

the states of anuran pectoral girdle architecture.

Herpetologica, 60: 84–97.

Kluge, A.G. 1989. A concern for evidence and

a phylogenetic hypothesis of relationships

among Epicrates (Boidae, Serpentes). Systemat-

ic Zoology, 38: 7–25.

Kluge, A.G. 1990. Species as historical individuals.

Biology and Philosophy, 5: 417–431.

Kluge, A.G. 1993. Aspidites and the phylogeny of

phythonine snakes. Records of the Australia

Museum Supplement, 19: 1–77.

Kluge, A.G. 2005. Taxonomy in theory and

practice, with arguments for a new phylogenetic

system of taxonomy. In M.A. Donnelly, B.I.

Crother, C. Guyer, M.H. Wake, and M.E.

White (editors), Ecology and evolution in the

tropics: a herpetological perspective: 7–47.

Chicago: University of Chicago Press.

Kluge, A.G., and J.S. Farris. 1969. Quantitative

phyletics and the evolution of anurans. System-

atic Zoology, 18: 1–32.

Kluge, A.G., and T. Grant. 2006. From conviction

to anti-superfluity: old and new justifications for

parsimony in phylogenetic inference. Cladistics,

22: 276–288.

Kneller, M. 1982. Fortpflanzung von Dendrobates

reticulatus im naturlichen Lebensraum und im

Terrarium. Aquarium, 153: 148–151.

Kneller, M., and K. Henle. 1985. Ein neuer

Blattsteiger-Frosch (Salientia: Dendrobatidae:

Phyllobates) aus Peru. Salamandra, 21: 62–69.

Kocher, T.D., W.K. Thomas, A. Meyer, S.V.

Edwards, S. Paabo, F.X. Villablanca, and A.C.

Wilson. 1989. Dynamics of mitochondrial DNA

evolution in Animals: amplification and se-

quencing with conserved primers Proceedings

of the National Academy of Science USA, 86:

6196–6200.

Kohler, J. 2000. Amphibian diversity in Bolivia:

a study with special reference to montane forest

regions. Bonner Zoologische Monographien,

48: 1–243.

Kok, P.J.R. 2000. A survey of the anuran fauna of

Montagne Belvedere, county of Saul, French

Guiana: Field list with comments on taxonomy

and ecology. British Herpetological SocietyBulletin, 71: 6–26.

Kok, P.J.R. 2001. Addenda to ‘A survey of theanuran fauna of Montagne Belvedere, county ofSaul, French Guiana’. British HerpetologicalSociety Bulletin, 73: 1.

Kuramoto, M. 1990. A list of chromosomenumbers of anuran amphibians. Bulletin of theFukuoka University of Education, 39: 83–127.

La Marca, E. 1985. A new species of Colostethus(Anura: Dendrobatidae) from the Cordillera deMerida, northern Andes, South America. Occa-sional Papers of the Museum of Zoology,University of Michigan, 710: 1–10.

La Marca, E. 1989. A new species of collared frog(Anura: Dendrobatidae: Colostethus) from Ser-rania de Portuguesa, Andes of Estado Lara,Venezuela. Amphibia–Reptilia, 10: 175–183.

La Marca, E. 1992. Catalogo taxonomico, biogeo-grafico y bibliografico de las ranas de Vene-zuela. Cuadernos Geograficos Universidad delos Andes, 9: 1–197.

La Marca, E. 1993. Phylogenetic relationships andtaxonomy of Colostethus mandelorum (Anura:Dendrobatidae), with notes on coloration,natural history, and description of the tadpole.Bulletin of the Maryland Herpetological Socie-ty, 29: 4–19.

La Marca, E. 1994. Descripcion de un nuevogenero de ranas (Amphibia: Dendrobatidae) dela Cordillera de Merida, Venezuela. Anuario deInvestigacion, 1991: 39–41.

La Marca, E. 1995. Biological and systematicsynopsis of a genus of frogs from northernmountains of South America (Anura: Dendro-batidae: Mannophryne). Bulletin of the Mary-land Herpetological Society, 31: 40–78.

La Marca, E. 1996 ‘‘1994’’. Taxonomy of the frogsof the genus Mannophryne (Amphibia; Anura;Dendrobatidae). Publicaciones de la Asociacionde Amigos de Donana, 4: 1–75.

La Marca, E. 1998 ‘‘1996’’. Ranas del generoColostethus (Amphibia: Anura; Dendrobatidae)de la Guayana venezolana, con la descripcion desiete especies nuevas. Publicaciones de la Aso-ciacion de Amigos de Donana, 9: 1–64.

La Marca, E. 2004. Systematic status of anenigmatic and possibly endangered dendrobatidfrog (Colostethus dunni) from the Valley ofCaracas, northern Venezuela. Herpetotropicos,1: 19–28.

La Marca, E., J. Manzanilla-Puppo, and A.Mijares-Urrutia. 2004. Revision taxonomicadel Colostethus del norte de Venezuela confun-dido durante largo tiempo con C. brunneus.Herpetotropicos, 1: 40–50.

La Marca, E., M. Vences, and S. Lotters. 2002.Rediscovery and mitochondrial relationships of


the dendrobatid frog Colostethus humilis suggest

parallel colonization of the Venezuelan Andes.

Studies on Neotropical Fauna and Environ-

ment, 37: 233–240.

Lakatos, I. 1978. The methodology of scientific

research programmes. Cambridge, UK: Cam-

bridge University Press.

Laurent, R.F. 1942. Note sur les procoeliens

firmisternes (Batrachia Anura). Bulletin du

Musee Royal d’Histoire Naturelle de Belgique,

18: 1–20.

Laurent, R.F. ‘‘1979’’ 1980. Esquisse d’une phylo-

genese des anoures. Bulletin de la Societe

Zoologique de France, 104: 397–422.

Lavilla, E.O. 1987. La larva de Rhinoderma

darwinii D. & B. (Anura: Rhinodermatidae).

Acta Zoologica Lilloana, 39: 81–88.

Lavilla, E.O., and P. Ergueta, S. 1995. Una nueva

especie de Telmatobius (Anura, Leptodactyli-

dae) de la ceja de montana de La Paz (Bolivia).

Alytes, 13: 45–51.

Lehtinen, R.M., M.J. Lannoo, and R.J. Wasser-

sug. 2004. Phytotelm-breeding anurans: past,

present and future research. Miscellaneous

Publications, Museum of Zoology, University

of Michigan, 193: 1–9.

Lehtinen, R.M., and R.A. Nussbaum. 2003.

Parental care: a phylogenetic perspective. In

B.G.M. Jamieson (editor), Reproductive biolo-

gy and phylogeny of Anura: 342–386. Enfield,

NH: Science Publishers, Inc.

Leon, P.E. 1970. Report of the chromosome

number of some Costa Rican anurans. Revista

de la Biologıa Tropical, 17: 119–124.

Lescure, J. 1975. Contribution a l’etude des

amphibiens de Guyane Francaise. III. Une

nouvelle espece de Colosthetus [sic] (Dendroba-

tidae): Colosthetus [sic] degranvillei n. sp. Bulle-

tin du Museum National d’Histoire Naturelle,

Zoologie, 293: 413–420.

Lescure, J. 1976. Etude de deux tetards de

Phyllobates (Dendrobatidae): P. femoralis (Bou-

lenger) et P. pictus (Bibron). Bulletin de la

Societe Zoologique de France, 101: 299–306.

Lescure, J. 1984. Las larvas de dendrobatidae [sic].

Reunion Iberoamericana de Conservacion y

Zoologıa de Vertebrados Actas II: 37–45.

Lescure, J., and R. Bechter. 1982. Le comporte-

ment de reproduction en captivite et le poly-

morphisme de Dendrobates quinquevittatus

Steindachner (Amphibia, Anura, Dendrobati-

dae). Revue Francaise d’Aquariologie, Herpe-

tologie, 8: 107–118.

Lescure, J., and C. Marty. 2000. Atlas des

Amphibiens de Guyane. Publications Scientifi-

ques du Museum National d’Histoire naturelle

Paris.

Liem, D.S., and W. Hosmer. 1973. Frogs of the

genus Taudactylus with descriptions of two new

species (Anura: Leptodactylidae). Memoires of

the Queensland Museum, 16: 435–457.

Lima, A.P., and J.P. Caldwell. 2001. A new

Amazonian species of Colostethus with sky blue

digits. Herpetologica, 57: 180–189.

Lima, A.P., J.P. Caldwell, and G.M. Biavati. 2002.

Territorial and reproductive behavior of an

Amazonian dendrobatid frog, Colostethus caer-

uleodactylus. Copeia, 2002: 44–51.

Lima, A.P., and C. Keller. 2003. Reproductive

characteristics of Colostethus marchesianus from

its type locality in Amazonas, Brazil. Journal of

Herpetology, 37: 754–757.

Lima, A.P., and W.E. Magnusson. 1998. Parti-

tioning seasonal time: interactions among size,

foraging activity and diet in leaf–litter frogs.

Oecologia, 116: 259–266.

Lima, A.P., and G. Moreira. 1993. Effects of prey

size and foraging mode on the ontogenetic

change in feeding niche of Colostethus stepheni

(Anura: Dendrobatidae). Oecologia, 95: 93–102.

Limerick, S. 1980. Courtship behavior and ovipo-

sition of the poison–arrow frog Dendrobates

pumilio. Herpetologica, 36: 69–71.

Lindquist, E.D., and T.E. Hetherington. 1998.

Tadpoles and juveniles of the Panamanian

golden frog, Atelopus zeteki (Bufonidae), with

information on development of coloration and

patterning. Herpetologica, 54: 370–376.

Liu, C.e.C. 1935. Types of vocal sac in the

Salientia. Proceedings of the Boston Society of

Natural History, 41: 19–40.

Lotters, S. 1988. Redefinition von Dendrobates

quinquevittatus (Steindachner, 1864) (Anura:

Dendrobatidae). Salamandra, 24: 72–74.

Lotters, S. 1996. The neotropical toad genus

Atelopus: checklist, biology, distribution. Koln:

M. Vences & F. Glaw.

Lotters, S., F. Castro Herrera, J. Kohler, and R.

Richter. 1997a. Notes on the distribution and

color variation of poison frogs of the genus

Phyllobates from western Colombia (Anura:

Dendrobatidae). Revue Francaise d’Aquariolo-

gie, Herpetologie, 24: 55–58.

Lotters, S., P. Debold, K. Henle, F. Glaw, and M.

Kneller. 1997b. Ein neuer Pfeigiftfrosch aus der

Epipedobates pictus-Gruppe vom Osthang der

Cordillera Azul in Peru. Herpetofauna, 19:

25–34.

Lotters, S., F. Glaw, J. Kohler, and F. Castro.

1999. On the geographic variation of the

advertisement call of Dendrobates histrionicus

Berthold, 1845 and related forms from north-

western South America (Anura: Dendrobati-

dae). Herpetozoa, 12: 23–38.


Lotters, S., K.-H. Jungfer, and A. Widmer. 2000.A new genus of aposematic poison frog(Amphibia: Anura: Dendrobatidae) from theupper Amazon basin, with notes on its re-productive behaviour and tadpole morphology.Jahreshefte der Gesellschaft fuer Naturkunde inWuttemberg, 156: 234–243.

Lotters, S., V.R. Morales, and C. Proy. 2003a.Another new riparian dendrobatid frog speciesfrom the upper Amazon basin of Peru. Journalof Herpetology, 37: 707–713.

Lotters, S., S. Reichle, and K.-H. Jungfer. 2003b.Advertisement calls of Neotropical poison frogs(Amphibia: Dendrobatidae) of the genera Co-lostethus, Dendrobates and Epipedobates, withnotes on dendrobatid call classification. Journalof Natural History, 37: 1899–1911.

Lotters, S., and M. Vences. 2000. Bermerkungenzur Nomenklatur und Taxonomie peruanischerPfeilgiftfrosche (Anura: Dendrobatidae: Den-drobates, Epipedobates). Salamandra, 36: 247–260.

Luddecke, H. 1976. Einige Ergebnisse aus Feld-beobachtungen an Phyllobates palmatus (Am-phibia, Ranidae) in Kolumbien. MitteilungenAus Dem Institut Colombo Aleman de Investi-gaciones Cientıficas, 8: 157–163.

Luddecke, H. 2000 ‘‘1999’’. Behavioral aspects ofthe reproductive biology of the Andean frogColostethus palmatus (Amphibia: Dendrobati-dae). Revista de la Academia Colombiana deCiencias Exactas, Fisicas y Naturales, 23:303–316.

Luddecke, H. 2003. Space use, cave choice, andspatial learning in the dendrobatid frog Colos-tethus palmatus. Amphibia Reptilia, 1: 37–46.

Lynch, J.D. 1969. Program. Final Ph.D. examina-tion, University of Kansas, Lawrence.

Lynch, J.D. 1971. Evolutionary relationships,osteology, and zoogeography of leptodactyloidfrogs. University of Kansas Museum of NaturalHistory Miscellaneous Publication, 53: 1–238.

Lynch, J.D. 1973. The transition from archaic toadvanced frogs. In J.L. Vial (editor), Evolution-ary biology of the anurans: contemporaryresearch on major problems: 133–182. Colum-bia: University of Missouri Press.

Lynch, J.D. 1978. A re-assessment of the telmato-biine leptodactylid frogs of Patagonia. Occa-sional Papers of the Museum of NaturalHistory, University of Kansas, 72: 1–57.

Lynch, J.D. 1979. A new genus for Elosia duidensisRivero (Amphibia, Leptodactylidae) fromsouthern Venezuela. American Museum Novi-tates, 2680: 1–8.

Lynch, J.D. 1982. Two new species of poison-dartfrogs (Colostethus) from Colombia. Herpetolo-gica, 38: 366–374.

Lynch, J.D. 1986. The definition of the MiddleAmerican clade of Eleutherodactylus based onjaw musculature (Amphibia: Leptodactylidae).Herpetologica, 42: 248–258.

Lynch, J.D. 1993. The value of the m. depresssormandibulae in phylogenetic hypotheses forEleutherodactylus and its allies (Amphibia:Leptodactylidae). Herpetologica, 49: 32–41.

Lynch, J.D., and W.E. Duellman. 1997. Frogs ofthe genus Eleutherodactylus in western Ecuador.Special Publication University of Kansas Mu-seum of Natural History, 23: 1–236.

Lynch, J.D., and P.M. Ruiz-Carranza. 1982. Anew genus and species of poison-dart frog(Amphibia: Dendrobatidae) from the Andes ofnorthern Colombia. Proceedings of the Biolog-ical Society of Washington, 95: 557–562.

Lynch, J.D., and P.M. Ruiz-Carranza. 1985. Unanueva especie de Colostethus (Amphibia: Den-drobatidae) de la Cordillera Occidental deColombia. Lozania, 54: 1–6.

Manzano, A., and E.O. Lavilla. 1995. Myologicalpeculiarities in Rhinoderma darwinii (Anura:Rhinodermatidae). Journal of Morphology,224: 125–129.

Manzano, A., S.A. Moro, and V. Abdala. 2003.The depressor mandibulae muscle in Anura.Alytes, 20: 93–131.

Marki, F., and B. Witkop. 1963. The venom of theColombian poison frog Phyllobates bicolor.Experientia, 19: 329–338.

Martin, W.F. 1972. Evolution of vocalization inthe genus Bufo. In W.F. Blair (editor), Evolu-tion in the genus Bufo: 279–309. Austin:University of Texas Press.

Martins, M. 1989. Nova especie de Colostethus daAmazonia central (Amphibia: Dendrobatidae)Revista Brasileira de Biologia, 49: 1009–1012.

Martins, M., and C.F.B. Haddad. 1988. Vocaliza-tion and reproductive behavior in the smithfrog, Hyla faber (Amphibia: Hylidae). Amphib-ia Reptilia, 9: 46–60.

Maxson, L.R., and C.W. Myers. 1985. Albuminevolution in tropical poison frogs (Dendrobati-dae): a preliminary report. Biotropica, 17: 50–56.

McClure, M.A., T.K. Vasi, and W.M. Fitch. 1994.Comparative analysis of multiple protein–se-quence alignment methods. Molecular Biologyand Evolution, 11: 571–592.

McDiarmid, R.W. 1971. Comparative morpholo-gy and evolution of frogs of the Neotropicalgenera Atelopus, Dendrophryniscus, Melano-phryniscus, and Oreophrynella. Bulletin of theLos Angeles County Museum of NaturalHistory Science, 12: 1–66.

McDiarmid, R.W., and R. Altig (editor) 1999.Tadpoles: the biology of anuran larvae. Chi-cago: University of Chicago Press.


Meinhardt, D.J., and J.R. Parmelee. 1996. A newspecies of Colostethus (Anura: Dendrobatidae)from Venezuela. Herpetologica, 52: 70–77.

Mendelson III, J.R., H.R. Da Silva, and A.M.Maglia. 2000. Phylogenetic relationships amongmarsupial frog genera (Anura: Hylidae: Hemi-phractinae) based on evidence from morphologyand natural history. Zoological Journal of theLinnean Society, 128: 125–148.

Mijares-Urrutia, A. 1991. Descripcion del rena-cuajo de Colostethus leopardalis Rivero conalgunos comentarios sobre su historia natural.Amphibia Reptilia, 12: 47–57.

Mijares-Urrutia, A., and A. Arends R. 1999. Anew Mannophryne (Anura: Dendrobatidae)from western Venezuela, with comments onthe generic allocation of Colostethus larandinus.Herpetologica, 55: 106–114.

Mijares-Urrutia, A., and E. La Marca. 1997.Tadpoles of the genus Nephelobates La Marca,1994 (Amphibia Anura Dendrobatidae), fromVenezuela. Tropical Zoology, 10: 133–142.

Mivart, S.G. 1869. On the classification of theanurous batrachians. Proceedings of the Zoo-logical Society of London, 1869: 280–295.

Morales, V.R. 1992. Dos especies nuevas deDendrobates (Anura: Dendrobatidae) para Peru.Caribbean Journal of Science, 28: 191–199.

Morales, V.R. 2002 ‘‘2000’’. Sistematica y biogeo-grafıa del grupo trilineatus (Amphibia, Anura,Dendrobatidae, Colostethus), con descripcion deonce nuevas especies. Publicaciones de laAsociacion de Amigos de Donana, 13: 1–59.

Morales, V.R., and P.M. Velazco. 1998. Unaespecie nueva de Epipedobates (Amphibia,Anura, Dendrobatidae) de Peru. Amphibia–Reptilia, 19: 369–376.

Moritz, C., C.J. Schneider, and D.B. Wake. 1992.Evolutionary relationships within the Ensatinaeschscholtzii complex confirm the ring speciesinterpretation. Systematic Biology, 41: 273–291.

Mortari, M.R., E.N.F. Schwartz, C.A. Schwartz,O.R. Pires, Jr., M.M. Santos, and C. Bloch, Jr.2004. Main alkaloids from the Brazilian Den-drobatidae frog Epipedobates flavopictus: pumi-liotoxin 251D, histrionicotoxin and decahydro-quinolines. Toxicon, 43: 303–310.

Mudrack, W. 1969. Pflege und Zucht einesBlattsteigerfrosches der Gattung Phyllobatesaus Ecuador. Salamandra, 5: 81–84.

Muller, K. 2004. PRAP—computation of Bremersupport for large data sets. Molecular Phyloge-netics and Evolution, 31: 780–782.

Myers, C.W. 1969. The ecological geography ofcloud forest in Panama. American MuseumNovitates, 2396: 1–52.

Myers, C.W. 1982. Spotted poison frogs: descrip-tions of three new Dendrobates from western

Amazonia, and resurrection of a lost speciesfrom ‘‘Chiriqui’’. American Museum Novitates,2721: 1–23.

Myers, C.W. 1987. New generic names for someneotropical poison frogs (Dendrobatidae). Pa-peis Avulsos de Zoologia, 36: 301–306.

Myers, C.W. 1991. Distribution of the dendrobatidfrog Colostethus chocoensis and description ofa related species occuring macrosympatrically.American Museum Novitates, 3010: 1–15.

Myers, C.W., and P.A. Burrowes. 1987. A newpoison frog (Dendrobates) from Andean Co-lombia, with notes on a lowland relative.American Museum Novitates, 2899: 1–17.

Myers, C.W., and J.W. Daly. 1976a. A new speciesof poison frog (Dendrobates) from AndeanEcuador, including an analysis of its skintoxins. Occasional Papers of the Museum ofNatural History, the University of Kansas, 59:1–12.

Myers, C.W., and J.W. Daly. 1976b. Preliminaryevaluation of skin toxins and vocalizations intaxonomic and evolutionary studies of poison-dart frogs (Dendrobatidae). Bulletin of theAmerican Museum of Natural History, 157:173–262.

Myers, C.W., and J.W. Daly. 1979. A name for thepoison frog of Cordillera Azul, eastern Peru,with notes on its biology and skin toxins(Dendrobatidae). American Museum Novitates,2674: 1–24.

Myers, C.W., and J.W. Daly. 1980. Taxonomy andecology of Dendrobates bombetes, a new Andeanfrog with new skin toxins. American MuseumNovitates, 2692: 1–23.

Myers, C.W., and J.W. Daly. 1983. Dart-poisonfrogs. Scientific American, 248: 120–133.

Myers, C.W., J.W. Daly, H.M. Garraffo, A.Wisnieski, and J.F. Cover, Jr. 1995. Discoveryof the Costa Rican poison frog Dendrobatesgranuliferus in sympatry with Dendrobatespumilio, and comments on taxonomic use ofskin alkaloids. American Museum Novitates,3144: 1–21.

Myers, C.W., J.W. Daly, and B. Malkin. 1978. Adangerously toxic new frog (Phyllobates) usedby Embera Indians of western Colombia, withdiscussion of blowgun fabrication and dartpoisoning. Bulletin of the American Museumof Natural History, 161: 307–366.

Myers, C.W., J.W. Daly, and V. Martınez. 1984.An arboreal poison frog (Dendrobates) fromwestern Panama. American Museum Novitates,2783: 1–20.

Myers, C.W., and M.A. Donnelly. 1997. A tepuiherpetofauna on a granitic mountain (Tama-cuari) in borderland between Venezuela andBrazil: report from the Phipps Tapirapeco


Expedition. American Museum Novitates, 3213:1–71.

Myers, C.W., and M.A. Donnelly. 2001. Herpeto-fauna of the Yutaje-Corocoro massif, Vene-zuela: Second report from the Robert G.Goelet American Museum-Terramar Expedi-tion to the northwestern tepuis. Bulletin of theAmerican Museum of Natural History, 261:1–85.

Myers, C.W., and W.E. Duellman. 1982. A newspecies of Hyla from Cerro Colorado, and othertree frog records and geographical notes fromwestern Panama. American Museum Novitates,2752: 1–32.

Myers, C.W., and L.S. Ford. 1986. On Atopo-phrynus, a recently described frog wronglyassigned to the Dendrobatidae. American Mu-seum Novitates, 2843: 1–15.

Myers, C.W., A. Paolillo O., and J.W. Daly. 1991.Discovery of a defensively malodorous andnocturnal frog in the family Dendrobatidae:phylogenetic significance of a new genus andspecies from the Venezuelan Andes. AmericanMuseum Novitates, 3002: 1–33.

Myers, C.W., L.O. Rodrıguez, and J. Icochea.1998. Epipedobates simulans, a new crypticspecies of poison frog from southeastern Peru,with notes on E. macero and E. petersi(Dendrobatidae). American Museum Novitates,3238: 1–20.

Narins, P.M., D.S. Grabul, K.K. Soma, P.Gaucher, and W. Hodl. 2005. Cross-modalintegration in a dart-poison frog. Proceedingsof the National Academy of Science USA, 102:2425–2429.

Narins, P.M., W. Hodl, and D.S. Grabul. 2003.Bimodal signal requisite for agonistic behaviorin a dart-poison frog, Epipedobates femoralis.Proceedings of the National Academy ofScience USA, 100: 577–580.

Navas, C.A. 1996a. The effect of temperature onthe vocal activity of tropical anurans: a compar-ison of high and low-elevation species. Journalof Herpetology, 30: 488–497.

Navas, C.A. 1996b. Thermal dependency of fieldlocomotor and vocal performance of high-elevation anurans in the tropical Andes. Journalof Herpetology, 30: 478–487.

Nicholls, G.E. 1916. The structure of the vertebralcolumn in the Anura Phaneroglossa and itsimportance as a basis of classification. Proceed-ings of the Linnaean Society of London, 128:80–92.

Nixon, K.C. 1999. The parsimony ratchet, a newmethod for rapid parsimony analysis. Cladistics,15: 407–414.

Nixon, K.C. 1999–2002. WinClada. Ver. 1.0000.Ithaca, NY: Published by the author.

Noble, G.K. 1922. The phylogeny of the Salientia

I. The osteology and the thigh musculature;

their bearing on classification and phylogeny.

Bulletin of the American Museum of Natural

History, 46: 1–87.

Noble, G.K. 1923. New batrachians from the

Tropical Research Station, British Guiana.

Zoologica, 3: 288–299.

Noble, G.K. 1926. The pectoral girdle of the

brachycephalid frogs. American Museum Novi-

tates, 230: 1–14.

Noble, G.K. 1927. The value of life history data in

the study of the evolution of the Amphibia.

Annals of the New York Academy of Sciences,

30: 31–128.

Noble, G.K. 1931. The biology of the Amphibia.

New York: McGraw-Hill.

Noble, G.K., and M.E. Jaeckle. 1928. The digital

pads of tree frogs: a study of the phylogenesis of

an adaptive tissue. Journal of Morphology and

Physiology, 45: 259–292.

Nuin, P.A.S. 2003. Description of the tadpole of

Megaelosia goeldii (Leptodactylidae, Hylodinae)

with natural history notes. Herpetological Re-

view, 34: 27–28.

Nunez, J.J., A.M. Zarraga, and J.R. Formas. 1999.

New molecular and morphometric evidence for

the validation of Eupsophus calcaratus and E.

roseus (Anura: Leptodactylidae) in Chile. Stud-

ies on Neotropical Fauna and Environment, 34:

150–155.

Orton, G.L. 1953. The systematics of vertebrate

larvae. Systematic Zoology, 2: 63–75.

Orton, G.L. 1957. The bearing of larval evolution

on some problems in frog classification. Sys-

tematic Zoology, 6: 79–86.

Palumbi, S.R., A. Martin, S. Romano, W.O.

McMillan, L. Stice, and G. Grabawski. 1991.

The Simple Fool’s Guide to PCR, Version 2.0.

Privately published, compiled by S. Palumbi,

University of Hawaii: Honolulu.

Parker, H.W. 1940. The Australasian frogs of the

family Leptodactylidae. Novitates Zoologicae.

Tring, 42: 1–106.

Parmelee, J.R. 1999. Trophic ecology of a tropical

anuran assemblage. Scientific Papers, Natural

History Museum, the University of Kansas, 11:

1–59.

Parsons, T.S., and E.E. Williams. 1962. The teeth

of Amphibia and their relation to amphibian

phylogeny. Journal of Morphology, 110:

375–389.

Pefaur, J.E. 1985. New species of Venezuelan

Colostethus (Dendrobatidae). Journal of Herpe-

tology, 19: 321–327.

Pefaur, J.E. 1993. Description of a new Colostethus

(Dendrobatidae) with some natural history


comments on the genus in Venezuela. Alytes, 11:

88–96.

Peracca, M. 1904. Rettili ed amfibii. Bolletino dei

Musei di Zoologia ed Anatomia Comparata

della R. Universita di Torino, 19: 1–41.

Peters, J.A. 1964. Dictionary of herpetology. New

York: Hafner Publishing Company.

Phillips, A., D. Janies, and W.C. Wheeler. 2000.

Multiple sequence alignment in phylogenetic

analysis. Molecular Phylogenetics and Evolu-

tion, 16: 317–330.

Polder, W.N. 1973. Over verzorging en voortplant-

ing in gevenschap van Dendrobates azureus en

enkele andere Dendrobatidae. Het Aquarium,

44: 16–22.

Polder, W.N. 1974. Over verzorging en voortplant-

ing in gevenschap van Dendrobates azureus en

enkele andere Dendrobatidae (2). Het Aquari-

um, 44: 186–191.

Pough, F.H., and T.L. Taigen. 1990. Metabolic

correlates of the foraging and social behaviour

of dart-poison frogs. Animal Behaviour, 39:

145–155.

Praderio, M.J., and M.D. Robinson. 1990. Re-

production in the toad Colostethus trinitatus [sic]

(Anura: Dendrobatidae) in northern Venezuela.

Journal of Tropical Ecology, 6: 331–341.

Pramuk, J.B., and B.I. Hiler. 1999. An investiga-

tion of obligate oophagy of Dendrobates pumilio

tadpoles. Herpetological Review, 30: 219–221.

Prohl, H. 2003. Variation in male calling behav-

iour and relation to male mating success in the

strawberry poison frog (Dendrobates pumilio).

Ethology, 109: 273–290.

Prohl, H., and O. Berke. 2001. Spatial distribu-

tions of male and female strawberry poison

frogs and their relation to female reproductive

resources. Oecologia, 129: 534–542.

Rada de Martınez, D. 1976. Cariotipo de Colos-

tethus trinitatis (Amphibia: Dendrobatidae).

Acta Biologica Venezuelica, 9: 213–220.

Ramos, C.W., N. Pimentel, and V. Martınez-

Cortes. 2002. Karyotype of the endemic golden

frog Atelopus zeteki (Dunn) from Panama.

Carribean Journal of Science, 38: 268–270.

Rasotto, M.B., P. Cardellini, and M. Sala. 1987.

Karyotypes of five species of Dendrobatidae

(Anura: Amphibia). Herpetologica, 43:

177–182.

Reig, O.A. 1958. Proposiciones para una nueva

macrosistematica de los anuros (nota prelimi-

nar). Physis, 21: 109–118.

Rivero, J.A. 1961. Salientia of Venezuela. Bulletin

of the Museum of Comparative Zoology, 126:

1–207.

Rivero, J.A. 1978 ‘‘1976’’. Notas sobre los anfibios

de Venezuela II. Sobre los Colostethus de los

Andes venezolanos. Memoria de la Sociedad de

Ciencias Naturales La Salle, 35: 327–344.

Rivero, J.A. 1979. Sobre el origen de la fauna

paramera de anfibios venezolanos. In S. La-

bouriau (editor), El Medio Ambiente Paramo:

165–175. Caracas: Ediciones C.E.A/IVIC/UN-

ESCO.

Rivero, J.A. 1980 ‘‘1978’’. Notas sobre los anfibios

de Venezuela III: nuevos Colostethus de los

Andes venezolanos. Memoria de la Sociedad de


Rivero, J.A. 1984. Una nueva especie de Colos-

tethus (Amphibia, Dendrobatidae) de la Cordil-

lera de la Costa, con anotaciones sobre otros

Colostethus, de Venezuela. Brenesia, 22: 51–56.

Rivero, J.A. 1984 ‘‘1982’’. Sobre el Colostethus

mandelorum (Schmidt) y el Colostethus inflexus

Rivero (Amphibia, Dendrobatidae). Memoria

de la Sociedad de Ciencias Naturales La Salle,

42: 9–16.

Rivero, J.A. 1990 ‘‘1988’’. Sobre las relaciones de

las especies del genero Colostethus (Amphibia:

Dendrobatidae). Memoria de la Sociedad de


Rivero, J.A. 1991a. New Colostethus (Amphibia,

Dendrobatidae) from South America. Breviora,

493: 1–28.

Rivero, J.A. 1991b. New Ecuadorean [sic] Colos-

tethus (Amphibia, Dendrobatidae) in the collec-

tion of National Museum of Natural History,

Smithsonian Institution. Caribbean Journal of

Science, 27: 1–22.

Rivero, J.A., and A. Almendariz. 1991. La

identificacion de los Colostethus (Amphibia,

Dendrobatidae) de Ecuador. Revista Politec-

nica, 26: 99–152.

Rivero, J.A., and H. Granados-Dıaz. 1990 ‘‘1989’’.

Nuevos Colostethus (Amphibia, Dendrobatidae)

del Departamento de Cauca, Colombia. Carib-

bean Journal of Science, 25: 148–152.

Rivero, J.A., L. Oliver, and M.d.l.A. Irizarry.

‘‘1987’’ 1989. Los discos digitales de tres

Eleutherodactylus (Anura, Leptodactylidae) de

Puerto Rico, con anotaciones sobre los meca-

nismos de adhesion en las ranas. Carribean

Journal of Science, 23: 226–237.

Rivero, J.A., and M.A. Serna. 1989 ‘‘1988’’. La

identificacion de los Colostethus (Amphibia,

Dendrobatidae) de Colombia. Caribbean Jour-

nal of Science, 24: 137–154.

Rivero, J.A., and M.A. Serna. 1991. Tres nuevas

especies de Colostethus (Anphibia [sic], Dendro-

batidae) de Colombia. Trianea, 4: 481–495.

Rivero, J.A., and M.A. Serna. 2000 ‘‘1995’’.

Nuevos Colostethus (Amphibia, Dendrobatidae)

del Departamento de Antioquia, Colombia, con

la descripcion del renacuajo de Colostethus


fraterdanieli. Revista de Ecologıa Latinoamer-icana, 2: 45–58.

Rodrıguez, L.O., and C.W. Myers. 1993. A newpoison frog from from Manu National Park,southeastern Peru (Dendrobatidae, Epipedo-bates). American Museum Novitates, 3068:1–15.

Rosa, C., O.J. Aguiar, A.A. Giaretta, and S.M.Recco-Pimentel. 2003. Karyotypic variation inthe genus Megaelosia (Anura, Hylodinae) withthe first description of a B-chromosome ina leptodactylid frog. Copeia, 3: 166–174.

Rozen, S., and H.J. Skaletsky. 2000. Primer3 onthe WWW for general users and for biologistprogrammers. In S. Krawetz and S. Misener(editors), Bioinformatics methods and proto-cols: methods in molecular biology: 365–386.Totowa, NJ: Humana Press.

Ruiz-Carranza, P.M., and M.P. Ramırez-Pinilla.1992. Una nueva especie de Minyobates (Anura:Dendrobatidae) de Colombia. Lozania, 61:1–16.

Ruthven, A.G. 1915. Description of a new taillessamphibian of the family Dendrobatidae. Occa-sional Papers of the Museum of Zoology,University of Michigan, 20: 1–3.

Ruthven, A.G., and H.T. Gaige. 1915. Thebreeding habits of Prostherapis subpunctatusCope. Occasional Papers of the Museum ofZoology, University of Michigan, 10: 1–5.

Ruvinsky, I., and L.R. Maxson. 1996. Phyloge-netic relationships among bufonoid frogs (An-ura: Neobatrachia) inferred from mitochondrialDNA sequences. Molecular Phylogenetics andEvolution, 5: 533–547.

Sankoff, D. 1975. Minimal mutation trees ofsequences. SIAM Journal on Applied Mathe-matics, 28: 35–42.

Sankoff, D., R.J. Cedergren, and G. Lapalme.1976. Frequency of insertion–deletion, transver-sion, and transition in evolution of 5S ribosomalRNA. Journal of Molecular Evolution, 7:133–149.

Santos, J.C., L.A. Coloma, and D.C. Cannatella.2003. Multiple, recurring origins of aposema-tism and diet specialization in poison frogs.Proceedings of the National Academy ofScience USA, 21335–21100.

Saporito, R.A., M.A. Donnelly, R.L. Hoffman,H.M. Garraffo, and J.W. Daly. 2003. Asiphonotid millipede (Rhinotus) as the sourceof spiropyrrolizidine oximes of dendrobatidfrogs. Journal of Chemical Ecology, 29: 2781–2786.

Saporito, R.A., H.M. Garraffo, M.A. Donnelly,A.L. Edwards, J.T. Longino, and J.W. Daly.2004. Formicine ants: an arthropod source forthe pumiliotoxin alkaloids of dendrobatid poi-

son frogs. Proceedings of the National Academyof Science USA, 101: 8045–8050.

Sarkar, S. 1998. Genetics and reductionism. NewYork: Cambridge University Press.

Savage, J.M. 1968. The dendrobatid frogs ofCentral America. Copeia, 1968: 745–776.

Savage, J.M. 1973. The geographic distribution offrogs: patterns and predictions. In J.L. Vial(editor), Evolutionary biology of the anurans:contemporary research on major problems:351–445. Columbia: University of MissouriPress.

Savage, J.M. 2002. The Amphibians and reptiles ofCosta Rica: a herpetofauna between two con-tinents, between two seas. Chicago: Universityof Chicago Press.

Savage, J.M., and W.R. Heyer. 1967. Variationand distribution in the tree-frog genus Phyllo-medusa in Costa Rica, Central America.Beitrage zur Neotropischen Fauna, 5: 111–131.

Savage, J.M., C.W. Myers, D.R. Frost, and T.Grant. In press. Dendrobatidae Cope, 1865(1850) (Amphibia, Anura): proposed con-servation. Bulletin of Zoological Nomencla-ture.

Scheltinga, D.M., and B.G.M. Jamieson. 2003.Spermatogenesis and the mature spermatozoon:form, function, and phylogenetic implications.In B.G.M. Jamieson (editor), Reproductivebiology and phylogeny of Anura: 119–251.Enfield, NH: Science Publishers.

Schlosser, G. 2002a. Development and evolutionof lateral line placodes in amphibians I. De-velopment. Zoology, 105: 119–146.

Schlosser, G. 2002b. Development and evolutionof lateral line placodes in amphibians II.Evolution. Zoology, 105: 177–193.

Schulte, R. 1989. Una nueva especie de ranavenenosa del genero Epipedobates registrada enla Cordillera Oriental, departamento de SanMartın. Boletin de Lima, 63: 41–46.

Schulte, R. 1990. Redescubrimiento y redefinicionde Dendrobates mysteriosus (Myers 1982) [sic]dela Cordillera del Condor. Boletin de Lima, 70:57–68.

Schulte, R. 1999. Pfeilgiftfrosche ‘‘Arteneil Peru’’.Waiblingen: INIBICO.

Senfft, W. 1936. Das Brutgeschaft des Baumstei-gerfrosches (Dendrobates auratus Girard) inGefangenschaft. Zoologische Garten Leipzig,8: 122–131.

Sexton, O.J. 1960. Some aspects of the behaviorand of the territory of a dendrobatid frog,Prostherapis trinitatis. Ecology, 41: 107–115.

Siddall, M.E., and A.G. Kluge. 1997. Probabilismand phylogenetic inference. Cladistics, 13:313–336.


Silva, A.P.Z., C.F.B. Haddad, and S. Kasahara.

2001. Cytogenetic analysis of Cycloramphus

boraceiensis Heyer (Anura, Leptodactylidae).

Revista Brasileira de Zoologia, 18: 111–

115.

Silverstone, P.A. 1970. An evolutionary study of

the poison-arrow frogs (Genus Dendrobates

Wagler). Ph.D. dissertation. University of

Southern California, Los Angeles, 199 pp.

Silverstone, P.A. 1971. Status of certain frogs of

the genus Colostethus, with descriptions of new

species. Los Angeles County Museum Contri-

butions in Science, 215: 1–8.

Silverstone, P.A. 1973. Observations on the

behavior and ecology of a Colombian poison-

arrow frog, the kokoe-pa (Dendrobates histrio-

nicus Berthold). Herpetologica, 29: 295–301.

Silverstone, P.A. 1975a. A revision of the poison–

arrow frogs of the genus Dendrobates Wagler.

Natural History Museum of Los Angeles

County Science Bulletin, 21: 1–55.

Silverstone, P.A. 1975b. Two new species of

Colostethus (Amphibia: Anura: Dendrobatidae)

from Colombia. Los Angeles County Museum

Contributions in Science, 268: 1–10.

Silverstone, P.A. 1976. A revision of the poison-

arrow frogs of the genus Phyllobates Bibron in

Sagra (family Dendrobatidae). Natural History

Museum of Los Angeles County Science Bulle-

tin, 27: 1–53.

