From Amazonia to the Atlantic Forest: Molecular Phylogeny Of

Home , Adelophryne, Adelophryne gutturosa, Phyzelaphryne

Molecular Phylogenetics and Evolution 65 (2012) 547–561

Contents lists available at SciVerse ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

From Amazonia to the Atlantic forest: Molecular phylogeny of Phyzelaphryninae frogs reveals unexpected diversity and a striking biogeographic pattern emphasizing conservation challenges ⇑ Antoine Fouquet a,b, , Daniel Loebmann c, Santiago Castroviejo-Fisher d, José M. Padial d, Victor G.D. Orrico e, Mariana L. Lyra e, Igor Joventino Roberto f, Philippe J.R. Kok g,h, Célio F.B. Haddad e, Miguel T. Rodrigues b a CNRS-Guyane – USR 3456, Immeuble Le Relais – 2, Avenue Gustave Charlery, 97300 Cayenne, French Guiana b Departamento de Zoologia, Universidade de São Paulo, Instituto de Biociências, Caixa Postal 11.461, CEP 05508-090 São Paulo, SP, Brazil c Laboratório de Vertebrados Terrestres, Universidade Federal do Rio Grande, Instituto de Ciências Biológicas, Av. Itália Km 8, Carreiros, CEP 96.203-900 Rio Grande, RS, Brazil d Department of Herpetology, American Museum of Natural History, Central Park West at 79th Street, New York, NY 10024-5192, United States e Departamento de Zoologia, Instituto de Biociências, Universidade Estadual Paulista Júlio de Mesquita Filho, Av. 24-A, 1515, Bela Vista, Caixa Postal 199, CEP 13506-900 Rio Claro, SP, Brazil f Departamento de Ciências Físicas e Biológicas, Laboratório de Zoologia, Universidade Regional do Cariri (URCA), Rua Cel. Antônio Luiz Pimenta, 1161, CEP 63105-000 Crato, Ceará, Brazil g Department of Vertebrates, Royal Belgian Institute of Natural Sciences, Rue Vautier 29, B-1000 Brussels, Belgium h Biology Department, Unit of Ecology and Systematics, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium article info abstract

Article history: Documenting the Neotropical amphibian diversity has become a major challenge facing the threat of glo- Received 12 May 2012 bal climate change and the pace of environmental alteration. Recent molecular phylogenetic studies have Revised 13 July 2012 revealed that the actual number of species in South American tropical forests is largely underestimated, Accepted 14 July 2012 but also that many lineages are millions of years old. The genera Phyzelaphryne (1 sp.) and Adelophryne (6 Available online 26 July 2012 spp.), which compose the subfamily Phyzelaphryninae, include poorly documented, secretive, and minute frogs with an unusual distribution pattern that encompasses the biotic disjunction between Amazo- Keywords: nia and the Atlantic forest. We generated >5.8 kb sequence data from six markers for all seven nominal Adelophryne species of the subfamily as well as for newly discovered populations in order to (1) test the monophyly of Amazonia Atlantic forest Phyzelaphryninae, Adelophryne and Phyzelaphryne, (2) estimate species diversity within the subfamily, Cryptic species and (3) investigate their historical biogeography and diversification. Phylogenetic reconstruction Neotropical diversity confirmed the monophyly of each group and revealed deep subdivisions within Adelophryne and Phyzel- Phyzelaphryne aphryne, with three major clades in Adelophryne located in northern Amazonia, northern Atlantic forest and southern Atlantic forest. Our results suggest that the actual number of species in Phyzelaphryninae is, at least, twice the currently recognized species diversity, with almost every geographically isolated population representing an anciently divergent candidate species. Such results highlight the challenges for conservation, especially in the northern Atlantic forest where it is still degraded at a fast pace. Molec- ular dating revealed that Phyzelaphryninae originated in Amazonia and dispersed during early Miocene to the Atlantic forest. The two Atlantic forest clades of Adelophryne started to diversify some 7 Ma minimum, while the northern Amazonian Adelophryne diversified much earlier, some 13 Ma minimum. This striking biogeographic pattern coincides with major events that have shaped the face of the South American continent, as we know it today. Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction clock for many biologists before this invaluable heritage vanishes. This is particularly critical in the tropics, which host the bulk of the Life is facing its 6th mass extinction (Barnosky et al., 2011), and diversity on Earth (Gaston and Williams, 1996) but still remain the description of the world’s biodiversity is a race against the largely under-documented regarding the actual magnitude of their biological diversity and the mechanisms responsible for its origin (Balakrishnan, 2005). Tropical forests of South America are crucial ⇑ Corresponding author at: CNRS-Guyane – USR 3456, Immeuble Le Relais – 2, Avenue Gustave Charlery, 97300 Cayenne, French Guiana. because they are believed to host more species than anywhere else E-mail address: [email protected] (A. Fouquet). in the world (Gaston and Williams, 1996; Myers et al., 2000;