Simmons, M.P. 2004. Independence of alignment

and tree search. Molecular Phylogenetics and

Evolution, 31: 874–879.

Simmons, M.P., and H. Ochoterena. 2000. Gaps

as characters in sequence-based phylogenetic

analyses. Systematic Biology, 49: 369–381.

Skinner, A. 2004. Hierarchy and monophyly.

Cladistics, 20: 498–500.

Slowinski, J.B. 1998. The number of multiple

alignments. Molecular Phylogenetics and Evo-

lution, 10: 264–266.

Sober, E. 1988. Reconstructing the past: parsimo-

ny, evolution, and inference. Cambridge, MA:

MIT Press.

Starrett, P.H. 1968. The phylogenetic significance

of the jaw musculature in anuran amphibians.

Ph.D. Dissertation, University of Michigan,

Ann Arbor, 179 pp.

Starrett, P.H. 1973. Evolutionary patterns in larval

morphology. In J.L. Vial (editor), Evolutionary

biology of the anurans: contemporary research

on major problems: 251–271. Columbia: Uni-

versity of Missouri Press.

Stebbins, R., and J. Hendrickson. 1959. Field

studies of amphibians in Colombia, South

America. Univesity of California Publications

in Zoology, 56: 497–540.

Summers, K. 1989. Sexual selection and intra–female competition in the green poison-dartfrog, Dendrobates auratus. Animal Behaviour,37: 797–805.

Summers, K. 1990. Paternal care and the cost ofpolygyny in the green dart-poison frog. Behav-ioral Ecology and Sociobiology, 27: 307–313.

Summers, K. 1992. Mating strategies in twospecies of dart-poison frogs: a comparativestudy. Animal Behaviour, 43: 907.

Summers, K. 1999. The effects of cannibalism onAmazonian poison frog egg and tadpole de-position and survivorship in Heliconia axilpools. Oecologia, 119: 557–564.

Summers, K. 2000. Mating and aggressive behav-iour in dendrobatid frogs from CorcovadoNational Park, Costa Rica: a comparativestudy. Behavior, 137: 7–24.

Summers, K., E. Bermingham, L. Weigt, S.McCafferty, and L. Dahlstrom. 1997. Pheno-typic and genetic divergence in three species ofdart-poison frogs with contrasting parentalbehavior. Journal of Heredity, 88: 8–13.

Summers, K., and D.J.D. Earn. 1999. The cost ofpolygyny and the evolution of female care inpoison frogs. Biological Journal of the LinneanSociety, 66: 515–538.

Summers, K., and C.S. McKeon. 2004. Theevolutionary ecology of phytotelmata use inNeotropical poison frogs. Miscellaneous Pub-lications, Museum of Zoology, University ofMichigan, 193: 55–73.

Summers, K., and R. Symula. 2001. Cannibalismand kin discrimination in tadpoles of theAmazonian poison frog, Dendrobates ventrima-culatus, in the field. Herpetological Journal, 11:17–21.

Summers, K., R. Symula, M. Clough, and T.Cronin. 1999a. Visual mate choice in poisonfrogs. Proceedings of the Royal Society ofLondon B, 266: 2141–2145.

Summers, K., L. Weigt, P. Boag, and E. Berming-ham. 1999b. The evolution of female parentalcare in poison frogs of the genus Dendrobates:Evidence from mitochondrial DNA sequences.Herpetologica, 55: 254–270.

Swofford, D. 1998–2002. PAUP*. PhylogeneticAnalysis Using Parsimony (*and Other Meth-ods). Ver. 4.0beta. Sunderland MA: SinauerAssociates.

Swofford, D., G.J. Olsen, P.J. Waddell, and D.M.Hillis. 1996. Phylogenetic inference. In D.M.Hillis, C. Moritz, and B.K. Mable (editors),Molecular systematics: 407–514. Sunderland,MA: Sinauer Associates.

Symula, R., R. Schulte, and K. Summers. 2001.Molecular phylogenetic evidence for a mimeticradiation in Peruvian poison frogs supports


a Mullerian mimicry hypothesis. Proceedings ofthe Royal Society of London B, 268: 2415–2421.

Symula, R., R. Schulte, and K. Summers. 2003.Molecular systematics and phylogeography ofAmazonian poison frogs of the genus Dendro-bates. Molecular Phylogenetics and Evolution,26: 452–475.

Takada, W., T. Sakata, S. Shimano, Y. Enami, N.Mori, R. Nishida, and Y. Kuwahara. 2005.Scheloribatid mites as the source of pumiliotox-ins in dendrobatid frogs. Journal of ChemicalEcology, 31: 2403–2415.

Taylor, E.H. 1951. Two new genera and a newfamily of tropical frogs. Proceedings of theBiological Society of Washington, 64: 33–40.

Test, F.H. 1954. Social aggressiveness in anamphibian. Science, 120: 140–141.

Test, F.H. 1956. Two new dendrobatid frogs fromnorthwestern Venezuela. Occasional Papers ofthe Museum of Zoology, University of Michi-gan, 577: 1–9.

Test, F.H., O.J. Sexton, and H. Heatwole. 1966.Reptiles of Rancho Grande and vicinity, EstadoAragua, Venezuela. Miscellaneous Publications,Museum of Zoology, University of Michigan,128: 1–63.

Thibaudeau, G., and R. Altig. 1999. Endotrophicanurans: development and evolution. In R.W.McDiarmid, and R. Altig (editors), Tadpoles:the biology of anuran larvae: 170–188. Chicago:University of Chicago Press.

Thompson, J.D., T.J. Gibson, F.J. Plewniak, F.,and D.G. Higgins. 1997. The ClustalX windowsinterface: flexible strategies for multiple se-quence alignment aided by quality analysistools. Nucleic Acids Research, 24: 4876–4882.

Thompson, J.D., D.G. Higgins, and T.J. Gibson.1994. CLUSTAL W: improving the sensitivityof progressive multiple sequence alignmentthrough sequence weighting, position specificgap penalties and weight matrix choice. NucleicAcids Research, 22: 4673–4680.

Tihen, J.A. 1965. Evolutionary trends in frogs.American Zoologist, 5: 309–318.

Titus, T.A., and A. Larson. 1996. Molecularphylogenetics of desmognathine salamanders(Caudata: Plethodontidae): A reevaluation ofevolution in ecology, life history, and morphol-ogy. Systematic Biolology, 45: 451–472.

Toft, C.A. 1980. Feeding ecology of thirteensyntopic species of anurans in a seasonaltropical environment. Oecologia, 45: 131–141.

Toft, C.A. 1995. Evolution of diet specialization inpoison-dart frogs (Dendrobatidae). Herpetolo-gica, 51: 202–216.

Toft, C.A., A.S. Rand, and M. Clark. 1982.Population dynamics and seasonal recruitment

in Bufo typhonius and Colostethus nubicola(Anura). In E.G. Leigh, A.S. Rand, and D.M.Windsor (editors), The ecology of a tropicalforest: seasonal rhythms and long–termchanges: 397–403. Washington, DC: Smithso-nian Institution Press.

Tokuyama, T., J.W. Daly, H.M. Garraffo, andT.F. Spande. 1992. Pyrrolizidine oximes: a novelclass of dendrobatid alkaloids. Tetrahedron, 48:4247–4258.

Toledo, L.F., L.D.A. Guimaraes, L.P. Lima, R.P.Bastos, and C.F.B. Haddad. 2004. Notes oncourtship, egg–laying site, and defensive behav-ior of Epipedobates flavopictus (Anura, Dendro-batidae) from two mountain ranges of centraland southeastern Brazil. Phyllomedusa, 3:145–147.

Trapido, H. 1953. A new frog from Panama,Dendrobates galindoi. Fieldiana Zoology, 34:181–187.

Trewavas, E. 1933. The hyoid and larynx of theAnura. Philosophical Transactions of the RoyalSociety of London, B, 222: 401–527.

Trueb, L. 1993. Patterns of cranial diversificationamong the Lissamphibia. In J. Hanken, andB.K. Hall (editors), The skull: 255–343. Chi-cago: University of Chicago Press.

Tschudi, J.J.v. 1838. Classification der Batrachier,mit Berucksichtigung der fossilien Thiere dieserAbtheilung der Reptilien. Neuchatel: Petitpierre.

Tyler, M.J. 1971. The phylogenetic significance ofvocal sac structure in hylid frogs. OccasionalPapers of the Museum of Natural History, theUniversity of Kansas, 19: 319–360.

van Wijngaarden, R., and F. Bolanos. 1992.Parental care in Dendrobates granuliferus (An-ura: Dendrobatidae), with a description of thetadpole. Journal of Herpetology, 26: 102–105.

Veiga-Menoncello, A.C.P., A.P. Lima, and S.M.Recco-Pimentel. 2003a. Cytogenetic analysis offour central Amazonian species of Colostethus(Anura – Dendrobatidae) with a diploid com-plement of 22 chromosomes. Hereditas, 139:189–198.

Veiga-Menoncello, A.C.P., A.P. Lima, and S.M.Recco-Pimentel. 2003b. Cytogenetics of twocentral Amazonian species of Colostethus (An-ura, Dendrobatidae) with nidicolous tadpoles.Caryologia, 56: 253–260.

Vences, M., J. Kosuch, R. Boistel, C.F.B. Haddad,E. La Marca, S. Lotters, and M. Veith. 2003a.Convergent evolution of aposematic colorationin Neotropical poison frogs: a molecular phy-logenetic perspective. Organisms Diversity andEvolution, 3: 215–226.

Vences, M., D.R. Vieites, F. Glaw, H. Brinkmann,J. Kosuch, M. Veith, and A. Meyer. 2003b.


Multiple overseas dispersal in amphibians.

Proceedings of the Royal Society of London.

Series B, Biological Sciences, 270: 2435–2442.

Vences, M., J. Kosuch, S. Lotters, A. Widmer,

K.H. Jungfer, J. Kohler, and M. Veith. 2000.

Phylogeny and classification of poison frogs

(Amphibia: Dendrobatidae), based on mito-

chondrial 16S and 12S ribosomal RNA gene

sequences. Molecular Phylogenetics and Evolu-

tion, 15: 34–40.

Vigle, G.O., and K. Miyata. 1980. A new species of

Dendrobates (Anura: Dendrobatidae) from the

lowland rain forests of western Ecuador. Bre-

viora, 459: 1–7.

Walker, C.F. 1938. The structure and systematic

relationships of the genus Rhinophrynus. Occa-

sional Papers of the Museum of Zoology,

University of Michigan, 372: 1–11.

Wang, L., and T. Jiang. 1994. On the complexity

of multiple sequence alignment. Journal of

Computational Biology, 1: 337–348.

Wells, J.G. 1994. Jewels of the rainforest: poison

frogs of the family Dendrobatidae. Neptune

City NJ: T.F.H. Publications.

Wells, K.D. 1978. Courtship and parental behav-

ior in a Panamanian poison-arrow frog (Den-

drobates auratus). Herpetologica, 34: 148–

155.

Wells, K.D. 1980a. Behavioral ecology and social

organization of a dendrobatid frog (Colostethus

inguinalis). Behavioral Ecology and Sociobiolo-

gy, 6: 199–209.

Wells, K.D. 1980b. Evidence for growth of

tadpoles during parental transport in Colos-

tethus inguinalis. Journal of Herpetology, 14:

428–430.

Wells, K.D. 1980c. Social behavior and commu-

nication of a dendrobatid frog (Colostethus

trinitatis). Herpetologica, 36: 189–199.

Weygoldt, P. 1980. Complex brood care and

reproductive behavior in captive poison-arrow

frogs, Dendrobates pumilio O. Schmidt. Behav-

ioral Ecology and Sociobiology, 7: 329–332.

Weygoldt, P. 1987. Evolution of parental care in

dart poison frogs (Amphibia: Anura: Dendro-

batidae). Zeitschrift fur Zoologische Systematik

und Evolutionsforschung, 25: 51–7.

Wheeler, W.C. 1994. Sources of ambiguity in

nucleic acid sequence alignment. In B. Schier-

water, B. Streit, G.P. Wagner, and R. DeSalle

(editors), Molecular ecology and evolution:

approaches and applications: 323–352. Basel:

Birkhauser.

Wheeler, W.C. 1995. Sequence alignment, param-

eter sensitivity, and the phylogenetic analysis of

molecular data. Systematic Biology, 44: 321–

331.

Wheeler, W.C. 1996. Optimization alignment: theend of multiple sequence alignment in phyloge-netics? Cladistics, 12: 1–9.

Wheeler, W.C. 1998. Alignment characters, dy-namic programing and heuristic solutions. In R.DeSalle, and B. Schierwater (editors), Molecularapproaches to ecology and evolution, 2nd ed:243–251. Basel: Birkhauser.

Wheeler, W.C. 1999. Fixed character states andthe optimization of molecular sequence data.Cladistics, 15: 379–385.

Wheeler, W.C. 2003a. Implied alignment: a syna-pomorphy–based multiple sequence alignmentmethod. Cladistics, 19: 261–268.

Wheeler, W.C. 2003b. Iterative pass optimizationof sequence data. Cladistics, 19: 254–260.

Wheeler, W.C. 2003c. Search-based optimization.Cladistics, 19: 348–355.

Wheeler, W.C., D. Gladstein, and J. De Laet.1996–2003. POY: Phylogeny reconstruction viaoptimization of DNA data. Ver. 3.0. ftp://ftp.amnh.org/pub/molecular/poy.

Widmer, A., S. Lotters, and K.-H. Jungfer. 2000.A molecular phylogenetic analysis of the neo-tropical dart-poison frog genus Phyllobates(Amphibia: Dendrobatidae). Naturwissenschaf-ten, 87: 559–562.

Wild, E.R. 1996. Natural history and resourceuse of four Amazonian tadpole assemblages.Occasional Papers of the Museum of NaturalHistory, the University of Kansas, 176: 1–59.

Wilkinson, J.A., M. Matsui, and T. Terachi. 1996.Geographic variation in in a Japanese tree frog(Rhacophorus arboreus) revelaed by PCR-aidedrestriction site analysis of mtDNA. Journal ofHerpetology, 30: 418–423.

Wilkinson, J.A., R.C. Drewes, and O.L. Tatum.2002. A molecular phylogenetic analysis of thefamily Rhacophoridae with an emphasis on theAsian and African genera. Molecular Phyloge-netics and Evolution, 24: 265–273.

Wollenberg, K., Noonan, B.P., and Lotters, S.2006. Polymorphism versus species richness.Systematics of large Dendrobates from theeastern Guiana Shield (Amphibia: Dendrobati-dae). Copeia, 2006.

Wyman, J. 1859a. On some unusual modes ofgestation. American Journal of Science ser. 2,27: 1–5.

Wyman, J. 1859b. September 16, 1857. Proceed-ings of the Boston Society of Natural History, 6:268–269.

Zimmermann, E. 1986. Breeding terrarium ani-mals: amphibians and reptiles. Neptune City,NJ: T.F.H. Publications, Inc.

Zimmermann, E. 1990. Behavioral signals andreproduction modes in the neotropical family


Dendrobtidae. In W. Hanke (editor), Biologyand physiology of amphibians: 61–73. NewYork: Gustav Fischer.

Zimmermann, H., and E. Zimmermann. 1981.Sozialverhalten, Fortpflanzungsverhalten undZucht der Farberfrosche Dendrobates histrioni-cus und D. lehmanni sowie einiger andererDendrobatiden. Zeitschrift des Kolner Zoo,24: 83–89.

Zimmermann, H., and E. Zimmermann. 1984.Durch Nachzucht erhalten: Baumsteigerfrosche.

Dendrobates quinquevittatus und D. reticulatus.Aquarien Magazin, 18: 35–41.

Zimmermann, H., and E. Zimmermann. 1985. Zurfortpflanzungsstrategie des PfeilgiftfroschesPhyllobates terribilis Myers, Daly & Malkin,1978 (Salientia: Dendrobatidae). Salamandra,21: 281–297.

Zimmermann, H., and E. Zimmermann. 1988.Etho-Taxonomie und zoogeographische arten-gruppenbildung bei pfeilgiftfroschen (Anura:Dendrobatidae). Salamandra, 24: 125–160.


APPENDIX 1CHRONOLOGY OF AVAILABLE SPECIES NAMES PROPOSED OR CURRENTLY IN DENDROBATOIDEA,

WITH ORIGINAL, CURRENT, AND PROPOSED GENERIC PLACEMENT

Name Authorship Status Original genus Current genusa

Revised

taxonomy

tinctorius Cuvier, 1797 Rana Dendrobates Dendrobates

nigerrima Spix, 1824 trivittatus Hyla — —

trivittata Spix, 1824 Hyla Epipedobates Ameerega

picta Tschudi, 1838 Hylaplesia Epipedobates Ameerega

bicolor Dumeril & Bibron,

1841

Phyllobates Phyllobates Phyllobates

obscurus Dumeril & Bibron,

1841

trivittatus Dendrobates — —

histrionicus Berthold, 1845 Dendrobates Dendrobates Oophaga

lateralis Guichenot, 1848 Batrachyla taeniata Dendrobates Batrachyla

(Ceratophryidae)

Batrachyla

(Ceratophryidae)

inhambanensis Bianconi, 1849 Phrynomantis

bifasciatus

Dendrobates Phrynomantis

(Microhylidae)

Phrynomantis

(Microhylidae)

auratus Girard, 1855 Dendrobates Dendrobates Dendrobates

lugubris O. Schmidt, 1857 Dendrobates Phyllobates Phyllobates

pumilio O. Schmidt, 1857 Dendrobates Dendrobates Oophaga

speciosus O. Schmidt, 1857 Dendrobates Dendrobates Oophaga

latimaculatus Gunther, 1859 ‘‘1858’’ auratus Dendrobates — —

truncatus Cope, 1861 Phyllobates Dendrobates Dendrobates

limbatus Cope, 1862 Phyllobates ‘‘Euhyas’’

(Brachycephalidae)

‘‘Euhyas’’

(Brachycephalidae)

glandulosus Fitzinger, 1863 Physalaemus olfersi Phyllobates Physalaemus

(Leptodactylidae)

Physalaemus

(Leiuperidae)

latinasus Cope, 1863 Phyllobates Colostethus Colostethus

braccatus Steindachner, 1864 Dendrobates Epipedobates Ameerega

cocteaui Steindachner, 1864 histrionicus Dendrobates — —

daudini Steindachner, 1864 tinctorius Dendrobates — —

eucnemis Steindachner, 1864 pictus Dendrobates — —

galactonotus Steindachner, 1864 Dendrobates Dendrobates Adelphobates

leucomelas Steindachner, 1864 Dendrobates Dendrobates Dendrobates

quinquevittatus Steindachner, 1864 Dendrobates Dendrobates Adelphobates

ridens Cope, 1866 Phyllobates ‘‘Eleutherodactylus’’

(Brachycephalidae)

‘‘Eleutherodactylus’’

(Brachycephalidae)

peruensis Steindachner, 1867 nomen dubium Phyllobates nomen dubium nomen dubium

typographus Keferstein, 1867 pumilio Dendrobates — —

inguinalis Cope, 1868 Prostherapis Colostethus Colostethus

chocoensis Posada Arango, 1869 bicolor Phyllobates — —

verruculatus W. Peters, 1870 Phyllobates Syrrhophus

(Brachycephalidae)

Syrrhophus

(Brachycephalidae)

bocagei Jimenez de la Espada, 1871 Hyloxalus Colostethus Hyloxalus

fuliginosus Jimenez de la Espada, 1871 Hyloxalus Colostethus Hyloxalus

pulchellum Jimenez de la Espada, 1871 Phyllodromus Colostethus Hyloxalus

betsileo Grandidier, 1872 Dendrobates Mantella

(Mantellidae)

Mantella

(Mantellidae)

madagascariensis Grandidier, 1872 Dendrobates Mantella (Mantellidae) Mantella

(Mantellidae)

chalceus W. Peters, 1873 Phyllobates ‘‘Eleutherodactylus’’

(Brachycephalidae)


(Brachycephalidae)

maculatus W. Peters, 1873 Dendrobates Epipedobates Ameerega

ignitus Cope, 1874 pumilio Dendrobates — —

labialis Cope, 1874 Dendrobates Epipedobates Ameerega



Revised

taxonomy

hylaeformis Cope, 1875 Phyllobates ‘‘Eleutherodactylus’’

Brachycephalidae


Brachycephalidae

talamancae Cope, 1875 Dendrobates Colostethus Allobates

cystignathoides Cope, 1877 Phyllobates Syrrhophus

(Brachycephalidae)

Syrrhophus

(Brachycephalidae)

ebenaui Boettger, 1880 Mantella betsileo Dendrobates Mantella (Mantellidae) Mantella

(Mantellidae)

parvulus Boulenger, 1882 Dendrobates Epipedobates Ameerega

whymperi Boulenger, 1882 Prostherapis Colostethus Hyloxalus

femoralis Boulenger, 1883 Prostherapis Epipedobates Allobates

hahneli Boulenger, 1883 Dendrobates Epipedobates Ameerega

fantasticus Boulenger, 1884 ‘‘1883’’ Dendrobates Dendrobates Ranitomeya

reticulatus Boulenger, 1884 ‘‘1883’’ Dendrobates Dendrobates Ranitomeya

trilineatus Boulenger, 1884 ‘‘1883’’ Phyllobates Colostethus Allobates

braccatus Cope, 1887 braccatus

Steindachner, 1864

Dendrobates — —

brunneus Cope, 1887 Prostherapis Colostethus Allobates

trinitatis Garman, 1887 Phyllobates Mannophryne Mannophryne

herminae Boettger, 1893 Prostherapis Mannophryne Mannophryne

vittatus Cope, 1893 Dendrobates Phyllobates Phyllobates

infraguttatus Boulenger, 1898 Phyllobates Colostethus Hyloxalus

opisthomelas Boulenger, 1899 Dendrobates Minyobates Ranitomeya

palmatus Werner, 1899 Phyllobates Colostethus Rheobates

pratti Boulenger, 1899 Phyllobates Colostethus Colostethus

subpunctatus Cope, 1899 Prostherapis Colostethus Hyloxalus

tricolor Boulenger, 1899 Prostherapis Epipedobates Epipedobates

variabilis Werner, 1899 subpunctatus Prostherapis — —

vertebralis Boulengeri, 1899 Phyllodromus Colostethus Hyloxalus

amoenus Werner, 1901 auratus Dendrobates — —

bolivianus Boulenger, 1902 Prostherapis Epipedobates Ameerega

alboguttatus Boulenger, 1903 Phyllobates Nephelobates Aromobates

festae Peracca, 1904 parvulus Prostherapis — —

flavopicta A. Lutz, 1925 Hylaplesia Epipedobates Ameerega

femoralis Barbour, 1905 boulengeri Prostherapis — —

equatorialis Barbour, 1908 Eleutherodactylus

unistrigatus

Prostherapis ‘‘Eleutherodactylus’’

(Brachycephalidae)


(Brachycephalidae)

boulengeri Barbour, 1909 Prostherapis Epipedobates Epipedobates

chocoensis Boulenger, 1912 Hylixalus Colostethus Hyloxalus

collaris Boulenger, 1912 Hylixalus Mannophryne Mannophryne

huigrae Fowler, 1913 ‘‘Eleutherodactylus’’

diastema

Hyloxalus ‘‘Eleutherodactylus’’

(Brachycephalidae)


(Brachycephalidae)

aurotaenia Boulenger, 1914 ‘‘1913’’ Dendrobates Phyllobates Phyllobates

coctaei Boulenger, 1914 ‘‘1913’’ histrionicus Dendrobates — —

paraensis Boulenger, 1914 ‘‘1913’’ galactonotus Dendrobates — —

walkeri Ruthven, 1915 Geobatrachus Geobatrachus

(Brachycephalidae)

Geobatrachus

(Brachycephalidae)

tarsalis Werner, 1916 subpunctatus Prostherapis — —

kingsburyi Boulenger, 1918 Phyllobates Colostethus Allobates

ranoides Boulenger, 1918 Dendrobates Colostethus Allobates

granuliventris Boulenger, 1919 palmatus Hylixalus — —

sylvaticus Barbour & Noble, 1920 Phyllobates Colostethus Hyloxalus

anthonyi Noble, 1921 Phyllobates Epipedobates Epipedobates

beatriciae Barbour & Dunn, 1921 lugubris Phyllobates — —

beebei Noble, 1923 Hyloxalus Colostethus Anomaloglossus

nubicola Dunn, 1924 Phyllobates Colostethus Silverstoneia



Revised

taxonomy

nigriventris A. Lutz, 1925 Hylaplesia ‘‘Eleutherodactylus’’

(Brachycephalidae)


(Brachycephalidae)

olfersioides A. Lutz, 1925 Eupemfix Colostethus Allobates

tetravittatus Miranda Ribeiro, 1926 trivittatus Dendrobates — —

brasiliensis Witte, 1930 Crossodactylus

gaudichaudi

Phyllobates Crossodactylus

(Hylodidae)

Crossodactylus

(Hylodidae)

flotator Dunn, 1931 Phyllobates Colostethus Silverstoneia

mandelorum Schmidt, 1932 Phyllobates Colostethus Allobates

panamensis Dunn, 1933 Hyloxalus Colostethus Colostethus

minutus Shreve, 1935 Dendrobates Minyobates Ranitomeya

ventrimaculatus Shreve, 1935 Dendrobates Dendrobates Ranitomeya

shrevei Dunn, 1940 minutus Dendrobates — —

vergeli Hellmich, 1940 Hyloxalus Colostethus Hyloxalus

bassleri Melin, 1941 Dendrobates Epipedobates Ameerega

igneus Melin, 1941 Dendrobates Dendrobates Ranitomeya

marchesianus Melin, 1941 Phyllobates Colostethus Allobates

peruvianus Melin, 1941 Phyllobates Colostethus Hyloxalus

intermedius Andersson, 1945 kingsburyi Phyllobates — —

riocasangae Andersson, 1945 pulchellus Phyllobates — —

taeniatus Andersson, 1945 pulchellus Phyllobates — —

galindoi Trapido, 1953 pumilio Dendrobates — —

bromelicola Test, 1956 Phyllobates Colostethus Allobates

confluens Funkhouser, 1956 histrionicus Dendrobates — —

espinosai Funkhouser, 1956 Phyllobates Epipedobates Epipedobates

neblina Test, 1956 Prostherapis Mannophryne Mannophryne

sylvaticus Funkhouser, 1956 Dendrobates Dendrobates Oophaga

granuliferus Taylor, 1958 Dendrobates Dendrobates Oophaga

machadoi Bokermann, 1958 tinctorius Dendrobates — —

dunni Rivero, 1961 Prostherapis Colostethus Aromobatidae incerta

sedis

shrevei Rivero, 1961 Prostherapis Colostethus Anomaloglossus

mertensi Cochran & Goin, 1964 Phyllobates Colostethus Colostethus

guayanensis Heatwole et al., 1965 pictus Phyllobates — —

riveroi Donoso-Barros, 1965 ‘‘1964’’ Prostherapis Mannophryne Mannophryne

alagoanus Bokermann, 1967 Phyllobates Colostethus Allobates

capixaba Bokermann, 1967 Phyllobates Colostethus Allobates

carioca Bokermann, 1967 Phyllobates Colostethus Allobates

azureus Hoogmoed, 1969 Dendrobates Dendrobates Dendrobates

ingeri Cochran & Goin, 1970 Dendrobates Epipedobates Ameerega

thorntoni Cochran & Goin, 1970 Phyllobates Colostethus Colostethus

walesi Cochran & Goin, 1970 subpunctatus Phyllobates — —

anthracinus Edwards, 1971 Colostethus Colostethus Hyloxalus

elachyhistus Edwards, 1971 Colostethus Colostethus Hyloxalus

fraterdanieli Silverstone, 1971 Colostethus Colostethus Colostethus

lehmanni Silverstone, 1971 Colostethus Colostethus Hyloxalus

ramosi Silverstone, 1971 Colostethus Colostethus Hyloxalus

steyermarki Rivero, 1971 Dendrobates Minyobates Minyobates

meridensis Dole & Durant, 1972 Colostethus Nephelobates Aromobates

sauli Edwards, 1974 Colostethus Colostethus Hyloxalus

abditaurantius Silverstone, 1975 Colostethus Colostethus Hyloxalus

altobueyensis Silverstone, 1975 Dendrobates Minyobates Ranitomeya

degranvillei Lescure, 1975 Colostethus Colostethus Anomaloglossus

fulguritus Silverstone, 1975 Dendrobates Minyobates Ranitomeya

goianus Bokermann, 1975 Colostethus Colostethus Allobates

imbricolus Silverstone, 1975 Colostethus Colostethus Colostethus



Revised

taxonomy

abditus Myers & Daly, 1976 Dendrobates Minyobates Ranitomeya

lehmanni Myers & Daly, 1976 Dendrobates Dendrobates Oophaga

occultator Myers & Daly, 1976 Dendrobates Dendrobates Oophaga

petersi Silverstone, 1976 Phyllobates Epipedobates Ameerega

pulchripectis Silverstone, 1976 Phyllobates Epipedobates Ameerega

smaragdinus Silverstone, 1976 Phyllobates Epipedobates Ameerega

viridis Myers & Daly, 1976 Dendrobates Minyobates Ranitomeya

zaparo Silverstone, 1976 Phyllobates Epipedobates Allobates

haydeeae Rivero, 1978 ‘‘1976’’ Colostethus Nephelobates Aromobates

orostoma Rivero, 1978 ‘‘1976’’ Colostethus Nephelobates Aromobates

terribilis Myers et al., 1978 Phyllobates Phyllobates Phyllobates

silverstonei Myers & Daly, 1979 Dendrobates Epipedobates Ameerega

bombetes Myers & Daly, 1980 Dendrobates Minyobates Ranitomeya

erythromos Vigle & Miyata, 1980 Dendrobates Epipedobates Ameerega

humilis Rivero 1980 ‘‘1978’’ Colostethus Colostethus Allobates

inflexus Rivero, 1980 ‘‘1978’’ alboguttatus Colostethus — —

leopardalis Rivero, 1980 ‘‘1978’’ Colostethus Nephelobates Aromobates

mayorgai River, 1980 ‘‘1978’’ Colostethus Nephelobates Aromobates

saltuensis Rivero 1980 ‘‘1978’’ Colostethus Colostethus Aromobates

myersi Pyburn, 1981 Dendrobates Epipedobates Allobates

andinus Myers & Burrowes, 1987 Dendrobates Epipedobates Ameerega

captivus Myers, 1982 Dendrobates Dendrobates Adelphobates

edwardsi Lynch, 1982 Colostethus Colostethus Hyloxalus

mysteriosus Myers, 1982 Dendrobates Dendrobates Dendrobatinae incerta

sedis

ruizi Lynch, 1982 Colostethus Colostethus Hyloxalus

syntomopus Lynch & Ruiz-Carranza, 1982 Atopophrynus Atopophrynus

(Brachycephalidae)

Atopophrynus

(Brachycephalidae)

vanzolinii Myers, 1982 Dendrobates Dendrobates Ranitomeya

olmonae Hardy, 1983 Colostethus Mannophryne Mannophryne

arboreus Myers et al., 1984 Dendrobates Dendrobates Oophaga

littoralis Pefaur, 1984 Colostethus Colostethus Hyloxalus

agilis Lynch & Ruiz-Carranza, 1985 Colostethus Colostethus Colostethus

azureiventris Kneller & Henle, 1985 Phyllobates Cryptophyllobates Hyloxalus

duranti Pefaur, 1985 Colostethus Nephelobates Aromobates

guatopoensis Dixon & Rivero Blanco, 1985 oblitteratus Colostethus — —

molinarii La Marca, 1985 Colostethus Nephelobates Aromobates

serranus Pefaur, 1985 Colostethus Nephelobates Aromobates

brachistriatus Rivero & Serna, 1986 Colostethus Colostethus Colostethus

breviquartus Rivero & Serna, 1986 Colostethus Colostethus Hyloxalus

imitator Schulte, 1986 Dendrobates Dendrobates Ranitomeya

nexipus Frost, 1986 Colostethus Colostethus Hyloxalus

oblitterata Rivero, 1986 ‘‘1984’’ Colostethus Mannophryne Mannophryne

peruviridis Bauer, 1986 Ameerega Epipedobates Ameerega

sanmartini Rivero, Langone, & Prigioni,

1986

Colostethus Colostethus Allobates

exasperatus Duellman & Lynch, 1988 Colostethus Colostethus Hyloxalus

mystax Duellman & Simmons, 1988 Colostethus Colostethus Hyloxalus

shuar Duellman & Simmons, 1988 Colostethus Colostethus Hyloxalus

variabilis Zimmermann & Zimmermann,

1988

Dendrobates Dendrobates Ranitomeya

ardens Jungfer, 1989 cainarachi Epipedobates — —

bilinguis Jungfer, 1989 Epipedobates Epipedobates Ameerega

cainarachi Schulte, 1989 Epipedobates Epipedobates Ameerega



Revised

taxonomy

stepheni Martins, 1989 Colostethus Colostethus Anomaloglossus

yustizi La Marca, 1989 Colostethus Mannophryne Mannophryne

alacris Rivero & Granados-Diaz, 1990

‘‘1989’’

Colostethus Colostethus Colostethus

castaneoticus Caldwell & Myers, 1990 Dendrobates Dendrobates Adelphobates

pinguis Rivero, Granados-Dias, 1990

‘‘1989’’

Colostethus Colostethus Hyloxalus

rufulus Gorzula, 1990 ‘‘1988’’ Dendrobates Epipedobates Allobates

betancuri Rivero & Serna, 1991 Colostethus Colostethus Hyloxalus

cevallosi Rivero, 1991 Colostethus Colostethus Hyloxalus

citreicola Rivero, 1991b nexipus Colostethus — —

faciopunctulatus Rivero, 1991a Colostethus Colostethus Hyloxalus

fallax Rivero, 1991b Colostethus Colostethus Hyloxalus

furviventris Rivero & Serna, 1991 Colostethus Colostethus Colostethus

idiomelus Rivero, 1991a Colostethus Colostethus Hyloxalus

jacobuspetersi Rivero, 1991 Colostethus Colostethus Colostethus

lacrimosus Myers, 1991 Colostethus Colostethus Anomaloglossus

larandina Yustiz, 1991 Colostethus Mannophryne Mannophryne

maculosus Rivero, 1991 Colostethus Colostethus Hyloxalus

marmoreoventris Rivero, 1991b Colostethus Colostethus Hyloxalus

mittermieri Rivero, 1991 Colostethus Colostethus Hyloxalus

nocturnus Myers et al., 1991 Aromobates Aromobates Aromobates

paradoxus Rivero, 1991 tricolor Colostethus — —

parcus Rivero, 1991b Colostethus Colostethus Hyloxalus

peculiaris Rivero, 1991 Colostethus Colostethus Hyloxalus

poecilonotus Rivero, 1991a Colostethus Colostethus Dendrobatoidea

incerta sedis

pumilus Rivero, 1991b Colostethus Colostethus Hyloxalus

sirensis Aichinger, 1991 Dendrobates Dendrobates Ranitomeya

tergogranularis Rivero, 1991 pulchellus Colostethus — —

torrenticola Rivero, 1991 jacobuspetersi Colostethus — —

yaguara Rivero & Serna, 1991 Colostethus Colostethus Colostethus

biolat Morales, 1992 Dendrobates Dendrobates Ranitomeya

lamasi Morales, 1992 Dendrobates Dendrobates Ranitomeya

mcdiarmidi Reynolds & Foster, 1992 Colostethus Colostethus Allobates

virolinensis Ruiz-Carranza & Ramırez-

Pinilla, 1992

Minyobates Minyobates Ranitomeya

argyrogaster Morales & Schulte, 1993 Colostethus Colostethus Hyloxalus

capurinensis Pefaur, 1993 Colostethus Colostethus Aromobates

fugax Morales & Schulte, 1993 Colostethus Colostethus Colostethus

macero Rodrıguez & Myers, 1993 Epipedobates Epipedobates Ameerega

chalcopis Kaiser et al., 1994 Colostethus Colostethus Allobates

juanii Morales, 1994 Colostethus Colostethus Allobates

utcubambensis Morales, 1994 Colostethus Colostethus Hyloxalus

awa Coloma, 1995 Colostethus Colostethus Hyloxalus

cordilleriana La Marca, 1995 ‘‘1994’’ Mannophryne Mannophryne Mannophryne

delatorreae Coloma, 1995 Colostethus Colostethus Hyloxalus

machalilla Coloma, 1995 Colostethus Colostethus Epipedobates

maquipucuna Coloma, 1995 Colostethus Colostethus Hyloxalus

toachi Coloma, 1995 Colostethus Colostethus Hyloxalus

parkerae Meinhardt & Parmelee, 1996 Colostethus Colostethus Anomaloglossus

vicentei Jungfer et al., 1996 Dendrobates Dendrobates Oophaga

atopoglossus Grant et al., 1997 Colostethus Colostethus Anomaloglossus

ayarzaguenai La Marca, 1997 ‘‘1996’’ Colostethus Colostethus Anomaloglossus



Revised

taxonomy

guanayensis La Marca, 1997 ‘‘1996’’ Colostethus Colostethus Anomaloglossus

murisipanesis La Marca, 1997 ‘‘1996’’ Colostethus Colostethus Anomaloglossus

parimae La Marca, 1997 ‘‘1996’’ Colostethus Colostethus Anomaloglossus

praderioi La Marca, 1997 ‘‘1996’’ Colostethus Colostethus Anomaloglossus

roraima La Marca, 1997 ‘‘1996’’ Colostethus Colostethus Anomaloglossus

rubriventris Lotters et al., 1997 Epipedobates Epipedobates Ameerega

ruthveni Kaplan, 1997 Colostethus Colostethus Colostethus

tamacuarensis Myers & Donnelly, 1997 Colostethus Colostethus Anomaloglossus

tepuyensis La Marca, 1997 ‘‘1996’’ Colostethus Colostethus Anomaloglossus

fascianiger Grant & Castro, 1998 Colostethus Colostethus Hyloxalus

lynchi Grant, 1998 Colostethus Colostethus Colostethus

planipaleae Morales & Velazco, 1998 Epipedobates Epipedobates Ameerega

amazonicus Schulte, 1999 Dendrobates Dendrobates Ranitomeya

baeobatrachus Boistel & de Massary, 1999 Colostethus Colostethus Anomaloglossus

caquetio Mijares-Urrutia & Arends R.,

1999

Mannophryne Mannophryne Mannophryne

duellmani Schulte, 1999 Dendorbates Dendrobates Ranitomeya

flavovittatus Schulte, 1999 Dendrobates Dendrobates Ranitomeya

intermedius Schulte, 1999 Dendrobates Dendrobates Ranitomeya

lamarcai Mijares-Urrutia & Arends R.,

1999

Mannophryne Mannophryne Mannophryne

pongoensis Schulte, 1999 Epipedobates Epipedobates Ameerega

rubrocephalus Schulte, 1999 Dendrobates Dendrobates Ranitomeya

yurimaguensis Schulte, 1999 imitator Dendrobates — —

borjai Rivero & Serna, 2000 ‘‘1995’’ Colostethus Colostethus Hyloxalus

cacerensis Rivero & Serna, 2000 ‘‘1995’’ inguinalis Colostethus — —

claudiae Junger et al., 2000 Dendrobates Dendrobates Ranitomeya

dysprosium Rivero & Serna, 2000 ‘‘1995’’ Colostethus Colostethus Colostethus

erasmios Rivero & Serna, 2000 ‘‘1995’’ Colostethus Colostethus Silverstoneia

excisus Rivero & Serna, 2000 ‘‘1995’’ Colostethus Colostethus Hyloxalus

picachos Ardila-Robayo et al., 2000

‘‘1999’’

Colostethus Colostethus Allobates

pseudopalmatus Rivero & Serna, 2000 ‘‘1995’’ Colostethus Colostethus Rheobates

ramirezi Rivero & Serna, 2000 ‘‘1995’’ Colostethus Colostethus Dendrobatidae incerta

sedis

simulans Myers et al., 2000 Epipedobates Epipedobates Ameerega

wayuu Acosta et al., 2000 ‘‘1999’’ Colostethus Colostethus Allobates

alessandroi Grant & Rodrıguez, 2001 Colostethus Colostethus Allobates

caeruleodactylus Lima & Caldwell, 2001 Colostethus Colostethus Allobates

melanolaemus Grant & Rodrıguez, 2001 Colostethus Colostethus Allobates

undulatus Myers & Donnelly, 2001 Colostethus Colostethus Allobates

saltuarius Grant & Ardila-Robayo, 2002 Colostethus Colostethus Hyloxalus

cepedai Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

conspicuus Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

crombiei Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

fratisenescus Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

fuscellus Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

gasconi Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

insperatus Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

masniger Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

ornatus Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

sumtuosus Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates

vanzolinius Morales, 2002 ‘‘2000’’ Colostethus Colostethus Allobates


APPENDIX 2CHRONOLOGY OF AVAILABLE DENDROBATOID GENUS-GROUP NAMES


Revised

taxonomy

nidicola Caldwell & Lima, 2003 Colostethus Colostethus Allobates

patitae Lotters et al., 2003 Colostethus Colostethus Hyloxalus

aeruginosus Duellman, 2004 Colostethus Colostethus Hyloxalus

craspedoceps Duellman, 2004 Colostethus Colostethus Allobates

eleutherodactylus Duellman, 2004 Colostethus Colostethus Hyloxalus

insulatus Duellman, 2004 Colostethus Colostethus Hyloxalus

leucophaeus Duellman, 2004 Colostethus Colostethus Hyloxalus

pulcherrimus Duellman, 2004 Colostethus Colostethus Hyloxalus

sordidatus Duellman, 2004 Colostethus Colostethus Hyloxalus

spilotogaster Duellman, 2004 Colostethus Colostethus Hyloxalus

nubeculosus Jungfer & Bohme, 2004 Dendrobates Dendrobates Dendrobates

pittieri La Marca et al., 2004 Colostethus Colostethus Allobates

triunfo Barrio-Amoros et al., 2005 Colostethus Colostethus Anomaloglossus

wothuja Barrio-Amoros et al., 2005 Colostethus Colostethus Anomaloglossus

chlorocraspedus Caldwell, 2005 Cryptophyllobates Cryptophyllobates Hyloxalus

a As discussed in Materials and Methods, there is no universally accepted ‘‘current’’ taxonomy of dendrobatoid frogs,

and this category is intended only to provide a general reference for the proposed changes. Nondendrobatoid taxonomy

follows Frost et al. (2006).