Primack and Corlett, 2005; Wilson, 1992). Despite being flagged Almeida et al., 2011). Such puzzling distribution led Hoogmoed among the famous ‘‘biodiversity hotspots’’ as a priority area for and Lescure (1984) to question the homogeneity of Phyzelaphryni- conservation some 15 years ago (Mittermeier et al., 1998), the nae, which has been later hypothesized to consist of remnants of a Atlantic forest of Brazil is still being degraded at a fast/steady pace, once more diverse and broadly distributed clade (Gonzalez-Voyer particularly in its northern part (Ribeiro et al., 2009). Even the et al., 2011). These frogs have been poorly represented in recent ef- immense Amazonia, with a large part of its surface remaining rel- forts to document phylogenetic relationships among Terrarana, atively intact, deeply suffers from human activities and is reduced and anurans in general (Hedges et al., 2008; Heinicke et al., in size at an extremely fast pace (Malhi et al., 2008). 2009); only one Adelophryne and one Phyzelaphryne samples have The biodiversity hosted by both areas is still so poorly docu- been so far included (Hedges et al., 2008; Heinicke et al., 2009). mented that we do not know what is lost with each exploited hect- The obvious reason for this is that it is challenging to gather a are of forest (Tuomisto et al., 1995; da Silva et al., 2005; Carnaval meaningful sampling because these frogs are very small et al., 2009). Moreover, in the last decades many studies have re- (SVL = 11 mm in A. pachydactyla to a maximum of 23 mm in A. pat- vealed a large underestimation of species number actually occur- amona), secretive (some species are locally very common, like A. ring in these regions (Giam et al., 2012). This has been gutturosa, but hard to find because of their microhabitat [Kok particularly striking for amphibians, with many species now recog- and Kalamandeen, 2008]) and with very restricted distributions nized as localized endemics or even isolated micro-endemics occur- (i.e. a few localized patches over an entire continent). These di- ring in small patches of these forests (e.g. Fouquet et al., 2012a, b; rect-developing frogs (Cassiano-Lima et al., 2011; MacCulloch Funk et al., 2012). Given that more than one third of the amphibian et al., 2008) are found exclusively in or under the forest litter (A. species are currently threatened by extinction – thus more than in maranguapensis breeds in bromeliads [Cassiano-Lima et al., any other vertebrate group (Stuart et al., 2004, 2008) – assessing the 2011]) from lowlands to mountain forests up to 1400 m a.s.l. Their actual biodiversity represented by South American amphibians is natural history is extremely poorly documented. Cassiano-Lima becoming a major challenge (Wake and Vredenburg, 2008). et al. (2011) recently provided some information on the reproduc- Estimating the South American biodiversity is not only a matter tion and development of A. maranguapensis, and MacCulloch et al. of counting species, but also of accounting for the depth of the rela- (2008) as well as Kok and Kalamandeen (2008) reported the obser- tionships among species, the so-called ‘‘phylogenetic diversity’’ vation of A. gutturosa laying a single very large egg. (Faith, 1992; Crozier, 1997; Purvis and Hector, 2000). Recent stud- Adelophryne and Phyzelaphryne (Phyzelaphryninae) can be dis- ies have revealed that many South American lineages are in fact tinguished from other Terrarana by the shape of terminal digits. millions of years old. Even among closely related South American However, the morphological distinction between these two genera amphibian species (e.g. Grant et al., 2006; Heinicke et al., 2007; is somewhat ambiguous (Hoogmoed and Lescure, 1984) and the Fouquet et al., 2012a) or within species (e.g. Fouquet et al., 2007, monophyly of Adelophryne has never been formally tested. Molec- 2012b; Funk et al., 2012) the prevalence of deep divergence re- ular analyses included only one Adelophryne and one Phyzelaph- vealed by molecular phylogenetics and phylogeography has been ryne, and these samples formed unambiguously a natural group – astonishing. The extent of unrecognized species that are geograph- i.e. Phyzelaphryninae (Hedges et al., 2008). However, considering ically restricted and could represent millions of years of indepen- the reduced number of terminals included and the frequent mor- dent evolution, and whether these undetected species are phological conservatism or parallel evolution in general morphol- threatened in the Atlantic forest and Amazonia are questions that ogy observed in other groups of Terrarana distributed in similar still cannot be answered. environments (e.g. Psychrophrynella, Phrynopus)(Hedges et al., A large proportion of South American frogs are terraranans, i.e. 2008; Gonzalez-Voyer et al., 2011), it would not be surprising to the New World direct-developing frogs, with more than 900 spe- find within Phyzelaphryninae relationships that contradict the cur- cies (Heinicke et al., 2007, 2009; Hedges et al., 2008). For example, rent taxonomy. the genus Pristimantis holds 400 nominal species and represents For example, some specimens from Colombia were allocated to the most species-rich genus among terrestrial vertebrates (Hei- the genus Phyzelaphryne in their original description (Heyer, 1977), nicke et al., 2007; Hedges et al., 2008), whereas other genera of and the advertisement call was described based on the Colombian Terrarana have very few and are restricted to very small areas material, but Hoogmoed and Lescure (1984) demonstrated that the despite being of similar age (Gonzalez-Voyer et al., 2011). Within call of P. miriamae described by Heyer (1977) in fact pertains to Terrarana, Eleutherodactylidae provides another striking example Adelophryne adiastola. Conversely, Lynch (2005) identified as A. adi- of such unbalance. This family holds more than 200 species, the astola specimens from Leticia that might actually belong to the genus Eleutherodactylus (including subgenera Syrrophus, Euhyas, genus Phyzelaphryne (see below). Moreover, the biogeographic pat- Peloruis, Schwartzius) holding more than 190 species and the tern observed within Adelophryne – i.e. occurring in northern remaining 16 species belonging to three other much smaller gen- Amazonia and the Atlantic forest, is mystifying given that the era: Diasporus (n = 9), Adelophryne (n = 6) and Phyzelaphryne (n = 1). Atlantic forest is separated from Amazonia by a northeast–south- The geographical distribution of these four genera provides a west belt of open or dry formations (Prado and Gibbs, 1993; da Sil- striking pattern given that they are all allopatric from southern va et al., 2004), which currently acts, and has acted in the past, as a USA to the Atlantic forest in Brazil. Eleutherodactylus has diversified barrier to biotic exchanges between these two forest blocks (Costa, in the Caribbean and the southern part of North America, while its 2003; Mori et al., 1981). Many ancient clades are endemic to one or sister group, Diasporus, occurs in Central America and Chocó the other of these regions, having very few species in common (Colombia). The two other genera are also found allopatrically (Duellman, 1999). The patterns provided by new insights in the and display an intriguing distribution. Phyzelaphryne is a mono- understanding of evolutionary relationships in many groups like typic genus previously known from only a few localities south of Dendrophryniscus/Amazophrynella (Fouquet et al., 2012b), Allobates the Amazon River (Heyer, 1977; Heyer and Gascon, 1995; De la (Santos et al., 2009), Leposoma (Pellegrino et al., 2011), and Vitreor- Riva et al., 2000), and Adelophryne contains six nominal species ana (Guayasamin et al., 2008, 2009) suggest that small vertebrates scattered in the northern periphery of Amazonia and in isolated distributed in Amazonia and the Atlantic forest, i.e. having frag- patches of the remnants of the northern part of the Atlantic forest mented range restricted to forest habitat, could display tens of mil- of Brazil. Most Adelophryne and Phyzelaphryne species are known lions of years of divergence. Moreover, highly conservative or only from their type locality and very few additional scattered parallel morphological evolution in small, dull-colored, terrestrial, populations (Loebmann et al., 2011; Ortega-Andrade, 2009; leaf litter-associated amphibian species has been repeatedly A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561 549 highlighted as a source of cryptic diversity (Fouquet et al., 2007; Nasikabatrachus/Sooglossus for POMC). A 345 bp portion of the Cytb Vieites et al., 2009). fragment remained missing for six Terrarana terminals and Gas- Therefore, given the conservation importance of amphibians in trotheca; and a portion of RAG-1 for seven Terrarana lineages. the region, and the numerous gaps in our understanding of the diversity and the origin of Phyzelaphryninae, we propose to (1) test 2.2. Data analyses the monophyly of Phyzelaphryninae within Terrarana and the monophyly of Adelophryne and Phyzelaphryne, (2) estimate species 2.2.1. Alignment diversity within these genera and (3) investigate their biogeogra- Most data consisted of coding regions, and alignment was phy and evolutionary history. unambiguous. We observed the insertion/deletion of one codon in RAG-1 fragment for Hyloidea/outgroup and several codon insertion/deletions in POMC, but none of them led to ambiguous align- 2. Materials and methods ment after checking the reading frame. For the 12S–16S fragment we performed alignment with MAFFT v6 (Katoh et al., 2002) under 2.1. Sampling default parameters except for the use of the L-INS-i strategy, which is adapted to sequences with one conserved domain and long gaps. Tissue samples of all nominal species within Phyzelaphryninae We obtained a final 5841 bp alignment. We used Bayesian analysis were taken from thigh muscle or liver (in one case from eggs) and and Maximum Parsimony to investigate phylogenetic relationships preserved in 95% ethanol (Table 1). Specimens were collected from among terminals. their type locality or from the closest localities possible, and are deposited in different collections (Table 1). We also collected tissue 2.2.2. Bayesian analysis samples from specimens of newly discovered populations that We divided the dataset into seven partitions: one for each co- were tentatively identified as belonging to one of the nominal don position of the mtDNA (1388/3 bp) and the nuDNA coding species within Phyzelaphryninae. Sequences of some specimens genes (2515 bp/3) and one for the 12S/16S fragment (1938 bp). of Phyzelaphryne miriamae and Adelophryne gutturosa were re- The choice of this partitioning was driven by the coding nature trieved from GenBank. Genomic DNA was extracted using Promega of mtDNA (Cytb, COI) and nuDNA (RAG1, POMC, TYR) loci and com- DNA extraction kit. A total of 31 Adelophryne and 14 Phyzelaphryne parable rates of evolution (Mueller, 2006; Hoegg et al., 2004; Fou- individuals were included. We follow the classification of Pyron quet et al., 2012a; see results). Many studies, indeed, reported that and Wiens (2011) for Terrarana. partitioning by both gene and codon position gave the best fit to We targeted three mitochondrial (Cytb; COI; 12S–16S) and the data (Caterino et al., 2001; Brandley et al., 2005). A more inclu- three nuclear loci (RAG1; POMC; TYR) that were already partly sive partitioning would have joined very different patterns of available for main Terrarana (n = 14), Hyloidea lineages (n = 14) molecular evolution, and greater partitioning would likely cause and five outgroups (Table 1), which were collated together with overparameterization (Marshall, 2010; Sullivan and Joyce, 2005). Phyzelaphryninae for a total of 80 terminals. Data not presented We used the software jModeltest version 0.1.1 (Posada, 2008; here for A. baturitensis from Serra de Baturité (type locality) and Guindon and Gascuel, 2003) to select the substitution model that Serra de Maranguape show that these populations are very similar best fits each of these partitions under Akaike’s Information Crite- to the ones included herein from Serra da Ibiapaba. rion (Akaike, 1974). The seven resulting models (Suppl. Mat.) were In order to reduce missing data for the other Terrarana and other employed in a Bayesian analysis with MrBayes 3.2 (Huelsenbeck Hyloidea terminals, we concatenated sequences from different spe- and Ronquist, 2001; Ronquist and Huelsenbeck, 2003). The Bayes- cies or even genera (Pelodryadinae, Centrolenidae, Sooglossus/ ian analysis consisted of two independent runs of 2.0 Â 107 gener- Nasikabatrachus) when monophyly of the group involved was ations, starting with random trees and 10 Markov chains (one unambiguous. The only early-diverging lineage within Terrarana cold), sampled every 1000 generations. We also performed sepa- that was not represented is Ceuthomantidae because we consid- rate runs for mtDNA and nuDNA using the same partitions and ered the available data too limited for nuclear DNA (missing POMC, models and 2.0 Â 107 generations for each run (Suppl. Mat.). Ade- TYR and most RAG-1) to be included in our analyses; the position of quate burn-in was determined by examining likelihood scores of Ceuthomantidae is also well supported (Heinicke et al., 2009). We the heated chains for convergence on stationarity, as well as the also completed the matrix directly from biological material (Table effective sample size of values in Tracer 1.5 (Rambaut and Drum- 1) and therefore produced sequence data for two loci or more for mond, 2007). We discarded 10% of the generations/trees. We con- most Terrarana terminals (Holoaden, Brachycephalus, Oreobates, sidered relationships strongly supported when posterior Haddadus, Barycholos and Eleutherodactylus) up to all the loci for probabilities were equal to or higher than 0.95. Euparkerella, for which no sequences were previously available. Fragments were amplified by standard PCR techniques; detailed 2.2.3. Maximum parsimony information about the primers is available in Table 2. Sequencing Of the 5841 total characters of the matrix, 2285 are constant, was performed using ABI Big Dye V3.1 (ABI, Foster City, USA) and 538 variable characters are parsimony-uninformative and 3018 resolved on an automated sequencer at IQUSP and Genomic are parsimony-informative using gaps as a fifth character state. Engenharia corp. (São Paulo, Brazil) and Macrogen Inc. (Korea). Se- The mtDNA partition totaled 3326 characters (1060 are constant, quences were edited and aligned with CodonCode Aligner v.3.5.2. 1974 parsimony informative) and the nuDNA partition totaled Novel sequences were deposited in GenBank (Table 1). 2515 characters (1225 are constant, 1044 parsimony informative). We generated 278 new sequences of terraranans (Table 1). We employed PAUP 4.0b10 (Swofford, 2002) to search for the Within Phyzelaphryninae some terminals harbor substantial miss- shortest tree with the heuristic search option, tree bisection– ing data. Nevertheless, preliminary analyses suggested that these reconnection (TBR) for branch swapping on 100 random-addition were not impeding resolution given that these missing data were sequence replicates. We subsequently computed 500 nonparamet- evenly distributed among main lineages (Lemmon et al., 2009; ric bootstrap pseudoreplicates (Efron, 1979; Felsenstein, 1985). We Wiens and Morrill, 2011; Wiens, 1998, 2003; Simmons, 2012). also performed separate runs for mtDNA and nuDNA using the For other terminals, missing data were limited to a maximum of same scheme (Suppl. Mat.). We considered relationships strongly two complete loci for Bryophryne (Cytb; RAG-1) and one complete supported when MP bootstrap percentages equaled or exceeded locus for three terminals (Hypodactylus and Phrynopus for COI; 70% (Hillis and Bull, 1993). We also ran a similar analysis treating Table 1 550 Sequence details including vouchers and accession numbers used for the Bayesian analysis.

Genus Species Cytb COI 12S 16S RAG1a RAG1b POMC TYR Locality Stat Lat. Long.

Adelophryne baturitensis CFBHT11100/ CFBHT11100/ CFBHT11100/ CFBHT11100/ CFBHT11100/JX298096 CFBHT11100/JX298197 Tiangua CE À3.709925 À40.934057 JX298317 JX298245 JX298277 JX298145 Adelophryne baturitensis CFBHT11101/ CFBHT11101/ CFBHT11101/ CFBHT11101/ CFBHT11101/JX298097 CFBHT11101/JX298198 Tiangua CE À3.709925 À40.934057 JX298318 JX298246 JX298278 JX298146 Adelophryne baturitensis CFBHT11110/ CFBHT11110/ CFBHT11110/ CFBHT11110/ CFBHT11110/JX298098 CFBHT11110/JX298199 Ibiapina CE À3.909126 À40.866494 JX298319 JX298247 JX298279 JX298147 Adelophryne baturitensis CFBHT11339/JX298377 CFBHT11339/ CFBHT11339/ CFBHT11339/ CFBHT11339/ CFBHT11339/ CFBHT11339/JX298101 CFBHT11339/JX298202 Ubajara CE À3.842332 À40.89323 JX298322 JX298250 JX298282 JX298150 JX298150 Adelophryne baturitensis MTR14012/JX298375 MTR14012/ MTR14012/JX298248 MTR14012/JX298280 MTR14012/ MTR14012/JX298148 MTR14012/JX298099 MTR14012/JX298200 Ibiapaba CE À5.078753 À40.933371 JX298320 JX298148 Adelophryne baturitensis MTR14013/JX298376 MTR14013/ MTR14013/JX298249 MTR14013/JX298281 MTR14013/ MTR14013/JX298100 MTR14013/JX298201 Ibiapaba CE À5.078753 À40.933371 JX298321 JX298149 Adelophryne sp. 2 PEU80/JX298379 PEU80/JX298323 PEU80/JX298283 PEU80/JX298151 PEU80/JX298103 PEU80/JX298204 Wenceslau BA À13.68852 À39.483175 Guimares Adelophryne sp. 3 MTR20222/JX298378 MTR20222/JX298102 MTR20222/JX298203 Rio Patipe, APA BA À13.32164 À39.016328 Guaibim CFBHT11716/ CFBHT11716/JX298104 CFBHT11716/JX298205 Caruaru PE À8.25498 À35.904408 Adelophryne sp. 1 CFBHT11716/JX298380 CFBHT11716/ CFBHT11716/