Name Authorship Type species

Hysaplesia Boie, 1826 Rana tinctoria

Dendrobates Wagler, 1830 Rana tinctoria

Phyllobates Dumeril and Bibron, 1841 Phyllobates bicolor

Eubaphus Bonaparte, 1850 Rana tinctoria

Colostethus Cope, 1866 Phyllobates latinasus

Prostherapis Cope, 1868 Prostherapis inguinalis

Phyllodromus Jimenez de la Espada, 1871’’1870’’ Phyllodromus pulchellum

Hyloxalus Jimenez de la Espada, 1871 ‘‘1870’’ Hyloxalus fuliginosus

Ameerega Bauer, 1986 Hyla trivittata

Minyobates Myers, 1987 Dendrobates steyermarki

Epipedobates Myers, 1987 Prostherapis tricolor

Phobobates Zimmermann and Zimmermann, 1988 Dendrobates silverstonei

Allobates Zimmermann and Zimmermann, 1988 Prostherapis femoralis

Pseudendrobates Bauer, 1988 Dendrobates silverstonei

Ranitomeya Bauer, 1988 Dendrobates reticulatus

Oophaga Bauer, 1988 Dendrobates pumilio

Aromobates Myers, Paolillo, and Daly, 1991 Aromobates nocturnus

Mannophryne La Marca, 1992 Colostethus yustizi

Nephelobates La Marca, 1994 Phyllobates alboguttatus

Paruwrobates Bauer, 1994 Dendrobates andinus

Cryptophyllobates Lotters, Jungfer, and Widmer, 2000 Phyllobates azureiventris

Adelphobates new genus Dendrobates castaneoticus

Anomaloglossus new genus Hyloxalus beebei

Rheobates new genus Phyllobates palmatus

Silverstoneia new genus Phyllobates nubicola


APPENDIX 3CHRONOLOGY OF DENDROBATOID FAMILY-GROUP NAMES

APPENDIX 4NUMBERS AND REFERENCES FOR SEQUENCES OBTAINED FROM GENBANK

Accession numbers, length (in base pairs; bp), and publication reference for GenBank sequencesincluded in this study.

Name Authorship

Phyllobatae Fitzinger, 1843

Eubaphidae Bonaparte, 1850

Eubaphina Bonaparte, 1850

Hylaplesidae Gunther, 1858

Hylaplesina Gunther, 1858

Dendrobatidae Cope, 1865

Colostethidae Cope, 1867

Hylaplesiina Gunther, 1868

Calostethina Mivart, 1869

Hylaplesiidae Cope, 1875

Phyllobatidae Parker, 1933

Allobatinae new subfamily

Anomaloglossinae new subfamily

Aromobatidae new family

Aromobatinae new subfamily

Hyloxalinae new subfamily

GenBank identification GenBank number Locus Length (bp) Reference

Colostethus awa AY364544 12S, tRNAval, 16S 2445 Santos et al., 2003

Colostethus baeobatrachus AY263231 16S 535 Vences et al., 2003a

Colostethus bocagei AY364545 12S, tRNAval, 16S 2435 Santos et al., 2003

Colostethus degranvillei AY263260 16S 506 Vences et al., 2003a



Colostethus elachyhistus AY364546 12S, tRNAval, 16S 2440 Santos et al., 2003

Colostethus fugax AY364547 12S, tRNAval, 16S 2442 Santos et al., 2003

Colostethus humilis AJ430673 16S 544 La Marca et al., 2002

Colostethus infraguttatus AY326028 12S, tRNAval, 16S 2418 Darst and Cannatella,

2004

Colostethus infraguttatus AY364548 12S, tRNAval, 16S 2433 Santos et al., 2003

Colostethus insperatus AY364557 12S, tRNAval, 16S 2434 Santos et al., 2003

Colostethus kingsburyi AY364550 12S, tRNAval, 16S 2457 Santos et al., 2003

Colostethus kingsburyi AY364549 12S, tRNAval, 16S 2446 Santos et al., 2003

Colostethus machalilla AY364551 12S, tRNAval, 16S 2444 Santos et al., 2003

Colostethus maculosus AY364552 12S, tRNAval, 16S 2436 Santos et al., 2003

Colostethus nexipus AY364553 12S, tRNAval, 16S 2444 Santos et al., 2003

Colostethus palmatus AY263228 16S 478 Vences et al., 2003a

Colostethus pratti AY263238 16S 499 Vences et al., 2003a

Colostethus pulchellus AY364554 12S, tRNAval, 16S 2443 Santos et al., 2003

Colostethus sauli AY364555 12S, tRNAval, 16S 2445 Santos et al., 2003

Colostethus sp MNHN1995-9454 AY263236 16S 546 Vences et al., 2003a



Colostethus sp QCAZ16490 AY364556 12S, tRNAval, 16S 2443 Santos et al., 2003





Colostethus stepheni AY263237 16S 487 Vences et al., 2003a

Colostethus subpunctatus AY263242 16S 448 Vences et al., 2003a

Colostethus toachi AY364563 12S, tRNAval, 16S 2444 Santos et al., 2003

Colostethus vertebralis AY364564 12S, tRNAval, 16S 2435 Santos et al., 2003

Dendrobates amazonicus AF482800 cytochrome b 268 Symula et al., 2003

Dendrobates amazonicus AF482785 16S 463 Symula et al., 2003

Dendrobates amazonicus AF482770 12S 280 Symula et al., 2003

Dendrobates auratus AY364370 16S 571 Biju and Bossuyt, 2003

Dendrobates auratus AY364349 12S, tRNAval, 16S 748 Biju and Bossuyt, 2003

Dendrobates auratus AY364395 rhodopsin 316 Biju and Bossuyt, 2003

Dendrobates biolat AF482809 cytochrome b 268 Symula et al., 2003

Dendrobates biolat AF482794 16S 506 Symula et al., 2003

Dendrobates biolat AF482779 12S 311 Symula et al., 2003

Dendrobates duellmani AY364566 12S, tRNAval, 16S 2456 Santos et al., 2003

Dendrobates fantasticus AF128624 cytochrome b 284 Clough and Summers,

2000

Dendrobates fantasticus AF128623 12S 361 Clough and Summers,

2000

Dendrobates fantasticus AF128622 16S 522 Clough and Summers,

2000

Dendrobates fantasticus DfTY26b AF412503 cytochrome b 272 Symula et al., 2003

Dendrobates fantasticus DfTY26b AF412475 16S 409 Symula et al., 2003

Dendrobates fantasticus DfTY26b AF412447 12S 282 Symula et al., 2003

Dendrobates imitator AF124118 16S 560 Vences et al., 2000

Dendrobates imitator AY263217 12S 354 Vences et al., 2003a

Dendrobates imitator AY263267 16S 492 Vences et al., 2003a

Dendrobates imitator DiTY26b AF412518 cytochrome b 282 Symula et al., 2003

Dendrobates imitator DiTY26b AF412490 16S 406 Symula et al., 2003

Dendrobates imitator DiTY26b AF412462 12S 282 Symula et al., 2003

Dendrobates lamasi AF482808 cytochrome b 268 Symula et al., 2003

Dendrobates lamasi AF482793 16S 499 Symula et al., 2003

Dendrobates lamasi AF482778 12S 311 Symula et al., 2003

Dendrobates quinquevittatus AY263253 16S 575 Vences et al., 2003a

Dendrobates sp. QCAZ16558 AY364568 12S, tRNAval, 16S 2459 Santos et al., 2003

Dendrobates steyermarki AY263244 16S 547 Vences et al., 2003a

Dendrobates sylvaticus AY364569 12S, tRNAval, 16S 2449 Santos et al., 2003

Dendrobates variabilis AF412463 12S 282 Symula et al., 2003

Dendrobates variabilis AY263249 16S 575 Vences et al., 2003a

Dendrobates ventrimaculatus

(French Guiana)

AY263248 16S Vences et al., 2003a

Epipedobates anthonyi QCAZ16591

(sensu Graham et al., 2004)

AY364576 12S, tRNAval, 16S 2443 Santos et al., 2003

Epipedobates azureiventris AY263255 16S 511 Vences et al., 2003a

Epipedobates azureiventris AF124124 16S 542 Vences et al., 2000

Epipedobates azureiventris AF128562 cytochrome b 283 Clough and Summers,

2000

Epipedobates azureiventris AF128561 12S 277 Clough and Summers,

2000

Epipedobates azureiventris AF128560 16S 516 Clough and Summers,

2000



Epipedobates bassleri AF128565 cytochrome b 275 Clough and Summers,

2000

Epipedobates bassleri AF128564 12S 358 Clough and Summers,

2000

Epipedobates bassleri AF128563 16S 519 Clough and Summers,

2000

Epipedobates bilinguis AY364571 12S, tRNAval, 16S 2430 Santos et al., 2003

Epipedobates bilinguis AF128559 cytochrome b 272 Clough and Summers,

2000

Epipedobates boulengeri AY364572 12S, tRNAval, 16S 2440 Santos et al., 2003

Epipedobates boulengeri AF128556 cytochrome b 278 Clough and Summers,

2000

Epipedobates hahneli (Bolivia) AF282246 16S 421 Lotters and Vences, 2000

Epipedobates hahneli QCAZ13325 AY364573 12S, tRNAval, 16S 2437 Santos et al., 2003

Epipedobates parvulus QCAZ16583 AY364574 12S, tRNAval, 16S 2438 Santos et al., 2003

Epipedobates pictus (sensu stricto) AF124126 16S 555 Vences et al., 2000

Epipedobates rubriventris AF282247 16S 566 Lotters and Vences, 2000

Epipedobates sp. QCAZ16589 AY364575 12S, tRNAval, 16S 2436 Santos et al., 2003

Epipedobates tricolor (sensu Graham

et al. 2004)

AY395961 12S, tRNAval, 16S 2393 Graham et al., 2004

Epipedobates zaparo QCAZ16601 AY364578 12S, tRNAval, 16S 2432 Santos et al. 2003

Epipedobates zaparo QCAZ16604 AY364579 12S, tRNAval, 16S 2433 Santos et al. 2003

Mannophryne collaris AJ430675 16S 534 La Marca et al., 2002

Mannophryne herminae AY263269 16S 506 Vences et al., 2003a

Mannophryne herminae AY263219 12S 358 Vences et al., 2003a

Mannophryne herminae AJ430676 16S 538 La Marca et al., 2002

Mannophryne sp. ULABG 4453 AY263221 16S 535 Vences et al., 2003a




Nephelobates molinarii AY263263 16S 505 Vences et al., 2003a

Nephelobates molinarii AY263216 12S 368 Vences et al., 2003a

Nephelobates molinarii AJ430678 16S 546 La Marca et al., 2002

Nephelobates sp. ULABG 4445 AY263229 16S 540 Vences et al., 2003a

Nephelobates sp. ULABG 4496 AJ430677 16S 543 La Marca et al., 2002

Phyllobates vittatus AY263265 16S 473 Vences et al., 2003a

Phyllobates vittatus AF124134 16S 556 Vences et al., 2000

Phyllobates vittatus AF128582 cytochrome b 284 Clough and Summers,

2000

Phyllobates vittatus AF128581 12S 360 Clough and Summers,

2000

Phyllobates vittatus AF128580 16S 517 Clough and Summers,

2000


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

ala

goanus

538

MR

T6031

Bra

zil:

Bah

ia:

Sao

Jose

da

Vit

ori

a,

Faze

nd

aU

naca

u,

15u0

99S

39u1

89W

DQ

502557

DQ

502126

DQ

502833

DQ

503232

DQ

502342

DQ

503093

4485

All

ophry

ne

ruth

veni

837

MA

D1512

Gu

yan

a:

Kab

oca

lica

mp

,

101

m,

4u1

7.1

09N

58u3

0.5

69W

AY

843564

AY

844538

AY

844766

2422

Als

odes

garg

ola

653

MA

CN

37942

Arg

enti

na:

Neu

qu

en:

Alu

min

e,

stre

am

10

km

WP

rim

ero

s

Pin

os

AY

843565

AY

844539

DQ

284118

AY

844767

AY

844197

4212

anth

onyi

838

RG

No

data

(cap

tive

bre

d)

DQ

502584

DQ

502151

DQ

502853

DQ

502355

DQ

503354

DQ

503104

4620

anth

onyi

1290

AM

CC

125660

Ecu

ad

or:

El

Oro

:7

km

W

Pasa

join

ban

an

a/c

aca

o

DQ

502215

608

arb

ore

us

340

CW

M

18636

Pan

am

a:

Ch

iriq

uı:

Co

nti

nen

tal

div

ide

ab

ove

up

per

Qu

ebra

da

de

Are

nam

1120

m

DQ

502467

DQ

502036

DQ

502763

DQ

503197

DQ

502306

DQ

503312

DQ

503060

DQ

502959

5714

Ate

lognath

us

pata

gonic

us

656

MA

CN

37905

Arg

enti

na:N

euq

uen

:C

ata

nL

il,

Lagu

na

del

Bu

rro

AY

843571

AY

844545

AY

844027

AY

844773

AY

844203

4409

Ate

lopus

spurr

ell

i

1275

MH

NU

C

273

Co

lom

bia

:C

ho

co:

Bah

ıa

So

lan

o,

Qu

ebra

da

Teb

ad

a,

160

m,

06u2

8.7

869N

77u2

0.6

789W

DQ

502200

DQ

502895

DQ

503380

3509

Ate

lopus

zete

ki

1039

UM

FS

11492

Cap

tive

bre

d,

Det

roit

Zo

o

(pare

nta

lst

ock

fro

mP

an

am

a:

Las

Fil

ipin

as,

nea

rS

ora

,

8u3

9.9

99N

80u0

.2499W

)

DQ

283252

DQ

502857

2177

aura

tus

Nic

ara

gu

a365

OM

NH

33270

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te

(ca.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502491

DQ

502060

DQ

502782

3471

aura

tus

Pan

am

a327

US

NM

313818

Pan

am

a:

Bo

cas

del

To

ro:

Lagu

na

de

Tie

rra

Osc

ura

,

3.7

km

So

fT

iger

Key

AY

843803

AY

843581

DQ

502751

AY

844554

DQ

284072

AY

844032

DQ

503304

AY

844781

AY

844211

6238

AP

PE

ND

IX5

TIS

SU

EA

ND

SE

QU

EN

CE

DA

TA

Bel

ow

we

giv

eth

esp

ecie

sid

enti

fica

tio

n(a

str

eate

din

the

text

an

dfi

gs.

70–76),

sam

ple

iden

tifi

cati

on

nu

mb

er(S

am

ple

ID),

sou

rce

(vo

uch

ero

rti

ssu

eid

enti

fica

tio

nn

um

ber

or,

ifu

navail

ab

le,

coll

ecto

ro

rb

reed

er),

loca

lity

,G

enB

an

kacc

essi

on

nu

mb

erfo

rea

chlo

cus,

an

dto

tal

nu

mb

ero

fb

ase

pair

s(b

p)

for

each

term

inal

seq

uen

ced

for

Faiv

ovic

het

al.

(2004),

Fro

stet

al.

(2006),

or

the

pre

sen

tst

ud

y.

Seq

uen

ces

pre

vio

usl

ysu

bm

itte

dto

Gen

Ban

kare

giv

enin

bo

ld(s

eeap

pen

dix

4fo

rad

dit

ion

al

seq

uen

ces

ob

tain

edfr

om

Gen

Ban

k).

See

Mate

rials

an

dM

eth

od

sfo

rco

llec

tio

nab

bre

via

tio

ns.

Lo

cus

ab

bre

via

tio

ns:

28S

(larg

en

ucl

ear

rib

oso

mal

sub

un

it),

CO

I(c

yto

chro

me

co

xid

ase

I),

cytb

(cyto

chro

me

b),

H1

(mit

och

on

dri

al

H-s

tran

dtr

an

scri

pti

on

un

it1),

H3

(his

ton

eH

3),

RA

G1

(rec

om

bin

ati

on

act

ivati

ng

gen

e1),

rho

do

psi

n(r

ho

do

psi

nex

on

1),

an

dS

IA(s

even

thin

ab

sen

tia).


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

aura

tus

Pan

am

a334

CW

M

17698(A

)

Pan

am

a:

Bo

cas

del

To

ro:

No

rth

sid

eIs

laP

ast

ore

s

DQ

502461

DQ

502030

DQ

502758

3471

aura

tus

Pan

am

a335

CW

M

17698(B

)

Pan

am

a:

Bo

cas

del

To

ro:

No

rth

sid

eIs

laP

ast

ore

s

DQ

502462

DQ

502031

DQ

502759

3471

auro

taenia

840

RG

No

data

(cap

tive

bre

d)

DQ

502586

DQ

502153

DQ

502855

DQ

502356

DQ

503356

DQ

503106

4618

‘‘A

yan

gan

na’’

607

RO

M39639

Gu

yan

a:

Mt.

Ayan

gan

na,

no

rth

east

pla

teau

,1490–

1550

m,

5u2

49N

59u5

79W

DQ

502560

DQ

502129

DQ

502836

DQ

503235

DQ

502345

DQ

503163

DQ

503344

DQ

503096

DQ

502993

6230

azu

reus

534

MR

T5089

Bra

zil

(no

oth

erd

ata

)D

Q502553

DQ

502122

DQ

502829

DQ

503228

DQ

502338

DQ

503338

DQ

503089

DQ

502989

5702

azu

reus

1330

BP

N977

Su

rin

am

e:S

ipali

win

i:fo

rest

isla

nd

on

Wsl

op

eo

fV

ier

Geb

rod

ers

Mts

.,S

ipali

win

i

savan

nah

,2u1

.49N

55u5

7.4

19W

DQ

502683

DQ

502251

DQ

502921

DQ

502386

DQ

503388

DQ

503135

DQ

503031

5386

baeo

batr

ach

us

14

PK

-437-1

Fre

nch

Gu

ian

a:

Pic

Mate

cho

,3u4

495

30N

3u2

9190W

DQ

502405

DQ

501980

DQ

502706

3448

baeo

batr

ach

us

42

PK

-437-2

Fre

nch

Gu

ian

a:

Pic

Mate

cho

,3u4

495

30N

3u2

9190W

DQ

502406

DQ

501981

DQ

502707

DQ

502275

3779

baeo

batr

ach

us

43

PK

-437-3

Fre

nch

Gu

ian

a:

Pic

Mate

cho

,3u4

495

30N

3u2

9190W

DQ

502407

DQ

501982

DQ

502708

3448

baeo

batr

ach

us

44

PK

-737-4

Fre

nch

Gu

ian

a:

Pic

Mate

cho

,3u4

495

30N

3u2

9190W

DQ

502408

DQ

501983

DQ

502709

DQ

502276

DQ

503281

DQ

503037

DQ

502936

5367

Batr

achyla

lepto

pus

655

MA

CN

38008

Arg

enti

na:

Ch

ub

ut:

Cu

sham

en,

Lago

Pu

elo

AY

843572

AY

844546

DQ

284119

AY

844028

AY

844774

AY

844204

4727

bee

bei

605

RO

M39631

Gu

yan

a:

Mo

un

t

Ayan

gan

na,

no

rth

east

pla

teau

,1490–1550

m,

5u2

49N

59u5

79W

DQ

502558

DQ

502127

DQ

502834

DQ

503233

DQ

502343

DQ

503342

DQ

503094

DQ

502991

5695

bee

bei

608

RO

M39632

Gu

yan

a:

Mo

un

tA

yan

gan

na,

no

rth

east

pla

teau

,1490–

1550

m,

5u2

49N

59u5

79W

DQ

502561

DQ

502130

DQ

502837

DQ

502346

3781

bic

olo

r1233

MB

No

data

(cap

tive

bre

d)

DQ

502617

DQ

502181

DQ

502884

DQ

502377

DQ

503377

DQ

503019

4989

bil

inguis

378

OM

NH

34125

Ecu

ad

or:

Su

cum

bıo

s:

Est

acı

on

Cie

nti

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502504

DQ

502073

2793


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

bil

inguis

401

OM

NH

34127

Ecu

ad

or:

Su

cum

bıo

s:

Est

acı

on

Cie

nti

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502527

DQ

502095

DQ

503216

DQ

502326

DQ

503157

DQ

503078

DQ

502978

5141

bil

inguis

1303

OM

NH

34126

Ecu

ad

or:

Su

cum

bıo

s:E

stacı

on

Cie

nti

fica

de

Un

iver

sid

ad

Cato

lica

nea

rR

eser

va

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S76u1

09W

DQ

502225

2411

bocagei

343

OM

NH

34070

Ecu

ad

or:

Su

cum

bıo

s:

Dan

tas

Tra

iln

ear

Res

erva

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S76u1

09W

DQ

502469

DQ

502038

DQ

502764

DQ

503199

DQ

502308

DQ

503314

DQ

503062

DQ

502961

5700

bocagei

344

LS

UM

Z

12908

Ecu

ad

or:

Su

cum

bıo

s:

Dan

tas

Tra

iln

ear

Res

erva

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S76u1

09W

DQ

502470

DQ

502039

DQ

502765

3462

bocagei

345

LS

UM

Z

12909

Ecu

ad

or:

Su

cum

bıo

s:

Dan

tas

Tra

iln

ear

Res

erva

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S76u1

09W

DQ

502471

DQ

502040

DQ

502766

3462

bocagei

1267

OM

NH

34072

Ecu

ad

or:

Su

cum

bıo

s:

Dan

tas

Tra

iln

ear

Res

erva

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S76u1

09W

DQ

502628

DQ

502192

DQ

502890

3460

boule

ngeri

280

UM

MZ

227952

Pet

trad

e,im

po

rted

fro

m

Ecu

ad

or

DQ

502447

DQ

283037

DQ

502742

DQ

283768

DQ

284063

DQ

282902

DQ

503301

DQ

282653

DQ

283461

6239

bra

ccatu

s537

MR

T5603

Bra

zil:

Mato

Gro

sso

:A

PM

Man

so,

15u2

79S

58u4

49W

DQ

502556

DQ

502125

DQ

502832

DQ

503231

DQ

502341

DQ

503161

DQ

503341

DQ

503092

5467

‘‘B

row

nsb

erg’’

1328

UT

AA

56469

Su

rin

am

e:B

rok

op

on

do

:

Bro

wn

sber

gN

atu

reP

ark

DQ

502681

DQ

502249

DQ

502919

DQ

503267

DQ

502385

DQ

503134

DQ

503029

5251

bru

nneus

352

OM

NH

34473

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502478

DQ

502047

2795

bru

nneus

612

MP

EG

11923

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502564

DQ

502132

DQ

503237

DQ

502348

DQ

503347

DQ

503098

DQ

502994

5049

bru

nneus

613

OM

NH

34460

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502565

DQ

502133

DQ

502840

3454

bru

nneus

1264

MP

EG

11921

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502625

DQ

502189

2795


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

bru

nneus

1271

OM

NH

34472

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502632

DQ

502196

2794

bru

nneus

1278

OM

NH

34468

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502638

DQ

502203

2795

bru

nneus

1280

OM

NH

34461

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502640

DQ

502205

2795

bru

nneus

1281

OM

NH

34471

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502641

DQ

502206

2796

bru

nneus

1286

OM

NH

34474

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502646

DQ

502211

2797

bru

nneus

1294

OM

NH

34470

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502650

DQ

502216

2796

bru

nneus

1316

OM

NH

34469

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502670

DQ

502238

DQ

502910

3454

caeru

leodacty

lus

406

MP

EG

13809

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502532

DQ

502100

DQ

502814

DQ

503218

DQ

502328

DQ

503329

DQ

503080

4929

caeru

leodacty

lus

621

OM

NH

37410

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502573

DQ

502141

DQ

502845

3453

caeru

leodacty

lus

1261

MP

EG

13808

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502622

DQ

502186

DQ

502887

3453

caeru

leodacty

lus

1287

OM

NH

37411

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502647

DQ

502212

DQ

502899

3453


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

cast

aneoti

cus

363

OM

NH

34517

Bra

zil:

Para

:101

km

S

an

d15

km

ES

an

tare

m

(nea

rR

ioC

uru

a-u

na),

3u9

9S

54u5

09W

DQ

502489

DQ

502058

DQ

502780

DQ

503202

DQ

502311

DQ

502964

4884

Centr

ole

ne

pro

soble

pon

827

SIU

C7053

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

AY

843574

AY

844776

AY

844206

3872

Cera

tophry

s

cra

nw

ell

i

539

JF929

Arg

enti

na:

San

taF

e:V

era,

‘‘L

as

Gam

as’

’

AY

843575

AY

843797

AY

844207

3470

Chacophry

s

pie

rott

ii

1184

AM

NH

A168435

No

data

(pet

trad

e)D

Q283328

2422

chlo

rocr

asp

edus

383

OM

NH

36170

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502509

DQ

502078

DQ

502798

DQ

503210

DQ

502320

DQ

503323

DQ

503072

DQ

502973

5713

chlo

rocr

asp

edus

385

MP

EG

13853

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502511

DQ

502080

DQ

502800

DQ

503212

DQ

502322

DQ

503153

DQ

503325

DQ

503074

DQ

502975

6245

chlo

rocr

asp

edus

623

OM

NH

36169

Bra

zil:

Acr

e:

Po

rto

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502575

DQ

502143

DQ

502846

3463

chlo

rocr

asp

edus

624

MP

EG

12511

Bra

zil:

Acr

e:

Po

rto

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502576

DQ

502144

DQ

502847

3463

chlo

rocr

asp

edus

1235

MP

EG

12505

Bra

zil:

Acr

e:

Po

rto

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502618

DQ

502182

DQ

502885

DQ

503260

DQ

502378

DQ

503378

DQ

503124

DQ

503020

5712

cla

udia

e323

US

NM

-FS

59979

Pan

am

a:

Bo

cas

del

To

ro:

Sen

d

Isla

Po

pa,

1k

mE

Su

mw

oo

d

Ch

an

nel

DQ

502452

DQ

502024

DQ

502747

3467

cla

udia

e324

US

NM

-FS

59980

Pan

am

a:

Bo

cas

del

To

ro:

Sen

d

Isla

Po

pa,

1k

mE

Su

mw

oo

d

Ch

an

nel

DQ

502453

DQ

283042

DQ

502748

DQ

283772

DQ

284071

DQ

503303

DQ

282654

DQ

283462

5703

cla

udia

e330

US

NM

-FS

51785

Pan

am

a:

Bo

cas

del

To

ro:

Isla

Co

lon

,Il

sa

Co

lon

,L

aG

ruta

DQ

502457

DQ

502027

DQ

502754

DQ

503307

DQ

503056

DQ

502956

5058


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

Coch

ranell

a

beja

ranoi

311

NK

A

5292

Bo

livia

:S

an

taC

ruz:

Cab

all

ero

:

San

Juan

Can

ton

,A

mb

oro

,

Nati

on

al

Park

,n

ear

San

Juan

del

Po

rtre

roo

nR

ıoC

erro

Bra

vo

,1800–2100

m,

nea

r

17u5

090

80S

64u2

392

30W

AY

843576

AY

844372

DQ

284066

AY

844029

AY

844372

AY

844777

AY

844208

4731

consp

icuus

614

MP

EG

12321

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502566

DQ

502134

DQ

502841

DQ

503238

DQ

502349

DQ

503348

DQ

503099

DQ

502995

5705

consp

icuus

615

OM

NH

35997

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502567

DQ

502135

DQ

502842

3448

Cro

ssodacty

lus

schm

idti

149

ML

PA

1414

Arg

enti

na:

Mis

ion

es:

Ari

sto

bu

lod

elV

all

e,

Baln

eari

oC

un

ap

iru

AY

843801

AY

843579

DQ

502738

AY

844552

DQ

284050

AY

844031

DQ

503298

AY

844780

AY

844210

6235

‘‘C

uru

a-U

na’’

353

OM

NH

34498

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502479

DQ

502048

DQ

502773

3455

‘‘C

uru

a-U

na’’

354

MP

EG

11961

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502480

DQ

502049

DQ

502774

3456

‘‘C

uru

a-U

na’’

517

MJH

3973

Bra

zil:

Para

:R

ioC

uru

a-u

na

DQ

502542

DQ

502110

DQ

502820

3455

‘‘C

uru

a-U

na’’

1260

OM

NH

34496

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502621

DQ

502185

DQ

502886

3456

‘‘C

uru

a-U

na’’

1268

OM

NH

34497

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502629

DQ

502193

DQ

502891

3455

‘‘C

uru

a-U

na’’

351

LS

UM

Z

15176

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-u

na),

3u9

9S54u5

09W

DQ

502477

DQ

502046

DQ

502772

DQ

503201

DQ

502310

DQ

503316

DQ

503064

DQ

502963

5712

‘‘C

uyab

eno

’’346

OM

NH

34086

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502472

DQ

502041

DQ

502767

DQ

503200

DQ

502309

DQ

503315

DQ

503063

DQ

502962

5706

‘‘C

uyab

eno

’’347

LS

UM

Z

12969

Ecu

ad

or:

Su

cum

bıo

s:

Palm

as

Ro

jas

Tra

iln

ear

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

,220

m,

0u0

9S76u1

09W

DQ

502473

DQ

502042

DQ

502768

3446

‘‘C

uyab

eno

’’348

LS

UM

Z

12970

Ecu

ad

or:

Su

cum

bıo

s:

Palm

as

Ro

jas

Tra

iln

ear

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

,220

m,

0u0

9S76u1

09W

DQ

502474

DQ

502043

DQ

502769

3447


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

‘‘C

uyab

eno

’’349

LS

UM

Z

12971

Ecu

ad

or:

Su

cum

bıo

s:

Res

erva

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S

76u1

09W

DQ

502475

DQ

502044

DQ

502770

3448

‘‘C

uyab

eno

’’350

LS

UM

Z

12972

Ecu

ad

or:

Su

cum

bıo

s:

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502476

DQ

502045

DQ

502771

3447

‘‘C

uyab

eno

’’402

LS

UM

Z

12938

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

nti

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502528

DQ

502096

DQ

502812

3445

‘‘C

uyab

eno

’’403

LS

UM

Z

12950

Ecu

ad

or:

Su

cum

bıo

s:

Dan

tas

Tra

iln

ear

Res

erva

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S76u1

09W

DQ

502529

DQ

502097

DQ

502813

3448

‘‘C

uyab

eno

’’1262

OM

NH

34085

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502623

DQ

502187

DQ

502888

3449

‘‘C

uyab

eno

’’1270

LS

UM

Z

12936

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

nti

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502631

DQ

502195

DQ

502893

3447

‘‘C

uyab

eno

’’1283

OM

NH

34079

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502643

DQ

502208

DQ

502897

3447

‘‘C

uyab

eno

’’1317

LS

UM

Z

12948

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502671

DQ

502239

DQ

502911

3447

Cycl

ora

mphus

bora

ceie

nsi

s

890

CF

BH

5757

Bra

zil:

Sao

Pau

lo:

Pic

ingu

ab

a:

Ub

atu

ba

DQ

502588

DQ

283097

DQ

502856

DQ

283813

DQ

284147

DQ

282924

DQ

503357

DQ

282675

DQ

283498

6222

deg

ranvil

lei

Gu

yan

a278

GB

Gu

yan

a:

Mer

um

e

Mo

un

tain

s

DQ

502445

DQ

502019

DQ

502740

DQ

503188

DQ

502296

DQ

503051

DQ

502951

5260

deg

ranvil

lei

Gu

yan

a279

GB

Gu

yan

a:

Mer

um

e

Mo

un

tain

s

DQ

502446

DQ

502020

DQ

502741

DQ

503189

DQ

502297

DQ

503300

DQ

503052

DQ

502952

5695


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

deg

ranvil

lei

Gu

yan

a1336

CP

I10209

Gu

yan

a:

Maza

run

i-P

oto

ro:

Mt.

Ro

raim

a,

1075

m

DQ

502689

DQ

502257

2800

del

ato

rreae

71

KU

220621

Ecu

ad

or:

Ch

arc

hi:

26.9

to

27.3

km

EM

ald

on

ad

oo

n

road

toT

ulc

an

DQ

502409

DQ

501984

DQ

502710

DQ

502277

DQ

503140

DQ

503282

DQ

503038

5150

Dendro

phry

nis

cus

min

utu

s

532

MJH

7095

Per

u:

Hu

an

uco

:R

ıo

Llu

llap

ich

is,

Pan

gu

an

a

AY

843804

DQ

502120

DQ

502828

AY

844555

DQ

284096

DQ

503337

4535

Edalo

rhin

apere

zi531

MJH

7082

Per

u:

Hu

an

uco

:R

ıo

Llu

llap

ich

is,

Pan

gu

an

a

AY

843585

AY

844558

DQ

284095

AY

844764

DQ

283474

4201

ela

chyhis

tus

Caja

marc

a

105

KU

212522

Per

u:

Caja

marc

a,

Ch

ota

,

4k

mW

Lla

ma,

2500

m

DQ

502417

DQ

501992

DQ

503181

DQ

502284

DQ

503288

DQ

502941

4637

ela

chyhis

tus

Caja

marc

a

106

KU

212523

Per

u:

Caja

marc

a,

Ch

ota

,4

km

WL

lam

a,

2500

m

DQ

502418

DQ

501993

2798

ela

chyhis

tus

Caja

marc

a

107

KU

212524

Per

u:

Caja

marc

a,

Ch

ota

,4

km

WL

lam

a,

2500

m

DQ

502419

DQ

501994

2799

ela

chyhis

tus

Piu

ra108

KU

219749

Per

u:

Piu

ra:

8.5

km

WC

an

chaq

ue

DQ

502420

DQ

501995

DQ

503182

DQ

502285

DQ

503289

DQ

502942

4640

ela

chyhis

tus

Piu

ra114

KU

212514

Per

u:

Piu

ra:

Ayaca

ba,

ca.

Ayaca

ba,

2750

m

DQ

502425

DQ

502000

DQ

503185

DQ

502288

DQ

503291

DQ

503046

DQ

502945

5037

ela

chyhis

tus

Piu

ra115

KU

212515

Per

u:

Piu

ra:

Ayaca

ba,

ca.

Ayaca

ba,

2750

m

DQ

502426

DQ

502001

DQ

502289

DQ

503292

DQ

503047

DQ

502946

4721

ela

chyhis

tus

Piu

ra116

KU

212516

Per

u:

Piu

ra:

Ayaca

ba,

ca.

Ayaca

ba,

2750

m

DQ

502427

DQ

502002

DQ

502722

3461

ela

chyhis

tus

Piu

ra117

KU

212517

Per

u:

Piu

ra:

Ayaca

ba,

ca.

Ayaca

ba,

2750

m

DQ

502428

DQ

502003

DQ

502723

3461

esp

inosa

i1139

AM

CC

125662

Ecu

ad

or:

Pic

hin

cha:

San

to

Do

min

go

de

los

Co

lora

do

s,

byp

ass

road

DQ

502594

DQ

502158,

DQ

502159

DQ

502862

2985

Eupso

phus

calc

ara

tus

657

MA

CN

37980

Arg

enti

na:

Neu

qu

en:

Hu

ilic

hes

:T

erm

as

de

Ep

ula

fqu

en

AY

843808

AY

843587

DQ

502852

AY

844560

DQ

284120

AY

844036

AY

844786

AY

844214

5796

fem

ora

lis

Cu

rua-U

na

395

OM

NH

34568

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-U

na),

3u9

9S54u5

09W

DQ

502521

DQ

502089

2790

fem

ora

lis

Cu

rua-U

na

396

OM

NH

34572

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-U

na),

3u9

9S54u5

09W

DQ

502522

DQ

502090

DQ

502809

3448

fem

ora

lis

Cu

rua-U

na

1298

MP

EG

12021

Bra

zil:

Para

:101

km

San

d

15

km

ES

an

tare

m(n

ear

Rio

Cu

rua-U

na),

3u9

9S54u5

09W

DQ

502654

DQ

502220

2819


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

fem

ora

lis

Cu

sco

Am

azo

nic

o

78

KU

215179

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

EP

uer

to

Mald

on

ad

o,

200

m

DQ

502415

DQ

501990

DQ

502716

DQ

503180

DQ

502283

4097

fem

ora

lis

Cu

sco

Am

azo

nic

o

128

KU

215177

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

EP

uer

to

Mald

on

ad

o,

200

m

DQ

502439

DQ

502014

DQ

502733

3452

fem

ora

lis

Cu

sco

Am

azo

nic

o

129

KU

215180

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

EP

uer

to

Mald

on

ad

o,

200

m

DQ

502440

DQ

502015

DQ

502734

3455

fem

ora

lis

Cu

yab

eno

399

OM

NH

34102

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,220

m,

0u0

9S

76u1

09W

DQ

502525

DQ

502093

2792

fem

ora

lis

Cu

yab

eno

400

OM

NH

34104

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502526

DQ

502094

2791

fem

ora

lis

Cu

yab

eno

1306

LS

UM

Z12798

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502661

DQ

502228

2792

fem

ora

lis

Pan

gu

an

a

526

MJH

7354

Per

u:

Hu

an

uco

:R

ıo

Llu

llap

ich

is,

Pan

gu

an

a

DQ

502549

DQ

502117

2793

fem

ora

lis

Po

rto

Walt

er

397

OM

NH

36066

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502523

DQ

502091

DQ

502810

3455

fem

ora

lis

Po

rto

Walt

er

398

OM

NH

36070

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502524

DQ

502092

DQ

502811

DQ

503215

DQ

502325

DQ

503156

DQ

503327

DQ

50307

75460

fem

ora

lis

Po

rto

Walt

er

1309

OM

NH

36073

Bra

zil:

Acr

e:P

ort

oW

alt

er,

9u3

493

8.9

0S72u4

693

7.1

0W

DQ

502664

DQ

502231

DQ

502904

3450

fem

ora

lis

Res

erva

Du

cke

520

MJH

3976

Bra

zil:

Am

azo

nas:

Res

erva

Flo

rest

al

Ad

olf

oD

uck

e

DQ

502545

DQ

502113

2790

fem

ora

lis

Rio

Fo

rmo

so

393

MP

EG

13415

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502519

DQ

502088

DQ

502808

3447

fem

ora

lis

Rio

Fo

rmo

so

394

LS

UM

Z

17552

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502520

DQ

283045

DQ

283774

DQ

284074

DQ

503326

DQ

282657

DQ

283465

5047


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

fem

ora

lis

Su

rin

am

e

1325

UT

AA

56478

Su

rin

am

e:S

ipali

win

i:

inth

evic

init

yo

fK

ayse

r

air

stri

p,

3u5

.79N

56u2

8.3

9W

DQ

502678

DQ

502246

DQ

502916

DQ

503264

DQ

502383

DQ

503385

DQ

503131

4956

flavopic

tus

536

MZ

US

P

111790

Bra

zil:

To

can

tin

s:P

ara

na,

12u3

69N

47u5

29W

DQ

502555

DQ

502124

DQ

502831

DQ

503230

DQ

502340

DQ

503160

DQ

503340

DQ

503091

5464

flota

tor

1143

SIU

C7657

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502597

DQ

502162

DQ

502864

DQ

503246

DQ

502361

DQ

503360

DQ

503001

5312

flota

tor

1145

SIU

C7664

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502599

DQ

502164

DQ

502866

DQ

503248

DQ

502363

DQ

503362

4534

frate

rdanie

li

Cald

as

1230

AR

AC

olo

mb

ia:

Cald

as:

Co

rdil

lera

Cen

tral,

Wsl

op

e,2400

m

DQ

502615

DQ

502179

DQ

502882

DQ

503259

DQ

502375

DQ

503375

DQ

503017

5310

frate

rdanie

li

Cau

ca

1226

MH

NU

C360

Co

lom

bia

:C

au

ca:

Po

payan

,

Haci

end

aL

aP

az,

2u2

8.7

049N

76u3

6.2

579W

DQ

502611

DQ

502175

DQ

502878

DQ

503256

DQ

502372

DQ

503371

DQ

503013

5304

frate

rdanie

li

Cau

ca

1227

MH

NU

C361

Co

lom

bia

:C

au

ca:

Po

payan

,

Haci

end

aL

aP

az,

2u2

8.7

049N

76u3

6.2

579W

DQ

502612

DQ

502176

DQ

502879

DQ

503257

DQ

502373

DQ

503372

DQ

503014

5305

frate

rdanie

li

Cau

ca

1228

MH

NU

C364

Co

lom

bia

:C

au

ca:

Po

payan

,

Haci

end

aL

aP

az,

2u2

8.7

049N

76u3

6.2

579W

DQ

502613

DQ

502177

DQ

502880

DQ

503258

DQ

502374

DQ

503373

DQ

503015

5304

fulg

uri

tus

499

MH

NU

C340

Co

lom

bia

:C

ho

co:

Bah

ıaS

ola

no

,S

ierr

aM

ecan

a,

260

m,

6u1

5.5

089N

77u2

1.3

369W

DQ

502538

DQ

502106

DQ

502817

DQ

503222

DQ

502332

DQ

503158

DQ

503333

DQ

503084

DQ

502983

6234

gala

cto

notu

s533

MR

T5088

Bra

zil:

loca

lity

un

kn

ow

n

DQ

502552

DQ

502121

DQ

503227

DQ

502337

DQ

503088

DQ

502988

4628

gala

cto

notu

s647

RG

No

data

(cap

tive

bre

d)

DQ

502583

DQ

502150

2823

gasc

oni

357

MP

EG

13003

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502483

DQ

502052

DQ

502777

3460

gasc

oni

358

MP

EG

12993

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502484

DQ

502053

2802

gasc

oni

1277

OM

NH

36637

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502637

DQ

502202

DQ

502896

3461


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

gasc

oni

1284

OM

NH

36636

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502644

DQ

502209

DQ

502898

3460

gra

nuli

feru

s339

CW

M19044

Co

sta

Ric

a:

Pu

nta

ren

as:

Ab

ou

t

6k

mair

lin

eE

Palm

ar

No

rte,

stre

am

dra

inin

gin

toR

io

Gra

nad

ad

eT

erra

ba

DQ

502466

DQ

502035

DQ

502762

DQ

503196

DQ

502305

DQ

503311

DQ

503059

DQ

502958

5703

hahneli

Cu

sco

Am

azo

nic

o

79

KU

215183

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

EP

uer

to

Mald

on

ad

o,

200

m

DQ

502416

DQ

501991

DQ

502717

3456

hahneli

Cu

sco

Am

azo

nic

o

109

KU

215185

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

EP

uer

to

Mald

on

ad

o,

200

m

DQ

502421

DQ

501996

DQ

502718

DQ

503183

DQ

502286

DQ

503142

DQ

503044

DQ

502943

5806

hahneli

Cu

sco

Am

azo

nic

o

110

KU

215184

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

EP

uer

to

Mald

on

ad

o,

200

m

DQ

502422

DQ

501997

DQ

502719

3462

hahneli

Let

icia

1354

ICN

50410

Co

lom

bia

:A

mazo

nas:

Let

icia

,L

ago

Yah

uarc

aca

DQ

502701

DQ

502270

DQ

502932

DQ

503276

DQ

502400

DQ

503174

4634

hahneli

Man

au

s386

OM

NH

37443

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502512

DQ

502081

DQ

502801

3457

hahneli

Man

au

s391

MP

EG

13849

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502517

DQ

502086

DQ

502806

3456

hahneli

Man

au

s392

OM

NH

37444

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502518

DQ

502087

DQ

502807

DQ

503214

DQ

502324

DQ

503155

DQ

503076

DQ

502977

5806

hahneli

Man

au

s1304

MP

EG

13844

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502659

DQ

502226

DQ

502902

3456

hahneli

Po

rto

Walt

er382

OM

NH

36088

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502508

DQ

502077

DQ

502797

DQ

503209

DQ

502319

DQ

503151

DQ

503071

DQ

502972

5799

hahneli

Po

rto

Walt

er388

MP

EG

12420

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502514

DQ

502083

DQ

502803

DQ

503213

DQ

502323

DQ

503154

DQ

503075

DQ

502976

5800

hahneli

Po

rto

Walt

er389

OM

NH

36092

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502515

DQ

502084

DQ

502804

3460


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

hahneli

Po

rto

Walt

er390

OM

NH

36090

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502516

DQ

502085

DQ

502805

3456

herm

inae

1141

CW

MV

enez

uel

a:

Tru

jill

o:

ab

ou

t

2k

m(a

irli

ne)

WL

aP

ena,

1920

m

DQ

502595

DQ

502160

2315

his

trio

nic

us

336

AM

NH

A122232

Co

lom

bia

:C

ho

co:

Qu

ebra

da

Vic

ord

o,

up

stre

am

fro

m

No

an

am

a

DQ

502463

DQ

502032

DQ

502760

3473

his

trio

nic

us

498

MH

NU

C344

Co

lom

bia

:C

ho

co:

Bah

ıaS

ola

no

,S

ierr

aM

ecan

a,

330

m,

6u1

5.5

819N

77u2

1.1

059W

DQ

502537

DQ

502105

DQ

502816

DQ

503221

DQ

502331

DQ

503332

DQ

503083

DQ

502982

5707

Hyali

nobatr

ach

ium

fleis

chm

anni

TJ9

7JA

C21365

Mex

ico

:O

axaca

:S

an

Jose

Pacı

fico

-Can

del

ari

aL

oxic

ha

Hw

y,

480m

DQ

283453

DQ

284043

DQ

283756

3804

Hyla

cin

ere

a207

MV

Z145385

US

A:

Tex

as:

Tra

vis

Co

.:A

ust

in,

mu

nic

ipal

go

lf

cou

rse

AY

843846

AY

549327

AY

844597

DQ

284057

AY

844063

AY

844816

AY

844241

5557

Hylo

des

phyll

odes

889

CF

BH

-T249

Bra

zil:

Sao

Pau

lo:

Pic

ingu

ab

a:

Ub

atu

ba

DQ

502587

DQ

283096

DQ

283812

DQ

284146

DQ

282923

DQ

282674

4378

Hylo

des

phyll

odes

1152

MC

L00015

Bra

zil

(no

ad

dit

ion

al

data

)D

Q502606

DQ

502171

DQ

502873

DQ

503253

DQ

502368

DQ

503367

DQ

503119

DQ

503009

5722

Hypsi

boas

boans

486

RW

M17746

Ven

ezu

ela:

Am

azo

nas:

Can

o

Agu

aB

lan

ca,

3.5

km

SE

Neb

lin

a

base

cam

po

nR

ıo

Bari

a

AY

843610

AY

844588

DQ

284086

AY

844055

AY

844809

AY

844231

4747

‘‘Ib

agu

e’’

1225

MU

J3564

Co

lom

bia

:T

oli

ma:

Ibagu

e,

El

To

tum

o,

fin

caL

aM

agn

oli

a,

Qu

ebra

da

El

Cu

ral,

1047

m

DQ

502610

DQ

502174

DQ

502877

DQ

503255

DQ

503370

DQ

503012

4967

‘‘Ib

agu

e’’

1345

AR

A2520

Co

lom

bia

:C

un

dim

am

arc

a:

La

Mes

a,

fin

caT

aca

rcu

na,

km

3vıa

Ch

ach

ipay,

1300

m

DQ

502693

DQ

502261

DQ

502924

DQ

503270

DQ

502394

DQ

503171

DQ

503396

5073

‘‘Ib

agu

e’’

1347

MA

R105

Co

lom

bia

:T

oli

ma:

Ibagu

e,

El

To

tum

o,

fin

caL

aM

agn

oli

a,

Qu

ebra

da

El

Cu

ral,

1047

m

DQ

502695

DQ

502264

DQ

502926

DQ

503397

3415

idio

melu

s77

KU

211885

Per

u:

Am

azo

nas:

Bo

ngara

,

Po

mach

och

as,

2150

m

DQ

502414

DQ

501989

DQ

502715

DQ

503179

DQ

502282

DQ

503287

DQ

503043

DQ

502940

5693


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

idio

melu

s120

KU

211908

Per

u:

San

Mart

in:

Rio

ja,

Esl

op

eA

bra

Pard

o

de

Mig

uel

,

2180

m

DQ

502431

DQ

502006

DQ

502726

DQ

502291

DQ

503294

DQ

503048

DQ

502948

5693

idio

melu

s121

KU

211109

Per

u:

San

Mart

in:

Rio

ja,

Esl

op

eA

bra

Pard

od

e

Mig

uel

,

2180

m

DQ

502432

DQ

502007

DQ

502727

DQ

502292

DQ

503295

DQ

503049

4623

idio

melu

s122

KU

211110

Per

u:

San

Mart

in:

Rio

ja,

Esl

op

eA

bra

Pard

od

e

Mig

uel

,

2180

m

DQ

502433

DQ

502008

DQ

502728

3460

idio

melu

s126

KU

211886

Per

u:

Am

azo

nas:

Po

mach

och

as,

2150

m

DQ

502437

DQ

502012

DQ

502294

DQ

503297

3565

imbri

colu

s1229

MH

NU

C257

Co

lom

bia

:C

ho

co:

Bah

ıaS

ola

no

,n

ear

Qu

ebra

da

Nab

uga,

15

m,

06u2

1.6

809N

77u2

0.4

329W

DQ

502614

DQ

502178

DQ

502881

DQ

503374

DQ

503122

DQ

503016

5065

inguin

ali

s1348

MU

J3247

Co

lom

bia

:C

ald

as:

La

Do

rad

a:

San

Ro

qu

e,R

eser

va

Natu

ral

Pri

vad

aR

iom

an

so,

255

m,

5u4

09N

74u4

79W

DQ

502696

DQ

502265

DQ

502927

DQ

502395

DQ

503398

4230

insu

latu

s124

KU

211877

Per

u:

Am

azo

nas:

6k

mW

Ped

roR

uiz

Gall

o,

1260

m

DQ

502435

DQ

502010

DQ

502730

DQ

502293

DQ

503296

DQ

502949

4978

insu

latu

s125

KU

211878

Per

u:

Am

azo

nas:

6k

mW

Ped

roR

uiz

Gall

o,

1260

m

DQ

502436

DQ

502011

DQ

502731

3457

juanii

1357

AR

A2394

Co

lom

bia

:M

eta:

Vil

lavic

enci

o,

Po

zo

Azu

l,560

m

DQ

502702

DQ

502271

DQ

502933

DQ

503277

DQ

502401

DQ

503403

4534

lehm

anni

Myer

s

an

dD

aly

338

CW

M19050

Co

lom

bia

:V

all

ed

elC

au

ca:

Gen

eral

regio

n

of

typ

elo

cali

ty

DQ

502465

DQ

502034

DQ

502761

DQ

503195

DQ

502304

DQ

503310

DQ

503058

DQ

502957

5709

Lepid

obatr

achus

laevis

939

AM

NH

A168407

No

data

(pet

trad

e)D

Q283152

DQ

283851

DQ

284191

DQ

282707

DQ

283543

4194

Lepto

dacty

lus

dis

codacty

lus

TJ8

1R

DS

Ecu

ad

or

(no

oth

erd

ata

)D

Q283433

DQ

284033

DQ

284410

DQ

282887

DQ

283742

4205

Lepto

dacty

lus

fusc

us

TJ5

0A

MN

H

A139088

Gu

yan

a:

So

uth

ern

Ru

pu

nu

ni

savan

na,

Ais

halt

on

(on

Ku

bab

aw

au

Cre

ek),

150

m,

2u2

893

10N

59u1

991

60W

DQ

283404

DQ

284015

DQ

284385

DQ

282862

DQ

283716

4745

Lepto

dacty

lus

hyla

edact

yla

529

MJH

3669

Per

u:

Hu

an

uco

:R

ıo

Llu

llap

ich

is,

Pan

gu

an

a

DQ

283063

DQ

283790

DQ

284093

3064


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

Lepto

dacty

lus

lineatu

s

632

AM

NH

A166426

Gu

yan

a:

Ber

ebic

eR

iver

cam

p

at

ca.

18

mi

(lin

ear)

SW

Kw

ak

wan

i(c

a2

mi

do

wn

river

fro

mK

uru

nd

iR

iver

con

flu

ence

),200

ft,

5u5

960N

58u1

491

40W

AY

843690

AY

844683

DQ

284112

AY

844129

AY

844303

4349

Lepto

dacty

lus

oce

llatu

s

568

MA

CN

38648

Arg

enti

na:

Bu

eno

sA

ires

:

Esc

ob

ar,

Lo

ma

Ver

de,

Est

ab

leci

mie

nto

‘‘L

os

Cip

rese

s’’

AY

843688

AY

844681

DQ

284104

AY

844302

3810

leucom

elas

645

RG

No

data

(cap

tive

bre

d)

DQ

502581

DQ

502149

DQ

502850

DQ

503242

DQ

502354

DQ

503353

DQ

503103

4948

Lim

nom

edusa

macr

oglo

ssa

681

MA

CN

38641

Arg

enti

na:

Mis

ion

es:

Ari

sto

bu

lod

elV

all

e:

Baln

eari

oC

un

ap

iru

AY

843689

AY

844682

DQ

284127

AY

844128

3595

lugubri

sN

icara

gu

a366

OM

NH

33325

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te(c

a.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502492

DQ

502061

DQ

502783

3453

lugubri

sP

an

am

a329

US

NM

-FS

195116

Pan

am

a:

Bo

cas

del

To

ro:

Sen

do

fIs

la

Po

pa,

1k

mE

Su

mw

oo

dC

han

nel

DQ

502456

DQ

283043

DQ

502753

DQ

503193

DQ

502302

DQ

503306

DQ

282655

DQ

283463

5303

mace

ro1133

LR

742

Per

u:

Mad

red

eD

ios:

Parq

ue

Naci

on

al

del

Man

u

DQ

502591

DQ

502155

DQ

502358

DQ

503167

DQ

503108

DQ

502997

4823

mach

ali

lla

73

KU

220631

Ecu

ad

or:

Man

ab

i:

38

km

NW

El

Carm

en,

caro

ad

toP

eder

nale

s

DQ

502410

DQ

501985

DQ

502711

DQ

503175

DQ

502278

DQ

503283

DQ

503039

DQ

502937

5708

mach

ali

lla

132

KU

220632

Ecu

ad

or:

Man

ab

i:

38

km

NW

El

Carm

en,

ca.

road

toP

eder

nale

s

DQ

502443

DQ

502018

DQ

502737

3461

‘‘M

agd

ale

na’’

1358

MU

J3520

Co

lom

bia

:C

ald

as:

La

Do

rad

a:

San

Ro

qu

e,

Res

erva

Natu

ral

Pri

vad

a

Rio

man

so,

280

m,

5u4

09N

74u4

69W

DQ

502703

DQ

502272

DQ

502934

DQ

503278

DQ

502402

DQ

503404

4530

‘‘M

an

au

s1’’

405

MP

EG

13826

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502531

DQ

502099

DQ

503217

DQ

502327

DQ

503328

DQ

503079

DQ

502979

5052


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

‘‘M

an

au

s1’’

620

MP

EG

13829

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502572

DQ

502140

2793

‘‘M

an

au

s1’’

1318

MP

EG

13827

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502672

DQ

502240

2792

Mannophry

ne

sp1322

WE

S1034

Ven

ezu

ela:

Est

ad

o

Mo

nagas:

Qu

ebra

da

qu

efl

uye

haci

ala

Cu

eva

del

Gu

ach

aro

(alt

ura

te

lad

ebo

per

oes

taalr

eded

or

de

los

1400

m),

cerc

ad

eC

ari

pe

DQ

502675

DQ

502243

DQ

502913

DQ

503263

DQ

502380

DQ

503382

DQ

503128

DQ

503024

5708

Mannophry

ne

sp1323

WE

S1035

Ven

ezu

ela:

Est

ad

o

Mo

nagas:

Qu

ebra

da

qu

efl

uye

haci

ala

Cu

eva

del

Gu

ach

aro

(alt

ura

te

lad

ebo

per

oes

taalr

eded

or

de

los

1400

m),

cerc

ad

e

Cari

pe

DQ

502676

DQ

502244

DQ

502914

DQ

502381

DQ

503383

DQ

503129

DQ

503025

5394

Mannophry

ne

sp1324

WE

S1036

Ven

ezu

ela:

Est

ad

o

Mo

nagas:

Qu

ebra

da

qu

efl

uye

haci

ala

Cu

eva

del

Gu

ach

aro

(alt

ura

te

lad

ebo

per

oes

taalr

eded

or

de

los

1400

m),

cerc

ad

e

Cari

pe

DQ

502677

DQ

502245

DQ

502915

DQ

502382

DQ

503384

DQ

503130

DQ

503026

5394

Megael

osi

agoel

dii

611

MZ

US

P

95879

Bra

zil:

Rio

de

Jan

eiro

:

Ter

eso

po

lis:

Rio

Bei

jaF

lor,

910

m,

22u2

49S

42u6

99W

DQ

502563

DQ

283072

DQ

502839

DQ

283797

DQ

284109

DQ

282911

DQ

503346

5069

Mela

nophry

nis

cus

kla

ppen

bachi

217

BB

216

Arg

enti

na:

Ch

aco

:

Pro

xim

idad

esd

eR

esis

ten

cia

DQ

502444

AY

843699

DQ

502739

DQ

283765

DQ

284060

DQ

503299

AY

844899

AY

844306

5689

min

utu

s1149

KR

L790

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502603

DQ

502168

DQ

502870

DQ

503251

DQ

502366

DQ

503365

DQ

503116

DQ

503006

5717

‘‘N

ebli

na

spec

ies’

’379

AM

CC

106112

Ven

ezu

ela:

Am

azo

nas:

Rio

Neg

ro:

Neb

lin

a

Base

Cam

po

nR

io

Maw

ari

nu

ma,

140

m,

0u5

09N

66u1

09W

DQ

502505

DQ

502074

DQ

502795

DQ

503207

DQ

502317

DQ

503149

DQ

503321

DQ

503069

DQ

502970

6222


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

‘‘N

ebli

na

spec

ies’

’380

AM

CC

106118

Ven

ezu

ela:

Am

azo

nas:

Rio

Neg

ro:

Neb

lin

a

Base

Cam

po

nR

io

Maw

ari

nu

ma,

140

m,

0u5

09N

66u1

09W

DQ

502506

DQ

502075

DQ

502796

3436

Nep

hel

obate

ssp

1321

WE

S626

Ven

ezu

ela:

Est

ad

oT

ruji

llo

:

Carr

eter

aH

um

oca

roB

ajo

-

Agu

a

de

Ob

isp

os,

2400

m

DQ

502674

DQ

502242

DQ

502379

DQ

503381

DQ

503127

DQ

503023

4730

nex

ipus

75

KU

211806

Per

u:

San

Mart

in:

Cata

rata

s

Ah

uash

iyacu

,14

km

NE

Tara

po

to,

730

m

DQ

502412

DQ

501987

DQ

502713

DQ

503177

DQ

502280

DQ

503285

DQ

503041

DQ

502939

5720

nex

ipus

123

KU

212486

Per

u:

San

Mart

in:

6k

mE

SE

Sh

ap

aja

,300

m

DQ

502434

DQ

502009

DQ

502729

3456

nex

ipus

130

KU

211807

Per

u:

San

Mart

in:

Cata

rata

s

Ah

uash

iyacu

,14

km

NE

Tara

po

to,

730

m

DQ

502441

DQ

502016

DQ

502735

3457

nex

ipus

131

KU

211808

Per

u:

San

Mart

in:

Cata

rata

s

Ah

uash

iyacu

,14

km

NE

Tara

po

to,

730

m

DQ

502442

DQ

502017

DQ

502736

3458

nid

icola

407

MP

EG

13821

Bra

zil:

Am

azo

nas:

Cast

an

ho

:

ca.

40

km

SM

an

au

s,at

km

12

on

road

toA

uta

zes,

3u3

791

0.4

0S59u8

697

8.4

0W

DQ

502533

DQ

502101

DQ

503219

DQ

502329

DQ

503330

DQ

503081

DQ

502980

5019

nid

icola

622

MP

EG

13820

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502574

DQ

502142

DQ

503241

DQ

502353

DQ

503352

DQ

503102

4245

nid

icola

1285

MP

EG

13819

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502645

DQ

502210

2768

noctu

rnus

1132

AM

NH

A130041

Ven

ezu

ela:T

ruji

llo

:A

bo

ut

2k

m(a

irli

ne)

ES

EA

gu

ad

e

Ob

isp

os,

2250

m,

94u2

9N

70u5

9W

DQ

502590

DQ

502154

DQ

502859

DQ

503243

DQ

502357

DQ

503107

DQ

502996

5270

noctu

rnus

1134

AM

NH

A130042

Ven

ezu

ela:T

ruji

llo

:A

bo

ut

2k

m(a

irli

ne)

ES

EA

gu

ad

e

Ob

isp

os,

2250

m,

94u2

9N

70u5

9W

DQ

502592

DQ

502156

DQ

502860

DQ

502359

DQ

503109

DQ

502998

4954


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

nu

bic

ola

1142

SIU

C7652

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502596

DQ

502161

DQ

502863

DQ

503245

DQ

503359

DQ

503111

DQ

503000

5387

nu

bic

ola

1146

SIU

C7663

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502600

DQ

502165

DQ

502867

DQ

503249

DQ

503113

DQ

503003

4950

‘‘nubic

ola

-sp

C’’

496

MH

NU

C

320

Co

lom

bia

:C

ho

co:

Bah

ıaS

ola

no

,

Qu

ebra

da

Teb

ad

a,

165

m,

06u2

8.9

249N

77u2

0.6

82W

DQ

502535

DQ

502103

DQ

503220

DQ

502330

DQ

503331

3878

‘‘nubic

ola

-sp

C’’

497

MH

NU

C

321

Co

lom

bia

:C

ho

co:

Bah

ıaS

ola

no

,

Qu

ebra

da

Teb

ad

a,

165

m,

06u2

8.9

249N

77u2

0.6

82W

DQ

502536

DQ

502104

DQ

503082

DQ

502981

3972

Odonto

phry

nus

ach

ale

nsi

s

869

ZS

M

733/2

000;

BB

1324

Arg

enti

na:

Co

rdo

ba:

Pam

pa

de

Ach

ala

;A

rgen

tin

a:

Co

rdo

ba:

pro

xim

ity

of

Pam

pil

la,

nea

rP

ara

do

rE

l

Co

nd

or

DQ

283247,

DQ

283248

DQ

283918

DQ

284273

DQ

282773

DQ

283611

4244

Odonto

phry

nus

am

eric

anus

309

JF1946

Arg

enti

na:

Bu

eno

sA

ires

:

Esc

ob

ar:

Lo

ma

Ver

de,

E‘‘

Lo

s

Cip

rese

s’’

AY

843704

AY

844695

AY

844901

AY

844309

3914

Ost

eocep

halu

s

tauri

nus

410

AM

NH

A131254

Ven

ezu

ela:

Am

azo

nas:

Neb

lin

a

base

cam

po

nR

io

Maw

ari

nu

ma

(5R

ıoB

ari

a),

140

M

AY

843709

AY

844700

DQ

284075

AY

844140

AY

844905

AY

844313

4735

palm

atu

s1346

MU

J5003

Co

lom

bia

:C

un

dim

am

arc

a:

La

Mes

a,

fin

caT

aca

rcu

na,

km

3vıa

Ch

ach

ipay,

1300

m

DQ

502694

DQ

502262,

DQ

502263

DQ

502925

DQ

503271

DQ

503172

3885

panam

ensi

s1150

SIU

C7666

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502604

DQ

502169

DQ

502871

DQ

503117

DQ

503007

4625

panam

ensi

s1223

CH

5546

Pan

am

a:

Dari

en:

Can

aD

Q502608

DQ

502172

DQ

502875

DQ

502370

DQ

503368

DQ

503120

DQ

503010

5393

Para

telm

ato

biu

ssp

891

CF

BH

-T240

Bra

zil:

Para

na:

Pir

aq

uara

DQ

283098

DQ

283814

DQ

284148

DQ

282925

DQ

282676

DQ

283499

4195

‘‘P

EG

-M1’’

359

OM

NH

37004

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502485

DQ

502054

DQ

502778

3454

‘‘P

EG

-M1’’

616

LS

UM

Z

17601

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502568

DQ

502136

DQ

502843

DQ

502350

DQ

503349

DQ

503100

4616


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

‘‘P

EG

-M1’’

1288

MP

EG

13397

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502648

DQ

502213

DQ

502900

3456

‘‘P

EG

-M2’’

617

OM

NH

36958

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502569

DQ

502137

DQ

503239

DQ

502351

DQ

503350

DQ

503101

4266

‘‘P

EG

-M2’’

618

MP

EG

13349

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502570

DQ

502138

2790

‘‘P

EG

-M2’’

1237

OM

NH

36959

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502620

DQ

502184

DQ

503262

DQ

503169

DQ

503379

DQ

503126

DQ

503022

5246

‘‘P

EG

-M2’’

1276

OM

NH

36985

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502636

DQ

502201

2790

‘‘P

EG

-M2’’

1282

MP

EG

13375

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502642

DQ

502207

2790

‘‘P

EG

-M2’’

1289

OM

NH

36993

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502649

DQ

502214

2790

‘‘P

EG

-M3’’

360

OM

NH

36988

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502486

DQ

502055

DQ

502779

3456

‘‘P

EG

-M3’’

619

MP

EG

13386

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502571

DQ

502139

DQ

502844

DQ

503240

DQ

502352

DQ

503164

DQ

503351

5073

‘‘P

EG

-M3’’

1263

MP

EG

13388

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502624

DQ

502188

2798

‘‘P

EG

-M3’’

1266

MP

EG

13385

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502627

DQ

502191

2798

‘‘P

EG

-M3’’

1272

MP

EG

13389

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502633

DQ

502197

2798

‘‘P

EG

-M3’’

1274

MP

EG

13384

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502635

DQ

502199

2797

‘‘P

EG

-M3’’

1279

MP

EG

13394

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502639

DQ

502204

2796


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

‘‘P

EG

-M3’’

1295

MP

EG

13387

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502651

DQ

502217

2798

‘‘P

EG

-M3’’

1296

MP

EG

13393

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502652

DQ

502218

2798

‘‘P

EG

-M3’’

1319

MP

EG

13383

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-M

irim

,

10u1

991

7.2

0S64u3

394

7.9

0W

DQ

502673

DQ

502241

DQ

502912

3456

pet

ers

i522

MJH

7041

Per

u:

Hu

an

uco

:R

ıo

Llu

llap

ich

is,

Pan

gu

an

a

DQ

502546

DQ

502114

DQ

502823

3454

pet

ers

i525

MJH

3715

Per

u:

Hu

mb

old

tD

Q502548

DQ

502116

DQ

502825

DQ

503225

DQ

502335

DQ

503159

DQ

503335

DQ

503087

DQ

502986

6236

Physa

laem

us

gra

cil

is

TJ6

5R

dS

788

Uru

gu

ay:

Flo

res

DQ

283417

DQ

284022

DQ

282875

DQ

283728

3889

pic

tus

Gu

yan

a

1331

BP

N1074

Gu

yan

a:

Maza

run

i-P

ota

ro:

Kart

ab

oP

t.(c

on

flu

ence

of

Cyu

ni

&M

aza

run

iR

iver

s)

DQ

502684

DQ

502252

DQ

502922

DQ

502387

DQ

503032

4553

Ple

uro

dem

a

bra

chyops

630

AM

NH

A139118

Gu

yan

a:

So

uth

ern

Ru

pu

nu

ni

Savan

na,

Ais

halt

on

(on

Ku

bab

aw

au

Cre

ek),

150

m,

2u2

893

10N

59u1

991

60W

AY

843733

AY

844721

AY

844926

3470

‘‘P

ort

o-

Walt

er1’’

381

MP

EG

12482

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502507

DQ

502076

DQ

503208

DQ

502318

DQ

503150

DQ

503322

DQ

503070

DQ

502971

5573

‘‘P

ort

o-

Walt

er1’’

625

OM

NH

36153

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502577

DQ

502145

2795

‘‘P

ort

oW

alt

er1’’

626

OM

NH

36148

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502578

DQ

502146

2794

‘‘P

ort

oW

alt

er1’’

1236

OM

NH

36152

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502619

DQ

502183

DQ

503261

DQ

503168

DQ

503125

DQ

503021

4813

‘‘P

ort

oW

alt

er1’’

1299

OM

NH

36147

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502655

DQ

502221

2794

‘‘P

ort

oW

alt

er1’’

1300

MP

EG

12480

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502656

DQ

502222

2795

‘‘P

ort

oW

alt

er1’’

1301

MP

EG

12486

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502657

DQ

502223

2794

‘‘P

ort

oW

alt

er1’’

1302

MP

EG

12488

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502658

DQ

502224

2794

‘‘P

ort

oW

alt

er1’’

1307

OM

NH

36149

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502662

DQ

502229

2794

‘‘P

ort

oW

alt

er1’’

1308

OM

NH

36151

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502663

DQ

502230

2794

‘‘P

ort

oW

alt

er2’’

355

MP

EG

12356

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502481

DQ

502050

DQ

502775

3455


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

‘‘P

ort

oW

alt

er2’’

356

MP

EG

12359

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502482

DQ

502051

DQ

502776

3455

‘‘P

ort

oW

alt

er2’’

1265

MP

EG

12360

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502626

DQ

502190

DQ

502889

3455

‘‘P

ort

oW

alt

er2’’

1269

OM

NH

36027

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502630

DQ

502194

DQ

502892

3455

‘‘P

ort

oW

alt

er2’’

1273

OM

NH

36026

Bra

zil:

Acr

e:P

ort

oW

alt

er,

8u1

593

1.2

0S72u4

693

7.1

0W

DQ

502634

DQ

502198

DQ

502894

3456

pra

deri

oi

1334

CP

I10198

Gu

yan

a:

Maza

run

i-P

oto

ro:

Mt.

Ro

raim

a,

1310

m

DQ

502687

DQ

502255

DQ

503391

DQ

503035

3996

pra

deri

oi

1335

CP

I10208

Gu

yan

a:

Maza

run

i-

Po

toro

:M

t.R

ora

ima,

1310

m

DQ

502688

DQ

502256

DQ

502923

DQ

502390

DQ

503392

DQ

503137

DQ

503036

5377

pra

tti

1144

SIU

C7658

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502598

DQ

502163

DQ

502865

DQ

503247

DQ

502362

DQ

503361

DQ

503112

DQ

503002

5700

‘‘p

ratt

i-li

ke’

’1224

CH

5524

Pan

am

a:

Dari

en:

Can

aD

Q502609

DQ

502173

DQ

502876

DQ

503254

DQ

502371

DQ

503369

DQ

503121

DQ

503011

5702

Pro

cer

ato

-

phry

sa

vel

inoi

306

JF1947

Arg

enti

na:

Mis

ion

es:

Gu

ara

ni:

San

Vic

ente

,C

am

po

An

exo

INT

A‘‘

Cu

art

elR

ıoV

icto

ria’’

DQ

283038,

DQ

283039

DQ

283769

DQ

282903

3262

Pse

udopalu

-

dic

ola

falc

ipes

650

MA

CN

38647

Arg

enti

na:

Co

rrie

nte

s:Y

ap

eyu

AY

843741

AY

844728

DQ

284117

AY

844168

AY

844930

3593

pulc

her

rim

us

118

KU

211947

Per

u:

Caja

marc

a,

imm

edia

te

vic

init

yo

fC

ute

rvo

,2620

m,

06u2

29S

78u4

99W

DQ

502429

DQ

502004

DQ

502724

DQ

503186

DQ

502290

DQ

503143

DQ

503293

DQ

502947

5840

pulc

her

rim

us

119

KU

211948

Per

u:

Caja

marc

a,

imm

edia

te

vic

init

yo

fC

ute

rvo

,2620

m,

06u2

29S

78u4

99W

DQ

502430

DQ

502005

DQ

502725

3456

pulc

hri

pec

tus

337

CW

M

19053

Bra

zil:

Am

ap

a:

Ser

ra

do

Navio

,0u5

99N

50u0

39W

DQ

502464

DQ

502033

DQ

503194

DQ

502303

DQ

503147

DQ

503309

DQ

503057

4808

pum

ilio

367

OM

NH

33297

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te(c

a.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502493

DQ

502062

DQ

502784

3475

pum

ilio

1313

OM

NH

33299

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te

(ca.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502668

DQ

502235

DQ

502907

3472


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

pum

ilio

1315

OM

NH

33300

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te(c

a.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502237

DQ

502909

3087

quin

quevit

tatu

s

Rio

Fo

rmo

so

370

OM

NH

37013

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-

Mir

im,

10u1

991

7.2

0S

64u3

394

7.9

0W

DQ

502496

DQ

502065

DQ

502787

3480

quin

quevit

tatu

s

Rio

Fo

rmo

so

371

OM

NH

37016

Bra

zil:

Ro

nd

on

ia:

Parq

ue

Est

ad

ual

Gu

aja

ra-

Mir

im,

10u1

991

7.2

0S

64u3

394

7.9

0W

DQ

502497

DQ

502066

DQ

502788

3480

quin

que-

vit

tatu

sR

ioIt

uxi

368

OM

NH

36665

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S

65u4

295

9.6

0W

DQ

502494

DQ

502063

DQ

502785

DQ

503203

DQ

502313

DQ

503066

DQ

502966

5285

quin

quevit

tatu

s

Rio

Itu

xi

369

MP

EG

13035

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S

65u4

295

9.6

0W

DQ

502495

DQ

502064

DQ

502786

3483

quin

quevit

tatu

s

Rio

Itu

xi

1312

MP

EG

13034

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502667

DQ

502234

DQ

502906

3482

Res

ervaD

uck

e524

MJH

3988

Bra

zil:

Am

azo

nas:

Res

erva

Flo

rest

al

Ad

olf

o

Du

cke

DQ

502547

DQ

502115

DQ

502824

3452

reti

cula

tus

528

MJH

3754

Per

u:

Lo

reto

:

Alp

ah

uayo

DQ

502551

DQ

502119

DQ

502827

DQ

503226

DQ

502336

DQ

503336

DQ

502987

5306

Rhaebo

gutt

atu

sT

J2A

MN

H

A141058

Gu

yan

a:

Du

bu

lay

Ran

cho

nB

erb

ice

Riv

er,

200

ft,

5u4

095

50N

57u5

193

20W

DQ

283375

DQ

283994

DQ

284361

DQ

283693

3824

Rhaebo

haem

ati

ticus

956

SIU

C7059

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

283167

DQ

283861

DQ

284205

DQ

282720

DQ

283557

4221

Rhin

oder

ma

darw

inii

1115

IZU

A3504

Ch

ile:

XR

egio

n:

Vald

ivia

,

Res

erva

Fo

rest

al

de

On

col

DQ

502589

DQ

283324

DQ

502858

DQ

283963

DQ

284320

DQ

282813

DQ

283654

5246

‘‘R

ioIt

uxi’

’404

MP

EG

12978

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502530

DQ

502098

2792

rora

ima

1337

CP

I10216

Gu

yan

a:

Maza

run

i-

Po

toro

:M

t.R

ora

ima,

1860–2350

m

DQ

502690

DQ

502258

DQ

502391

DQ

503393

3557


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

rora

ima

1338

CP

I10217

Gu

yan

a:

Maza

run

i-

Po

toro

:M

t.R

ora

ima,

1860–2350

m

DQ

502691

DQ

502259

DQ

502392

DQ

503394

DQ

503138

3954

rora

ima

1339

Tad

po

le

(un

tagged

)

Gu

yan

a:

Maza

run

i-

Po

toro

:M

t.R

ora

ima,

1860–2350

m

DQ

502692

DQ

502260

DQ

503269

DQ

502393

DQ

503395

DQ

503139

4275

salt

uensi

s1360

MU

J3726

Co

lom

bia

:B

oyaca

:C

ub

ara

,

Fati

ma,

Qu

ebra

da

Gra

lan

day,

1560

m

DQ

502705

DQ

502274

DQ

502935

DQ

503280

DQ

502404

DQ

503406

4543

‘‘S

ao

Fra

nci

sco

’’516

MJH

3909

Bra

zil:

Am

azo

nas:

Face

nd

aS

ao

Fra

nci

sco

,

2k

mN

km

49

on

Man

au

s-M

an

aca

pu

ruro

ad

DQ

502541

DQ

502109

2795

Scyth

rophry

s

saw

ayae

892

CF

BH

6072

Bra

zil:

Para

na:

Pir

aq

uara

DQ

283099

DQ

283815

DQ

284149

DQ

282926

DQ

283500

4338

silv

ers

tonei

646

RG

No

data

(cap

tive

bre

d)

DQ

502582

DQ

283073

DQ

502851

DQ

283798

DQ

284116

DQ

503165

DQ

282663

DQ

283479

5037

speci

osu

s341

CW

M17826(D

)P

an

am

a:

Ch

iriq

uı:

Co

nti

nen

tal

div

ide

ab

ove

up

per

Qu

ebra

da

de

Are

nam

1250–1400

m

DQ

502468

DQ

502037

DQ

503198

DQ

502307

DQ

503313

DQ

503061

DQ

502960

5055

stepheni

514

MJH

3928

Bra

zil:

Am

azo

nas:

Res

erva

Flo

rest

al

Ad

olf

oD

uck

e

DQ

502539

DQ

502107

DQ

502818

DQ

503223

DQ

502333

DQ

503334

DQ

503085

DQ

502984

5691

stepheni

515

MJH

3950

Bra

zil:

Am

azo

nas:

Res

erva

Flo

rest

al

Ad

olf

oD

uck

e

DQ

502540

DQ

502108

DQ

502819

3455

subpunct

atu

s1359

MU

J5212

Co

lom

bia

:B

ogo

ta,

D.C

.,ca

mp

us

of

Un

iver

sid

ad

Naci

on

al

de

Co

lom

bia

DQ

502704

DQ

502273

DQ

503279

DQ

502403

DQ

503405

3889

sylv

ati

cus

Barb

ou

ran

dN

ob

le

76

KU

219756

Per

u:

Piu

ra:

Ayaca

ba,

12.7

km

EE

lT

am

bo

,

2820

m

DQ

502413

DQ

501988

DQ

502714

DQ

503178

DQ

502281

DQ

503141

DQ

503286

DQ

503042

5463

sylv

ati

cus

Barb

ou

ran

dN

ob

le

113

KU

219757

Per

u:

Piu

ra:

Ayaca

ba,

12.7

km

EE

lT

am

bo

,

2820

m

DQ

502424

DQ

501999

DQ

502721

3455

sylv

ati

cus

Fu

nk

ho

use

r

364

LS

UM

Z

14730

Ecu

ad

or:

San

to

Do

min

go

DQ

502490

DQ

502059

DQ

502781

DQ

502312

DQ

503317

DQ

503065

DQ

502965

5395

‘‘T

afe

lber

g’’

1326

UT

A

A55758

Su

rin

am

e:S

ipali

win

i:

ca.