JX298284 547–561 (2012) 65 Evolution and Phylogenetics Molecular / al. et Fouquet A. JX298324 JX298251 Adelophryne maranguapensis CFBHT14103/ CFBHT14103/ CFBHT14103/ CFBHT14103/ CFBHT14103/JX298105 CFBHT14103/JX298206 Maranguape CE À3.89029 À38.712502 JX298325 JX298252 JX298285 JX298152 Adelophryne maranguapensis CFBHT14119/JX298381 CFBHT14119/ CFBHT14119/ CFBHT14119/ CFBHT14119/ CFBHT14119/ CFBHT14119/JX298106 CFBHT14119/JX298207 Maranguape CE À3.89029 À38.712502 JX298326 JX298253 JX298286 JX298153 JX298153 Adelophryne sp. 5 CFBHE234/JX298383 CFBHE234/ CFBHE234/JX298254 CFBHE234/JX298288 CFBHE234/ CFBHE234/JX298155 CFBHE234/JX298108 CFBHE234/JX298209 Mariana MG À20.3663 À43.444848 JX298328 JX298155 Adelophryne sp. 5 CFBHE235/ CFBHE235/JX298255 CFBHE235/JX298289 CFBHE235/JX298109 CFBHE235/JX298210 Mariana MG À20.3663 À43.444848 JX298329 Adelophryne sp. 5 MTR17521/JX298382 MTR17521/ MTR17521/JX298287 MTR17521/ MTR17521/JX298107 MTR17521/JX298208 PE Rio Doce, MG À19.70991 À42.729192 JX298327 JX298154 Marliéria Adelophryne sp. 5 MTR21918/ MTR21918/ MTR21918/JX298110 MTR21918/JX298211 Serra do Cipo MG À19.54879 À43.550606 JX298330 JX298156 Adelophryne sp. 4 MTR13570/JX298384 MTR13570/ MTR13570/JX298256 MTR13570/JX298290 MTR13570/ MTR13570/JX298111 MTR13570/JX298212 Faz. Nova BA À16.531111 À39.118056 JX298331 JX298157 Alegria, Trancoso Adelophryne sp. 6 MTR15919/JX298385 MTR15919/ MTR15919/JX298291 MTR15919/ MTR15919/JX298158 MTR15919/JX298112 MTR15919/JX298213 Serra Bonita, BA À15.3901 À39.5630 JX298332 JX298158 Camacan Adelophryne sp. 6 CFBH23672/JX298386 CFBH23672/ CFBH23672/ CFBH23672/ CFBH23672/ CFBH23672/ CFBH23672/JX298113 CFBH23672/JX298214 Una BA À15.2716 À39.069843 JX298333 JX298257 JX298292 JX298159 JX298159 Adelophryne pachydactyla MTR16244/JX298388 MTR16244/ MTR16244/JX298259 MTR16244/JX298294 MTR16244/ MTR16244/JX298161 MTR16244/JX298115 MTR16244/JX298216 Serra das BA À15.1833 À39.3452 JX298335 JX298161 Lontras, Arataca Adelophryne pachydactyla MTR5988/JX298387 MRT5988/ MTR5988/JX298258 MRT5988/JX298293 MRT5988/ MRT5988/JX298160 MRT5988/JX298114 MRT5988/JX298215 Serra do BA À15.210915 À39.480972 JX298334 JX298160 Teimoso, Jussari Adelophryne sp. 7 MTR13808/JX298389 MTR13808/ MTR13808/JX298295 MTR13808/ MTR13808/JX298162 MTR13808/JX298116 MTR13808/JX298217 Serra do Navio AP 0.912857 À52.007933 JX298336 JX298162 Adelophryne patamona PK1875/JX298164 PK1875/JX298118 PK1875/JX298219 Mount Gu 5.219169 À60.575209 Maringma Adelophryne patamona PK1969/JX298390 PK1969/JX298337 PK1969/JX298260 PK1969/JX298296 PK1969/JX298163 PK1969/JX298163 PK1969/JX298117 PK1969/JX298218 Mount Gu 5.219169 À60.575209 Maringma Adelophryne patamona ROM43035/ ROM43035/ ROM43035/ ROM43035/ ROM43035/JX298120 Mount Gu 5.089576 À59.827538 (Paratype) JX298339 JX298262 JX298298 JX298166 Wokomung Adelophryne patamona ROM43034/ ROM43034/ ROM43034/ ROM43034/ ROM43034/JX298119 ROM43034/JX298220 Mount Gu 5.089576 À59.827538 (Holotype) JX298338 JX298261 JX298297 JX298165 Wokomung ROM39578/ ROM39578/GQ345262 ROM39578/EU186772 Mount Gu 5.395223 À59.962406 Adelophryne patamona ROM39578/GQ345201 ROM39578/ ROM39578/ GQ345296 Ayanganna EU186679 EU186679 Adelophryne adiastola AJC2463JX298391 AJC2463/ AJC2463/JX298263 AJC2463/JX298299 AJC2463/ AJC2463/JX298167 AJC2463/JX298121 AJC2463/JX298221 Com. Puerto Col. 1.198056 À70.281389 JX298340 JX298167 Vaupes Adelophryne gutturosa PK1168/JX298393 PK1168/JX298342 PK1168/JX298266 PK1168/JX298302 PK1168/JX298169 PK1168/JX298123 PK1168/JX298223 Muri Muri Gu 5.27729 À59.432316 creek, KaieteurNP Adelophryne gutturosa PK2231/JX298392 PK2231/JX298341 PK2231/JX298264 PK2231/JX298300 PK2231/JX298168 PK2231/JX298122 PK2231/JX298222 La Escalera, Ven 6.014069 À61.449308 Bolivar state Adelophryne gutturosa PK1362/JX298170 PK1362/JX298124 PK1362/JX298224 Elinkwa creek, Gu 5.27729 À59.432316 KaieteurNP Adelophryne gutturosa ROM44051/ ROM44051/ Meamu River Gu 6.232029 À60.619926 JX298265 JX298301 Phyzelaphryne miriamae SMS629/JX298394 SMS629/JX298343 SMS629/JX298267 SMS629/JX298303 SMS629/JX298171 SMS629/JX298125 SMS629/JX298225 Com. São AM À3.78943 À59.034048 Sebastião dos Bargas Phyzelaphryne miriamae MTR19141/JX298397 MTR19141/ MTR19141/JX298307 MTR19141/ MTR19141/JX298175 MTR19141/JX298129 MTR19141/JX298229 Moio Bamba, AM À4.720095 À62.133036 JX298347 JX298175 Margem D Purus Phyzelaphryne miriamae MTR19437/JX298396 MTR19437/ MTR19437/JX298306 MTR19437/ MTR19437/JX298174 MTR19437/JX298128 MTR19437/JX298228 Moio Bamba, AM À4.720095 À62.133036 JX298346 JX298174 Margem D Purus Phyzelaphryne miriamae MTR12700/JX298395 MTR12700/ MTR12700/JX298268 MTR12700/JX298304 MTR12700/ MTR12700/JX298172 MTR12700/JX298126 MTR12700/JX298226 Igarapé Açu, AM À4.344167 À58.635 JX298344 JX298172 Rio Abacaxis Phyzelaphryne miriamae MTR12789/ MTR12789/JX298305 MTR12789/ MTR12789/JX298173 MTR12789/JX298127 MTR12789/JX298227 São Sebastião, AM À4.30889 À58.63639 JX298345 JX298173 Rio Abacaxis Phyzelaphryne miriamae LSUMZ16935/ LSUMZ16935/ LSUMZ16935/ 40 km S AM À3.61944 À60.44633 EU186689 EU186689 EU186774 Manaus Phyzelaphryne sp. 1a MTR19206/JX298399 MTR19206/ MTR19206/JX298270 MTR19206/JX298309 MTR19206/ MTR19206/JX298177 MTR19206/JX298131 MTR19206/JX298231 Terra AM À4.702148 À62.309074 JX298349 JX298177 Vermelha, Marg. E Purus Phyzelaphryne sp. 1a FS054/JX298398 FS054/JX298348 FS054/JX298269 FS054/JX298308 FS054/JX298176 FS054/JX298130 FS054/JX298230 RDS do Uacari, AM À5.762570 À67.89884 Com. Anaxiqui, margem E Rio Juruá Phyzelaphryne sp. 1b AndesA(JMP2283)/ AndesA(JMP2283)/ AndesA(JMP2283)/ AndesA(JMP2283)/ Senda ZaﬁreÀ Col. JX298350 JX298310 JX298132 JX298232 Takana al Norte de la carretera Leticia Phyzelaphryne sp. 1b AndesA832/ AndesA832/ AndesA832/ AndesA832/ AndesA832/JX298134 AndesA832/JX298235 Leticia, Col. À4.111667 À69.960833 JX298354 JX298274 JX298314 JX298181 Kilometro 13 AndesA915/ AndesA915/JX298133 AndesA915/JX298234 Leticia, km 9– Col. À4.124167 À69.941389 Phyzelaphryne sp. 1b AndesA915/ AndesA915/ AndesA915/ JX298180 10 carretera