4.0

km

No

fT

afe

lber

g

air

stri

p

DQ

502679

DQ

502247

DQ

502917

DQ

503265

DQ

503386

DQ

503132

DQ

503027

4478


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

tala

mancae

Nic

ara

gu

a

361

OM

NH

33236

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te(c

a.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502487

DQ

502056

2792

tala

mancae

Nic

ara

gu

a

362

OM

NH

33237

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te(c

a.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502488

DQ

502057

2794

tala

mancae

Nic

ara

gu

a

408

OM

NH

33238

Nic

ara

gu

a:

Rıo

San

Juan

:N

ear

Isla

de

Dia

man

te(c

a.

15

km

SE

El

Cast

illo

on

Rıo

San

Juan

),10u5

69N

84u1

89W

DQ

502534

DQ

502102

DQ

502815

3449

tala

mancae

Pan

am

a325

US

NM

-FS

52055

Pan

am

a:

Bo

cas

del

To

ro:

Cayo

Nan

cy

AY

843799

AY

843577

DQ

502749

AY

844550

DQ

502300

AY

844373

AY

844778

DQ

502954

5701

tala

mancae

Pan

am

a326

US

NM

-FS

59757

Pan

am

a:

Bo

cas

del

To

ro:

Sen

do

fIs

la

Po

pa,

1k

mE

Su

mw

oo

dC

han

nel

DQ

502454

DQ

502025

DQ

502750

3450

tala

mancae

Pan

am

a1147

SIU

C7667

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502601

DQ

502166

DQ

502868

DQ

503250

DQ

502364

DQ

503363

DQ

503114

DQ

503004

5704

Telm

ato

biu

sja

huir

a313

AM

NH

A165110

Bo

livia

:L

aP

az:

Bau

tist

aS

aaved

ra:

Ch

ara

zan

iC

an

ton

,S

trea

m4,

15u7

9490S

68u5

391

70W

DQ

502448

DQ

283040

DQ

502743

DQ

283770

3787

Telm

ato

biu

s

marm

ora

tus

315

AM

NH

A165114

Bo

livia

:L

aP

az:

Bau

tist

aS

aaved

ra:

Ch

ara

zan

i

Can

ton

,st

ream

2700–2750

m,

15u7

9490S

68u5

391

70W

AY

843769

AY

844757

DQ

284068

AY

844952

AY

844355

4193

Telm

ato

biu

ssp

314

AM

NH

A165130

Bo

livia

:L

aP

az:

Bau

tist

aS

aaved

ra:

Ch

ara

zan

i

Can

ton

,st

ream

4,

15u7

9490S

68u5

391

70W

DQ

283041

DQ

283771

DQ

284067

3070

tepuyen

sis

606

RO

M

39637

Gu

yan

a:

Mo

un

tA

yan

gan

na,

no

rth

east

pla

teau

,1490

m,

5u2

49N

59u5

79W

DQ

502559

DQ

502128

DQ

502835

DQ

503234

DQ

502344

DQ

503162

DQ

503343

DQ

503095

DQ

502992

6228

terr

ibil

is1135

AM

NH

A118566

Co

lom

bia

:C

au

ca:

Qu

ebra

da

Gu

an

gu

ı,0.5

km

ab

ove

Rio

Pati

a(u

pp

erS

aij

ad

rain

age)

,

100–200

m

DQ

502593

DQ

502157

DQ

502861

DQ

503244

DQ

502360

DQ

503358

DQ

503110

DQ

502999

5706


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

terr

ibil

is1232

MB

No

data

(cap

tive

bre

d)

DQ

502616

DQ

502180

DQ

502883

DQ

502376

DQ

503376

DQ

503123

DQ

503018

5388

‘‘T

ho

masi

ng’’

1332

UT

A

A56709

Gu

yan

a:

Maza

run

i-

Po

taro

:M

t.T

ho

masi

ng

(,2

km

NIm

baim

ad

ai)

5u4

492

2.8

0N60u1

795

1.3

0W

DQ

502685

DQ

502253

DQ

502388

DQ

503389

DQ

503033

4326

‘‘T

ho

masi

ng’’

1333

UT

A

A56710

Gu

yan

a:

Maza

run

i-

Po

taro

:M

t.T

ho

masi

ng

(,2

km

NIm

baim

ad

ai)

5u4

492

2.8

0N60u1

795

1.3

0W

DQ

502686

DQ

502254

DQ

502389

DQ

503390

DQ

503136

DQ

503034

4722

Thoro

pa

mil

iari

s1186

CF

BH

3239

Bra

zil:

Sao

Pau

lo:

Pic

ingu

ab

a:

Ub

atu

ba

DQ

502607

DQ

283331

DQ

502874

DQ

502369

3796

tinct

ori

us

535

MR

T5087

Bra

zil:

loca

lity

un

kn

ow

n

DQ

502554

DQ

502123

DQ

502830

DQ

503229

DQ

502339

DQ

503339

DQ

503090

DQ

502990

5702

tinct

ori

us

1327

UT

A

A56495

Su

rin

am

e:S

ipali

win

i:

ca.

1.0

km

No

fT

afe

lber

g

air

stri

p

DQ

502680

DQ

502248

DQ

502918

DQ

503266

DQ

502384

DQ

503387

DQ

503133

DQ

503028

5702

tril

inea

tus

74

KU

215172

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

E

Pu

erto

Mald

on

ad

o,

200

m

DQ

502411

DQ

501986

DQ

502712

DQ

503176

DQ

502279

DQ

503284

DQ

503040

DQ

502938

5713

tril

inea

tus

112

KU

215175

Per

u:

Mad

red

eD

ios:

Cu

sco

Am

azo

nic

o,

15

km

E

Pu

erto

Mald

on

ad

o,

200

m

DQ

502423

DQ

501998

DQ

502720

DQ

503184

DQ

502287

DQ

503290

DQ

503045

DQ

502944

5712

tril

inea

tus

527

MJH

7477

Per

u:

Hu

an

uco

:R

ıo

Llu

llap

ich

is,

Pan

gu

an

a

DQ

502550

DQ

502118

DQ

502826

3457

trin

itati

s609

MV

Z

199828

Tri

nid

ad

an

dT

ob

ago

:

Nari

va

Pari

sh:

Ch

aru

ma

Ward

,

Tam

an

aC

ave

DQ

502562

DQ

502131

DQ

502838

DQ

503236

DQ

502347

DQ

503345

DQ

503097

4928

triv

itta

tus

Balb

ina

519

MJH

3907

Bra

zil:

Am

azo

nas:

Base

2is

lan

din

rese

rvo

iro

fU

atu

ma

river

,

8k

mN

Wre

pre

sad

eB

alb

ina

DQ

502544

DQ

502112

DQ

502822

3456

triv

itta

tus

Let

icia

1350

ICN

50437

Co

lom

bia

:A

mazo

nas:

Let

icia

,K

m11

(Let

icia

-

Tara

paca

)

DQ

502698

DQ

502267

DQ

502929

DQ

503273

DQ

502397

DQ

503173

DQ

503400

5066

triv

itta

tus

Man

au

s627

OM

NH

37455

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502579

DQ

502147

DQ

502848

3458


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

triv

itta

tus

Man

au

s628

OM

NH

37453

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502580

DQ

502148

DQ

502849

3458

triv

itta

tus

Pan

gu

an

a518

MJH

7483

Per

u:

Hu

an

uco

:R

ıo

Llu

llap

ich

is,

Pan

gu

an

a

DQ

502543

DQ

502111

DQ

502821

DQ

503224

DQ

502334

DQ

503086

DQ

502985

5267

triv

itta

tus

Po

rt

Walt

er

1297

MP

EG

12447

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502653

DQ

502219

DQ

502901

3456

triv

itta

tus

Po

rt

Walt

er

1305

MP

EG

12468

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502660

DQ

502227

DQ

502903

3454

triv

itta

tus

Po

rto

Walt

er

384

MP

EG

12504

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S,

72u4

693

7.1

0W

DQ

502510

DQ

502079

DQ

502799

DQ

503211

DQ

502321

DQ

503152

DQ

503324

DQ

503073

DQ

502974

6237

triv

itta

tus

Po

rto

Walt

er

387

MP

EG

12450

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502513

DQ

502082

DQ

502802

3456

triv

itta

tus

Su

rin

am

e1329

BP

N910

Su

rin

am

e:P

ara

:

Para

mari

bo

-Ap

ura

road

DQ

502682

DQ

502250

DQ

502920

DQ

503268

DQ

503170

DQ

503030

5075

triv

itta

tus

Tam

bo

pata

319

US

NM

268846

Per

u:

Mad

red

eD

ios:

Pu

erto

Mald

on

ad

o:

Exp

lore

r’s

Inn

,30

km

(air

lin

e)S

SW

of

Tam

bo

pata

Res

erve

DQ

502449

DQ

502021

DQ

502744

DQ

503190

DQ

502298

DQ

503145

DQ

503053

5029

triv

itta

tus

Tam

bo

pata

322

US

NM

269052

Per

u:

Mad

red

eD

ios:

Pu

erto

Mald

on

ad

o:

Exp

lore

r’s

Inn

,30

km

(air

lin

e)S

SW

of

Tam

bo

pata

Res

erve

DQ

502451

DQ

502023

DQ

502746

DQ

503191

DQ

502299

DQ

503146

DQ

503302

DQ

503054

DQ

502953

6234

truncatu

s1151

RG

No

data

(cap

tive

bre

d)

DQ

502605

DQ

502170

DQ

502872

DQ

503252

DQ

502367

DQ

503366

DQ

503118

DQ

503008

5712

truncatu

s1351

ICN

48474

Co

lom

bia

:C

ord

ob

a:

Pu

eblo

nu

evo

,h

aci

end

a

Pra

ga

DQ

502699

DQ

502268

DQ

502930

DQ

503274

DQ

502398

DQ

503401

4550

truncatu

s1352

ICN

48477

Co

lom

bia

:C

ord

ob

a:

Pu

eblo

nu

evo

,fi

nca

Em

baja

da

DQ

502700

DQ

502269

DQ

502931

DQ

503275

DQ

502399

DQ

503402

4550

undula

tus

331

AM

NH

A159141

Ven

ezu

ela:

Am

azo

nas:

Cer

ro

Yu

taje

,1700

m,

5u4

69N

66u8

9W

DQ

502458

DQ

502028

DQ

502755

3449

undula

tus

332

AM

NH

A159139

Ven

ezu

ela:

Am

azo

nas:

Cer

roY

uta

je,

1700

m,

5u4

69N

66u8

9W

DQ

502459

DQ

283044

DQ

502756

DQ

283773

DQ

284073

DQ

503308

DQ

282656

DQ

283464

5702


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

undula

tus

333

AM

NH

A159140

Ven

ezu

ela:

Am

azo

nas:

Cer

roY

uta

je,

1700

m,

5u4

69N

66u8

9W

DQ

502460

DQ

502029

DQ

502757

3451

vanzoli

nii

372

OM

NH

36035

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502498

DQ

502067

DQ

502789

DQ

503204

DQ

502314

DQ

503318

DQ

503067

DQ

502967

5708

vanzoli

nii

373

OM

NH

36037

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502499

DQ

502068

DQ

502790

3472

vanzoli

nii

1314

OM

NH

36036

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502669

DQ

502236

DQ

502908

3471

ventr

imacu

latu

s

Let

icia

1349

JDL

24489

Co

lom

bia

:A

mazo

nas:

Let

icia

,K

m11

(Let

icia

-

Tara

paca

)

DQ

502697

DQ

502266

DQ

502928

DQ

503272

DQ

502396

DQ

503399

4547

ventr

imacu

latu

s

Man

au

s

1310

OM

NH

37440

Bra

zil:

Am

azo

nas:

Cast

an

ho

:ca

.40

km

S

Man

au

s,at

km

12

on

road

to

Au

taze

s,3u3

791

0.4

0S

59u8

697

8.4

0W

DQ

502665

DQ

502232

DQ

502905

3466

ventr

imacu

latu

s

Po

mp

eya

374

OM

NH

34091

Ecu

ad

or:

Su

cum

bıo

s:

Est

aci

on

Cie

ntı

fica

de

Un

iver

sid

ad

Cato

lica

nea

r

Res

erva

Fau

nıs

tica

Cu

yab

eno

,

220

m,

0u0

9S76u1

09W

DQ

502500

DQ

502069

DQ

502791

DQ

503205

DQ

502315

DQ

503319

DQ

502968

5307

ventr

imacu

latu

s

Po

rto

Walt

er

375

MP

EG

12394

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502501

DQ

502070

DQ

502792

3466

ventr

imacu

latu

s

Po

rto

Walt

er

1311

OM

NH

36062

Bra

zil:

Acr

e:P

ort

o

Walt

er,

8u1

593

1.2

0S

72u4

693

7.1

0W

DQ

502666

DQ

502233

2810

ventr

imacu

latu

s

Rio

Itu

xi

376

OM

NH

36666

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502502

DQ

502071

DQ

502793

DQ

503206

DQ

502316

DQ

503148

DQ

503320

DQ

503068

DQ

502969

6232

ventr

imacu

latu

s

Rio

Itu

xi

377

OM

NH

36667

Bra

zil:

Am

azo

nas:

Rio

Itu

xi:

Sch

effe

rM

ad

eire

ira,

8u2

894

5.8

0S65u4

295

9.6

0W

DQ

502503

DQ

502072

DQ

502794

3468

vic

ente

i1148

KR

L789

Pan

am

a:

Co

cle:

El

Co

pe,

Parq

ue

Naci

on

al

Gen

eral

de

Div

isio

n‘‘

Om

ar

To

rrij

os

Her

rera

’’

DQ

502602

DQ

502167

DQ

502869

DQ

502365

DQ

503364

DQ

503115

DQ

503005

5406

vit

tatu

s839

RG

No

data

(cap

tive

bre

d)

DQ

502585

DQ

502152

DQ

502854

DQ

503166

DQ

503355

DQ

503105

4821


Sp

ecie

sS

am

ple

IDS

ou

rce

Lo

cali

tyC

yto

chro

me

bH

1C

OI

Rh

od

op

sin

His

ton

eH

3T

yro

sin

ase

RA

G1

SIA

28S

To

tal

bp

zaparo

321

US

NM

546404

Ecu

ad

or:

Past

aza

:C

oca

,

130

km

So

fN

uev

o

Go

lan

dri

na,

on

trail

W

tow

ard

Rio

Cu

rara

y,

300

m,

1u0

79S

76u5

79W

DQ

502450

DQ

502022

DQ

502745

3452

zaparo

328

US

NM

546405

Ecu

ad

or:

Past

aza

:C

oca

,

130

km

So

fN

uev

o

Go

lan

dri

na,

on

trail

W

tow

ard

Rio

Cu

rara

y,

300

m,

1u0

79S

76u5

79W

DQ

502455

DQ

502026

DQ

502752

DQ

503192

DQ

502301

DQ

503305

DQ

503055

DQ

502955

5713

‘‘zaparo

’’127

KU

221841

Per

u:

Lo

reto

:S

an

Jaci

nto

,

175

m

DQ

502438

DQ

502013

DQ

502732

DQ

503187

DQ

502295

DQ

503144

DQ

503050

DQ

502950

5801


APPENDIX 6SPECIMENS EXAMINED

The following list of specimens examined includesmaterial used explicitly to score the character statesreported in appendix 7. The extensive materialexamined in the course of species identification andtransformation series individuation is not listed.Species are listed following the new taxonomy pro-posed above. See Materials and Methods for collec-tion abbreviations.

OUTGROUP TAXA

Atelopus spurrelli: COLOMBIA: Choco: QuesadaRiver, Atrato River, AMNH 13597–98. Puerto Utrıa,AMNH 50983–84. Serranıa de Baudo, northern baseAlto del Buey, Quebrada Mutata, 200 m, AMNH102065–68.

Atelopus zeteki: PANAMA: El Valle, AMNH44687–91. Probably El Valle region, AMNH 45995–96. Cocle: El Valle de Anton, 2000 ft, AMNH 55533–43. N side of El Valle de Anton, AMNH 83920–22.Panama: Laguna [probably 9 km E El Valle], AMNH55544.

Crossodactylus schmidti: ARGENTINA: Misiones:Aristobulo del Valle, Balneario Cunapiru, JF 832,850.

Cycloramphus boraceiensis: BRAZIL: Sao Paulo:Boraceia, AMNH 54546.

Cycloramphus fuliginosus: BRAZIL: Rio de Ja-neiro: Rio de Janeiro, Tijuca, KU 92789 (C&S).

Dendrophryniscus minutus: ECUADOR: Morona-Santiago: 320 m, 2u409S 77u429W, Cusuime, RıoCusuime [60 km airline SE Macas], AMNH 93804–72.

Eupsophus roseus: CHILE: Cerro Caracol, Cordil-lera de la Costa, Concepcion, 25–137 m, AMNH13979. Corral, AMNH 22102, 22126, 23988, 22104(skeleton). Ancud, Chiloe Island, AMNH 22142,22151. Valdivia, AMNH 23959. Valdivia, BosqueSan Martın, KU 207501 (C&S).

Hylodes phyllodes: BRAZIL: Sao Paulo: Serra doMar, Estacao Biologica de Boraceia, 850 m, AMNH103850–95,103945–46 (larvae).

Megaelosia goeldii: BRAZIL: Rio de Janeiro:Petropolis, AMNH 70249. Near SE edge Teresopolis,950 m, AMNH 103947–53.

Melanophryniscus stelzneri: ARGENTINA: Cor-doba: Achiras, AMNH 51883–92. San Luis: Sierras ofSan Luis, 1000 m, AMNH 76121–23, 77710 (skele-ton).

Rhinoderma darwinii: CHILE: Concepcion,AMNH 7567. Arauco, AMNH 14441–45, 37848–50.Valdivia, AMNH 37813–14. Isla Maullın, 41u359S73uW, AMNH 45331. Received from ConcepcionZoo, AMNH 58082–91.

Telmatobius jahuira: BOLIVIA: La Paz: BautistaSaavedra, Charazani Canton, Stream 4, 15u79490S68u539170W, AMNH 165110.

Thoropa lutzi: BRAZIL: Rio de Janeiro: Rio deJaneiro, Sumare, KU92850 (C&S).

Thoropa miliaris: BRAZIL: Rio de Janeiro:AMNH 36275–76. Rio de Janeiro, AMNH 509.

Voita de Almoco, Serra do Itatiaya, AMNH 17043–46, 17048–49, 17059. Mountains near Rio de Janeiro,AMNH 20251–52, 20254. Teresopolis, AMNH 20861,70141. Teresopolis, Parque Nacional da Serra dosOrgaos, 2700 ft, AMNH 52186.

AROMOBATIDAE

Allobates femoralis: COLOMBIA: Putumayo:Santa Rosa de Sucumbıos (Kofan Indian Village),upper Rıo San Miguel, AMNH 116149. Ca. 10 km(airline) S Mocoa, AMNH 85258 (C&S), 85260(C&S). BRAZIL: Amapa: Serra do Navio, 100–200 m, 0u599N 50u39W, AMNH 140633–49.GUYANA: Iwokrama, Cowfly Camp, AMNH164053. Iwokrama, cutline, AMNH 164054. PERU:Loreto: 3 km NE Pebas on Rıo Amazonas, 3u89S71u499W, AMNH 103581 (C&S). SURINAME:Brokopondo: 65 km airline SSE Paramaribo onAfobaka Rd., AMNH 87680. Saramacca: RaleighCataracts, Coppename River, AMNH 87681–86.Marowijne: Loe Creek, Camp Hofwijks VIII, 56 km(airline) SSW Oelemari, AMNH 90930–32.

Allobates insperatus: ECUADOR: Napo: SantaCecilia, 340 m, KU 109310 (C&S), 149663–90,149691 (C&S), 149692–70, 149671 (C&S), 149672–707. Rıo Yasuni, 150 km upstream from Rıo Napo,KU 175165, 175168–69. Dureno, 320 m, KU 175485.Limoncocha, 243 m, KU 182124.

Allobates juanii: COLOMBIA: Cundinamarca:Medina, vereda Choapal, 6–7 km NNW Medina,580–630 m, ICN 15644–45. Meta: Acacias, Porta-chuelo (crest of divide between Guayabetal andAcacias, ca. 1500 m), ICN 5097 (C&S). Cubarral, ElDorado, ICN 39494–95.

Allobates kingsburyi: PERU: Chanchamayo,AMNH 42282–83, 43604, 43606. ECUADOR: Ma-poto, 1300 m, UMMZ 89063–64. Pastaza: AbitaguaNapo-Pastaza, 1200 m, UMMZ 90373 (3 specimens),90374 (8 specimens), 90375, 217617 (C&S). Napo-Pastaza Mera Oriente, 1000 m, UMMZ 90376. NearMera Oriente, UMMZ 90377 (2 specimens).

Allobates ‘‘Magdalena’’: COLOMBIA: Caldas: LaDorada: San Roque: Reserva Natural Privada Rio-manso, 280 m, 5u409N 74u469W, MUJ 3519–34, 3544.Santander: Cimitarra: Los Indios: El Triangulo, FincaLas Camelias, 240 m, MUJ 2900, 2917. PuertoAraujo, Hacienda El Manantial, 180 m MUJ 2927–28.

Allobates ‘‘Neblina species’’: VENEZUELA: Ama-zonas: Rıo Negro: Neblina Base Camp on RıoMawarinuma, 140 m, 0u509N 66u109W, AMNH118650–64, 118667 (C&S), 118669 (C&S), 118670,118673 (larvae), 118674–83, 118684 (C&S), 118685–86, 118687 (C&S), 118688–90.

Allobates olfersioides: BRAZIL: Rio de Janeiro:Guanabara, Tijuca, AMNH 72445–47, UMMZ127922 (3 specimens), UMMZ 217618 (C&S), KU93161 (C&S).

Allobates talamancae: COLOMBIA: Valle delCauca: Buenaventura, Bajo Calima, MUJ 808 (+larvae). Risaralda: Pueblo Rico, Santa Cecilia yalrededores, ICN 47972. COSTA RICA: Puntarenas:Osa Peninsula, Corcovado National Park, about


6 km E Sirena Biological Station on Trail to LosPatos Ranger Station, UMMZ 193379 (C&S). PAN-AMA: Bocas del Toro: east slope of Cerro Miramar,(ca. 1.5 km S of Miramar), AMNH 113893–901. IslaBastimentos, AMNH 118380–81 (C&S). Isla Basti-mentos, hills W Short Cut, 3.1 km SE Toro Point,AMNH 124225–33. South side Cayo de Agua,AMNH 124234–39 (+ uncataloged carcasses).

Allobates trilineatus: PERU: Madre de Dios:Parque Nacional del Manu, Cocha Cashu BiologicalStation, 11u519S 71u199W, 380 m, AMNH 153038–39.Puerto Maldonado, 280 m, USNM 343061.

Allobates undulatus: VENEZUELA: Amazonas:Cerro Yutaje, 1750 m, 5u469S 66u89W, AMNH159118–40, 159141–42 (C&S).

Allobates zaparo: ECUADOR: AMNH 52881–82(C&S). Morona-Santiago: Ashuara village on RıoMacuma, ca. 10 km above Rıo Morona [ca. 83 kmESE Macas], 300 m, AMNH, 94562–68. Pastaza:Coca, 130 km S of Nueva Golandrina, on trail Wtoward Rıo Curaray, 300 m, 1u079S 76u579W, USNM546404–405.

Anomaloglossus ‘‘Ayanganna’’: GUYANA: Mt.Ayanganna, northeast plateau, 1490–1550 m, 5u249N59u579W, ROM 39639.

Anomaloglossus baeobatrachus: BRAZIL: Amapa:Serra do Navio, AMNH 140650–73, AMNH 140674(larvae). FRENCH GUIANA: Saul, Mantagne Bel-vedere, 3u379N 53u129W, IRSNB-KBIN 12662, 12976,12977.

Anomaloglossus beebei: GUYANA: near KaieteurFalls, AMNH 18683. Kaieteur National Park,5u118N 59u298W, UMMZ 218880, 221371–74. Mt.Ayanganna, northeast plateau, 1490–1550 m, 5u249N59u579W, ROM 39629–32.

Anomaloglossus ‘‘Brownsberg’’: SURINAME:Brokopondo: Brownsberg Nature Park, UTA 56469.

Anomaloglossus degranvillei: SURINAME: Maro-wijn: Central Lely Mountains, headwaters of DjoekaCreek (Suralco Camp V), 620 m, AMNH 90871–76,90878–92.

Anomaloglossus praderioi: GUYANA: Mazaruni-Potoro: Mt. Roraima, 1310 m, CPI 10198–205,101207–208.

Anomaloglossus roraima: GUYANA: Mazaruni-Potoro: Mt. Roraima, 1860–2350 m, CPI 10212–17 (+untagged larvae).

Anomaloglossus stepheni: BRAZIL: Amazonas:Ducke Reserve, 150 m, KU 129988–130008. DuckeReserve, KU 130009–14. Sudam Floral Reserve,74 km E Santarem, KU 129987, 130144–45.

Anomaloglossus ‘‘Tafelberg’’: SURINAME:Sipaliwini: ca. 4.0 km N of Tafelberg airstrip, UTA55758.

Anomaloglossus tepuyensis: GUYANA: Mt. Ayan-ganna, northeast plateau, 1490 m, 5u249N 59u579W,ROM 39637. VENEZUELA: Bolıvar: Auyantepui,Camp 1, 1700 m, 5u519N 62u329W, AMNH 164817–22. Auyantepui Camp 3, 1850 m, 5u538N 62u388W,AMNH 164823. Auyantepui Camp 4, 1600 m, 5u589N62u339W, AMNH 168424–33.

Anomaloglossus ‘‘Thomasing’’: GUYANA: Ma-zaruni-Potaro: Mt. Thomasing (,2 km N Imbaima-dai) 5u44922.80N 60u17951.30W, UTA 56708–10.

Aromobates molinarii: VENEZUELA: Merida:Cascada de Bailadores, 1800 m, UMMZ 176207(C&S), 176208–11, 176220, 176222.

Aromobates nocturnus: VENEZUELA: Trujillo:about 2 km (airline) ESE Agua de Obispos, 2250 m,9u429N 70u059W, AMNH 129940 (C&S), 130006–13,130014 (C&S), 130016–21, 130026–33, 130036–38,130041 (C&S), 130047 (skeleton).

Aromobates saltuensis: COLOMBIA: Norte deSantander: Bucarsica, ICN 42512–16, 33587.

Aromobates sp.: VENEZUELA: Trujillo: about2 km (airline) ESE Agua de Obispos, 2250 m, 9u429N70u059W, AMNH 129958–74.

Mannophryne collaris: VENEZUELA: near Mer-ida, Rıo Albirregas, AMNH 10512–16. Merida,UMMZ 217615 (C&S). Trujillo: Between Niquitoand La Columna, USNM 291062–64.

Mannoprhyne herminae: VENEZUELA: Aragua:Rancho Grande, near Maracay, AMNH 70761–87,116941–977, 116978–79 (larvae). Parque NacionalHenri Pittier, Rancho Grande, below toma del agua,1100 m, USNM 259176 (larvae). Rıo Ocumare,110 m, UMMZ 210143–44 (C&S). Carabobo: SanEsteban, UMMZ 139774–75 (larvae).

Mannophryne trinitatis: TRINIDAD AND TO-BAGO: Trinidad: Arima Valley, Spring Hill Estate,on trail to Guacharo Cave, USNM 166302–04. St.George, Maracas, on trail to Maracas waterfall,USNM 166305–37. St. George, Mount El Tucuche,USNM 166338–42. 7 mi N Arima, 800 ft, UMMZ167465 (C&S), 167469, 167471, 167474. NorthernRange, ca. 8 km (airline) N Arima, 560 m, AMNH118384 (C&S), 118389 (C&S).

Rheobates palmatus: COLOMBIA: Magdalena,UMMZ 149232, 149233 (skeletons). Cundinamarca:Anolaima, AMNH 13472. Honda [data ambiguous,possibly from Paramo del Verjon, E of Bogota],AMNH 20359–63. Paramo del Verjon, E of Bogota,AMNH 20364–65, 20367–69. Formeque, 20409–18.Choachi, AMNH 20425–33, 20436–37. Aguadita,AMNH 22610–15. Ca. 25 mi N Villavicencio, UTA8028–32, 39725. Meta: Serranıa de la Macarena,UTA 4929. Los Micos (16 km S San Juan), ca.4 hours S of this location at Canon Joel, UTA39711. Cano Cristalina, 8.0 hours S Los Micos, UTA39712. Ca. 30 km WSW Vista Hermosa, CanoSardinita, UTA 39713, 39715–23, 39737, 39738–40(larvae). Ca. 32 km WSW Vista Hermosa, CanoSardinita, UTA 39714. Sierra de La Macarena, ca.30 km WSW Vista Hermosa, UTA 39724. Santander:Virolın, 2500 m, MUJ 5003. Tolima: 20.3 mi WNWCajamarca [on way from Cali to Bogota], UTA39728–29.

DENDROBATIDAE

Ameerega bassleri: PERU: San Martın: Cainarachi,AMNH 42313. Pachiza, Rıo Huallaga, AMNH42327, 42333, 43402 (C&S). Chasuta, AMNH42867, 42944.

Ameerega bilinguis: COLOMBIA: Putumayo: ca.10 km (airline) S Mocoa, 700–800 m, AMNH 85200–208, 85210–14, 85215 (C&S), 85216, 85219 (C&S),85221 (C&S), 85224, 85226–27.


Ameerega braccatus: BRAZIL: Mato Grosso:Estacao Ecologica Serra das Araras, USNM 505750(larvae).

Ameerega flavopicta: BRAZIL: Goias: Minacu:Upper Tocantins River, Serra da Mesa, 13u509130S48u199280W, AMNH 158104–05. Minas Gerais: Jabo-ticatubas, Serra do Cipo, km 114, AMNH 88642.Santana do Riacho, USNM 505751 (larvae).

Ameerega hahneli: PERU: Loreto: Yagua IndianVillage, headwaters of Rıo Loretoyacu [100+ km NWLeticia], AMNH 96185–96. 5 rd km NE Previsto,near Boqueron del Padre Abad, upper Rıo Aguaytıa,500 m, AMNH 118421 (C&S). BRAZIL: Amazonas:Igarape Belem, near Rio Solimoes, ca. (70 km ELeticia), AMNH 96751–54. Presidente Figueiredo,USNM 505752 (larvae). COLOMBIA: Amazonas:Leticia, boca a Los Lagos (Yahuarcaca), ICN 53105(larvae).

Ameerega macero: PERU: Madre de Dios: Westside Rıo Manu across from Cocha Cashu BiologicalStation, Parque Nacional del Manu, ca. 380 m,11u559S 71u19W, AMNH 129473–74, 133205, 133207(larvae), 134159–63.

Ameerega petersi: PERU: Junın: Valle de Perene,1200m, AMNH 17257. Otica, Rıo Tambo, AMNH111000. San Martın: Achinamisa (Rıo Huallaga),AMNH 42179, 42505–07, 42546. Chasuta (RıoHuallaga), AMNH 42790, 42945. Huanuco: MonteAlegre, Rıo Pachitea, AMNH 43016 (C&S).

Ameerega pulchripectus: BRAZIL: Amapa: Serrado Navio, AMNH 137280–137293.

Ameerega picta: BOLIVIA: Buenavista, AMNH22637–38. Santa Cruz: Buenavista, 500 m, AMNH33959, 34075, 39562–63. 4.5 km N 1.5 km E CerroAmboro, Rıo Pitasama, 620 m, 17u459S 63u409W,AMNH 153546. 3 km N 13.5 km W San Rafael deAmboro, Rıo Saguayo, 400 m, AMNH 153547–49.Beni: Rıo Benicito, Chacabo, AMNH 70151. RıoMamor’ ca. 4 km below Santa Cruz, 11u109S, AMNH79196–211. 45 km N of Yacuma, 400 m, 14u429S67u49W, AMNH 153550–51. 6 km W of Casarabe,400 m, AMNH 153552–73. Prov. Ballivian, Lago DelGringo, 10 km N of Puerto Salinas, 1 km from BeniRiver, 14u108S 67u408W, UMMZ 184099 (C&S).

Ameerega rubriventris: PERU: Ucayali: El Bo-queron del Padre Abad, edge of road connectingTingo Marıa and Pucallpa, ca. 1000 m, AMNH168494–97.

Ameerega silverstonei: PERU: Huanuco: about30 km NE Tingo Marıa, Cordillera Azul, [5 5 kmby road SW high point (1640 m) on Tingo Marıa-Pucallpa Rd], 1330 m, AMNH 91845–46, 91847(C&S), 91848 (C&S), 91849 (C&S), 91851, 94803–05.

Ameerega trivittata: PERU: Cachiyacu (East ofBalsapuerto), AMNH 42576. Madre de Dios: 30 km(airline) SSW of Puerto Maldonado, TambopataReserve, Explorer’s Inn, 280 m, 12u509S 69u179W,USNM 268845–47. San Martın: Achinamisa, AMNH42183–84, 42539–43, 42545, 43204. SURINAME:Jodensavanne, Kamp 8, AMNH 77450. Brokopondo:Brownsberg Nature Park, near Mazaroni Top, ca.450 m, AMNH 118431 (C&S) Marowijne: CentralLely Mountains, headwaters of Djoeka Creek (Sur-alco Camp V), 620 m, AMNH 90916–22, 90977–78.Airstrip, Lely Mountains, 680 m, AMNH 90923–29,

90979. Saramacca: Raleigh Cataracts, CoppenameRiver, 50 m, AMNH 118428 (C&S).

Colostethus fraterdanieli: COLOMBIA: Antioquia:ca. 10 km airline [17 km by rd] NW of Bolıvar onQuibdo road, 2050 m, AMNH 104361–68. Caldas:5.5–6 km by rd southeastward Villa Marıa, 2320 m,AMNH 104375–92, 104399 (male + larvae), 104400(male + larvae), 104401 (male + larvae). 13 km by rdsoutheastward Villa Maria, 2400 m, AMNH 104397.

Colostethus fugax: ECUADOR: Pastaza: Cabe-ceras del Rıo Bobonaza, 2250 ft., USNM 282831.

Colostethus imbricolus: COLOMBIA: Choco: Ser-ranıa de Baudo, N slope Alto del Buey, 970 m,AMNH 102082. Serranıa de Baudo, northern baseAlto del Buey, Quebrada Mutata, 200 m, AMNH102083–85.

Colostethus inguinalis: COLOMBIA: Caldas: LaDorada, San Roque, Reserva Natural Privada Rio-manso, 255 m, 5u 409780N 74u 479780W, MUJ 3247.Choco: River Truando, USNM 4349. Upper RıoNapipı, 45 min by canoe below mouth of RıoMerendo (tributary of Rıo Napipı), ca. 60–90 m,LACM 42325–42332; trail between Rıo Merendo andCerro Los Hermanos, LACM 42333; upper RıoNapipı, forested hills near river on left bank, 45 minby canoe below mouth of Rıo Merendo, 60–200 m,LACM 42334–42340, 43955; upper Rıo Napipı,forested hills near river on right bank, LACM42341–42344; upper Rıo Opogado, ca. 1 hr 45 minby canoe above mouth of Rıo Merendo, LACM42345–42490. Camino de Yupe, LACM 72009–10.

Colostethus panamensis: COLOMBIA: Choco:Parque Nacional Natural Los Katios, IAvH [IND-AN] 3337–3370, 6206, 6208–6209. PANAMA: Cocle:Continental Divide N El Cope, 600–800 m, 80u368W,AMNH 98317 (female + larvae), 98318 (female +larvae). El Valle, Rıo Anton, 6 650 m, AMNH 87293(females + larvae). Veraguas: 6–12 km N Santa Fe Nof Altopiedra and Agricultural School in montanearea called Buenos Aires, UMMZ 167459 (C&S).

Colostethus pratti: COLOMBIA: Risaralda: PuebloRico, Santa Cecilia y alrededores, ICN 47973–74,47976, 47978. PANAMA: Cocle: 12 km N El Cope,continental divide at sawmill, UMMZ 167503 (C&S).Darien: Rıo Jaque, 1.5 km above Rıo Imamado,AMNH 118364 (C&S), 118365–67, 118369–370,118371 (C&S). Veraguas: 5–6 mi NW (via road)Santa Fe (Pacific drainage), 1700–2000 ft, AMNH108339. Cerro Delgadito, 2–4 mi W Santa Fe, ca4000 ft, AMNH 162528. 6–12 km N Santa Fe N ofAltopiedra and Agricultural School in mt area calledBuenas Aires, UMMZ 167460. 6–12 km N Santa FeN of Altopiedra and Agricultural School in mt areacalled Buenas Aires, 3000–3500 ft, UMMZ 167506,167512, 167514–15.

Colostethus ‘‘pratti-like’’: PANAMA: Darien:Cana, CH4052–47, 4650, 4702–03, 5524–25, 5598(larvae), 5601–02.

Dendrobates auratus: COSTA RICA: Puntarenas:8 km ENE Palmar Norte, AMNH 118524 (C&S).PANAMA: Bocas del Toro: 8.9 km (airline) WSWChiriquı Grande, 100 m, AMNH 113904. 4.9 km(airline) WSW Chiriquı Grande, 60 m, AMNH113905–06. East slopes Cerro Miramar (ca. 1.5 kmS of Miramar), 340 m, AMNH 113907–12. Cocle:


Continental divide N El Cope’ 700 m, AMNH 97874.Continental divide N El Cope’ 600–800 m, AMNH98325–40. East shoulder Cerro Caracol (above ElValle de Anton), 870 m, AMNH 114588. Panama:Isla Tobago, AMNH 118528 (C&S) (+20 uncatalogedskinned carcasses).

Dendrobates azureus: SURINAME: Nickerie: Si-paliwini Savannah, AMNH 88626–88631 (+ uncata-loged AMNH specimens).

Dendrobates leucomelas: BRAZIL: Roraima: Serrado Tepequ’m, 500–600 m, 3u458N 61u458W, AMNH137308 (larvae), 137309–11. VENEZUELA: Amazo-nas: Rıo Pescado, Mt. Duida Region, 325 ft, AMNH23179. Rıo Pescado, foothills camp, 750 ft, AMNH23202, 23206. Cano Pescado, AMNH 23235. Bolıvar:Mt. Auyan-tepui, 460 m, AMNH 46045–47, 46051.San Felix, edge of Rıo Orinoco, AMNH 75789. GuriDam, ca. 300 m, AMNH 81455. El Manteco, AMNH90203–04. Canaima Falls (near Auyan Tepui),AMNH 90998.