JX298353 JX298273 JX298313 547–561 (2012) 65 Evolution and Phylogenetics Molecular / al. et Fouquet A. Phyzelaphryne sp. 1b AndesA(JMP2058)/ AndesA(JMP2058)/ AndesA(JMP2058)/ AndesA(JMP2058)/ AndesA(JMP2058)/ AndesA(JMP2058)/ AndesA(JMP2058)/ AndesA(JMP2058)/ Leticia, Col. À4.111667 À69.960833 JX298400 JX298355 JX298275 JX298315 JX298182 JX298182 JX298135 JX298236 Kilometro 13 Phyzelaphryne sp. 1b AndesA970/ AndesA970/ AndesA970/ AndesA970/ Leticia, Varzea Col. À04.11949 À69.95104 JX298351 JX298271 JX298311 JX298178 del arroyo Huallar ka ka junto a Tanimboca Phyzelaphryne sp. 1b AndesA971/ AndesA971/ AndesA971/ AndesA971/ AndesA971/JX298233 Leticia, Varzea Col. À04.11949 À69.95104 JX298352 JX298272 JX298312 JX298179 del arroyo Huallar ka ka junto a Tanimboca Euparkerella brasiliensis Eubra/JX298402 Eubra/JX298357 Eubra/JX298276 Eubra/JX298316 Eubra/JX298185 Eubra/JX298185 Eubra/JX298137 Eubra/JX298237 PN Floresta da RJ Tijuca Holoaden bradei/ AF1600/JX298403 AF1600/JX298358 MZUSP131872/ MZUSP131872/ AF1600/JX298186 AF1600/JX298186 AF1600/JX298138 USNM207945/ luedwadlti EU186728 EU186710 EU186779 Brachycephalus ephippium/ CFBH16828/HM216367 MCL98/JX298359 DMH#2/AY326008 DMH#2/AY326008 MCL98/JX298187 USNM207716/ USNM207716/ MCL98/JX298238 nodoterga GQ345290 GQ345256 MTR19385/JX298188 KU218150/AY819093 KU215462/EU186764 Oreobates quixensis/ MTR19385/JX298404 MTR19385/ KU215462/ KU215462/ KU218150/ DQ679273 cruralis JX298360 EU186666 EU186666 Haddadus binotatus AF1463/JX298405 AF1463/JX298361 USNM303077/ USNM303077/ AF1463/JX298189 USNM303077/ USNM303077/ CFBH5813/DQ282918 EF493361 EF493361 GQ345293 GQ345259 Barycholos ternetzi/pulcher MTR14750/JX298401 CFBHT23511/ KU217781/ CFBHT306/ CFBHT3227/ MTR14750/JX298184 MTR14750/JX298136 CFBHT306/DQ282921 JX298356 EU186727 DQ283094 JX298183 Eleutherodactylus portoricensis/ MSB: Herp: 77538/ CZACC: n.a./ USNM305421/ USNM305421/ 287MC/JX298190 287MC/JX298190 USNM326784HQ831999 USNM326784/ auriculatus/ HM229995 GQ357659 GQ345176 GQ345176 EF493455 coqui/ johnstonei/cooki Diasporus diastema MVZ203844/GQ345200 USNM572456/ MVZ203844/ MVZ203844/ MVZ203844/ MVZ203844/GQ345261 MVZ203844/EU186773 FJ766810 EU186682 EU186682 GQ345279 Strabomantis biporcatus/ CVULA7073/GQ345204 USNM572433/ CVULA7073/ CVULA7073/ CVULA7073/ KU179076/ CVULA7073/GQ345265 CVULA7073/EU186775 bufoniformis/ FJ766633 EU186691 EU186691 GQ345283 GQ345299 necerus Psychrophrynella usurpator/ KU17349/GQ345205 567/JX298362 KU183049/EU186696 KU183049/EU186696 KU183049/ KU183049/ KU183049/GQ345266 KU183049/EU186776 iatamasi/ GQ345284 GQ345300 wettsteini Craugastor crassidigitus/ UCR16900/EF629465 KRL0743/ DMH86-112/ DMH86-112/ MVZ12020/ MVZ12020GQ345292 MVZ12020/GQ345258 MVZFC13463/ ﬁtzingeri/ FJ766646 AY326001 AY326001 GQ345277 EF493481 podiciferus Ischnocnema guentheriaff./ ??/GQ345196 CFBHT11170/ ??/EF493533 ??/EF493533 ??/GQ345276 ??/GQ345291 ??/GQ345257 ??/EF493510 lactea JX298364 sp./gaigeae/ CFBHT5732/JX298406 KRL1196/ KU217871/EF493513 KU217871/EF493513 KU217869/ KU217869/ CFBHT5732/JX298140 356MC/JN691911 Pristimantis curtipes/ FJ766789 DQ679272 DQ679272 zeuctotylus Hypodactylus brunneus KU178258/GQ345203 KU178258/EF493357 KU178258/EF493357 KU178258/ KU178258/ KU178258/GQ345264 KU178258/EF493484 GQ345282 GQ345298 Phrynopus bracki USNM286919/ USNM286919/ USNM286919/ USNM286919/ USNM286919/ USNM286919EF493507 GQ345202 EF493709 EF493709 GQ345281 GQ345297 Bryophryne sp./cophites MNCN20992/ KU173497/EF493537 KU173497/EF493537 MNCN20992/JX298139 KU173497/EF493508 JX298363 Gastrotheca cornuta/cf. KRL799/AY843811 KRL1163/ KRL799/AY843591 KRL799/AY843591 AMNH107251/ MNK5286/AY844380 AMNH107251/ KRL799/AY844040 marsupiata FJ766705 DQ679280 DQ679314 Telmatobius sanborni/ MNCN43526/GU060614 AMNHA165110/ WED53381/ WED53381/ KU212455/ AMNHA165114/ KU212455/AY819097 FB-2006/DQ347182 verrucosus/ DQ502743 AY326018 AY326018 DQ679271 AY844529

vellardi/truebae 551 Ceratophrys sp./cornuta/ AF2125/JX298407 AF2125/JX298365 WED55587AY326014 WED55587AY326014 KU215537/ KU215537/ KU215537/AY819091 VUB1006/DQ347168 (continued on next page) 552 Table 1 (continued)

Genus Species Cytb COI 12S 16S RAG1a RAG1b POMC TYR Locality Stat Lat. Long.

ornata DQ679269 DQ679269 Odontophrynus americanus AF665/FJ685666 AF665/JX298366 JF1891/AY843704 JF1891/AY843704 AF665/JX298191 JF1891/AY844480 AF665/JX298141 AF665/JX298239 Thoropa miliaris/ AF1434/FJ685662 CFBH3239/ CFBH3239/ CFBH3239/ AF1434/JX298192 USNM209318/ USNM209318/ AF1434/JX298241 taophora DQ502874 DQ283331 DQ283331 GQ345301 GQ345271 Rhinella margaritifera/ ROM40103/JX298409 ROM40103/ USNM268828/ USNM268828/ USNM268828/ MACN38639/ KU215143/AY819080 MRT6313/JN692075 arenarum JX298367 DQ158490 DQ158490 DQ158407 AY844370 Allobates femoralis/granti/ AfemSapoiv10a/ OMNH36070/ LSUMZ17436/ LSUMZ17436/ AF519/JX298193 UTAA56478/ KU220660/AY819088 OMNH36070/ trilineatus DQ523152 DQ502811 EU342537 EU342537 DQ503385 DQ503156 Centrolenidae uranoscopa/ MTR15819/JX298412 KRL0852/ CFBH5729/AY843595 CFBH5729/AY843595 MTR15819/ UCR17418/EU663519 MTR15819/JX298142 MNK5242/AY844029 colymbiphyllum/ FJ766714 JX298194 eurygnatha/ valerioi/ bejaranoi Paratelmatobius mantiqueira/ ITH0938/JX298413 ITH0938/ CFBH240/EU224408 CFBH240/EU224408 ITH0938/ ITH0938/JX298195 ITH0938/JX298143 ITH0938/JX298242 cardosoi JX298372 JX298195 Leptodactylus knudseni/ 396MC/JX298414 396MC/JX298373 FC13095/AY326017 FC13095/AY326017 1890T/JX298196 1890T/JX298196 396MC/JX298144 109MC/JX298243 pentadactylus/ myersi

Amazoprynella bokermanni/ MTR10040/JX298410 MTR10176/ QCAZ883/DQ158420 QCAZ883/DQ158420 QCAZ883/ MJH7095/DQ503337 KU221827/AY819081 3035T/JX298240 547–561 (2012) 65 Evolution and Phylogenetics Molecular / al. et Fouquet A. minutus/sp. JX298368 DQ158346 Hyla japonica/ IABHU6123/AB303949 IABHU6123/ LSUMZH-230/ LSUMZH-230/ ??/FJ227068 LSUMZH-230/ PB42-7/HM152465 LSUMZH-230/ / arenicolororeades/ CHUNB56875/ AB303949KRL0917/ AY843633WED55380/ AY843633WED55380/ KU221949/ AY844420MJH7076/AY844497 KU221949/AY819153AY844078 MJH7076/AY844157 PhyllomedusaAgalychnis callidryas/ GQ365966 FJ766570 AY326045 AY326045 EF174319 tomopterna Litoria/Cyclorana aurea/caerulea/ AM52744/ ??/AY835904 DMH/AY326038 DMH/AY326038 TWR1007/ SAMA17215/ AM52744/GQ366037 AM52744/AY844130 meiriana AY843937+manyaDLSN- EF174310 AY844475 72386/EF125030 Rana nigromaculata NC_002805 NC_002805 NC_002805 NC_002805 KUHE32995/ temporaria??/ KUHE32995/AB526647 FMNH232879/ AB526661 AY323776 DQ282932 Kaloula pulchra/ cplGEN/AY458595 cplGEN/AY458595 cplGEN/AY458595 cplGEN/AY458595 SIH-09/AY323772 SIH-09/AY323772 ZCMV11017/HM998968 AB611924 taprobanica Microhyla heymonsi/sp. cplGEN/AY458596 cplGEN/AY458596 cplGEN/AY458596 cplGEN/AY458596 MVZ236751/ MVZ236751/ MNCN-DNA28462/ MVZ236751/EF395979 EF396095 EF396095 HM998967 Calyptocephallela geayi JN3/JX298415 JN3/JX298374 AMNHA168414/ AMNHA168414/ MNCN8002/ MNCN8002/ ??/AY819090 JN3/JX298244 DQ283439 DQ283439 AY583337 AY583337 Nasikabatrachus/ thomasseti/ ?/AY341742 ??/GU136124 UMMZ(#15)/ RAN25162/ MNHN2003.3412/ MNHN2003.3412/ UMMZ(#15)/ Sooglossus sahyadrensis/ DQ28344 DQ283452 DQ872921 DQ872921 DQ283028 sechellensis Acronyms for newly added material: CFBHT = Celio F. B. Haddad Tissue collection; MTR, PEU, AF, FS, MCL, ITH = Miguel Trefaut Rodrigues field number; PK = Philippe Kok field numbers; SMS = Sergio Marques de Souza field numbers; ROM = Royal Ontario Museum; AJC = Andrew J. Crawford field numbers; T = François Catzeflis field numbers; JN = José J. Nuñes field numbers; LSUMZ = Lousiana State University Museum of Zoology; MNCN = Museo Nacional de Ciencias Naturales; AndesA = Universidad de los Andes; MC = Christian Marty field numbers; JMP = Jose M. Padial field numbers. A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561 553

Table 2 Primer details including primer name, sequences and authors.

Gene Primers Sequences Authors Cytb CYB-05L GCCAACGGCGCATCCTTCTTCTT Meyer (1993) Cytb LGL765 GAAAAACCAYCGTTGTWATTCAACT Bickham et al. (1995) Cytb CbR2 GTGAAGTTRTCYGGGTCYCC Fouquet et al. (2012) COI dgLCO1490 GGTCAACAAATCATAAAGAYATYGG Meyer (1993) COI dgHCO2198 TAAACTTCAGGGTGACCAAARAAYCA Meyer (1993) 12S t-Phe-frog ATAGCRCTGAARAYGCTRAGATG Wiens et al. (2005) 12S t-Val-frog TGTAAGCGARAGGCTTTKGTTAAGCT Wiens et al. (2005) 12S MVZ59 ATAGCACTGAAAAYGCTDAGATG Graybeal (1997) 12S tRNAval GGTGTAAGCGAGAGGCTT Goebel et al. (1999) 16S 16Sbr-H CCGGTCTGAACTCAGATCACGT Palumbi et al. (1991) 16S 16SC-16L GTRGGCCTAAAAGCAGCCAC Darst and Cannatella (2004) RAG1a MartFL1 AGCTGGAGYCARTAYCAYAARATG Hoegg et al. (2004) RAG1a Ad2R ATTGGCTCTCCATGTTTCATAG This paper RAG1a AMPF2 ACNGGNMGICARATCTTYCARCC Hoegg et al. (2004) RAG1a RAG1C GGAGATGTTAGTGAGAARCAYGG Biju and Bossuyt (2003) RAG1a Ad1R CTTCACGCACCAACTTTTCATC This paper RAG1b Amp F1 ACAGGATATGATGARAAGCTTGT Hoegg et al. (2004) RAG1b Mart R6 GTGTAGAGCCARTGRTGYTT Hoegg et al. (2004) POMC POMC1 GAATGTATYAAAGMMTGCAAGATGGWCCT Wiens et al. (2005) POMC POMC2 TAYTGRCCCTTYTTGTGGGCRTT Wiens et al. (2005) TYR TYR1E GAGAAGAAAGAWGCTGGGCTGAG Bossuyt and Milinkovitch (2000) TYR TYR1C GGCAGAGGAWCRTGCCAAGATGT Bossuyt and Milinkovitch (2000) TYR TYR1H ACACTTCTGGGCATCTCTCC Bossuyt and Milinkovitch (2000)

gaps as missing data to check whether it could have a signiﬁcant We eventually kept 38 terminals and 889 bp for RAG1, 41 inﬂuence. terminals and 544 bp for POMC and 39 terminals and 532 bp for TYR. Because some haplotype groups were not connected to each 2.2.4. Single nuDNA locus networks and mtDNA genetic distances other within the 95% limit of probability of parsimony, we at- In order to support our proposed species delineation we also tempted to connect them by increasing the connection threshold computed a statistical parsimony network for each nuDNA locus up to a maximum of 30 steps. using TCS 1.21 (Clement et al., 2000) with a 95% connection limit. Genetic distances (p distance) were also computed for mito- The original alignments used previously were reduced to Phyzel- chondrial loci using MEGA 5.1 (Tamura et al., 2011) and are shown aphryninae and were also trimmed in order to reduce missing data. in Table 3.

Table 3 p distances calculated among Phyzelaphryninae species using (a) Cytb; (b) 16S (the 463 bp ending).

(a) Cytb A. baturitensis A. sp. 3 0.222 A. sp. 2 0.231 0.202 A. sp. 1 0.242 0.236 0.250 A. maranguapensis 0.226 0.192 0.196 0.240 A. sp. 5 0.273 0.307 0.255 0.305 0.297 A. sp. 4 0.295 0.284 0.291 0.289 0.274 0.238 A. sp. 6 0.289 0.298 0.289 0.287 0.270 0.250 0.266 A. pachydactyla 0.273 0.286 0.281 0.296 0.283 0.244 0.229 0.227 A. sp. 7 0.256 0.275 0.272 0.283 0.261 0.263 0.260 0.263 0.270 A. patamona 0.278 0.253 0.267 0.269 0.253 0.285 0.258 0.277 0.265 0.226 A. adiastola 0.276 0.284 0.300 0.276 0.274 0.277 0.253 0.264 0.279 0.236 0.215 A. gutturosa 0.261 0.273 0.276 0.281 0.271 0.245 0.267 0.259 0.255 0.233 0.203 0.184 P. miriamae 0.295 0.282 0.307 0.283 0.285 0.308 0.281 0.265 0.285 0.245 0.281 0.243 0.258 P. sp. 1a 0.304 0.304 0.281 0.324 0.286 0.289 0.291 0.303 0.294 0.267 0.259 0.266 0.271 0.204 (b) 16S A. baturitensis A. sp. 2 0.137 A. sp. 1 0.118 0.144 A. maranguapensis 0.128 0.170 0.138 A. sp. 5 0.209 0.236 0.202 0.220 A. sp. 4 0.184 0.195 0.180 0.209 0.138 A. sp. 6 0.208 0.232 0.187 0.210 0.143 0.117 A. pachydactyla 0.206 0.227 0.197 0.203 0.162 0.115 0.123 A. sp. 7 0.195 0.219 0.192 0.212 0.171 0.132 0.156 0.161 A. patamona 0.173 0.204 0.181 0.195 0.172 0.155 0.159 0.158 0.141 A. adiastola 0.200 0.218 0.201 0.240 0.204 0.192 0.196 0.202 0.182 0.134 A. gutturosa 0.207 0.194 0.184 0.233 0.204 0.172 0.191 0.172 0.183 0.125 0.117 P. miriamae 0.197 0.234 0.210 0.239 0.231 0.205 0.217 0.204 0.204 0.184 0.199 0.184 P. sp. 1a 0.213 0.235 0.246 0.262 0.245 0.231 0.237 0.206 0.231 0.197 0.202 0.211 0.125 554 A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561

2.2.5. Molecular dating sister group of Eleutherodacylinae. Adelophryne is confirmed as To estimate timing of diversification within Phyzelaphryninae, being monophyletic and as the sister group of Phyzelaphryne. we undertook molecular dating with Beast 1.6.2 (Drummond and Moreover, Adelophryne is represented by three deeply divergent Rambaut, 2007). We used two approaches: (1) concatenated and well-sustained clades that are geographically circumscribed dataset method and (2) multilocus species tree method (ÃBEAST; to Northern Amazonia Clade (NAMC), Northern Atlantic Forest Heled and Drummond, 2010), both with a matrix focusing on Clade (NAFC; from Ceará to Bahia) and Southern Atlantic forest Eleutherodactylidae. Clade (SAFC; from Bahia to Minas Gerais) (Figs. 1a and 2). Each Preliminary analyses revealed close affinity among some termi- of these four major clades harbors deep subdivisions. nals for which sequence data were incomplete, and were thus dis- Terrarana is well supported as monophyletic in all methods. carded. In other cases some closely related terminals were Relationships among main Terrarana clades are mostly similar to complementary and were combined to represent only one termi- Hedges et al. (2008) and slightly different from Pyron and Wiens nal. With this strategy we were able to obtain an almost complete (2011). Relationships among Brachycephalidae, Eleutherodactyli- matrix representing all the main lineages in Phyzelaphryninae dae and Craugastoridae remain poorly resolved using the total except one from Valença (Bahia, Brazil; MTR20222). This lineage dataset. However, within Craugastoridae the interrelationships represents a candidate species (see results), but we preferred to among subfamilies are relatively different from those shown in discard this terminal because of missing data and because it is Pyron and Wiens (2011) with (1) Craugastorinae strongly sus- not needed to evaluate the broad temporal aspects of the diversifi- tained as the sister group to the other Craugastoridae, (2) Hypo- cation of the group. Nevertheless, the overall diversity within dactylus forming a strongly sustained clade with Pristimantinae Phyzelaphryninae is well represented. and Strabomantinae, (3) this last clade being the sister group of Both analyses were calibrated on the crown age of Eleuthero- Holoadeninae with Euparkerella as the sister group of Holoaden dactylinae (31.3 Ma; 44.7–21.1) estimated by Heinicke et al. and (4) Strabomantinae weakly sustained as nested within (2007, 2009) based on a large sequence dataset and fossil/biogeo- Pristimantinae. graphic calibrations. This prior was set as a normal distribution with mean and sd equal to the estimation from Heinicke et al. 3.2. Species diversity/candidate species (2009). Monophyly of Eleutherodacylinae and Phyzelaphryninae was enforced considering previous results. All partitions were By combining evidence from tentative identification of the considered underestimated uncorrelated lognormal rates. The tree specimens, references, phylogenetic position, time of divergence prior used the Birth and Death Process, with a UPGMA generated and geographical locations, we identify ‘‘cryptic species’’ and flag starting tree and the auto optimize option for operators. We these lineages as candidate species. We use the term ‘‘cryptic’’ in computed 108 generations, sampled every 1000 generations. The a relaxed definition given that we did not examine thoroughly concatenated analysis used the same partitioning as previously; the morphological differences that may exist between the segre- i.e. seven partitions (coding mtDNA by codons, 12S–16S, coding gated entities; that is why we use the term candidate species. Nev- nuDNA by codons) each under GTR + G with linked tree prior for ertheless, we argue that these differences are very subtle, which is all trees. The multilocus dataset, however, was based on the four emphasized by the misidentifications already documented (Heyer, loci with unlinked tree prior for all trees (mtDNA, RAG1, POMC 1977; Hoogmoed and Lescure, 1984; Hoogmoed et al., 1994; Lynch, and TYR) with each nuDNA coding gene sub-partitioned by codon 2005), and that the term ‘‘cryptic’’ can be used in a previous position. Each partition was considered under GTR + G, estimated definition: ‘‘two or more distinct species previously classified as base frequencies and four gamma categories. a single one due to overall morphological similarity that prevents We examined convergence on stationarity using Tracer 1.5 immediate obvious distinction’’ (Bickford et al., 2007; Pfenninger (Rambaut and Drummond, 2007). For both analyses effective and Schwenk, 2007). We provide details below and in discussion sample sizes were >200 for all parameters except where prior justifying our species delineation. and posterior jumped between alternative values. A few relative Phyzelaphryne (Southern Amazonian Clade, SAMC) is subdivided substitution rates were also with low ESS, jumping from high val- in two well-supported clades. One is distributed on the east side ues to zero (probably because no substitutions of these types are from the right margin of Purus River to Abacaxis River (Fig. 2). observed) rendering the prior on the rate invalid (Drummond The other one is situated on the west from the left margin of the et al., 2002). Therefore, we computed additional 108 generations Purus River to Leticia at the border between Brazil, Colombia and run with the prior distribution of these relative rates changed from Peru. This genus is for the first time reported from these two latter a gamma to a uniform distribution bounded between 10À5 and 1. countries. Actually, Lynch (2005) previously found the species in This made ESS > 150 for all parameters except alpha for CP1 and Colombia, but erroneously referred it to Adelophryne adiastola 2 for TYR and RAG1 and their respective tree likelihoods for the instead of Phyzelaphryne. Levels of divergence between the two multilocus analysis. Nevertheless, the time estimates were similar Phyzelaphryne clades and the absence of allelic sharing in the three among runs. The maximum clade credibility trees were computed nuDNA loci (Fig. 3) strongly suggest additional specific subdivi- with Tree Annotator 1.6.2. sions. We tentatively associate the eastern clade to the nominal species given that the type locality lies in the Madeira river catch- ment. This clade appears strongly structured across catchments of 3. Results the Madeira, Purus (right margin) and Abacaxis rivers with well- differentiated pairs of lineages. 3.1. Monophyly of Phyzelaphryninae and genera The western clade associated to a candidate species is even more deeply subdivided between the Colombian populations and Relationships are well resolved within Eleutherodactylidae with the two Brazilian ones (Juruá, Purus left margin) with no allelic only two poorly sustained nodes as well as among other Terrarana sharing in any of the three nuDNA loci (Fig. 3). Nevertheless, con- with only three poorly sustained nodes (Fig. 1a). However, deeper sidering the absence of other lines of evidence that could corrobo- relationships among Hyloidea remain ambiguous. Both methods rate the hypothesis of additional species in this clade, we (BA and MP) reveal very deep divergence among well-defined conservatively assign them to a single species. groups within Phyzelaphryninae using the three matrix configura- The Northern Amazonian Clade (NAMC) is recovered as the tions (Fig. 1a). This subfamily is itself strongly supported as a clade, sister group of all the Adelophryne representatives of the Atlantic A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561 555

(a)

(b)