Dendrobates tinctorius: GUYANA: Shudikar-Wan, AMNH 49301–28. BRAZIL: Amapa: Serra doNavio, 100–200 m, AMNH 140675. Serra do Navio,ca. 100 m, AMNH 140676–87. Serra do Navio, KU93147 (C&S).

Dendrobates truncatus: COLOMBIA: Antioquia:Santa Rosa de Oso, 2640 m [doubtful locality],AMNH 38820–21. Medellın, AMNH 39087. Cundi-namarca: Fusagasuga, AMNH 40309–12. Caldas: LaDorada, San Roque, Reserva Natural Privada Rio-manso, 255 m, 5u 409780N 74u 479780W, MUJ 3088(larvae). Cesar: El Roncon, ca. 10–12 km E Becerril(foothills of Sierra de Perija), 250–280 m, AMNH84381–83. Huila: Neiva, Tamarindo, Alto La Tri-buna, Reserva Hocol, 570 m, 3u49N 75u22.38W, MUJ3607. Tolima: Shore of Rıo Gualı, 1–2 km aboveMariquita, 530 m, AMNH 85229–36, 118401 (C&S),118403 (C&S). Magdalena: Sierra Nevada de SantaMarta, El Pueblito, Parque Nacional Tayrona, 230–290 m, AMNH 88578–79.

Epipedobates anthonyi: ECUADOR: Azuay: ca.10 km (airline) W Santa Isabel, Rıo Jubones drainage,1490 m, AMNH 104903–17. El Oro: 10 km SEMachala, 20 m, AMNH 118499 (C&S), 118502(C&S).

Epipedobates boulengeri: COLOMBIA: Cauca: IslaGorgona, USNM 145248 (larvae), 145249–252,145253 (C&S), 145254–300 (topotypes), AMNH50970–72 (topotypes).

Epipedobates espinosai: ECUADOR: Pichincha:Rıo Baba, 5–10 km SSW Santo Domingo de losColorados, 500 m, AMNH 89668–87, 118411 (C&S),118417 (C&S). Santo Domingo de los Colorados,Bypass Road, AMNH 162663. Centro Cientıfico RıoPalenque, 170 m, AMNH 104869–98, 162662. Ca.2 km S Santo Domingo de los Colorados AMNH162664.

Epipedobates machalilla: ECUADOR: Chimbo,BMNH 1898.3.1.4–7. Guayas: 3 km N Naranjal,30 m, AMNH 89525–36, 89537 (male with 19tadpoles). Manabı: 38 km NW El Carmen, ca. roadto Pedernales KU 220631–33.

Epipedobates tricolor: ECUADOR: Bolivar-Coto-paxi border, ca. 7 km (airline) SSW of El Corazon,

AMNH 104946–54. Cotopaxi: 11 km E (by road)Moraspungo, USNM 286082–83.

Hyloxalus awa: ECUADOR: Pichincha: ca. 15 kmSE Santo Domingo de los Colorados, at Tinalandia,800 m, AMNH 111541–44. 8 km SE Santo Domingode los Colorados, Hacienda Delta, UMMZ 217614(C&S).

Hyloxalus azureiventris: PERU: San Martın: Achi-namisa, AMNH 42186.

Hyloxalus bocagei: ECUADOR: Napo: Rıo Oya-cachi at Quito-Lago Agrio Road, about 20 km NNEBaeza, 1550 m, AMNH 89570–71. Morona-Santiago:Cusuime, Rıo Cusuime [60 km airline SE Macas],320 m, AMNH 94043–73. Pastaza: Hills N of Mera,78u88W, 1u268S UMMZ 182465 (C&S).

Hyloxalus delatorreae: Carchi: 14 km (airline) SEMaldonado, 2500 m, KU 182197. 18.5 km E Mal-donado, ca. Maldonado-Tulcan rd, 2420 m, KU220618.

Hyloxalus elachyhistus: ECUADOR: Rıo Linoma,AMNH 16262–303, 16305–13, 16315, 16317, 16321.Loja: 2150 m, KU 120543 (C&S).

Hyloxalus ‘‘Ibague’’: COLOMBIA: Cundinamarca:La Mesa, Sector de la gran vıa, Finca Tacarcuna, 3kmvıa Cahipay, 1200 m. ARA 2343–45, 2347–57, 2443–44. Ibague, El Totumo, Finca La Magnolia, quebradaEl Cural, 1047 m, MUJ 3545–48, 3551–3553, (+untagged larvae, to be deposited at MUJ).

Hyloxalus infraguttatus: COLOMBIA: Narino:4 km NW Junın, 1170 m, AMNH 85031 (male +larvae). ECUADOR: Azuay: About 16 km (airline) WSanta Isabel, Rıo Jubones drainage, 1000 m, AMNH89563–65, 91824. Ca. 10 km (airline) W Santa Isabel,Rıo Jubones drainage, 1490 m, AMNH 104846–48.Rıo Minas, 8 km (airline) W of Santa Isabel, RıoJubones drainage, 1440 m, AMNH 104849. Ca. 9 km(airline) E Pasaje, 100 m, AMNH 104041–45. El Oro:about 20 km (airline) E Pasaje, 240 m, AMNH 91823,104838–40.

Hyloxalus nexipus: PERU: Amazonas: S of Ain-tami entse on the Rıo Cenepa, 4u289S 78u109W,USNM 317147–53. Vicinity of San Antonio, on theRıo Cenepa, 4u309S 78u109W, USNM 317154–60,USNM 317609 (larvae). Vicinity of Sua, on the RıoCenepa, 4u329S 78u119W, USNM 317161–63. Huam-pami, Quebrada Sasa, on the Rıo Cenepa, 210 m,4u289S 78u109W, USNM 317164–72. Shimpunts,vicinity of, on the lower Rıo Alto Cenepa (tributaryof the Rıo Cenepa), 4u259S 78u129W, USNM 317173–74. Paagat, on the lower Rıo Alto Cenepa (tributaryof the Rıo Cenepa), 4u259S 78u129W, USNM 317175.Vicinity of Kagka, at the confluence of Rıo Kagkaand Rıo Comaina (tributary of the Rıo Cenepa),4u279S 78u139W, USNM 317176. Vicinity of Tseasim,on the upper Rıo Huampami (tributary of the RıoCenepa), 4u239S 78u109W, USNM 3171777–79. Vi-cinity of Shaim, on the Rıo Alto Comaina (tributaryof the Rıo Cenepa), 4u159S 78u229W, USNM 317180–85.

Hyloxalus pulchellus: COLOMBIA: Putumayo:4 km (airline) SE San Francisco, 2320 m, AMNH85018–21. ECUADOR: Napo: Rıo Azuela, Quito-Lago Agrio Road, eastern base Volcan Reventador,1700 m, AMNH 89538 (C&S).


Hyloxalus sauli: COLOMBIA: Putumayo: ca.10 km (airline) S Mocoa, 700–800 m, AMNH85029. ECUADOR: Napo: Rıo Nachiyacu S ofVenecia, 77u429W, 1u79S UMMZ 182477(C&S),182478–79. Rıo Cotopino, UMMZ 195745.

Hyloxalus subpunctatus: COLOMBIA: Boyaca:Chita, El Arbolito, carretera Chita-La Salina, ICN3990. Duitama, Paramo de la Rusia, ICN 4468.Duitama, Paramo de La Rusia, km 19–21 carreteraDuitama-Charala, 3650 m, ICN 26963. Pajarito,hacienda Comijoque, 2015 m, ICN 7196, 7235, 7237.Susacon, Paramo Guantiva, km 3–4 via Onzaga,3170 m, ICN 10361. Ramiriqui, km 11–12 carreteraRamiriquı-Zetaquira, 3060m, ICN 11020. Ventaque-mada, vereda El Boqueron, Paramo Albarracın,3120m, ICN 11044. Paramo de Pisba, ICN 31699.Cundinamarca: Bogota, Ciudad Universitaria (Uni-versidad Nacional), ICN 27024, 45777 (larvae).Bogota, near Monserrate, 3000 m, UMMZ 221158–59 (C&S). Usme, Laguna Chisaca, 3700m, ICN11868, 45779 (larvae). Bogota, Universidad de LosAndes, 2500 m, ICN 33686, 45778 (larvae). Guasca,ICN 35672, 45780 (larvae).

Hyloxalus sylvaticus: PERU: summit Cordillerabetween Chanchaque and Huancabamba, 3100 m,KU 138071–79. 29.3 km SW Huancabamba, 3010 m,KU 181667–73. 31 km SW Huancabamba, 3080 m,KU 181674–79. SW slope Abra de Porculla, 1850 m,KU 164093 (C&S).

Hyloxalus toachi: ECUADOR: Pichincha: RıoBaba, 5–10 km SSW Santo Domingo de los Color-ados, 500 m, AMNH 89550–61, 89562 (male +larvae). Ca. 15 km SE Santo Domingo de los Color-ados, at Tinalandia, 800 m, AMNH 111539–40.

Hyloxalus vertebralis: ECUADOR: Cinincay,8300 ft, AMNH 17458, 17604–08, 140977–141011.Azuay: Cuenca, 8365 ft, USNM 282308–16, 282352–58. Canar: about 8 km SE Canar, Pan-Amer Hwy,3000 m, AMNH 89569 (male + larvae). Cuenca,2540 m, KU 120633–34 (C&S). Chimborazo: ChunchiEl Tambo Road 20 km N of Gun Junction ca.9600 ft, UMMZ 217621 (C&S).

Adelphobates castaneoticus: BRAZIL: Para: nearCachoeira Jurua, Rio Xingu, 3u229S 51u519W,AMNH 133451–55 (paratypes).

Adelphobates galactonotus: BRAZIL: Para: Ca-choeira do Limao, Rio Tapajos, AMNH 128232–33.

Adelphobates quinquevittatus: BRAZIL: Rondonia:Alto Paraiso, AMNH 124068–71, 124072 (larvae).

Oophaga arborea: PANAMA: Bocas del Toro andChiriquı: ca. 7 km airline W of Chiriquı Grande, 20 m(bocas del Toro), and continental divide (Chiriquı),AMNH 116771–80 (paratypes). Chiriquı: continentaldivide above upper Quebrada de Arena, 1120 m,82u129310W, AMNH 116725–60 (paratopotypes),116761–68 (C&S; paratopotypes). Continental divideabove upper Quebrada de Arena, 1200–1300 m,82u139W, AMNH 116769–70 (paratypes).

Oophaga granulifera: COSTA RICA: 4.5 km WRincon de Osa, 40 m, KU 110223 (C&S). Puntarenas:about 6 km airline E Palmar Norte, stream draininginto Rıo Grande de Terruba, AMNH 134069,134071–81, 118408–409. 8 km ENE Palmer Norte,90 m, AMNH 86631.

Oophaga histrionica: COLOMBIA: Choco: Westbank of lower Rıo San Juan at Pangala, ca. 40 km (byboat) N Palestina, AMNH 88242–82. Risaralda: ca.7 km (airline) SE Santa Cecilia, upper Rıo San Juan,500–600 m, AMNH 118458 (C&S), 118461–62 (C&S).

Oophaga lehmanni: COLOMBIA: Valle del Cauca:ca. 13 km W Dagua, Rıo Anchicaya drainage, 850–1200 m, AMNH 88154–95 (topoparatypes), 88231–34(C&S; topoparatypes), 118435–37, 118438 (C&S),118439, 118441, 118442 (C&S), 118443–45.

Oophaga pumilio: PANAMA: Bocas del Toro: IslaBastimentos, near Bastimentos, AMNH 102256–63.East end Isla Escudo de Veraguas, AMNH 118510(C&S). Isla Bastimentos AMNH 118514 (C&S).7.1 km (airline) WSW Chiriquı Grande, 70–100 m,AMNH 161952 (larvae). (+ several hundred un-cataloged skinned carcasses from Bocas del Toro atAMNH.)

Oophaga speciosa: PANAMA: Chiriquı: Continen-tal divide above upper Quebrada de Arena, 1140–1410 m, 82u139400W, AMNH 124279–321. Continen-tal divide above upper Quebrada de Arena, 1300 m,82u139400W, AMNH 124322–31. Continental divideabove upper Quebrada de Arena, 1250–1400 m,82u139400W, AMNH 124332–34, 118447 (C&S),118454 (C&S). Continental divide above upperQuebrada de Arena, 1250–1410 m, 82u139400W,AMNH 124335–48, 124349 (larvae). Continentaldivide above upper Quebrada de Arena, 1250–1410 m, 82u139300W, AMNH 161120, 161122–23.

Oophaga sylvatica: COLOMBIA: Narino: Guaya-cana, 500 m, AMNH 85048–158, 86635–40. ECUA-DOR: Pichincha: Rıo Baba, 5–10 km SSW SantoDomingo de los Colorados, 500 m, AMNH 89589–601. About 10 mi S of Santo Domingo de losColorados, in banana plantation, AMNH 88225–26(C&S).

Oophaga vicentei: PANAMA: Cocle: ContinentalDivide N El Cope, 600–800 m, AMNH 98344–50,98351–53 (C&S), 98354 (larvae). East shoulder CerroCaracol (above El Valle de Anton), 870 m, AMNH114583–84, 114586, 114587 (C&S).

Minyobates steyermarki: VENEZUELA: Amazo-nas: SW sector Cerro Yapacana, 900 m, AMNH100760–99, 118579 (C&S), 118572 (C&S), 118575–76(C&S), 118581 (C&S).

Phyllobates aurotaenia: COLOMBIA: Choco: 2 kmabove Playa de Oro, Rıo San Juan, 210 m, AMNH85238–45. Vicinity of Playa de Oro, upper Rıo SanJuan, ca. 200 m, AMNH 85246 (male + larvae), 85247(male + larvae), 85248 (male + larvae), 85249 (male +larvae), 87167 (male + larvae) AMNH 87168 (male +larvae), 161108 (C&S), 161109–111.

Phyllobates bicolor: COLOMBIA: Risaralda:about 7 km (airline) SE Santa Cecilia, mountain sideabove north bank Rıo San Juan, 500–600 m, AMNH98209–236, 98256 (C&S).

Phyllobates lugubris: PANAMA: Bocas del Toro:ca. 5 km W Almirante, 30–40 m, AMNH 86642 (male+ larvae). Rıo Changuinola, near Quebrada ElGuabo, 16 km airline W Almirante, 200 m, AMNH107237 (larva from AMNH 107231). East slope CerroMiramar (ca. 1.5 km S of Miramar), 340 m, AMNH113936. Penınsula Valiente, near Punta Valiente, 1–5 m, AMNH 113937. Northwest side Isla San


Cristobal, 5–10 m, AMNH 113938–39. North sideIsla Pastores, 5–15 m, AMNH 113940. Mainland ca.1 km (airline) NNW Isla Split Hill, 20–30 m, AMNH113941–43. Mainland ca. 1 km (airline) NNW IslaSplit Hill, 20–30 m, AMNH 124350–53, 55–56. Ca.5 km W Almirante, 40 m, AMNH 118554 (C&S),118557 (C&S).

Phyllobates terribilis: COLOMBIA: Cauca: Queb-rada Guanguı, about 0.5 km above junction with RıoPatia, in upper Rıo Saija drainage, 100–200 m,AMNH 86319–24 (C&S; paratopotypes), 118563(skeleton; paratopotype), 125831–35 (C&S). Captivebred, lab-reared F1 from paratopotypic parents(AMNH A118562-67, A125826-830 series), AMNH162738–43.

Phyllobates vittatus: Western Zoological Supply(received by John W. Daly, National Institutes ofHealth), AMNH 114041. COSTA RICA: Puntarenas:8 km ENE Palmar Norte, 90 m, AMNH 82257,86643–45, 118542–45 (C&S), 118546, 118547–50(C&S), 118551.

Ranitomeya biolat: PERU: Cuzco: San Martin-3,ca. 5 km N of the Camisea River, 474 m, USNM537557–58. Cashiriari-3, S of the Camisea River,690 m, USNM 537559–565. Madre de Dios: ParqueNacional del Manu, Cocha Cashu Biological Station,ca. 380 m, AMNH 143908. Pakitza, Reserve Zone,Manu National Park, ca. 57 km (airline) NW ofmouth of Rıo Manu, on Rıo Manu, 350 m, USNM342882 (larvae).

Ranitomeya claudiae: PANAMA: Bocas del Toro:Isla Colon, near La Gruta, AMNH 102307–68,103514–23 (C&S). Isla Colon, about 2.5 km airlineN La Gruta, 40 m, AMNH 124255–65.

Ranitomeya fulgurita: PANAMA: Panama: km12.8 on El Llano-Cartı Rd, 290 m, AMNH 89435–37. Km 14.6 on El Llano-Cartı Rd, 370 m, AMNH89438–47, 89448–53 (C&S).

Ranitomeya imitator: PERU: San Martın: Km 33,Carretera Tarapoto-Yurimagaus, Valle del Rıo Cai-narache, 500–650 m, AMNH 127991–99, 128003–06,162723–27, KU 209412–13 (C&S). Km 48, road toYurimaguas from Tarapoto AMNH 162728–30. Km

10 road to Chamilla from Tarapoto, AMNH 162731–32.

Ranitomeya minuta: PANAMA: Panama: CerroCampana, 2800 ft, AMNH 59660–62. Cerro Cam-pana, 900–950 m, AMNH 84896–900. S slope CerroCampana, 900–950 m, AMNH 87310. Km 14.6 on ElLlano-Cartı Rd, 370 m, AMNH 89426–32. CerroCampana, 800 m, AMNH 118132.

Ranitomeya reticulata: PERU: Loreto: 3 km airlineSSW Mishana, on Rıo Nanay, 150 m, AMNH103619–30, 103638–73; AMNH 103676 (C&S),103680–81 (C&S).

Ranitomeya vanzolinii: BRAZIL: Acre: PortoWalter, Rio Jurua, 8u169S 72u469W, AMNH 108332(paratopotype). PERU: Loreto: mouth of Rıo Sepa-hua (Rıo Urubamba), AMNH 43597–98.

Ranitomeya ventrimaculata: COLOMBIA: Amazo-nas: Leticia, Km 7 (Leticia-Tarapaca), ICN 47609.Leticia, Km 9 (Leticia-Tarapaca), ICN 47330–32,47334–35. Leticia, km 11 (Leticia-Tarapaca), ICN53027 (larvae), 53034 (larvae). PERU: Loreto: 3 kmNE Pebas on Rıo Amazonas, 100 m, 103603–04(C&S).

Silverstoneia flotator: PANAMA: Colce: El Vallede Anton, 2000 ft, AMNH 55509, 116781–83. ElValle, Rıo Anton, 6650 m, AMNH 87300–01,124210–15. Continental Divide N El Cope’ 98323.Km 13.4 on El Llano-Cartı Rd, 300 m, AMNH104229 (larvae). Barro Colorado Island, KU 77678(C&S).

Silverstoneia nubicola: PANAMA: Chiriquı: upperRıo Chiriquı, Fortuna Dam Site, 1000 m, AMNH94846–48, 94849 (larvae). Upper Rıo Chiriquı, nearmouth Rıo Hornito, 1020 m, AMNH 114574–77. RıoFrijoles, UMMZ 145585 (C&S).Continental divideabove upper Quebrada de Arena, 1660–1270 m,AMNH 124249.

Silverstoneia ‘‘nubicola-spC’’: COLOMBIA:Choco: Serranıa de Baudo, northern base Alto delBuey, Quebrada Mutata, 200 m, AMNH 102092–95.Bahıa Solano, Quebrada Tebada, 165 m, 06u28.9249N77u20.6829W, MHNUC 320–21.


APPENDIX 7PHENOTYPIC DATA

Phenotypic data are reported here in FASTAformat. The dataset may be downloaded from http://research.amnh.org/herpetology/downloads.html asa Hennig 86 matrix. Inapplicable characters areshown with a dash (-); uncoded characters (missingdata) are shown with a question mark (?).

Atelopus spurrelli3000000----0000000000-0120100-0-----15457645510000200000--0003411100000000--10-0-------????????????????????????????101100??0----????1001???0-002???0------------------------1?Atelopus zeteki3000000----0000000000-0120100-0-----05456544500000[03 ]00000- -0000000100000000- -10-0 - - - - - - -131001232110????????????????101100??0----???????????0-002???0------------------------12Crossodactylus schmidti111101100000010100000-0101100-1011101221422111000040000101000[04]000?0?000001002100-------101110232011????????0?0-00-100??210112020?1?????????1000????0------------------------?4Cycloramphus boraceiensis3011010----0011110000-00-0100-0-----142456445100004 0 0 0 0 0 - - 0 0 1 5 3 0 0 1 ? 0 ? 0 0 0 0 1 1 0 1 0 - 0 - - - - - - -100000232??0??????1011?-0??1?001232011020011210000001010??????????????????????????????4Dendrophryniscus minutus3011000----0000000000-00-0-00-0-----04114312300000400000--0004233102100000--00-0-------100001232010??????10000-00-2100112?????????4????????????????0------------------------0?Eupsophus calcaratus1011010----0000000000-0120100-0-----00000000000000400000--000530[01]102000001101100--- - - - -11000000?100??????100?0003-100112120110000001101100010101000??????????????????????????6Hylodes phyllodes111101112221111111100-0110100-12222112123221110000400001100003411102000[01]01002100-------101010232011?0?00?10000-00-2100021211100201201001001101010000------------------------?4Megaelosia goeldii111103112221111111100-00-0100-12222213234432212000400000--0006144?020000010000-0-------101000232011????????????00-11010????1??????8?10010011010????0------------------------?[56]Melanophryniscus klappenbachi3011010----0000000000-00-0100-0-----0323433440000010000[01][01]10005333102100000--10-0-------10110123201???????10000-00-1100112010--- -?07001012110-002000100111111100010000010000002Rhinoderma darwinii1011000----0000000010-00-0100-0-----02134545401000 4 0 0 0 2 0 - - 0 0 0 4 2 3 3 1 0 2 1 0 0 0 0 0 0 0 1 1 0 0 - - - - - - -1011000[12]2110??????1?110-?3-1?0?1010011020?02001000010-002100??????????????????????????4Telmatobius jahuira2011030----0000000000-0110100-1-----13236544410000100000--0004233?0??00001101100-------??????????????

????1???0-???00001????????????????????1?10????????--????--?-?-??????????4Thoropa miliaris1010030----0000000000-0100100-0-----00000000000000400000--0005300?001000010010-0-------??????232??????????0110-?????000232110000?121100000110101000??????????????????????????4alagoanus110022100000000000001100-0101200111000000110001001400011020002000?00111011111110-------?11?0023201???????????10??????100001110011140011111111010000??????????????????????????3anthonyi010113111110000000001200-0101200111[01]00012110001000[23]10001100015333102111011111110-------?????????????0010?20?312???2111000001200101500001111110100001011101001011100000010100??arboreus010100102320000000000-00-0101300112100000000000000[12]00000--0005333002011011111110-------200100011211?100100012???12121?10001110000116001012110-01[01][01][01]11001100011010000000000001??auratus010101103330000000000-00-010100022210000000000100000[03]000--0003411002111011111110-------20000023211?00???0001110[01]1-2111000011[01][01]0[01]006000012110-0100[01][01]1011111111111110011110001?0aurotaenia0101131[01]1110000000000-00-0101202222100000000001000203000--0004252000111011111110-------2010002320102?????????10???211100000120010050001-0001110100001111001010000000000000010??awa11010[12]111120000000010-00-01011012222000001100010004000011[01]1004[134]33100111011111110-------10??0023201???000???131200-11?100011120010140001111110011000???????????????????????????azureiventris210102111220000000000-00-?10120122210000000001000003021000004[34][13][13]000?11011111110-------10100023201?20000100111000-21???????????????????????1?0?????0------------------------??azureus010101102220000000010-00-0001300111000000000000000100000--0004252002111011111110-------?0000023211?????????11101??211???????????????????????????0001010101001010100001000001??baeobatrachus110022111110010100001300-01012011111121232211110004000011100060001001110111111111101011101000222010???????????????21???????????????????????1?0????????????????????????????????bassleri210113100100000000001100-010130000000000000000000020000110000?2?2?0?1??????????0-------?????????????????????110????11?000011[12]00111600001111110100001011101101011100010000000??beebei110100112220000000000-00-0101001111102112310000000403000--0000000120111011111111000101??????????????0?00020211012021???????????????????????1?0????????????????????????????????bicolor


010103111110000000000-00-0101200[12][12][12]000000000001000[12]01000--0004000000111011111110-------????????????2?????0??010???21110000112001005000111111101001011110010110000000000010100?bilinguis210113100110000000001100-010110011110000000000000011000100000[35][34][13][13]100111011111110--------??????????????????2???10???21110000112001014000011111101[01]000101000100000000000000000???biolat010100103330000000000-00-0100-0022220000000000000000010-100003411?0?111011111110-------101100232110????????????1??21???????????????????????0-0????????????????????????????????bocagei010101122220010100000-00-0101000011014245[56]43[45]0100040000[01]01001[25][01]1[04]100111011111110-------???????????????????????????1[01]?100011120010160100111111010000???????????????????????????boulengeri110113111110000000001200-01012001111000001100000004000111000125[13][13]00?111011111110-------101000232010??????????10???21??0???????????-00001001110100000------------------------??‘‘Tafelberg’’11002310000011010001?????01?11001220111333200022001400000--000?4?0?0?1??????????1110101?????????????????????????????????????????????????????1?0????????????????????????????????‘‘Brownsberg’’11001212222001010001?????01?1101222012123221112200140000110000?0?0???1??????????1??0????????????????????????????????????????????????????????1?0????????????????????????????????‘‘Thomasing’’11011111111001010000?????01?1102222214244433311100402001010?1?0?01?????????????1110101????????????????????????????11???????????????????????1?0????????????????????????????????braccatus????????????????????????????????????????????????????????????????????????????????-------101000232010??????????10???2???????????????????????????????????????????????????????????caeruleodactylus1100231????0000000000-00-010120?????000123100000004000110200000001?01??????????0-------001200232010?0000000111000?21???????????????????????1?0???????????????????????????????2castaneoticus010101103330000000000-00-0001300111000000000000010100000--0005333000111011111110-------201000232[01]1?????????????11-211??????????????????????0-0?????1010001100001000000000001??chlorocraspedus21????11?22????????????????????????????????????0002?3010--???????0??????????????-------????????????????????????????1???????????????????????????????0-----------------------???claudiae010000101[12][12]0000000000-00-010130001100000000000000050300100000[35][34][13][13]022111011111110-------???????????????????????????21?1000111[12][01]01016001012110-011[01]011001101001000100010000001??collaris

110102111110000000000-00-01010011111041344323110004030010[01]0102124100111011111110-------101000232011??1???????10???111100001120010020100????1101?100???????????????????????????degranvillei110102100110010100[01]10-00-0101100222113123221112001403001100006144100011011111111110101?12-----000?0???????????10-3?11??????????????3????????1?0-???????????????????????????????delatorreae110???1????0000000000-00-0101?0??????00000000000004000111000?2?301?0???????????0-------10100023201???????????100??21???????????????????????1?0-???????????????????????????????elachyhistus1101011[01]1110010100010-00-010110000001[34]234[23][23][12]11200040000110101[26][013]3[034]100111011111110-------10100023201??????????110???11??00000120010?50000111111011000???????????????????????????espinosai110113111110000000001200-01012001110000[01]01100000014000[01]10000025[13][13]112111011111110--------10100023201???????????10???21?1000011200101500001111110100001011100001010100000000101??fantasticus???????????????????????????????????????????????????????????????????????????????0-------?????????????????????110121211??????????????????????0-0?????1011101000000000000000000??femoralis210113122220000000000-00-0101202222200000110001000[15]00120--00042111[01]0111011111110-------10100023201???????????12[01]??2111000001200101300011111110100000------------------------?3flavopictus210103100000000000001100-0101100000000023310000000500001100013411?00?11011111110-------101000232010????????1?10???11???????????????????????0?0?????1011011000000000000000000?3flotator110123111110000000001200-01012002221000001100010014000211000-[02]000112111011111110-------110010002010??????????10???21?100000120011?400000111110100000------------------------??fraterdanieli110103111110000000011200-0101100222000000110001000400001100013[14]00102111011111110-------10000023201???????????1000-21???????????????????????1?01????0------------------------03fugax1101131????0000000001200-010120???????????????0001400011100000000???????????????????????????????????????-??????????????????????????????????????????????????????????-????????????fulguritus010100102220000000000-00-000130001100000000000000050300100000[35][34][13][13]022111011111110-------????????????0????100?1101??21?10000112001015000111110-0100011001100001000000000000001??galactonotus010101103330000000010-00-0101200111000000000001000100000--0004252000111011111110-------???????????????????????????211??????????????????????0-0??0001??1??1?????????????????1??


granuliferus210100102[23]20000000000-00-0101300112000000000001000100000--0004252002111011111110-------20010011211?10000?01[12]1111212111000011??001?6001012110-0111011011101111010110110000001?1hahneli210103111110000000001100-0101200222100000000000000110011100004233?02111011111110-------101100122010?????????112???21?100001110010060001111111011000101000100000000000000010??3herminae110102111220010100010-00-01010022222131[13][23]33[02][02][01]0000403011000112024100111011111110--------101000232010??????????10???1??100000110010030100111111011000??????????????????????????3histrionicus010101112[23][23]0000000000-00-0101301222100000000000000[12]00000--0005333002111011111110-------20010011211?1?01000122111212111000010----106000012110-011[01][01]11011101111010110111000001?1humilis1101?11111200?0?000?0-00-0101001222?1???????????00400001020014?0?1??????????1??0-------???????????????????????????21???????????????????????1?0????????????????????????????????‘‘Ibague’’110112111110000000000-00-011120011100000000000100040001110000202010011101?111??0-------101000232010??????????1????21??????????????????????????????????????????????????????????idiomelus1101?31000000000000?0-00-0101?0??????00000000??0004000011010061441?0????????1??0-------10100023201???????????1000-11???????????????????????1?0????????????????????????????????imbricolus110102111110010100011200-0101001111113134321012000413001010005333100?11011111110-------???????????????????????????11???????????????????????1?0??-???0------------------------0?imitator010000103330000000000-00-0100-00122200000000000000000[01][02]0--000341100[02]111011111110-------???????????????????????????21110000110000016000112110-011000101001111001100010000000???infraguttatus110102100110000000000-00-01011000110031[23][34][23]2000100040000110101[25][13]33100111011111110--------101000232010??????????10????????????????????????????1?0?????0------------------------0?inguinalis1101131112[12]00101?00[01]1200-010100222221[23]2[23][34]3[23]2[12]1200040002100000405010111101111-1110-------???????????????????????????11??????????????????????????????????????????????????????????insperatus11002311[01]110000000001100-0101200122100012110001001403021020002020100111011111110-------10100023201???????????100??21?1000111000111300001111110100000------------------------??insulatus110??31?????????????0-00-010110?????0221411000000?4000011010???001?0???????????0-------10100023201?????????????00-11??????????????????????????????????????????????????????????

juanii110[01]13111110000000000-00-0101201222200012110001001403021020016000100111011111110-------???????????????????????????21?100001120110150100001010010000???????????????????????????kingsburyi110103111110010100000-00-010120122220000000000100040302100000[25]130100111011111110-------101100232011??????????1????11?100010110010150001100011011000???????????????????????????lehmanni Myers and Daly01010111[12]220000000000-00-0101300111000000000000000200000--000[45][23]33001111011111110-------????????????1????001?211121211100001[01]0003104000[01]12110-0111[01]11011110101111110110000001??leucomelas010101102220000000000-00-000[01]300111000000000000000000000--000[45][23]5[23]000111011111110-------2010002321102???????111000-211??????0???????????????0-0??0001011100100001100000000001??lugubris010103111110000000000-00-0101201222100000000000000203000--00042[13][13]000111011111110-------1010002320102?????????101??211100011110010160000011111010000110111100001000001000000??3macero210113111110000000000-00-010100122210000000000000010000100000[35][34][13][13]102111011111110--------1011002320102?????????1000-21???????????????0???????1?011000101010110100000000001000???machalilla110113100110000000001200-0101200011000012110000001400021100000000102111011111110-------101000232010??????????100?????1000101?01111?00??????1?0??000???????????????????????????‘‘Magdalena’’010[01]23111110000000001200-0101200[12]220000123100000014030210200040[25]010011101?111??0-------???????????????????????????21??????????????????????????????????????????????????????????minutus010000101[12][12]0000000000-00-010130011100000000000000010300[01]0[01]0005333?22111011111110--------10110023211?0?????????1?1??21?1000101100101[56]001012110-011[01]011001101101010100000000001??molinarii110101111110010100000-00-01010012221131343[23]2211000403000--001[25]333100111011111110-------10100023201?????????????00-1??100000120000030100000011010000???????????????????????????‘‘Neblina species’’210023111110000000001100-0101212222200000110002001403010--0000000100011011111110-------10100023201???????????10???21?100001120010130000111111010000???????????????????????????Nephelobates sp 1321110103100000000000011100-0101100111000000000001000403000--0002033100111011111110-------???????????????????????????21???????????????????????????????0------------------------0?nexipus010001111220010100000-00-01000012221142455345100004030110[01]000212310011101?111??0-------????????????2?????????10???1[01]??????????????????????????????????????????????????????????


nidicola1100231????0000000000-00-010120???????????????00014000110200020201?0??? ? ?? ?11??0- - - - - - -12 - - - -001100????????1?0-23-21???????????????????????1?0???????????????????????????????2nocturnus110101111110010100000-00-0101001111114338677512000403001010006144100111[01]11111110-------101000232010????????????00-00?1000011110100201011000110110000------------------------0?nubicola110113111110010100001200-0101201[12][12]2100000110001001400021100004000112111011111110-------110010[12]02010??????????10???21?1000001210101500001111110100000------------------------??‘‘nubicola-spC’’110113111110010100001200-010120122210000011000100140001110000341111?111011111110-------???????????????????????????21???????????????????????1?0??-??????????????????????????????palmatus110101111110010100010-00-01010012[12]21142466[46][46]5120004000[12]10[01]0015[013][04]0100111011111110-------101000232011?1011?00111000-111?0???????????2?100100011011000??????????????????????????3panamensis110103111110010100011200-010100122211[23]12432221200[01]400021000000400102111011111110-------?01000232010?1010?2?1?1100-1111000001200101500011111110110100------------------------13parvulus??????????????????????????????????????????????????????????-?????????????????????0-------????????????2?????????10???21???????????????????????????????101100100100000000000000???petersi210113111110000000001100-010100012210000000000100010000110000[35][34][13][13]100111011111110-------10100023201???????????12???????0?????????????????1111?011000101100100100000000000000???pictus210113100000000000001100-0101100111000000000000000-1001110000[45]333102111011111110-------10100023201???????????10???211100000120010140000111111010000101110010101000000111000??3praderioi110102111110000000001300-0101201222113134320001001403000--000[26]0[04]0?10111011111111010101????????????????????????????21??????????????????????????????????????????????????????????pratti110113111110000000001200-010120122210000000000100140302100000600[04]110111011111110-------?0100023201???????????11???21?100001120010140001111111011000???????????????????????????‘‘pratti-like’’110113111110000000001200-0101201222100000000001001403021000006000?10111011111110-------?01000232010??????????10????1???????????????????????1?0????????????????????????????????pulchellus110111100000000000000-00-010120012220000000000100040000110[01]0-5[03]1[01]100111011111110-------?0100023201???????????100??2??100000120010130001111111011000???????????????????????????

pulcherrimus010??31?????????????0-00-010110?????0000000000000?4000011000?33331?0???????????0-------???????????????????????????11??????????????????????????????????????????????????????????pulchripectus210113111[12]10000000000-00-0101100111000000000001000100011100004411012111011111110-------101000232010?????????110???211??????????????????????0-0?????101100100101000000001000???pumilio010100102330000000000-00-0001300111000000000000000[0123]00000--000[034][024][015][012]002111011111110-------2001001121101000010113111212111000011000[02]106001012110-0111[01]1101111111111111011011000101quinquevittatus010101102330000000000-00-0001300122000000000000010000120--0003211002111011111110-------200000232111????????????1??21???????????????????????0-0?????1011111111010000000000001??reticulatus010000101220000000000-00-0101300111000000000000000000000--0003411022111011111110-------????????????1????????3101[12]0211100011120010[01]6001012110-01000[01]1011101010001100000000001??roraima110101111110000000001100-0101200111000000000000000400000--001[46]1[24]4110?11011111111000101?101000232010????????????1??21???????????????????????1?0????????????????????????????????‘‘Ayanganna’’110102111110000000001300-?101201122100000000000001403000--0002?2??10????????1??1000101????????????????????????????21???????????????????????1?0????????????????????????????????rubriventris210113111110000000000-00-010100011100000000000100011000110000533310?111011111110-------1010002211???????????????????1?????????????????????1?0????????????????????????????????saltuensis1101031111100[01]0[01]00000-00-010100122211212332111000040300101000512010011101?111??0-------???????????????????????????21??????????????????????????-???????????????????????????????sauli110101111110010100000-00-0101101122113234332212000400001[01]00002020100111011111110-------?0100023201?????????????00?11?1000101211101501000-000110100000------------------------??silverstonei210113111110000000000-00-0101000000000000000000000[12]10000--001425[02]100111011111110-------10100023201?3????????11000-2111000001??010?300000000110100001001100101010000001000100??speciosus010101102220000000000-00-01013001[12][12]100000000000000300000--0000000002111011111110-------2001001121101????100?3111212111000111000[01]106000112110-0111011011101111011110010000001??stepheni110[01]23100000010100001300-0101100000012123221112001400001100006140?0?111011111111110101?12--


---00011??0?10?2?100-23-21?????????????????????????????????????????????????????????3steyermarki010002100220000000000-00-0101300011000000000000000000000--001614402[02]111011111110-------??????????????????2???10???21?1000001[01]00[01]0[01]6001011110-01[01]0001001100000001100000000001??subpunctatus110111100000000000000-00-0101200[01][01][01]0000000000010004000011000120[13][13]100111011111110--------101000232010????????111000-21?1000011200110310011011110100000------------------------?3sylvaticus Barbour and Noble11010?1?????????????0-00-0101?0???????????????000040000010[01]0000001?0????????1??0-------10100023201???????????1000-11?1000001??010?50101000011010000???????????????????????????sylvaticus Funkhouser010101112220000000000-00-0101300111100000000000000[123]00000--000[35][34][13][13]002111011111110-------????????????1??????????????21?1000010----104000012110-0111011011101111010100101000001?1talamancae110113111110010100000-00-0101201222100012110001000403020--0004[01][02]0100111011111110-------101000232010??????????11???21?1000[01]11200[12][01]1300011111110100000------------------------03tepuyensis110101122220010100000-00-010100222211424544331200040000101000[56][13][34][034]1001110111111111101011101-?0232011??-??-??-???00?11--?????????????????----1?0?-??????????????????????????????terribilis010103111110000000000-00-0101200[12][12][12]000000000001000301000--0002000000111011111110-------1010002320102???????0010???2111000011210?016000[01][01]11111010000110000001000000000000101-??tinctorius010101113330000000010-00-000130011100000000000000010[03]000--0003411000111011111110-------20000023211??????1??1110???211?000011????0?6000012110-0110[01]01011101111010100001010001??toachi110113111110000000000-00-01012012221000[01][02][01][01]0001000400011100006000102111011111110--------101000232010??????????10???21???????????????????????1?0????????????????????????????????tricolor110113111110000000001200-010120[01][01]11[01]00012110001000210001100015333?02111011111110-------???????????????????????????21???????????????????????1?0?????100110000100010000000010???trilineatus110[01]23110010000000001200-0101200122100000110001001403021020002000100111011111110-------

101000232010??????????12???21?????????????????????????0????????????????????????????????trinitatis110101111110000000000-00-010100112210[23]112110000000403001000102020100111011111110-------?????????????01?????111000-1111000011200101301001111110100000------------------------03trivittatus2101131111100000000[01]0-00-0101[23]011111000[01][02][01][01]000[01]000[23]000[01]1100004233112111011111110-------10100023201?3????????11000-21110000112001014000110110-010000101000100100000001001000-?3truncatus010101102330000000010-00-0101300111000000000001000203020--0005433002111011111110-------200000232111????????1110[01]??21110000110001016000012110-0100[01]0101000110101010000000000??0undulatus110113111110000000001110-0101201222100000110001001400000--0012000100111011111110-------101000232011????????????00-21?100000120011130001111111010000???????????????????????????vanzolinii010000103330000000000-00-0101300022200000000000000000000--0003233002111011111110-------????????????????????2?101[12]021???????????????????????0-0?????101000101010000001000000??1ventrimaculatus010100103330000000000-00-01013000[12]220000000000000000012[01]00000[35][34]11022111011111110--------200000232110????????21101[012]021110001110002116000012110-0110001011101110011000000000001??vertebralis110101100000000000000-00-010110000000000000000100040000110[01]0020[12]0110111011111110-------101000232010????????1?100??21?100001110011050000111111011000??????????????????????????3vicentei010100102330000000000-00-0001300000000000000000000[01]00000--0004000022111011111110-------?00100??211010000100111112121?1000110----106000112110-0101011001100001010100010000001??vittatus010103111110000000010-00-0101201111100000000001000103010--0004252000111011111110-------101000232010??????0??1101??21?10000012[01]010[01]30000111111010000110001111100000100010000-??zaparo210113112210000000000-00-0101201222100012110001000--0[01]20--0004233100111011111110-------???????????????????????????21??????????????????????????????????????????????????????????


APPENDIX 8

DIAGNOSTIC DNA SEQUENCE TRANSFORMATIONS FOR NAMED CLADES

Below we report all unambiguous DNA sequence transformations to diagnose the named clades addressed in the

monophyletic taxonomy presented above; unambiguous morphological transformations are reported in the text.

Insertions and deletions are depicted by gap states (-). Locus abbreviations: 28S (large nuclear ribosomal

subunit), COI (cytochrome c oxidase I), cytb (cytochrome b), H1 (mitochondrial H-strand transcription unit 1),

H3 (histone H3), RAG1 (recombination activating gene 1), rhodopsin (rhodopsin exon 1), SIA (seventh in

absentia), and tyr (tyrosinase). Other abbreviations: Anc (ancestral state), Des (descendant state), frag. (DNA

sequence fragment), Pos (aligned position).

Taxon/Locus Pos Anc Des

Centrolenidae

H1 frag. 2 150 T C

H1 frag. 2 174 T C

H1 frag. 2 196 T C

H1 frag. 2 316 T C

H1 frag. 2 545 A T

H1 frag. 2 707 T A

H1 frag. 3 110 T C

H1 frag. 3 177 A C

H1 frag. 3 251 A G

H1 frag. 4 14 T A

H1 frag. 4 130 A C

H1 frag. 5 42 A G

H1 frag. 6 34 T C

H1 frag. 6 36 T C

H1 frag. 6 61 — C

H1 frag. 6 187 — T

H1 frag. 6 279 T A

H1 frag. 6 287 T C

H1 frag. 6 295 — C

H1 frag. 6 328 A C

H1 frag. 6 359 T C

H1 frag. 6 386 T C

H1 frag. 6 446 T C

H1 frag. 6 472 — C

H1 frag. 6 525 A T

H1 frag. 6 572 A C

H1 frag. 8 224 T C

H1 frag. 8 592 T G

H1 frag. 9 70 A G

H1 frag. 9 75 T A

H1 frag. 9 158 T —

H1 frag. 9 191 A T

H1 frag. 9 200 — A

H1 frag. 10 37 T C

H1 frag. 12 104 G A

H1 frag. 12 446 T A

H1 frag. 13 121 — A

H1 frag. 13 195 T A

H1 frag. 15 9 A G

Cruciabatrachia

H3 286 G A


H1 frag. 2 31 — C

H1 frag. 2 399 C T

H1 frag. 2 576 G A

H1 frag. 2 577 A G

H1 frag. 3 118 C —

H1 frag. 5 73 A T

H1 frag. 6 128 A T

H1 frag. 8 532 T A

H1 frag. 8 577 C T

H1 frag. 9 299 — T

H1 frag. 9 371 A T

H1 frag. 10 99 C T

H1 frag. 12 225 C A

H1 frag. 13 80 A T

H1 frag. 13 180 A T

rhodopsin 106 A G

rhodopsin 296 T C

rhodopsin 299 T C

tyr frag. 1 9 T C

tyr frag. 2 188 C T

Leptodactylidae

28S frag. 1 510 T A

H3 169 G T

H1 frag. 2 444 A G

H1 frag. 2 496 T A

H1 frag. 2 507 C A

H1 frag. 2 541 T C

H1 frag. 2 581 C A

H1 frag. 2 681 — C

H1 frag. 2 697 A C

H1 frag. 3 182 A T

H1 frag. 3 185 C —

H1 frag. 3 190 T A

H1 frag. 4 128 A T

H1 frag. 4 191 A C

H1 frag. 4 204 T A

H1 frag. 5 38 A T

H1 frag. 6 44 T C

H1 frag. 6 213 — T

H1 frag. 6 214 — T

H1 frag. 6 422 G A

H1 frag. 8 41 A T


H1 frag. 8 116 A C

H1 frag. 8 247 A T

H1 frag. 8 263 A T

H1 frag. 8 531 — A

H1 frag. 9 48 A C

H1 frag. 9 91 T A

H1 frag. 9 115 T A

H1 frag. 9 238 T C

H1 frag. 10 238 A —

H1 frag. 12 263 — T

H1 frag. 12 378 T —

H1 frag. 12 408 A C

H1 frag. 13 32 A G

H1 frag. 13 78 — C

H1 frag. 13 203 T C

H1 frag. 13 284 T C

rhodopsin 135 — T

tyr frag. 1 71 T C

tyr frag. 1 179 G C

tyr frag. 2 59 C T

tyr frag. 2 149 C T

Chthonobatrachia

H3 175 T G

H1 frag. 2 409 T A

H1 frag. 2 492 C T

H1 frag. 3 56 A C

H1 frag. 3 129 A T

H1 frag. 4 35 A T

H1 frag. 4 206 C T

H1 frag. 6 157 A T

H1 frag. 6 446 T A

H1 frag. 6 471 — T

H1 frag. 8 188 A G

H1 frag. 8 555 C T

H1 frag. 9 96 C T

H1 frag. 9 125 C T

H1 frag. 9 160 A C

H1 frag. 9 355 A C

H1 frag. 9 378 C T

H1 frag. 10 3 C T

H1 frag. 12 112 C T

H1 frag. 12 249 C T


H1 frag. 12 387 C T

H1 frag. 13 158 C A

rhodopsin 10 G A

SIA frag. 1 189 A G

SIA frag. 2 12 T C

tyr frag. 1 151 T C

tyr frag. 2 110 T C

Ceratophryidae

28S frag. 1 181 C —

28S frag. 1 385 C —

28S frag. 1 596 C T

28S frag. 1 606 C A

H3 290 G C

H1 frag. 2 316 T C

H1 frag. 2 447 G A

H1 frag. 2 489 — A

H1 frag. 2 526 C T

H1 frag. 3 123 C T

H1 frag. 4 52 C T

H1 frag. 6 116 C T

H1 frag. 6 218 A T

H1 frag. 6 241 — T

H1 frag. 6 256 — T

H1 frag. 8 283 — A

H1 frag. 8 538 T A

H1 frag. 8 565 T G

H1 frag. 8 593 — A

H1 frag. 8 595 — C

H1 frag. 9 79 A G

H1 frag. 9 103 A T

H1 frag. 9 109 A T

H1 frag. 9 191 A C

H1 frag. 10 18 T A

H1 frag. 12 109 A T

H1 frag. 12 110 A T

H1 frag. 12 187 T A

H1 frag. 12 280 — A

H1 frag. 12 401 — C

Batrachylinae

28S frag. 1 238 G C

28S frag. 1 387 C G

28S frag. 1 402 — C



28S frag. 1 567 — C

28S frag. 1 598 — A

28S frag. 1 610 C A

28S frag. 1 617 C A

H1 frag. 2 421 — C

H1 frag. 2 496 T A

H1 frag. 2 563 T C

H1 frag. 2 615 A C

H1 frag. 3 10 G A

H1 frag. 6 79 A T

H1 frag. 6 157 T C

H1 frag. 6 170 — T

H1 frag. 6 438 T C

H1 frag. 6 474 A T

H1 frag. 6 524 A T

H1 frag. 6 570 T A

H1 frag. 7 7 C A

H1 frag. 8 45 T A

H1 frag. 8 259 C T

H1 frag. 8 530 — T

H1 frag. 8 587 T A

H1 frag. 9 70 A G

H1 frag. 9 148 — C

H1 frag. 9 158 T C

H1 frag. 9 172 A C

H1 frag. 9 264 A T

H1 frag. 10 2 C A

H1 frag. 10 70 C T

H1 frag. 10 262 A T

H1 frag. 12 17 G A

H1 frag. 12 377 — C

H1 frag. 12 411 T —

H1 frag. 12 425 T C

H1 frag. 13 14 C A

rhodopsin 96 C T

rhodopsin 98 T A

rhodopsin 185 C G

rhodopsin 271 C G

rhodopsin 299 C G

SIA frag. 1 3 T C

SIA frag. 2 54 C T

Ceratophryinae

28S frag. 1 182 C —

28S frag. 1 241 — T

28S frag. 1 386 — T

28S frag. 1 418 A —

28S frag. 1 419 T C

28S frag. 1 478 G A

28S frag. 1 596 T —

28S frag. 1 606 A —

28S frag. 1 657 — C

H1 frag. 1 24 A G

H1 frag. 2 90 A G

H1 frag. 2 254 A T

H1 frag. 2 314 — C

H1 frag. 2 321 T C


H1 frag. 2 497 T A

H1 frag. 2 632 A C

H1 frag. 2 707 T C

H1 frag. 3 129 T A

H1 frag. 4 145 T A

H1 frag. 4 191 A C

H1 frag. 6 20 T C

H1 frag. 6 218 T —

H1 frag. 6 267 — C

H1 frag. 6 328 A C

H1 frag. 6 352 T A

H1 frag. 6 445 A T

H1 frag. 6 469 T A

H1 frag. 6 506 A G

H1 frag. 7 10 T C

H1 frag. 8 52 A C

H1 frag. 8 206 A T

H1 frag. 8 215 A T

H1 frag. 8 297 A C

H1 frag. 8 319 C T

H1 frag. 8 420 T C

H1 frag. 9 149 C A

H1 frag. 9 198 T A

H1 frag. 9 238 T C

H1 frag. 9 264 A G

H1 frag. 9 265 T A

H1 frag. 10 1 C T

H1 frag. 10 18 A C

H1 frag. 10 96 C A

H1 frag. 12 221 A T

H1 frag. 15 9 A G

H1 frag. 15 26 T C

rhodopsin 124 T C

Telmatobiinae

H3 6 A G

H3 102 C T

H3 123 — C

H3 127 T —

H3 314 C T

H1 frag. 2 390 A C

H1 frag. 2 492 T C

H1 frag. 3 184 A T

H1 frag. 3 307 C T

H1 frag. 4 137 A T

H1 frag. 5 73 T C

H1 frag. 6 141 T C

H1 frag. 6 241 T C

H1 frag. 6 270 C T

H1 frag. 6 319 — G

H1 frag. 6 377 A C

H1 frag. 6 449 T C

H1 frag. 6 471 T C

H1 frag. 6 565 G A

H1 frag. 6 574 C T

H1 frag. 6 578 C T

H1 frag. 7 37 A C


H1 frag. 8 107 C T

H1 frag. 8 160 T A

H1 frag. 8 167 A G

H1 frag. 8 177 A G

H1 frag. 8 197 T C

H1 frag. 8 233 A T

H1 frag. 8 253 A T

H1 frag. 8 328 T C

H1 frag. 8 364 T C

H1 frag. 8 369 C T

H1 frag. 8 422 A C

H1 frag. 8 451 A T

H1 frag. 8 454 A C

H1 frag. 8 532 A C

H1 frag. 8 577 T A

H1 frag. 9 26 T A

H1 frag. 9 160 C A

H1 frag. 9 213 G T

H1 frag. 9 242 T A

H1 frag. 9 248 A G

H1 frag. 9 286 A C

H1 frag. 9 313 A T

H1 frag. 9 371 T A

H1 frag. 9 378 T C

H1 frag. 9 383 T C

H1 frag. 9 384 G A

H1 frag. 9 407 C A

H1 frag. 10 26 A T

H1 frag. 10 37 T C

H1 frag. 10 51 T A

H1 frag. 10 127 C T

H1 frag. 11 14 T C

H1 frag. 11 16 C A

H1 frag. 12 69 A G

H1 frag. 12 101 G T

H1 frag. 12 151 A G

H1 frag. 12 160 T A

H1 frag. 12 235 T C

H1 frag. 12 249 T A

H1 frag. 12 408 A T

H1 frag. 12 461 A T

H1 frag. 12 463 T C

H1 frag. 13 20 A T

H1 frag. 13 49 — C

H1 frag. 13 88 C T

H1 frag. 13 97 T A

H1 frag. 13 180 T A

H1 frag. 13 190 T C

H1 frag. 13 203 T C

H1 frag. 13 204 T C

rhodopsin 308 T A

Hesticobatrachia

28S frag. 1 325 — G

28S frag. 1 350 — C

28S frag. 1 603 — C

cytb frag. 2 34 C A


cytb frag. 2 47 A C

cytb frag. 2 159 T C

cytb frag. 2 190 G A

cytb frag. 2 223 C T

cytb frag. 3 11 C A

H1 frag. 2 707 T C

H1 frag. 3 308 C T

H1 frag. 6 31 G A

H1 frag. 6 232 T C

H1 frag. 6 328 A T

H1 frag. 8 319 C A

H1 frag. 8 369 C T

H1 frag. 8 527 C T

H1 frag. 9 213 G T

H1 frag. 9 286 A T

H1 frag. 10 1 C T

H1 frag. 10 108 A T

H1 frag. 12 113 A T

H1 frag. 12 408 A T

H1 frag. 13 102 — A

H1 frag. 13 223 A T

rhodopsin 263 G C

Cycloramphidae

28S frag. 1 224 — T

28S frag. 1 276 — T

28S frag. 1 317 — G

28S frag. 1 319 — G

28S frag. 1 359 — G

28S frag. 1 361 — T

28S frag. 1 397 — G

COI frag. 1 108 A T

cytb frag. 2 41 C T


cytb frag. 2 180 T A


H3 15 A C

H1 frag. 2 492 T A

H1 frag. 2 579 — A

H1 frag. 6 19 C T

H1 frag. 6 100 — C

H1 frag. 6 247 — A

H1 frag. 6 248 — A

H1 frag. 6 476 T A

H1 frag. 8 259 C A

H1 frag. 8 565 T A

H1 frag. 12 184 T —

H1 frag. 12 213 T A

H1 frag. 12 274 — C

H1 frag. 13 80 T —

H1 frag. 13 88 C A

H1 frag. 13 220 — A

rhodopsin 168 A G

rhodopsin 271 C G

tyr frag. 2 197 T C

tyr frag. 2 275 C T

Calamitophrynia



28S frag. 1 571 — T

28S frag. 1 578 — C

28S frag. 1 584 — C

28S frag. 1 587 — C

28S frag. 1 588 — C

28S frag. 1 624 — G

28S frag. 1 626 — G

28S frag. 1 633 — G

28S frag. 1 634 — G

28S frag. 1 636 — G

H1 frag. 2 150 T C

H1 frag. 2 374 T C

H1 frag. 2 423 A T

H1 frag. 2 545 A T

H1 frag. 2 581 C A

H1 frag. 3 207 C T

H1 frag. 6 195 — C

H1 frag. 6 469 T A

H1 frag. 8 206 A —

H1 frag. 8 208 A —

H1 frag. 10 55 A T

H1 frag. 10 71 C T

H1 frag. 10 96 C —

H1 frag. 13 67 G A

H1 frag. 13 180 T A

rhodopsin 87 A G

rhodopsin 116 T C

Leiuperidae

H3 6 A G

H3 136 C T

H3 175 G T

H3 199 G A

H1 frag. 2 632 A T

H1 frag. 3 56 C A

H1 frag. 3 110 T C

H1 frag. 3 250 A G

H1 frag. 3 251 A G

H1 frag. 4 135 A T

H1 frag. 4 204 T A

H1 frag. 6 141 T A

H1 frag. 6 314 — C

H1 frag. 6 422 G —

H1 frag. 6 445 A T

H1 frag. 8 116 A C

H1 frag. 8 155 G A

H1 frag. 8 188 G T

H1 frag. 8 224 T C

H1 frag. 8 278 A T

H1 frag. 8 281 T C

H1 frag. 8 313 A T

H1 frag. 8 399 A T

H1 frag. 8 592 T C

H1 frag. 8 601 A G

H1 frag. 9 208 — T

H1 frag. 9 283 T A

H1 frag. 9 371 T A

H1 frag. 10 23 — G


H1 frag. 10 26 A T

H1 frag. 10 47 — A

H1 frag. 12 160 T C

H1 frag. 12 378 T C

H1 frag. 13 22 T C

H1 frag. 13 160 A C

H1 frag. 13 204 T C

Agastorophrynia

28S frag. 1 214 C —

28S frag. 1 305 G —

28S frag. 1 385 C G

28S frag. 1 589 — G

28S frag. 1 628 — C

H1 frag. 1 24 A G

H1 frag. 2 399 T A

H1 frag. 2 424 A —

H1 frag. 3 15 A T

H1 frag. 3 255 G A

H1 frag. 4 172 T A

H1 frag. 6 181 G —

H1 frag. 6 290 A —

H1 frag. 6 315 T C

H1 frag. 6 538 A T

H1 frag. 9 160 C —

H1 frag. 10 3 T C

H1 frag. 10 78 — C

H1 frag. 11 14 T A

H1 frag. 12 285 C A

rhodopsin 54 A G

rhodopsin 299 C G

Bufonidae

28S frag. 1 181 C —

28S frag. 1 192 G A

28S frag. 1 381 G C

28S frag. 1 413 A C

28S frag. 1 492 — T

28S frag. 1 499 — C

28S frag. 1 597 C G

COI frag. 1 47 A C

COI frag. 1 123 T C

COI frag. 1 258 C T

COI frag. 1 299 C T

COI frag. 1 326 A T

COI frag. 2 209 T A

COI frag. 2 235 T C

COI frag. 2 284 C T

cytb frag. 1 9 C A

cytb frag. 1 10 G C

cytb frag. 1 13 C A

cytb frag. 2 56 T C

cytb frag. 2 95 C T





cytb frag. 3 14 C T

H3 114 A C


H3 130 G A

H1 frag. 1 69 A C

H1 frag. 2 27 T —

H1 frag. 2 319 T C

H1 frag. 2 542 T C

H1 frag. 2 682 C T

H1 frag. 2 695 C T

H1 frag. 3 92 C A

H1 frag. 3 300 G A

H1 frag. 3 307 C A

H1 frag. 3 327 C T

H1 frag. 6 16 A G

H1 frag. 6 33 T C

H1 frag. 6 125 — A

H1 frag. 6 232 C A

H1 frag. 6 371 C T

H1 frag. 6 521 T A

H1 frag. 6 535 T C

H1 frag. 7 36 T A

H1 frag. 8 15 C T

H1 frag. 8 39 T A

H1 frag. 8 45 T C

H1 frag. 8 120 G A

H1 frag. 8 200 — A

H1 frag. 8 282 — A

H1 frag. 8 344 A T

H1 frag. 8 556 — A

H1 frag. 9 185 A C

H1 frag. 9 204 — T

H1 frag. 9 342 A C

H1 frag. 12 185 — A

H1 frag. 12 304 — T

H1 frag. 13 112 T A

RAG1 frag. 1 181 G A

RAG1 frag. 2 76 C T

rhodopsin 39 A G

rhodopsin 85 A C

rhodopsin 182 A G

rhodopsin 188 A G

rhodopsin 221 G C

rhodopsin 230 T C

SIA frag. 1 57 C G

SIA frag. 2 30 T C

SIA frag. 2 99 G C

Nobleobatia

28S frag. 1 193 — T

28S frag. 1 199 — C

28S frag. 1 225 C G

28S frag. 1 561 — C

28S frag. 1 562 — C

28S frag. 1 592 — C

28S frag. 1 593 — T

28S frag. 1 605 — G

28S frag. 1 611 — G

28S frag. 1 623 — G

28S frag. 1 637 — C

COI frag. 1 132 T C


COI frag. 1 153 T A

COI frag. 1 216 C T

COI frag. 1 307 T C

COI frag. 2 10 T A

COI frag. 2 98 T A

COI frag. 2 278 T C



H3 277 C T

H1 frag. 2 90 A —

H1 frag. 2 138 G A

H1 frag. 2 492 T C

H1 frag. 2 697 A T

H1 frag. 3 248 G A

H1 frag. 6 195 C T

H1 frag. 6 240 C T

H1 frag. 6 462 T A

H1 frag. 9 62 T C

H1 frag. 9 75 T C

H1 frag. 9 103 A T

H1 frag. 9 176 — C

H1 frag. 9 191 A T

H1 frag. 9 387 A C

H1 frag. 10 99 T C

H1 frag. 12 149 A —

H1 frag. 12 191 — C

H1 frag. 12 253 A T

H1 frag. 12 311 A C

H1 frag. 13 21 A T

H1 frag. 13 102 A —

H1 frag. 13 223 T A

rhodopsin 291 G A

Hylodidae

28S frag. 1 192 G C

28S frag. 1 202 — A

28S frag. 1 223 C —

28S frag. 1 413 A G

28S frag. 1 495 — C

28S frag. 1 497 — A

28S frag. 1 498 — A

28S frag. 1 528 C G

28S frag. 1 556 — G

28S frag. 1 570 C G

28S frag. 1 571 T A

28S frag. 1 579 — T

28S frag. 1 591 G T

28S frag. 1 606 C G

28S frag. 1 625 — C

28S frag. 1 634 G C

28S frag. 1 636 G C

28S frag. 1 638 — A

28S frag. 1 639 — A

28S frag. 1 661 G C

28S frag. 2 69 A G

28S frag. 2 164 — A

COI frag. 1 74 A C

COI frag. 1 84 T C



COI frag. 1 279 T G

COI frag. 2 198 T C

COI frag. 2 200 A T

COI frag. 2 318 A G

COI frag. 2 345 C T

cytb frag. 1 10 G T

cytb frag. 1 13 C T

cytb frag. 1 56 C T

cytb frag. 2 87 G A


H3 69 C G

H3 84 G C

H3 193 G C

H3 253 C T

H1 frag. 1 12 G A

H1 frag. 1 24 G —

H1 frag. 2 374 C T

H1 frag. 2 409 A C

H1 frag. 2 508 C A

H1 frag. 2 574 C T

H1 frag. 2 581 A C

H1 frag. 3 217 A G

H1 frag. 3 308 T C

H1 frag. 4 137 A T

H1 frag. 4 144 A C

H1 frag. 4 194 A T

H1 frag. 4 206 T A

H1 frag. 5 63 — T

H1 frag. 6 203 A C

H1 frag. 6 255 — C

H1 frag. 6 478 A T

H1 frag. 6 504 A —

H1 frag. 6 596 G A

H1 frag. 7 13 A C

H1 frag. 8 79 G A

H1 frag. 8 199 — C

H1 frag. 8 345 — G

H1 frag. 8 377 A T

H1 frag. 8 532 A T

H1 frag. 8 612 A T

H1 frag. 9 147 T A

H1 frag. 9 328 G T

H1 frag. 9 357 C T

H1 frag. 10 55 T A

H1 frag. 10 95 A C

H1 frag. 10 108 T A

H1 frag. 10 238 A C

H1 frag. 11 14 A C

H1 frag. 12 108 A T

H1 frag. 12 221 A T

H1 frag. 12 362 A T

H1 frag. 12 399 A T

H1 frag. 15 16 C T

RAG1 frag. 1 51 T C

RAG1 frag. 1 127 A T

RAG1 frag. 2 106 C T

RAG1 frag. 2 133 T C


RAG1 frag. 2 175 T C

rhodopsin 0 A G

rhodopsin 39 A C

rhodopsin 60 T C

rhodopsin 124 T C

rhodopsin 185 C G

rhodopsin 191 A G

rhodopsin 195 T C

rhodopsin 299 G T

SIA frag. 2 54 C A

SIA frag. 2 57 C T

SIA frag. 2 69 C T

tyr frag. 1 19 T A

tyr frag. 1 41 T C

tyr frag. 1 74 T C

tyr frag. 1 103 A G

tyr frag. 1 113 C T

tyr frag. 1 210 T C

tyr frag. 2 164 T G

tyr frag. 2 178 G C

tyr frag. 2 188 T C

tyr frag. 2 213 C T

tyr frag. 2 275 C T

Dendrobatoidea

28S frag. 1 139 C A

28S frag. 1 140 A C

28S frag. 1 219 — T

28S frag. 1 238 G C

28S frag. 1 356 — G

28S frag. 1 408 — A

28S frag. 1 410 C A

28S frag. 1 485 C T

28S frag. 1 538 — A

28S frag. 1 560 G C

28S frag. 1 566 G C

28S frag. 1 594 — A

28S frag. 1 595 — C

28S frag. 1 648 C —

28S frag. 1 649 C G

28S frag. 2 67 A C

COI frag. 1 63 C T

COI frag. 1 108 A T

COI frag. 1 150 T C

COI frag. 1 181 A C

COI frag. 2 44 C T

COI frag. 2 53 C A

COI frag. 2 130 T C

COI frag. 2 131 T C

cytb frag. 1 11 T C

cytb frag. 1 12 A C

cytb frag. 1 33 A T

cytb frag. 1 61 T A

cytb frag. 1 63 A C

cytb frag. 1 68 T G

cytb frag. 2 34 A C

cytb frag. 2 41 C T

cytb frag. 2 59 C T



cytb frag. 2 128 A T





H3 175 G T

H1 frag. 1 41 C T

H1 frag. 2 58 C A

H1 frag. 2 148 T C

H1 frag. 2 149 A T

H1 frag. 2 173 A C

H1 frag. 2 195 A G

H1 frag. 2 232 G C

H1 frag. 2 316 T C

H1 frag. 2 432 A C

H1 frag. 2 488 — A

H1 frag. 2 496 T C

H1 frag. 2 545 T A

H1 frag. 2 572 G A

H1 frag. 2 613 C T

H1 frag. 2 618 C T

H1 frag. 3 177 A —

H1 frag. 3 182 A —

H1 frag. 3 184 A —

H1 frag. 3 185 C —

H1 frag. 3 208 C T

H1 frag. 3 214 A T

H1 frag. 3 219 C A

H1 frag. 3 242 G A

H1 frag. 4 30 C T

H1 frag. 4 103 T C

H1 frag. 4 116 A T

H1 frag. 4 130 A C

H1 frag. 6 123 T C

H1 frag. 6 141 T C

H1 frag. 6 225 - C

H1 frag. 6 297 A T

H1 frag. 6 311 A T

H1 frag. 6 336 T C

H1 frag. 6 422 G A

H1 frag. 6 445 A T

H1 frag. 7 37 A T

H1 frag. 8 13 A G

H1 frag. 8 25 C T

H1 frag. 8 88 C A

H1 frag. 8 141 A G

H1 frag. 8 148 A T

H1 frag. 8 169 C T

H1 frag. 8 230 A —

H1 frag. 8 233 A —

H1 frag. 8 263 A —

H1 frag. 8 265 A —

H1 frag. 8 281 T —

H1 frag. 8 366 G A

H1 frag. 8 422 A C

H1 frag. 8 451 A T


H1 frag. 8 541 A T

H1 frag. 8 577 T A

H1 frag. 8 609 A G

H1 frag. 9 26 T A

H1 frag. 9 31 — C

H1 frag. 9 55 A —

H1 frag. 9 70 A —

H1 frag. 9 128 — A

H1 frag. 9 193 — C

H1 frag. 9 245 T C

H1 frag. 9 258 T A

H1 frag. 9 262 T C

H1 frag. 9 299 T A

H1 frag. 9 395 A C

H1 frag. 10 4 A —

H1 frag. 10 8 G T

H1 frag. 10 83 C T

H1 frag. 10 145 C T

H1 frag. 10 196 A G

H1 frag. 10 247 A T

H1 frag. 12 38 A C

H1 frag. 12 46 G A

H1 frag. 12 49 T C

H1 frag. 12 159 A C

H1 frag. 12 167 C —

H1 frag. 12 174 T —

H1 frag. 12 176 G A

H1 frag. 12 349 A C

H1 frag. 12 430 T A

H1 frag. 12 455 A T

H1 frag. 12 461 A C

H1 frag. 13 33 C T

H1 frag. 13 34 A G

H1 frag. 13 124 A C

RAG1 frag. 1 4 A C

RAG1 frag. 1 7 T G

RAG1 frag. 1 21 A C

RAG1 frag. 1 60 C T

RAG1 frag. 1 79 T C


RAG1 frag. 1 128 C G

RAG1 frag. 1 138 A C




RAG1 frag. 2 163 A G

rhodopsin 48 A C

rhodopsin 54 G T

rhodopsin 87 G A

rhodopsin 116 C T

rhodopsin 240 G A

rhodopsin 308 T G

SIA frag. 1 18 T C

SIA frag. 1 33 A G

SIA frag. 1 39 T C

SIA frag. 2 75 A G

tyr frag. 1 9 C T



tyr frag. 1 11 C A

tyr frag. 1 27 C T

tyr frag. 1 29 T A

tyr frag. 1 44 T C

tyr frag. 1 71 T A

tyr frag. 1 73 C T

tyr frag. 1 192 A G

tyr frag. 1 201 C T

tyr frag. 1 214 G A

tyr frag. 2 11 G C

tyr frag. 2 22 C G

tyr frag. 2 32 G C

tyr frag. 2 41 C G

tyr frag. 2 209 C A

Aromobatidae

28S frag. 1 297 — C

COI frag. 1 25 G C

COI frag. 1 194 A C

COI frag. 2 81 C T

COI frag. 2 210 G A



cytb frag. 3 26 A C

H1 frag. 2 290 C —

H1 frag. 2 297 T A

H1 frag. 2 298 G A

H1 frag. 2 449 T C

H1 frag. 2 555 A C

H1 frag. 3 56 C A

H1 frag. 3 244 A T

H1 frag. 3 320 C A

H1 frag. 4 107 — T

H1 frag. 4 227 C T

H1 frag. 5 38 A G

H1 frag. 5 45 G A

H1 frag. 5 91 C T

H1 frag. 6 13 A T

H1 frag. 6 19 C T

H1 frag. 6 38 T C

H1 frag. 6 49 A C

H1 frag. 6 158 — A

H1 frag. 6 223 — C

H1 frag. 6 361 T A

H1 frag. 6 416 T A

H1 frag. 6 471 T A

H1 frag. 6 485 C T

H1 frag. 6 527 — C

H1 frag. 6 572 A C

H1 frag. 7 42 T C

H1 frag. 8 52 A C

H1 frag. 8 146 C T

H1 frag. 8 147 A T

H1 frag. 8 384 A T

H1 frag. 8 420 T C

H1 frag. 8 463 T —

H1 frag. 8 532 A C

H1 frag. 8 568 A T


H1 frag. 9 125 T A

H1 frag. 9 129 — C

H1 frag. 9 202 A C

H1 frag. 9 238 T C

H1 frag. 9 259 G A

H1 frag. 9 346 A C

H1 frag. 9 371 T A

H1 frag. 9 382 G C

H1 frag. 10 3 C T

H1 frag. 10 54 A C

H1 frag. 10 95 A T

H1 frag. 10 183 T C

H1 frag. 10 262 A C

H1 frag. 10 266 T C

H1 frag. 12 8 T C

H1 frag. 12 76 G T

H1 frag. 12 191 C G

H1 frag. 12 311 C T

H1 frag. 12 425 T A

RAG1 frag. 2 61 G A



rhodopsin 72 G A

SIA frag. 1 12 T G

SIA frag. 2 51 T C

SIA frag. 2 72 T C

tyr frag. 2 177 A C

tyr frag. 2 182 T C

tyr frag. 2 206 A G

Anomaloglossinae

COI frag. 1 18 A C

COI frag. 1 74 A T

COI frag. 1 197 T C

COI frag. 1 225 T C

COI frag. 1 288 A T

COI frag. 2 206 T C

COI frag. 2 209 T C

cytb frag. 1 83 T C

cytb frag. 1 92 T C

cytb frag. 2 1 A T

cytb frag. 2 10 C T

cytb frag. 2 53 C T



cytb frag. 3 14 C A

H1 frag. 2 447 G A

H1 frag. 2 526 C T

H1 frag. 2 697 T C

H1 frag. 5 94 T C

H1 frag. 6 133 T C

H1 frag. 6 197 — A

H1 frag. 8 215 A T

H1 frag. 8 543 A G

H1 frag. 8 577 A C

H1 frag. 9 109 A T

H1 frag. 9 131 T C

H1 frag. 9 286 T C


H1 frag. 9 395 C T

H1 frag. 9 396 A T

H1 frag. 10 2 C T

H1 frag. 10 11 — C

H1 frag. 10 14 T C

H1 frag. 12 430 A C

H1 frag. 13 80 A G

H1 frag. 13 223 A T

H1 frag. 13 274 C T

H1 frag. 14 3 G A

Anomaloglossus

COI frag. 1 11 C T

COI frag. 1 59 C T

COI frag. 1 147 T A

COI frag. 1 171 T C

COI frag. 1 228 A C

COI frag. 1 320 T A

COI frag. 2 10 A C

COI frag. 2 122 T C

COI frag. 2 244 C A

COI frag. 2 270 C T

COI frag. 2 285 C T

COI frag. 2 351 C T

cytb frag. 1 15 A C

cytb frag. 1 33 T A

cytb frag. 1 48 A C

cytb frag. 2 26 A C

cytb frag. 2 135 C A




H1 frag. 2 11 T A

H1 frag. 2 113 — A

H1 frag. 2 117 T C

H1 frag. 2 172 A C

H1 frag. 2 232 C T

H1 frag. 2 241 T C

H1 frag. 2 359 — C

H1 frag. 2 390 A C

H1 frag. 2 451 A T

H1 frag. 2 476 T C

H1 frag. 2 545 A C

H1 frag. 6 328 T C

H1 frag. 6 372 T C

H1 frag. 6 446 C T

H1 frag. 6 450 — C

H1 frag. 6 463 A T

H1 frag. 6 469 A C

H1 frag. 6 474 A T

H1 frag. 8 41 T C

H1 frag. 8 113 T A

H1 frag. 8 166 A C

H1 frag. 8 170 T C

H1 frag. 8 241 T A

H1 frag. 8 344 A C

H1 frag. 8 365 A G

H1 frag. 8 527 T C


H1 frag. 8 587 T A

H1 frag. 9 161 — C

H1 frag. 9 284 T A

H1 frag. 10 79 — T

H1 frag. 10 108 T C

H1 frag. 10 250 — A

H1 frag. 12 349 C A

H1 frag. 12 378 T C

H1 frag. 12 434 A C

H1 frag. 13 33 T C

H1 frag. 13 67 A —

H1 frag. 13 103 — A

H1 frag. 13 112 T A

H1 frag. 13 138 T C

H1 frag. 13 154 G A

H1 frag. 13 182 C T

H1 frag. 13 185 C T

H1 frag. 13 192 A C

H1 frag. 13 197 G A

rhodopsin 103 C T

rhodopsin 170 C A

tyr frag. 1 32 A G

tyr frag. 1 134 T C

tyr frag. 1 179 G C

tyr frag. 1 228 G A

tyr frag. 2 86 G A

tyr frag. 2 87 C A

tyr frag. 2 89 T G

tyr frag. 2 144 A C

tyr frag. 2 262 A G

Rheobates

H1 frag. 12 34 T C

H1 frag. 12 66 T C

H1 frag. 12 67 C T

H1 frag. 12 108 A T

H1 frag. 12 160 T C

H1 frag. 12 181 C T

H1 frag. 12 261 — A

H1 frag. 12 265 T A

H1 frag. 12 305 — C

H1 frag. 12 306 — T

H1 frag. 12 341 — T

H1 frag. 12 342 — T

H1 frag. 12 362 A G

H1 frag. 12 415 T C

H1 frag. 12 465 A T

H1 frag. 13 22 T C

H1 frag. 13 32 A T

H1 frag. 13 57 — T

H1 frag. 13 58 — T

H1 frag. 13 148 C —

H1 frag. 13 165 G A

H1 frag. 13 167 T C

H1 frag. 13 173 C T

H1 frag. 13 203 T C

H1 frag. 13 204 A C

H1 frag. 13 206 A C



H1 frag. 15 1 T C

H1 frag. 16 11 A G

Aromobatinae

COI frag. 1 53 C T

COI frag. 2 1 C T

COI frag. 2 10 A T

COI frag. 2 154 C T

COI frag. 2 206 T A

cytb frag. 1 13 C A

cytb frag. 1 15 A T

cytb frag. 1 27 C T

cytb frag. 2 19 C T

cytb frag. 2 41 T C

cytb frag. 2 86 T A



H3 302 C A

H3 314 C T

H1 frag. 2 211 A T

H1 frag. 2 236 A T

H1 frag. 2 369 — T

H1 frag. 2 476 T C

H1 frag. 2 545 A T

H1 frag. 3 207 T —

H1 frag. 3 255 A G

H1 frag. 3 322 C T

H1 frag. 3 324 A T

H1 frag. 4 30 T C

H1 frag. 4 32 G A

H1 frag. 6 47 A C

H1 frag. 6 116 C T

H1 frag. 6 123 C A

H1 frag. 6 162 — T

H1 frag. 6 277 — T

H1 frag. 6 371 C T

H1 frag. 6 412 C T

H1 frag. 6 538 T C

H1 frag. 8 12 A G

H1 frag. 8 123 T C

H1 frag. 8 220 — C

H1 frag. 8 352 — T

H1 frag. 8 353 — T

H1 frag. 8 541 T A

H1 frag. 8 612 A T

H1 frag. 9 65 T A

H1 frag. 9 156 — T

H1 frag. 9 185 A —

H1 frag. 9 247 A T

H1 frag. 9 345 G A

H1 frag. 10 53 — T

H1 frag. 10 92 — C

H1 frag. 10 254 A T

H1 frag. 11 16 T A

H1 frag. 12 52 T C

H1 frag. 12 108 A T

H1 frag. 13 117 — A

SIA frag. 2 81 A G


Aromobates

28S frag. 1 325 G —

cytb frag. 1 51 C A

cytb frag. 1 63 C T

cytb frag. 1 80 C T

cytb frag. 2 53 C T





H1 frag. 2 11 T C

H1 frag. 2 107 C T

H1 frag. 2 117 T A

H1 frag. 2 127 C T

H1 frag. 2 206 A T

H1 frag. 2 510 A C

H1 frag. 2 522 T C

H1 frag. 3 281 T A

H1 frag. 4 52 C T

H1 frag. 4 113 A T

H1 frag. 4 152 T A

H1 frag. 4 191 A C

H1 frag. 6 128 T A

H1 frag. 6 161 — T

H1 frag. 6 352 T A

H1 frag. 6 394 A T

H1 frag. 6 448 — T

H1 frag. 6 506 A G

H1 frag. 6 598 T C

H1 frag. 7 7 C T

H1 frag. 8 29 G A

H1 frag. 8 52 C T

H1 frag. 8 85 C T

H1 frag. 8 90 C T

H1 frag. 8 415 A T

H1 frag. 8 538 C A

H1 frag. 9 149 C T

H1 frag. 9 248 C T

H1 frag. 9 282 T —

H1 frag. 9 287 — A

H1 frag. 9 312 A G

H1 frag. 9 371 A T

H1 frag. 9 387 C A

H1 frag. 9 395 C T

H1 frag. 10 91 A C

H1 frag. 10 156 A T

H1 frag. 12 46 A G

H1 frag. 12 169 A T

H1 frag. 12 172 A T

H1 frag. 12 253 C A

H1 frag. 12 282 — T

H1 frag. 12 380 — C

H1 frag. 13 36 T A

H1 frag. 13 165 G A

H1 frag. 13 173 C T