Fig. 1. (a) Phylogenetic reconstruction based on Bayesian analysis of concatenated loci. Bootstrap supports from MP analyses are also indicated after posterior probabilities. Ã stand for pp values higher than 0.99 and bootstraps % higher than 99. Hyphens (-) indicate nodes not recovered with MP and values in red indicate poorly supported nodes i.e. pp < 0.95 and node not recovered with MP. (b) Topologies obtained from Bayesian analysis and MP for main Phyzelaphryninae lineages.

forest from Bayesian analysis, but under MP this clade is sustained southwards to the Brazilian state of Minas Gerais. Given the incon- as the sister group of the SAFC instead (Fig. 1b and Suppl. Mat.). sistency in morphological identification, level of divergence, Surprisingly, the population identified as A. gutturosa from Serra absence of shared nuDNA alleles (Fig. 3) and geographical loca- do Navio (Amapá, Brazil) is in fact recovered as the sister group tions, we call these populations a candidate species. We tentatively of all the other species of this clade and thus renders A. gutturosa assign the population from Serra das Lontras and Serra do Teimoso paraphyletic. We therefore refer the Serra do Navio population to to the nominal species (A. pachydactyla) given that their morpho- an undescribed species of Adelophryne. Another noteworthy result logical characteristics agree with Hoogmoed et al. (1994), and is that the previously published sequences of A. gutturosa (Heinicke the geographical proximity to the type locality. et al., 2009) obtained from an individual from Mount Ayanganna (Guyana) in fact correspond to A. patamona. DNA sequences 3.3. Molecular dating obtained from the holotype (ROM 43034) and the paratypes of A. patamona are included herein (Table 1), allowing us to be certain Both ‘‘concatenated’’ and ‘‘multilocus’’ approaches led to similar that the previous identification was erroneous. Interestingly, the topologies, notably supporting the NAMC as the sister group of Colombian species A. adiastola is recovered nested within this NAFC + SAFC with high posterior probabilities. Time estimates NAMC as the sister species of A. gutturosa, both forming a clade are, however, younger from the multilocus analysis than from grouped with A. patamona. the concatenated one. Phyzelaphryninae crown age is recovered The Northern Atlantic Forest Clade (NAFC) gathers A. baturiten- between 40.5 My old (concatenated) and 27.4 My old (multilocus) sis, A. maranguapensis (two species described from elevationally thus originating during late Eocene/early Oligocene. Major clades isolated moist forests in Ceará state, northeastern Brazil), one of Adelophryne are recovered to have diverged between 25.8 Ma population from the state of Pernambuco that has been identified (concatenated) and 16.5 Ma (multilocus) thus during early as A. baturitensis (Loebmann et al., 2011), and two populations from Miocene. The NAMC diversified earlier than the other major clades Bahia state. Relationships among species within that group remain given that the four species originated between 20 (concatenated) largely unresolved. The two neighbor populations from Bahia state and 13.4 Ma (multilocus) while NAFC, SAFC and SAMC diversified are highly divergent and form a strongly supported clade. More- later between 14.6–12.8 (concatenated) and 7.2–8 Ma over, they do not share any alleles for the nuDNA loci (Fig. 3). (multilocus). The southern Atlantic Forest Clade (SAFC) includes the last The fact that the concatenated analysis yielded older divergence nominal species A. pachydactyla and no less than three additional times than did the multilocus analysis is consistent with an expec- highly divergent lineages corresponding to newly discovered pop- tation that the average coalescence time for the various gene lin- ulations, extending the range of the genus ca. 650 km straight line eages should exceed somewhat the divergence times of the 556 A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561

Fig. 2. Map of the distribution of sampling localities (circles), and type localities (stars). Additional records from the literature are illustrated in Ecuador, Brasil and Colombia for A. adiastola (Ortega-Andrade, 2009), Brazil ES for A. cf. pachydactyla (Almeida et al., 2011) and in Bolivia ad Brazil PA for P. miriamae – (from Heyer, 1977; precise locality not mentioned, De la Riva et al., 2000).

population lineages (Liu et al., 2009). An independent test of our species is even more surprising considering the few and scattered divergence-time estimates is to ask whether they predict reason- localities that have so far been sampled; they represent only a tiny able rates of evolution for the mitochondrial Cytb gene, whose evo- portion of the potential distribution of these two genera. This esti- lutionary rate has been calibrated in many prior studies of mate matches previous DNA-based attempts to evaluate the actual vertebrates. An expected evolutionary of 2.1% sequence divergence species richness in tropical amphibians (Fouquet et al., 2007; per million years has been obtained by comparing multiple pairs of Vieites et al., 2009; Jansen et al., 2011; Funk et al., 2012) and also sister species whose separation was caused by formation of the matches sudden increases in species richness of several Terrarana Isthmus of Panama (reviewed by Reece et al., 2010). Using the data genera (e.g. Brachycephalus Pombal, 2010). Our species delineation in Table 3, we compare our estimated divergence times with those is based on the convergence of evidence from identification of the obtained using the Cytb calibration for the eight interspecific specimens, references, phylogenetic position, time of divergence branching events in Fig. 4 that we estimate to be less than 20 mil- and geographical location. Fine-tuned species delineation would lion years; these are the cases for which substitutional saturation greatly benefit from an ‘‘integrative taxonomy’’ approach (Dayrat, of Cytb should be minimal. For the five nodes within the NAFC 2005; Will et al., 2005; Padial et al., 2010), but this approach lies and SAFC, divergence times estimated from the Cytb calibration beyond the scope of our paper given that thorough examination are very close to our estimates from the concatenated analysis, dif- of the morphological variation as well as vocalization comparisons fering by no more than 8%. For the three nodes within the NAMC would require material not yet at hand. Nonetheless, in all cases and SAMC, divergences estimated by the Cytb calibration are closer the levels of divergence and concordance among several unlinked to the multilocus estimated dates, being identical in one case, 21% loci leave little doubt that these populations correspond to previ- higher in a second case, and 17% lower in the remaining case. These ously undetected species. results support the fidelity of our estimated divergence times as The divergence time between nominal species (e.g. between A. being consistent with the expected evolutionary rate of Cytb. adiastola and A. gutturosa or between A. maraguapensis and A. baturitensis) is similar to or lower than that between our candidate species (Fig. 4). The case of NAMC is particularly compelling given that 4. Discussion Adelophryne sp. 7 (Serra do Navio, Amapá) is the sister lineage to all the other species of the clade with a TMRCA estimated between 20 4.1. Cryptic diversity and conservation (concatenated) and 13.4 Ma (multilocus). Hoogmoed et al. (1994) already noticed that the animals from Serra do Navio that they With up to eight candidate species detected, our results indicate assigned to A. gutturosa are slightly different from the type material a >100% increase in the species diversity of the group, which likely from Guyana. Within NAFC, relationships among the species re- still remains underestimated. This high number of candidate main unclear, but divergences are similar among them (Fig. 4) with A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561 557

Fig. 3. Statistical parsimony networks for each nuDNA locus and each major clade. Haplotypes are shown as circles proportional in size to haplotype frequency. Each nominal and candidate species are delimited by a color ﬁlled rectangle. a minimum estimate around 6.5 Ma (multilocus), which is much nominal species: Adelophryne pachydactyla. The latest divergence higher than the divergence generally observed among sister spe- is estimated around 10.5 My old (concatenated) and 5.4 My old cies of frogs (e.g. Fouquet et al., 2007; Vences et al., 2005; Vieites (multilocus); all the highly divergent lineages are recovered on et al., 2009). both mt and nuDNA, and at least the populations from Minas Moreover, in addition to being highly divergent from both A. Gerais are morphologically different (Felipe Leite pers. com.). maranguapensis and A. baturitensis, the isolated population from Therefore, the three additional lineages undoubtedly represent Pernambuco is more than 500 km away from any nominal Adeloph- candidate species. The 13.2–8 My separating the two Phyzelaph- ryne population (Loebmann et al., 2011), and the two populations ryne clades are also compelling evidence for the existence of from Bahia are more than 1000 km away from any sampled distinct species. Subdivision of the western candidate species of nominal Adelophryne population of the same clade. Therefore, we Phyzelaphryne into several species-level entities is also very likely argue that all three lineages represent candidate species, herein given the estimated 3 My of divergence and the lack of nuDNA called Adelophryne sp. 1–3. The distinction between Adelophryne allele sharing. sp. 2 and 3 is, however, more arbitrary given that we miss data In addition to the newly detected lineages/species, the old diver- for Adelophryne sp. 3, but based on available mt and nuDNA se- gence times between Phyzelaphryne and Adelophryne (40–30 Ma), quences, divergence is also very deep (20% with Cb) (Fig. 1a; Table among the three Adelophryne major clades (25–16 Ma), and among 4). The populations clustering into the SAFC comprise only one the species within the different clades (all >6 Ma) – particularly in 558 A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561

(a) (b)

Fig. 4. Bayesian time-calibrated, maximum clade-credibility tree using relaxed clock with (a) concatenated partitioned dataset (b) multilocus species tree (ÃBeast). Calibration point is indicated with yellow circle. Posterior probabilities are indicated above the nodes, while the median of the posterior distributions of the ages of the nodes are indicated below. Ninety-ﬁve percent credibility intervals are indicated with blue bars. (c) Posterior distribution of the mean rate of substitution of each locus from the multilocus species tree analysis. (d) Simpliﬁed tree of the topology obtained from each locus from the multilocus species tree analysis.

NAMC – are striking. Such results highlight the inherent difficulty in in Amazonia and Atlantic forest, that dates back to the Eocene, studying amphibian diversity and evolutionary trends based on some 46 Ma according to Heinicke et al. (2009), and estimated be- morphology alone, because it can be extremely conserved (Cherry tween 44.2 (concatenated) and 32.8 (multilocus) Ma in this work. et al., 1977, 1978; Emerson, 1986) and is often homoplastic (Boss- This divergence has already been discussed by Heinicke et al. uyt and Milinkovitch, 2000; Parra-Olea and Wake, 2001; Guayas- (2007) and was attributed to an ancient overseas dispersal from amin et al., 2008). South America towards Middle America and the proto Caribbean. Species from the Atlantic forest are also characterized by extre- Two subsequent events are most noteworthy: the basal split of mely old divergences among populations previously considered (1) Phyzelaphryninae and of (2) Adelophryne. (1) The divergence conspecific, and it is likely that more species remain to be between the genera Phyzelaphryne and Adelophryne dates back to discovered in that biome as well as in Amazonia. Actually, the doc- 40–30 Ma, which corresponds to the Eocene/Oligocene boundary umented record of A. cf. pachydactyla from Espírito Santo, Brazil by i.e. one of the major extinction events related to an abrupt cooling Almeida et al. (2011), as well as the single population identified as of the global climate (Prothero, 1994). Given a likely northern A. adiastola from Ecuador by Ortega-Andrade (2009) deserve spe- Amazonian origin of Adelophryne (see below) and southern Amazo- cial attention, as they could correspond to additional candidate nian origin for Phyzelaphryne, this split likely originates from a species. north/south fragmentation of the range due to climate change. This Revealing such remarkable diversity in a clade morphologically is a period of southern uplift of the Andes (Hoorn et al., 2010a,b), highly homogeneous stresses the challenge for conservation that isolation of Antarctica and the creation of a circumpolar current, we are facing, given that all of these species have highly restricted dramatic drop of the sea level, and major climatic changes (Or- distributions, sometimes in isolated highlands like Ceará and tiz-Jaureguizar and Cladera, 2006). This period also corresponds Pernambuco, and that human impact or climate change is a real to the prevalence of large grazing herbivores and ‘modernization’ threat for such species/populations (Corlett, 2012). The situation of other faunal aspects during the mid-Cenozoic, reflecting adapta- in the northern Atlantic forest being particularly worrying (Ribeiro tion to major environmental changes, including increased aridity et al., 2009), the northern fragments deserve prime conservation and cooling (Flynn and Wyss, 1998). Late Eocene–early Oligocene priority (Carnaval et al., 2009; Ribeiro et al., 2009). also witnessed the spread of open vegetation at the expense of the rainforest that previously dominated the southern South Amer- 4.2. Biogeography ican continent (Roig-Juñent et al., 2006; Romero, 1986). Such conditions were unlikely favorable to forest-restricted frogs with The biogeographic pattern in Eleutherodactylidae is particularly direct development in the forest litter and may be responsible striking with a first split between Eleutherodactylinae, occurring in for the initial disjunction within the group that today occurs on Middle America and the Caribbean, and Phyzelaphryninae, found opposite sides of the Amazon River. Interestingly, the origin of A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561 559

the bufonid ‘‘range expansion phenotype’’, as coined by Van Bocxl- aer et al. (2010), corresponds to this period of habitat modiﬁcation.