H1 frag. 13 190 C T

H1 frag. 15 20 C A


RAG1 frag. 1 78 T C

SIA frag. 1 54 T C

SIA frag. 1 180 A G

SIA frag. 2 72 C T

SIA frag. 2 204 T G

Mannophryne

cytb frag. 1 6 C T

cytb frag. 1 92 T A

cytb frag. 2 17 G A

cytb frag. 2 18 C T

H1 frag. 3 92 C A

H1 frag. 3 124 T C

H1 frag. 3 248 A G

H1 frag. 4 21 A C

H1 frag. 4 28 C T

H1 frag. 4 34 G A

H1 frag. 4 37 G A

H1 frag. 4 41 T C

H1 frag. 4 43 T G

H1 frag. 4 143 A C

H1 frag. 5 45 A T

H1 frag. 6 23 G T

H1 frag. 6 43 C G

H1 frag. 6 44 T C

H1 frag. 6 45 A C

H1 frag. 6 141 C —

H1 frag. 6 147 C T

H1 frag. 6 189 A T

H1 frag. 6 195 T C

H1 frag. 6 287 T C

H1 frag. 6 336 C A

H1 frag. 6 393 T G

H1 frag. 6 408 A C

H1 frag. 6 462 A T

H1 frag. 6 463 A T

H1 frag. 6 469 A C

H1 frag. 6 527 C T

H1 frag. 6 552 G C

H1 frag. 7 1 C T

H1 frag. 7 45 A G

H1 frag. 8 259 C T

H1 frag. 8 277 — C

H1 frag. 8 527 T —

H1 frag. 8 543 A C

H1 frag. 8 555 T G

H1 frag. 8 568 T C

H1 frag. 9 30 A T

H1 frag. 9 109 A —

H1 frag. 9 125 A G

H1 frag. 9 202 C T

H1 frag. 9 209 C A

H1 frag. 9 382 C T

H1 frag. 9 384 G T

H1 frag. 9 408 T A

H1 frag. 10 17 A C

H1 frag. 10 32 G T

H1 frag. 10 37 C T


H1 frag. 10 48 A —

H1 frag. 10 72 C T

H1 frag. 10 99 C T

H1 frag. 10 194 T C

H1 frag. 10 228 T C

H1 frag. 10 247 T A

H1 frag. 10 259 A C

H1 frag. 12 34 T C

H1 frag. 12 51 A G

H1 frag. 12 159 C T

H1 frag. 12 168 — T

H1 frag. 12 295 T —

H1 frag. 12 311 T A

H1 frag. 12 357 A —

H1 frag. 12 362 A —

H1 frag. 13 20 A C

H1 frag. 13 26 A T

H1 frag. 13 88 C T

H1 frag. 13 124 C T

Allobatinae/Allobates

COI frag. 1 108 T C

COI frag. 1 279 T A

COI frag. 2 227 T A

COI frag. 2 315 A T

COI frag. 2 318 A C

COI frag. 2 329 G A



H1 frag. 2 220 C T

H1 frag. 2 317 T C

H1 frag. 2 399 A T

H1 frag. 2 423 T C

H1 frag. 2 466 T C

H1 frag. 2 541 T C

H1 frag. 2 682 C —

H1 frag. 3 200 T C

H1 frag. 3 309 A —

H1 frag. 4 184 T A

H1 frag. 6 157 T A

H1 frag. 6 218 A C

H1 frag. 6 328 T C

H1 frag. 6 359 T C

H1 frag. 6 364 A C

H1 frag. 6 394 A C

H1 frag. 6 511 A C

H1 frag. 8 148 T A

H1 frag. 8 272 — T

H1 frag. 8 527 T A

H1 frag. 9 129 C T

H1 frag. 9 381 A C

H1 frag. 10 145 T A

H1 frag. 12 169 A G

H1 frag. 12 184 T C

H1 frag. 12 249 T A

H1 frag. 12 349 C —

H1 frag. 12 391 A —

H1 frag. 12 436 G T



H1 frag. 12 446 T C

H1 frag. 12 448 T C

H1 frag. 12 458 T C

H1 frag. 15 29 A C

rhodopsin 33 G A

rhodopsin 302 G A

SIA frag. 1 12 G A

SIA frag. 1 57 C G

SIA frag. 2 60 C T

Dendrobatidae

28S frag. 1 345 G C

28S frag. 1 547 — T

28S frag. 1 550 — T

28S frag. 1 551 — T

COI frag. 1 94 C G

COI frag. 1 234 T A

COI frag. 2 138 T C

H3 75 C T

H3 196 A G

H1 frag. 2 16 — C

H1 frag. 2 17 A T

H1 frag. 2 127 C T

H1 frag. 2 196 T C

H1 frag. 2 292 — T

H1 frag. 2 390 A T

H1 frag. 2 414 C A

H1 frag. 4 178 A C

H1 frag. 4 191 A T

H1 frag. 4 204 T A

H1 frag. 4 206 T C

H1 frag. 6 31 A G

H1 frag. 6 359 T C

H1 frag. 8 166 A T

H1 frag. 8 167 A G

H1 frag. 8 224 T A

H1 frag. 8 270 A T

H1 frag. 9 120 A T

H1 frag. 9 134 — A

H1 frag. 9 153 — T

H1 frag. 9 185 A —

H1 frag. 9 322 C T

H1 frag. 9 408 T C

H1 frag. 10 15 — A

H1 frag. 12 74 G A

H1 frag. 12 188 — G

H1 frag. 12 198 — T

H1 frag. 12 208 — T

H1 frag. 12 249 T C

H1 frag. 12 454 G A

H1 frag. 13 206 A C

H1 frag. 13 272 C T

tyr frag. 1 57 A C

tyr frag. 1 151 C T

tyr frag. 2 17 A G

tyr frag. 2 254 A G

Colostethinae

28S frag. 1 226 — T


28S frag. 1 312 — G

28S frag. 1 313 — G

COI frag. 1 98 T —

COI frag. 1 102 — A

COI frag. 1 270 T C

COI frag. 1 307 C A

COI frag. 2 83 A T

COI frag. 2 132 C T

COI frag. 2 209 T A

COI frag. 2 244 C T

COI frag. 2 287 A T

COI frag. 2 348 A C

cytb frag. 1 27 C T

cytb frag. 1 42 C T


cytb frag. 3 8 T C

H3 51 C T

H3 187 C T

H1 frag. 2 15 A T

H1 frag. 2 137 A T

H1 frag. 2 212 C T

H1 frag. 2 397 C A

H1 frag. 2 507 C A

H1 frag. 3 116 A T

H1 frag. 3 279 A C

H1 frag. 3 308 T —

H1 frag. 4 35 A C

H1 frag. 4 41 T A

H1 frag. 4 46 T A

H1 frag. 4 181 A T

H1 frag. 5 34 A T

H1 frag. 6 51 C T

H1 frag. 6 134 — C

H1 frag. 6 225 C T

H1 frag. 6 244 T C

H1 frag. 6 270 C A

H1 frag. 6 318 — T

H1 frag. 6 338 — T

H1 frag. 6 449 T C

H1 frag. 6 470 A C

H1 frag. 6 496 G A

H1 frag. 6 596 G A

H1 frag. 7 48 C T

H1 frag. 8 14 G A

H1 frag. 8 22 C T

H1 frag. 8 121 C T

H1 frag. 8 189 — C

H1 frag. 8 344 A T

H1 frag. 8 405 T A

H1 frag. 8 545 — C

H1 frag. 9 231 T A

H1 frag. 9 382 G A

H1 frag. 9 383 T C

H1 frag. 10 14 T A

H1 frag. 10 262 A T

H1 frag. 11 14 A T

H1 frag. 12 52 T C


H1 frag. 12 110 A T

H1 frag. 12 279 A T

H1 frag. 12 329 — C

H1 frag. 12 362 A G

H1 frag. 12 368 A C

H1 frag. 12 442 — T

H1 frag. 13 20 A T

H1 frag. 13 71 A G

H1 frag. 13 171 A T

RAG1 frag. 1 21 C T

RAG1 frag. 1 42 G A

RAG1 frag. 1 57 A G






SIA frag. 1 102 C T

tyr frag. 1 113 C T

tyr frag. 1 184 A G

tyr frag. 1 202 A C

tyr frag. 2 86 G T

tyr frag. 2 99 A G

tyr frag. 2 128 G C

tyr frag. 2 263 G A

Ameerega

28S frag. 1 225 G A

28S frag. 1 350 C T

28S frag. 1 387 C T

COI frag. 1 34 C A

COI frag. 1 37 A C

COI frag. 1 47 A C

COI frag. 1 74 A C

COI frag. 1 84 T C

COI frag. 1 129 A T

COI frag. 1 243 C T

COI frag. 1 291 T C

COI frag. 2 80 T C

COI frag. 2 83 T C

COI frag. 2 98 A T

COI frag. 2 179 T A

COI frag. 2 188 T C

COI frag. 2 328 A C

COI frag. 2 331 T C

COI frag. 2 348 C G

cytb frag. 1 24 T C

cytb frag. 1 68 G A






H3 105 T G

H3 111 A G

H3 220 C T

H1 frag. 1 24 G C

H1 frag. 2 44 A C


H1 frag. 2 322 A T

H1 frag. 2 447 G A

H1 frag. 2 496 C T

H1 frag. 2 526 C T

H1 frag. 2 551 C T

H1 frag. 2 576 G A

H1 frag. 2 577 A G

H1 frag. 2 613 T C

H1 frag. 2 614 C T

H1 frag. 2 665 A —

H1 frag. 2 698 — C

H1 frag. 3 281 T A

H1 frag. 4 43 T C

H1 frag. 4 128 A G

H1 frag. 4 173 — C

H1 frag. 4 191 T C

H1 frag. 4 242 T C

H1 frag. 6 24 A C

H1 frag. 6 157 T C

H1 frag. 6 244 C —

H1 frag. 6 251 A T

H1 frag. 6 261 C T

H1 frag. 6 270 A T

H1 frag. 6 338 T C

H1 frag. 6 352 T C

H1 frag. 6 371 C T

H1 frag. 6 462 A T

H1 frag. 6 476 T C

H1 frag. 7 37 T C

H1 frag. 8 36 C T

H1 frag. 8 49 G A

H1 frag. 8 90 A C

H1 frag. 8 102 G A

H1 frag. 8 147 A T

H1 frag. 8 148 T C

H1 frag. 8 167 G A

H1 frag. 8 189 C T

H1 frag. 8 229 T C

H1 frag. 8 235 T C

H1 frag. 8 270 T C

H1 frag. 8 369 T C

H1 frag. 8 400 A T

H1 frag. 8 415 A C

H1 frag. 8 451 T C

H1 frag. 8 454 A C

H1 frag. 8 463 T C

H1 frag. 8 538 C T

H1 frag. 8 545 C A

H1 frag. 8 592 T C

H1 frag. 9 18 C T

H1 frag. 9 26 A T

H1 frag. 9 38 A G

H1 frag. 9 54 T C

H1 frag. 9 128 A —

H1 frag. 9 172 A —

H1 frag. 9 193 C T

H1 frag. 9 286 T A



H1 frag. 9 346 A C

H1 frag. 9 387 C A

H1 frag. 9 395 C T

H1 frag. 9 405 A C

H1 frag. 9 407 C A

H1 frag. 10 3 C T

H1 frag. 10 17 A C

H1 frag. 10 25 A C

H1 frag. 10 26 T C

H1 frag. 10 83 T —

H1 frag. 10 145 T C

H1 frag. 10 183 T C

H1 frag. 10 194 T C

H1 frag. 10 238 A T

H1 frag. 12 46 A G

H1 frag. 12 51 A G

H1 frag. 12 87 T C

H1 frag. 12 112 T C

H1 frag. 12 221 A —

H1 frag. 12 255 — C

H1 frag. 12 295 T —

H1 frag. 12 363 — C

H1 frag. 12 425 T C

H1 frag. 13 20 T C

H1 frag. 13 22 T A

H1 frag. 13 36 T A

H1 frag. 13 39 A —

H1 frag. 13 67 A T

H1 frag. 13 88 C A

H1 frag. 13 112 T A

H1 frag. 13 142 C T

H1 frag. 13 152 T A

H1 frag. 13 158 T C

H1 frag. 13 165 G A

H1 frag. 13 173 C T

H1 frag. 13 201 T C

RAG1 frag. 2 37 A C

SIA frag. 1 78 A C

SIA frag. 2 165 T G

SIA frag. 2 186 C T

Colostethus

28S frag. 1 233 C —

28S frag. 1 345 C —

COI frag. 1 15 A T

COI frag. 1 111 C T

COI frag. 1 150 C T

COI frag. 1 216 T C

COI frag. 1 219 A T

COI frag. 2 102 C T

COI frag. 2 113 T C

COI frag. 2 227 T A

COI frag. 2 293 C T

COI frag. 2 345 C T

cytb frag. 1 71 A C

cytb frag. 1 80 C T

cytb frag. 2 87 G A



cytb frag. 2 199 A G

cytb frag. 3 20 T C

H1 frag. 1 41 T C

H1 frag. 2 196 C T

H1 frag. 4 130 C T

H1 frag. 4 152 T C

H1 frag. 5 94 T C

H1 frag. 6 195 T A

H1 frag. 6 317 — A

H1 frag. 6 361 T —

H1 frag. 7 14 T A

H1 frag. 8 259 C A

H1 frag. 8 359 — C

H1 frag. 9 240 T C

H1 frag. 9 418 T C

H1 frag. 10 51 C A

H1 frag. 10 104 T C

rhodopsin 185 C G

rhodopsin 215 T C

Epipedobates

cytb frag. 2 10 C T

cytb frag. 2 44 A T

cytb frag. 2 87 G C

cytb frag. 2 98 C T


H1 frag. 1 24 G A

H1 frag. 2 41 A T

H1 frag. 2 48 T C

H1 frag. 2 107 C A

H1 frag. 2 148 C A

H1 frag. 2 316 C T

H1 frag. 2 397 A —

H1 frag. 2 666 — T

H1 frag. 2 667 — T

H1 frag. 2 668 — T

H1 frag. 2 682 C T

H1 frag. 2 690 C T

H1 frag. 3 187 T —

H1 frag. 3 193 A —

H1 frag. 3 201 T A

H1 frag. 3 287 A G

H1 frag. 3 324 A G

H1 frag. 4 128 A T

H1 frag. 5 50 — A

H1 frag. 5 73 C A

H1 frag. 6 24 A T

H1 frag. 6 134 C T

H1 frag. 6 203 A T

H1 frag. 6 218 A C

H1 frag. 6 240 T C

H1 frag. 6 315 C T

H1 frag. 6 352 T A

H1 frag. 6 431 — T

H1 frag. 6 525 A T

H1 frag. 6 538 T C

H1 frag. 6 570 T A

H1 frag. 7 15 T A


H1 frag. 7 30 C T

H1 frag. 8 41 T C

H1 frag. 8 52 A G

H1 frag. 8 270 T A

H1 frag. 8 355 A C

H1 frag. 8 538 C T

H1 frag. 8 543 A T

H1 frag. 8 577 A G

H1 frag. 8 612 A G

H1 frag. 9 40 A T

H1 frag. 9 65 T C

H1 frag. 9 92 — C

H1 frag. 9 109 A C

H1 frag. 9 174 T —

H1 frag. 9 202 A C

H1 frag. 9 214 — C

H1 frag. 9 245 C T

H1 frag. 9 248 C T

H1 frag. 9 298 A G

H1 frag. 9 305 T A

H1 frag. 9 351 — C

H1 frag. 10 32 G A

H1 frag. 10 51 C T

H1 frag. 10 72 C T

H1 frag. 10 85 T C

H1 frag. 10 99 C T

H1 frag. 12 66 T C

H1 frag. 12 112 T A

H1 frag. 12 160 T A

H1 frag. 12 253 T C

H1 frag. 12 295 T A

H1 frag. 12 368 C G

H1 frag. 12 439 — A

H1 frag. 13 97 T —

H1 frag. 13 112 T C

H1 frag. 13 275 T C

H1 frag. 13 278 A G

H1 frag. 14 2 A G

Silverstoneia

28S frag. 1 575 — C

28S frag. 1 585 — T

COI frag. 1 59 C T

COI frag. 1 78 T C

COI frag. 1 180 T C

COI frag. 1 217 C T

COI frag. 2 180 T C

COI frag. 2 215 T A

COI frag. 2 290 C T

cytb frag. 1 69 C T

cytb frag. 1 92 T A

cytb frag. 2 56 T C

H1 frag. 2 541 T C

H1 frag. 2 542 T A

H1 frag. 2 545 A C

H1 frag. 4 30 T C

H1 frag. 5 42 G A

H1 frag. 6 34 C T


H1 frag. 6 338 T A

H1 frag. 6 535 T C

H1 frag. 8 349 — A

H1 frag. 8 378 T C

H1 frag. 8 391 A G

H1 frag. 8 463 T —

H1 frag. 8 541 T A

H1 frag. 9 131 T A

H1 frag. 9 176 C T

H1 frag. 9 238 T C

H1 frag. 9 408 C T

H1 frag. 10 1 T —

H1 frag. 10 95 A T

H1 frag. 10 108 T A

H1 frag. 10 266 T A

H1 frag. 12 349 C T

H1 frag. 12 378 T C

H1 frag. 13 39 A T

H1 frag. 13 138 T A



Hyloxalinae/Hyloxalus

28S frag. 1 246 C —

COI frag. 1 18 A C

COI frag. 1 68 T C

COI frag. 1 138 C T

COI frag. 1 159 T C

COI frag. 1 191 G C

COI frag. 1 320 T A

COI frag. 1 330 A G

COI frag. 2 101 T C

COI frag. 2 198 T C

COI frag. 2 245 T C

COI frag. 2 247 A T

cytb frag. 1 15 A T


H3 78 C A

H1 frag. 2 16 C A

H1 frag. 2 31 C T

H1 frag. 2 54 G A

H1 frag. 2 345 A T

H1 frag. 2 397 C T

H1 frag. 2 448 T C

H1 frag. 2 492 C A

H1 frag. 2 697 T C

H1 frag. 2 718 A T

H1 frag. 3 15 T C

H1 frag. 4 41 T C

H1 frag. 6 204 — T

H1 frag. 6 233 — T

H1 frag. 6 261 A T

H1 frag. 7 14 T C

H1 frag. 7 49 C T

H1 frag. 8 55 C T

H1 frag. 8 79 G A

H1 frag. 8 197 T A

H1 frag. 8 297 A C



H1 frag. 8 337 — A

H1 frag. 9 165 A T

H1 frag. 9 248 C T

H1 frag. 9 286 T C

H1 frag. 10 228 T C

H1 frag. 10 271 A G

H1 frag. 12 89 C T

H1 frag. 12 102 G A

H1 frag. 12 279 A C

H1 frag. 12 387 T A

H1 frag. 13 172 T G

H1 frag. 13 178 T A

H1 frag. 13 274 C T

H1 frag. 14 3 G A


SIA frag. 2 30 T C

SIA frag. 2 51 T C

Dendrobatinae

28S frag. 1 334 G C

COI frag. 1 317 C G

cytb frag. 1 11 C T

cytb frag. 1 12 C A

cytb frag. 2 156 A C

H1 frag. 1 74 T A

H1 frag. 2 82 A C

H1 frag. 2 107 C —

H1 frag. 2 147 C T

H1 frag. 2 172 A T

H1 frag. 2 322 A T

H1 frag. 2 438 T C

H1 frag. 2 447 G A

H1 frag. 2 451 A C

H1 frag. 2 511 A T

H1 frag. 2 526 C T

H1 frag. 3 123 C T

H1 frag. 4 135 A C

H1 frag. 4 210 C T

H1 frag. 5 92 A C

H1 frag. 5 94 T C

H1 frag. 6 148 — T

H1 frag. 6 530 — T

H1 frag. 7 47 C A

H1 frag. 8 41 T C

H1 frag. 8 52 A G

H1 frag. 8 148 T —

H1 frag. 8 289 — T

H1 frag. 8 538 C A

H1 frag. 8 587 T —

H1 frag. 9 224 A C

H1 frag. 9 235 — A

H1 frag. 9 247 A C

H1 frag. 9 378 T C

H1 frag. 10 135 A C

H1 frag. 10 139 A G

H1 frag. 10 198 C T

H1 frag. 10 266 T A

H1 frag. 12 15 A T


H1 frag. 12 48 G A

H1 frag. 12 104 A G

H1 frag. 12 235 T A

H1 frag. 12 336 C T

H1 frag. 12 378 T C

H1 frag. 12 451 — A

H1 frag. 13 33 T C

H1 frag. 13 97 T —

RAG1 frag. 1 47 C T

rhodopsin 0 A T

rhodopsin 66 T C

SIA frag. 1 15 T C

SIA frag. 1 78 A G

SIA frag. 2 165 T C

tyr frag. 1 155 C T

tyr frag. 1 189 G A

tyr frag. 2 11 C T

tyr frag. 2 41 G A

tyr frag. 2 275 C T

Adelphobates

28S frag. 1 239 — C

COI frag. 1 34 C T

COI frag. 1 44 C G

COI frag. 1 78 T C

COI frag. 1 117 A G

COI frag. 1 150 C T

COI frag. 1 317 G T

COI frag. 2 22 T C

COI frag. 2 44 T C

COI frag. 2 56 A G

COI frag. 2 68 C T

COI frag. 2 98 T A

COI frag. 2 128 A C

COI frag. 2 130 C T

COI frag. 2 138 C T

COI frag. 2 179 C A

COI frag. 2 213 C T

COI frag. 2 265 A G

COI frag. 2 296 C A

COI frag. 2 318 T C

COI frag. 2 345 C T

cytb frag. 1 12 A C

cytb frag. 1 28 A C

cytb frag. 2 72 A T

cytb frag. 2 73 T C





H1 frag. 1 18 G A

H1 frag. 2 54 G A

H1 frag. 2 209 C T

H1 frag. 2 212 C A

H1 frag. 2 214 G A

H1 frag. 2 292 T C

H1 frag. 2 297 T G

H1 frag. 2 321 C T


H1 frag. 2 414 A T

H1 frag. 2 423 T A

H1 frag. 2 454 C T

H1 frag. 2 469 G A

H1 frag. 2 471 C T

H1 frag. 2 488 A T

H1 frag. 2 492 C T

H1 frag. 2 515 A G

H1 frag. 2 695 C T

H1 frag. 3 76 A T

H1 frag. 3 242 A G

H1 frag. 3 281 T A

H1 frag. 3 329 G A

H1 frag. 4 4 C T

H1 frag. 4 28 C T

H1 frag. 4 178 C A

H1 frag. 4 199 — T

H1 frag. 4 210 T —

H1 frag. 5 94 C T

H1 frag. 6 48 T C

H1 frag. 6 157 G C

H1 frag. 6 203 A C

H1 frag. 6 215 T C

H1 frag. 7 24 A T

H1 frag. 7 30 C T

H1 frag. 7 34 T C

H1 frag. 8 43 — T

H1 frag. 8 46 — A

H1 frag. 8 90 A T

H1 frag. 8 93 A G

H1 frag. 8 141 G A

H1 frag. 8 144 T C

H1 frag. 8 211 — A

H1 frag. 8 253 T —

H1 frag. 8 378 T C

H1 frag. 8 391 A G

H1 frag. 8 607 T C

H1 frag. 8 614 C A

H1 frag. 9 1 A G

H1 frag. 9 26 A T

H1 frag. 9 59 — T

H1 frag. 9 146 — A

H1 frag. 9 176 C A

H1 frag. 9 213 T A

H1 frag. 9 237 T C

H1 frag. 9 247 C A

H1 frag. 9 260 A G

H1 frag. 9 262 C T

H1 frag. 9 298 A G

H1 frag. 9 336 C A

H1 frag. 9 369 — C

H1 frag. 10 281 G A

H1 frag. 12 40 C A

H1 frag. 12 113 C T

H1 frag. 12 184 T C

H1 frag. 12 227 — C

H1 frag. 12 257 — C


H1 frag. 12 272 — A

H1 frag. 12 336 T C

H1 frag. 12 411 A T

H1 frag. 13 14 A C

H1 frag. 13 52 — T

H1 frag. 13 53 — T

H1 frag. 13 278 G A

H1 frag. 14 8 T C

SIA frag. 1 129 T C

SIA frag. 2 105 A G

SIA frag. 2 171 A G

Dendrobates

28S frag. 1 193 T A

28S frag. 1 329 — C

COI frag. 1 11 C T

COI frag. 1 37 G A

COI frag. 1 159 T C

COI frag. 1 252 A T

COI frag. 2 50 T A

COI frag. 2 53 A T

COI frag. 2 125 C T

COI frag. 2 212 C A

COI frag. 2 244 G T

COI frag. 2 278 C T

COI frag. 2 285 C T

COI frag. 2 302 T C

cytb frag. 1 71 C T

cytb frag. 2 26 A C

cytb frag. 2 41 T C





cytb frag. 3 11 A T

H1 frag. 2 60 C T

H1 frag. 2 70 T C

H1 frag. 2 196 T C

H1 frag. 2 449 C T

H1 frag. 2 506 G A

H1 frag. 3 17 A T

H1 frag. 3 123 T A

H1 frag. 3 201 T C

H1 frag. 3 288 C A

H1 frag. 4 157 T C

H1 frag. 5 73 T C

H1 frag. 6 47 A C

H1 frag. 6 50 T C

H1 frag. 6 128 T —

H1 frag. 6 148 C A

H1 frag. 6 352 T A

H1 frag. 6 365 — C

H1 frag. 6 390 — C

H1 frag. 6 391 — C

H1 frag. 6 417 — A

H1 frag. 6 446 C —

H1 frag. 6 530 T C

H1 frag. 7 14 T C



H1 frag. 7 30 C A

H1 frag. 8 11 T C

H1 frag. 8 41 C T

H1 frag. 8 66 T A

H1 frag. 8 86 T C

H1 frag. 8 124 A G

H1 frag. 8 422 C A

H1 frag. 8 601 A G

H1 frag. 9 151 C T

H1 frag. 9 174 T C

H1 frag. 9 176 C T

H1 frag. 9 238 T —

H1 frag. 9 240 T C

H1 frag. 9 299 T C

H1 frag. 10 2 C T

H1 frag. 10 15 A T

H1 frag. 10 48 A T

H1 frag. 10 130 A T

H1 frag. 10 138 A C

H1 frag. 11 11 — A

H1 frag. 11 12 — T

H1 frag. 12 34 T C

H1 frag. 12 200 — A

H1 frag. 13 16 A C

H1 frag. 13 48 A C

H1 frag. 13 71 A G

H1 frag. 13 138 T C

H1 frag. 13 225 A T

H1 frag. 13 283 C T


RAG1 frag. 2 94 A G


RAG1 frag. 2 214 A T

rhodopsin 85 A C

rhodopsin 106 G A

rhodopsin 291 A G

SIA frag. 1 12 T A

SIA frag. 1 36 T C

SIA frag. 1 69 A G

SIA frag. 1 123 A G

SIA frag. 1 171 T C

SIA frag. 2 93 T C

SIA frag. 2 108 T C

SIA frag. 2 132 A C

SIA frag. 2 135 A C

SIA frag. 2 204 T G

Minyobates/M. steyermarki

H1 frag. 12 51 A G

H1 frag. 12 109 A T

H1 frag. 12 110 A G

H1 frag. 12 184 T C

H1 frag. 12 197 — T

H1 frag. 12 249 C T

H1 frag. 12 307 — T

H1 frag. 12 311 C T

H1 frag. 12 373 A G

H1 frag. 12 405 — C


H1 frag. 12 406 — C

H1 frag. 12 408 T C

H1 frag. 12 425 T —

H1 frag. 12 447 T G

H1 frag. 12 448 T C

H1 frag. 13 22 T C

H1 frag. 13 36 T C

H1 frag. 13 46 — C

H1 frag. 13 71 A G

H1 frag. 13 111 — A

H1 frag. 13 149 — T

H1 frag. 13 166 C T

H1 frag. 13 172 T —

H1 frag. 13 181 G T

H1 frag. 13 201 T C

H1 frag. 15 6 T C

H1 frag. 15 16 C T

H1 frag. 15 29 A G

Oophaga

28S frag. 1 181 C T

28S frag. 1 403 T A

COI frag. 1 53 C T

COI frag. 1 68 T C

COI frag. 1 103 C T

COI frag. 1 191 G A

COI frag. 1 194 A C

COI frag. 1 216 T C

COI frag. 1 234 A T

COI frag. 1 336 A G

COI frag. 2 33 T C

COI frag. 2 86 C T

COI frag. 2 89 A C

COI frag. 2 116 T C

COI frag. 2 148 A G

COI frag. 2 154 C T

COI frag. 2 197 T C

COI frag. 2 262 T C

COI frag. 2 269 C A

COI frag. 2 270 C T

COI frag. 2 284 C T

COI frag. 2 287 A T

COI frag. 2 290 C T

COI frag. 2 318 T A

cytb frag. 1 55 T G

cytb frag. 1 68 G T

cytb frag. 2 1 A T

cytb frag. 2 10 C T

cytb frag. 2 25 T C

cytb frag. 2 31 A T

cytb frag. 2 42 C T

cytb frag. 2 65 C T

cytb frag. 2 87 G A




cytb frag. 3 8 T C

cytb frag. 3 23 T G


H1 frag. 1 36 C A

H1 frag. 1 47 A T

H1 frag. 2 95 A T

H1 frag. 2 102 A C

H1 frag. 2 120 — C

H1 frag. 2 152 G A

H1 frag. 2 174 T C

H1 frag. 2 217 T A

H1 frag. 2 361 — T

H1 frag. 2 366 — T

H1 frag. 2 397 C T

H1 frag. 2 398 T A

H1 frag. 2 402 T A

H1 frag. 2 409 T A

H1 frag. 2 500 A T

H1 frag. 2 508 C T

H1 frag. 2 511 T A

H1 frag. 2 541 T C

H1 frag. 2 551 C A

H1 frag. 2 580 A G

H1 frag. 2 606 C T

H1 frag. 2 659 — T

H1 frag. 2 690 C A

H1 frag. 2 707 C —

H1 frag. 3 81 T C

H1 frag. 3 162 T C

H1 frag. 3 166 A G

H1 frag. 3 200 T C

H1 frag. 3 209 T A

H1 frag. 3 249 A T

H1 frag. 4 130 C A

H1 frag. 4 154 T A

H1 frag. 4 202 T C

H1 frag. 6 27 C T

H1 frag. 6 315 C T

H1 frag. 6 336 C T

H1 frag. 6 445 T G

H1 frag. 6 455 G A

H1 frag. 6 474 A G

H1 frag. 6 496 G A

H1 frag. 6 565 G A

H1 frag. 6 570 T C

H1 frag. 6 596 G A

H1 frag. 7 37 T A

H1 frag. 8 22 C T

H1 frag. 8 52 G A

H1 frag. 8 102 G A

H1 frag. 8 229 C T

H1 frag. 8 241 G A

H1 frag. 8 270 T A

H1 frag. 8 289 T A

H1 frag. 8 303 A T

H1 frag. 8 319 C T

H1 frag. 8 463 T C

H1 frag. 9 30 A C

H1 frag. 9 120 T A

H1 frag. 9 183 T C


H1 frag. 9 224 C A

H1 frag. 9 258 A T

H1 frag. 9 302 T A

H1 frag. 9 336 C T

H1 frag. 9 407 C T

H1 frag. 10 29 T A

H1 frag. 10 76 A T

H1 frag. 10 262 A C

H1 frag. 12 109 A C

H1 frag. 12 173 T C

H1 frag. 12 311 C T

H1 frag. 12 427 T C

H1 frag. 13 124 T C

H1 frag. 13 128 — A

H1 frag. 13 159 A T

H1 frag. 13 227 T C



rhodopsin 12 A T

rhodopsin 215 T C

rhodopsin 263 C T

SIA frag. 1 57 C G

SIA frag. 1 85 A C

SIA frag. 1 180 A G

SIA frag. 2 60 A C

SIA frag. 2 69 C G

SIA frag. 2 111 T C

SIA frag. 2 183 T C

Phyllobates

28S frag. 1 197 — C

28S frag. 1 198 — C

28S frag. 1 203 — G

28S frag. 1 204 — T

28S frag. 1 260 — T

28S frag. 1 351 — G

28S frag. 1 352 — G

28S frag. 1 353 — G

COI frag. 1 21 T C

COI frag. 1 25 G C

COI frag. 1 56 C T

COI frag. 1 94 G C

COI frag. 1 114 C T

COI frag. 1 144 T C

COI frag. 1 184 A T

COI frag. 1 197 T C

COI frag. 1 267 C T

COI frag. 1 288 A T

COI frag. 2 1 C T

COI frag. 2 10 A T

COI frag. 2 27 G A

COI frag. 2 47 C A

COI frag. 2 68 C T

COI frag. 2 83 A G

COI frag. 2 86 C T

COI frag. 2 157 T A

COI frag. 2 180 T C

COI frag. 2 188 T G



COI frag. 2 210 G A

COI frag. 2 223 T C

COI frag. 2 232 T C

COI frag. 2 269 C A

COI frag. 2 285 C T

COI frag. 2 293 C A

COI frag. 2 346 C T

cytb frag. 2 44 A C

cytb frag. 2 69 T A





cytb frag. 3 26 A T

H3 60 G A

H1 frag. 1 63 C A

H1 frag. 2 3 C T

H1 frag. 2 13 A —

H1 frag. 2 119 — A

H1 frag. 2 161 A T

H1 frag. 2 233 — T

H1 frag. 2 368 — T

H1 frag. 2 374 C T

H1 frag. 2 423 T A

H1 frag. 2 432 C A

H1 frag. 2 481 T —

H1 frag. 2 492 C T

H1 frag. 2 496 C T

H1 frag. 2 507 C A

H1 frag. 2 547 A T

H1 frag. 2 563 T C

H1 frag. 2 572 A T

H1 frag. 2 682 C T

H1 frag. 2 707 C T

H1 frag. 3 201 T A

H1 frag. 3 279 A T

H1 frag. 3 305 T —

H1 frag. 3 306 A —

H1 frag. 4 130 C A

H1 frag. 4 181 A C

H1 frag. 4 191 T A

H1 frag. 4 202 T A

H1 frag. 5 20 G A

H1 frag. 5 97 C T

H1 frag. 6 7 G A

H1 frag. 6 11 A G

H1 frag. 6 27 C A

H1 frag. 6 141 C T

H1 frag. 6 225 C A

H1 frag. 6 328 T C

H1 frag. 6 359 C T


H1 frag. 6 361 T A

H1 frag. 6 408 A G

H1 frag. 6 416 T C

H1 frag. 6 457 A C

H1 frag. 6 462 A C

H1 frag. 6 469 A G

H1 frag. 6 558 A C

H1 frag. 6 568 C T

H1 frag. 7 30 C T

H1 frag. 8 42 A T

H1 frag. 8 58 G A

H1 frag. 8 73 G T

H1 frag. 8 81 A C

H1 frag. 8 93 A G

H1 frag. 8 142 — C

H1 frag. 8 146 C T

H1 frag. 8 253 T A

H1 frag. 8 344 A —

H1 frag. 8 414 A C

H1 frag. 8 524 — A

H1 frag. 8 555 T A

H1 frag. 9 9 A G

H1 frag. 9 38 A G

H1 frag. 9 54 T C

H1 frag. 9 62 C A

H1 frag. 9 75 C T

H1 frag. 9 134 A T

H1 frag. 9 149 C T

H1 frag. 9 177 — A

H1 frag. 9 183 T A

H1 frag. 9 250 T A

H1 frag. 9 260 A —

H1 frag. 9 265 C A

H1 frag. 9 283 C A

H1 frag. 9 367 — C

H1 frag. 9 371 T C

H1 frag. 9 386 A T

H1 frag. 9 421 T C

H1 frag. 10 3 C A

H1 frag. 10 95 A T

H1 frag. 10 108 T C

H1 frag. 10 124 T C

H1 frag. 10 249 T C

H1 frag. 10 262 A C

H1 frag. 12 110 A C

H1 frag. 12 181 C T

H1 frag. 12 327 T C

H1 frag. 12 443 — T

H1 frag. 13 20 A C

H1 frag. 13 65 — T

H1 frag. 13 112 T —


H1 frag. 13 195 T C

H1 frag. 13 204 T A

H1 frag. 13 223 A T

H1 frag. 13 284 T C

RAG1 frag. 1 43 G A


RAG1 frag. 2 85 C T


rhodopsin 9 C T

rhodopsin 60 T C

rhodopsin 93 C T

rhodopsin 173 G A

rhodopsin 203 T C

SIA frag. 2 105 A C

Ranitomeya

COI frag. 1 28 A C

COI frag. 1 59 C T

COI frag. 1 105 A T

COI frag. 1 129 A C

COI frag. 1 181 C T

COI frag. 1 200 A T

COI frag. 2 163 A C

cytb frag. 1 15 A T

cytb frag. 2 56 T C


cytb frag. 2 133 A G



H1 frag. 2 16 C —

H1 frag. 2 232 C A

H1 frag. 2 254 C T

H1 frag. 2 290 C T

H1 frag. 2 374 C A

H1 frag. 2 590 G A

H1 frag. 2 604 C T

H1 frag. 2 697 T C

H1 frag. 3 15 T C

H1 frag. 3 201 T C

H1 frag. 3 308 T C

H1 frag. 4 181 A T

H1 frag. 4 207 A T

H1 frag. 5 4 T C

H1 frag. 6 16 A C

H1 frag. 6 45 A C

H1 frag. 6 147 C A

H1 frag. 6 370 A C

H1 frag. 6 371 C T

H1 frag. 6 474 A T

H1 frag. 6 521 T A

H1 frag. 6 544 T A

H1 frag. 6 554 A C


H1 frag. 8 9 C T

H1 frag. 8 22 C T

H1 frag. 8 36 C T

H1 frag. 8 49 G A

H1 frag. 8 86 T C

H1 frag. 8 101 A C

H1 frag. 8 180 A C

H1 frag. 8 229 C —

H1 frag. 8 328 T A

H1 frag. 8 344 A C

H1 frag. 8 588 — C

H1 frag. 9 96 T A

H1 frag. 9 115 T C

H1 frag. 9 152 — T

H1 frag. 9 165 A C

H1 frag. 9 183 T C

H1 frag. 9 191 T C

H1 frag. 9 193 C A

H1 frag. 10 1 T C

H1 frag. 10 17 A T

H1 frag. 10 25 A T

H1 frag. 10 31 A T

H1 frag. 10 76 A C

H1 frag. 10 91 A C

H1 frag. 10 99 C —

H1 frag. 10 117 — G

H1 frag. 10 200 T C

H1 frag. 10 260 — C

H1 frag. 12 38 T C

H1 frag. 12 39 T C

H1 frag. 12 46 A G

H1 frag. 12 82 A G

H1 frag. 12 235 A T

H1 frag. 12 311 C —

H1 frag. 12 415 T A

H1 frag. 13 74 — G

H1 frag. 13 114 — C

H1 frag. 13 115 — C

H1 frag. 13 116 — A

H1 frag. 13 152 T A

H1 frag. 13 277 C T

RAG1 frag. 1 198 T G

SIA frag. 1 3 T C

SIA frag. 1 12 T G

SIA frag. 1 147 C T

SIA frag. 2 30 T C


phylogenetic systematics of dart poison frogs

Documents