1 Moreover, the late Eocene–early Oligocene period matches divergence times in the higher taxon Terrarana major clades that are al- /JX298216 JX298213 X298236 most exclusively associated with forest habitat (Heinicke et al., 2009), with some of these clades being endemic to either the E234/JX298209

R19206/JX298231 Atlantic forest or Amazonia. CFBHT14119/ JX298207 CFBHT11716/ JX298205 (2) Despite somewhat conflicting signal among loci (Figs. 1a; 4d), both concatenated and multilocus approaches favored unambiguously NAMC as the sister group of the other Adelophryne. Such pattern shown by the Atlantic forest Adelophryne – being actually nested within otherwise Amazonian Phyzelaphryninae i.e. that Phyzelaphryne descends from an Amazonian lineage and is the sister taxon to Adelophryne occurring both in Amazonia and in the Atlantic forest – is a pattern never recovered previously. Given that CFBHT14119/ JX298106 JX298104 MTR14013/JX298100 MTR14013/JX298201 Eleutherodactylinae likely originated by dispersal from northwest- ern South America (Heinicke et al., 2007) and that the Atlantic forest Adelophryne are nested within Phyzelaphryninae, it seems likely that Adelophryne originally was situated in northern Amazo- nia and subsequently dispersed to the Atlantic forest some 23– 16 Ma. Nonetheless, we acknowledge that such short internodes at the base of Adelophryne allied with the conflicting results found CFBHT14119/ JX298153 CFBHT11339/ JX298150 by MP, call for a deeper investigation based on a larger number of unlinked nuclear loci and other sources of evidence. Nonetheless, the split between the NAMC and the Atlantic forest Adelophryne matches a period when the Purus Arch connected the Guiana Shield and the Brazilian Shield (Hoorn et al., 2010b). Later (mid Miocene), the Pebas system and the flowing paleo Amazon river have most likely prevented any possible route to the southeast MTR19437/JX298174 MTR19437/JX298174 MTR19437/JX298128 MTR19437/JX298228 JX298153 JX298150 for such small-bodied terrestrial and direct-developing frogs. Moreover, the 20–15 My window corresponds to a period of higher temperature (Zachos et al., 2001). Such conditions may have al- lowed Adelophryne to disperse rapidly given the short internode between two Atlantic forest clades today in contact on each bank of the Rio de Contas (Bahia). Nonetheless, it is striking that Adel- ophryne could have dispersed over great distances between the Guiana Shield and southern Atlantic forest in such a short time EU186689/ LSUMZ16935 frame (<2 My). Similarly, the north vs. south Atlantic forest pattern observed within Adelophryne is concordant with several studies of vicariant forms whose limits are more or less coincident with the Rio Doce valley (northern Espírito Santo state; Carnaval et al., 2009; Costa, 2003; Pellegrino et al., 2005; Pinto-da-Rocha et al., 2005; da Silva et al., 2004). Several plant taxa are restricted to either one of these LSUMZ16935 CFBHT14119/JX298253 CFBHT14119//JX298286 CFBHT14119/ CFBHT11716/JX298251 CFBHT11716//JX298284 CFBHT11716/ areas, producing a strong floristic differentiation between the northern and southern Atlantic forests (Oliveira-Filho and Fontes, 2000). This pattern strikingly matches what is observed in other taxa like Dendrophryniscus (Fouquet et al., 2012b) and Leposoma (Pellegrino et al., 2011). In these examples, divergence times also indicate that areas of environmental stability lasted for 20 My in the Guiana Shield and in several parts of the Atlantic forest from CFBHT14119/ JX298326 CFBHT11716/ JX298324 Miocene to Quaternary, a much longer time period than that mod- eled by Carnaval and Moritz (2008).

Acknowledgments

We are grateful to the many people and institutions that made this study possible and Allan Larson (MPE AE) as well as the two anonymous reviewers for their sound comments on the manu- PK1969/JX298390PK2231/JX298392 PK1969/JX298337MTR19437/JX298396 MTR19437/JX298346 PK2231/JX298341 ROM39578/EU186679 EU186689/ ROM39578/EU186679 PK2231/JX298264 PK1969/JX298163 PK2231/JX298300 PK1969/JX298163 PK1969/JX298117 PK2231//JX298168 PK1969/JX298218 PK2231/JX298122 PK2231/JX298222 AJC2463/JX298391 AJC2463/JX298340 AJC2463/JX298263 AJC2463/JX298299 AJC2463/JX298167 AJC2463/JX298167 AJC2463/JX298121 AJC2463/JX29822 MTR16244/JX298388 MTR16244/JX298335 MTR16244/JX298259 MTR16244/JX298294 MTR16244/JX298161 MTR16244/JX298161 MTR16244/JX298115 MTR16244 CFBHT14119/ JX298381 JX298380 MTR14013/JX298376 MTR14013/JX298321 MTR14013/JX298249 MTR14013/JX298281 CFBHT11339/ script. Thanks to Renato Recoder, Marco A. Sena, Mauro Teixeira Jr., José Cassimiro da Silva, Agustin Camacho, Dante Pavan, Gabriel Skuk, Vanessa Verdade, Roberta Damasceno, Renata Amaro, Sergio Marques de Souza, Francisco dal Vechio, José Mario Guellere, Tami Mott, Pedro M. S. Nunes, H. Bonﬁm, Sonia Machado, Felipe Leite patamona gutturosa adiastola pachydactyla baturitensis maranguapensis miriamae . sp. 7. . MTR13808/JX298389 MTR13808/JX298336 MTR13808/JX298295 MTR13808/JX298162 MTR13808/JX298162 MTR13808/JX298116 MTR13808/JX298217 . . . sp. 5. sp. 4. sp. 6 CFBHE234/JX298383 MTR13570/JX298384 MR17521/JX298327 MR15919/JX298385 MTR13570/JX298331 CFBHE234/JX298254 MTR13570/JX298256 MR15919/JX298332 CFBHE234/JX298288 MTR13570/JX298290 CFBH23672/JX298257 MR15919/JX298291 CFBHE234/JX298155 MTR13570/JX298157 CFBHE234/JX298155 MR15919/JX298158 CFBHE234/JX298108 MR15919/JX298158 CFBH MR15919/JX298112 MR15919/ MTR13570/JX298111 MTR13570/JX298212 . . sp. 2. sp. 1 PEU80/JX298379 CFBHT11716/ PEU80/JX298323 PEU80/JX298283 PEU80/JX298151 PEU80/JX298103 PEU80/JX298204 . . sp. 1b JMP2058/JX298400 JMP2058/JX298355 JMP2058/JX298275 JMP2058/JX298315 JMP2058/JX298182 JMP2058/JX298182 JMP2058/JX298135 JMP2058/J . sp. 1a MTR19206/JX298399 MTR19206/JX298349 MTR19206/JX298270 MTR19206/JX298309 MTR19206/JX298177 MTR19206/JX298177 MTR19206/JX298131 MT . P A A A P A P A A A A A A A SpeciesA Cb COI 12S 16Sand Luciana Fusinatto RAG1a for help RAG1b in the ﬁeld and/or POMC for collected TYR

Table 4 Sequence details including vouchers and accession numbers used for the molecular dating. invaluable material. We also thank Erney Plessman de Camargo, 560 A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561

José Maria da Silva, and Admilson Torres for their invaluable help Corlett, R.T., 2012. Climate change in the tropics: the end of the world as we know during fieldwork in Amapá. We are also grateful to Manuel An- it? Biol. Conserv. 151, 22–25. Costa, L.P., 2003. The historical bridge between the Amazon and the Atlantic forest tunes Jr., Sabrina Baroni and Maira Concistré (Instituto de Biociên- of Brazil: a study of molecular phylogeography with small mammals. J. cias da Universidade de São Paulo) who assisted with lab work. Biogeogr. 30, 71–86. Instituto Chico Mendes de Conservação da Biodiversidade (ICM- Crozier, R.H., 1997. Preserving the information content of species: genetic diversity, phylogeny, and conservation worth. Annu. Rev. Ecol. Syst. 28, 243–268. Bio), Conselho Nacional de Desenvolvimento Científico and Tec- da Silva, J.M.C., Rylands, A.B., Da Fonseca, G.A.B., 2005. The fate of the Amazonian nológico (CNPq DL. Doctoral fellowship 140226/2006-0) and areas of endemism. Conserv. Biol. 19 (3), 689–694. Parque Nacional da Serra do Cipó granted collection permits and daSilva, J.M.C., da Sousa, M.C., de Castelletti, C.H.M., 2004. Areas of endemism for passerine birds in the Atlantic forest, South America. Global Ecol. Biogeogr. 13, assisted in the field. Funding was provided by CNPq, Fundação de 85–92. Amparo à Pesquisa do Estado de São Paulo (FAPESP, 2003/10335- Dayrat, B., 2005. Towards integrative taxonomy. Biol. J. Linn. Soc. 85, 407–415. 8, 2011/50146-6 and 2010/51071-7; A. Fouquet PD scolarship De la Riva, I., Köhler, J., Lötters, S., Reichle, S., 2000. Ten years of research on Bolivian amphibians: updated checklist, distribution, taxonomic problems, literature 2007/57067-9, V.D. Orrico PhD scholarship FAPESP 2007/57067- and iconography. Rev. Esp. Herpetol. 14, 19–164. 9, M.L. Lyra PD scholarship 2010/50124-0), Fundação Cearense de Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by Apoio ao Desenvolvimento Científico e Tecnológico (FUNCAP) pro- sampling trees. BMC Evol. Biol. 7, 214. vided I.J. Roberto’s scholarship. The work of S. Castroviejo-Fisher Drummond, A.J., Nicholls, G.K., Rodrigo, A.G., Solomon, W., 2002. Estimating mutation parameters, population history and genealogy simultaneously from was financed by a Universidad de los Andes (2009–2010) and a temporally spaced sequence data. Genetics 161, 1307–1320. Fulbright/Spanish Ministry of Education (2010–2012) post-doc- Duellman, W.E., 1999. Patterns of Distribution of Amphibians: A Global Perspective. toral research contracts, and J. M. Padial’s research is funded by a Johns Hopkins University Press, Baltimore, Maryland. Efron, B., 1979. The 1977 Rietz lecture. Bootstrap methods: another look at the Gerstner Post-doctoral Fellowship at the American Museum of jackknife. Ann. Stat. 7, 1–26. Natural History. PJRK’s fieldwork in Guyana was made possible Emerson, S.B., 1986. Convergence and morphological constraint in frogs: variation thanks to the financial support of the Belgian Directorate-General in postcranial morphology. Fieldiana Zool. 43, 1–19. Faith, D.P., 1992. Conservation evaluation and phylogenetic diversity. Biol. Conserv. of Development Cooperation with additional support from the King 61 (1), 1–10. Léopold III Fund for Nature Exploration and Conservation; speci- Felsenstein, J., 1985. Confidence limits on phylogenies – an approach using the mens from Guyana were collected under Permit Numbers bootstrap. Evolution 39, 783–791. Flynn, J.J., Wyss, A.R., 1998. Recent advances in South American mammalian 030605BR006 and 160107BR068, and exported under Permit Num- paleontology. Trends Ecol. Evol. 13, 11449–11454. bers 191205SP01, 040406SP014, 040706SP0171 and 191207SP018 Fouquet, A., Gilles, A., Vences, M., Marty, C., Blanc, M., Gemmell, N.J., 2007. issued by the Guyana Environmental Protection Agency. Underestimation of species richness in neotropical frogs revealed by mtDNA analyses. PLoS ONE 2 (10), e1109, doi: 1110.1371/journal.pone.0001109. Fouquet, A., Noonan, B.P., Rodrigues, M.T., Pech, N., Gilles, A., Gemmell, N.J., 2012a. Appendix A. Supplementary material Multiple quaternary refugia in the eastern Guiana Shield revealed by comparative phylogeography of 12 frog species. Syst. Biol. 61 (3), 461–489. Fouquet, A., Recoder, R., Teixeira, M., Cassimiro da Silva, J., Amaro, R.C., Camacho Supplementary data associated with this article can be found, in Guerrero, A., Damasceno, R., Carnaval, A.C., Moritz, C., Rodrigues, M.T., 2012b. the online version, at http://dx.doi.org/10.1016/j.ympev.2012. Molecular phylogeny and morphometric analyses reveal deep divergence 07.012. between Amazonia and Atlantic forest species of Dendrophryniscus. Mol. Phylogenet. Evol. 62, 826–838. Funk, W.C., Caminer, M., Ron, S.R., 2012. High levels of cryptic species diversity References uncovered in Amazonian frogs. Proc. Roy. Soc. Lond. B. 279 (1734), 1806–1814. Gaston, K.J., Williams, P.H., 1996. Spatial patterns in taxonomic diversity. In: Gaston, K.J. (Ed.), Biodiversity: A Biology of Numbers and Difference. Blackwell Science, Akaike, H., 1974. A new look at the statistical-model identification. IEEE Trans. Mississauga, Ontario, Canada, pp. 202–229. Automat. Contr. 19 (6), 716–723. Giam, X., Scheffers, B.R., Sodhi, N.S., Wilcove, D.S., Ceballos, G., Ehrlich, P.R., 2012. Almeida, A.P., Gasparini, J.L., Peloso, P.L.V., 2011. Frogs of the state of Espírito Santo, Reservoirs of richness: least disturbed tropical forests are centres of southeastern Brazil – the need for looking at the ‘coldspots’. Check List 7 (4), undescribed species diversity. Proc. Roy. Soc. B 279, 67–76. 542–560. Grant, T., Frost, D.R., Caldwell, J.P., Gagliardo, R., Haddad, C.F.B., Kok, P.J.R., Means, Balakrishnan, R., 2005. Species concepts, species boundaries and species D.B., Noonan, B.P., Schargel, W.E., Wheeler, W.C., 2006. Phylogenetic identification: a view from the tropics. Syst. Biol. 54 (4), 689–693. systematics of dart-poison frogs and their relatives (Amphibia: Barnosky, A.D., Matzke, N., Tomiya, S., Wogan, G.O., Swartz, B., Quental, T.B., Athesphatanura: Dendrobatidae). Bull. Am. Mus. Nat. Hist. 299, 1–262. Marshall, C., McGuire, J.L., Lindsey, E.L., Maguire, K.C., Mersey, B., Ferrer, E.A., Gonzalez-Voyer, A., Padial, J.M., Castroviejo-Fisher, S., de la Riva, I., Vila, C., 2011. 2011. Has the earth’s sixth mass extinction already arrived? Nature 471 (7336), Correlates of species richness in the largest neotropical amphibian radiation. J. 51–57. Evol. Biol. 24 (5), 931–942. Bickford, D., Lohman, D.J., Sodhi, N.S., Ng, P.K.L., Meier, R., Winker, K., Ingram, K.K., Guayasamin, J.M., Castroviejo-Fisher, S., Ayarzagüena, J., Trueb, L., Vilà, C., 2008. Das, I., 2007. Cryptic species as a window on diversity and conservation. Trends Phylogenetic relationships of glassfrogs (Centrolenidae) based on mitochondrial Ecol. Evol. 22, 148–155. and nuclear genes. Mol. Phylogenet. Evol. 48, 574–595. Bossuyt, F., Milinkovitch, M.C., 2000. Convergent adaptive radiations in Madagascan Guayasamin, J.M., Castroviejo-Fisher, S., Trueb, L., Rada, M., Vilà, C., 2009. and Asian ranid frogs reveal covariation between larval and adult traits. Proc. Phylogenetic systematics of glassfrogs (Amphibia: Centrolenidae) and their Natl. Acad. Sci. USA 97, 6585–6590. sister taxon Allophryne ruthveni. Zootaxa 2100, 1–97. Brandley, M.C., Schmitz, A., Reeder, T.W., 2005. Partitioned Bayesian analyses, Guindon, S., Gascuel, O., 2003. A simple, fast, and accurate algorithm to estimate partition choice, and the phylogenetic relationships of scincid lizards. Syst. Biol. large phylogenies by maximum likelihood. Syst. Biol. 52, 696–704. 54, 373–390. Hedges, S.B., Duellman, W.E., Heinicke, M.P., 2008. New World direct-developing Carnaval, A.C., Moritz, C., 2008. Historical climate modeling predicts patterns of frogs (Anura: Terrarana): molecular phylogeny, classification, biogeography, current biodiversity in the Brazilian Atlantic forest. J. Biogeogr. 35, 1187–1201. and conservation. Zootaxa 1737, 1–182. Carnaval, A.C., Hickerson, M.J., Haddad, C.F.B., Rodrigues, M.T., Moritz, C., 2009. Heinicke, M.P., Duellman, W.E., Hedges, S.B., 2007. Major Caribbean and Central Stability predicts genetic diversity in the Brazilian Atlantic forest hotspot. American frog faunas originated by ancient oceanic dispersal. Proc. Natl. Acad. Science 323, 785–789. Sci. USA 104 (24), 10092–10097. Cassiano-Lima, D., Borges-Nojosa, D.M., Cascon, P., Cechin, S.Z., 2011. The Heinicke, M.P., Duellman, W.E., Trueb, L., Means, D.B., MacCulloch, R.D., Hedges, S.B., reproductive mode of Adelophryne maranguapensis Hoogmoed, Borges & 2009. A new frog family (Anura: Terrarana) from South America and an Cascon, 1994, (Anura, Eleutherodactylidae) an endemic and threatened expanded direct-developing clade revealed by molecular phylogeny. Zootaxa species from Atlantic forest remnants in northern Brazil. North-Western J. 2211, 1–35. Zool. 7 (1), 92–97. Heled, J., Drummond, A.J., 2010. Bayesian inference of species trees from multilocus Caterino, M.S., Reed, R.D., Kuo, M.M., Sperling, F.A., 2001. A partitioned likelihood data. Mol. Biol. Evol. 27 (3), 570–580. analysis of swallowtail butterfly phylogeny (Lepidoptera: Papilionidae). Syst. Heyer, W.R., 1977. Taxonomic notes on frogs from the Madeira and Purus rivers, Biol. 50, 106–127. Brazil. Pap. Avul. de Zool. 31 (8), 141–162. Cherry, L.M., Case, S.M., Wilson, A.C., 1977. Relative rates of morphological Heyer, W.R., Gascon, C., 1995. Collection notes and call characteristics for evolution in primates, carnivores and frogs. Am. Zool. 17, 910. Ischnocnema quixensis and Phyzelaphryne miriamae (Amphibia: Anura: Cherry, L.M., Case, S.M., Wilson, A.C., 1978. Frog perspective on morphological Leptodactylidae). J. Herpetol. 29 (2), 304–307. difference between humans and chimpanzees. Science 200, 209–211. Hillis, D.M., Bull, J.J., 1993. An empirical-test of bootstrapping as a method for Clement, M., Posada, D., Crandall, K.A., 2000. TCS: a computer program to estimate assessing confidence in phylogenetic analysis. Syst. Biol. 42, 182–192. gene genealogies. Mol. Ecol. 9, 1657–1660. A. Fouquet et al. / Molecular Phylogenetics and Evolution 65 (2012) 547–561 561

Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic Pombal, J.R., 2010. A posição taxonômica das ‘‘variedades’’ de Brachycephalus trees. Bioinformatics 17, 754–755. ephippium (Spix, 1824) descritas por Miranda-Ribeiro, 1920 (Amphibia, Anura, Hoegg, S., Vences, M., Brinkmann, H., Meyer, A., 2004. Phylogeny and comparative Brachycephalidae). Bol. Mus. Nac. 526, 1–12. substitution rates of frogs inferred from sequences of three nuclear genes. Mol. Posada, D., 2008. JModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25, Biol. Evol. 21 (7), 1188–1200. 1253–1256. Hoogmoed, M.S., Lescure, J., 1984. A new genus and two new species of minute Purvis, A., Hector, A., 2000. Getting the measure of biodiversity. Nature 405 (6783), leptodactylid frogs from northern South America, with comments upon 212–219. Phyzelaphryne (Amphibia: Anura: Leptodactylidae). Zool. Med. Leiden 58 (6), Prado, D.E., Gibbs, P.E., 1993. Patterns of species distributions in the dry seasonal 85–115. forests of South America. Ann. Missouri Bot. Gard. 80, 902–927. Hoogmoed, M.S., Borges, D.M., Cascon, P., 1994. Three new species of the genus Primack, R., Corlett, R.T., 2005. Tropical Rain Forests: An Ecological and Adelophryne (Amphibia: Anura: Leptodactylidae) from northeastern Brazil, with Biogeographical Comparison. Blackwell Publishing, Malden, Massachusetts. remarks on the other species of the genus. Zool. Med. Leiden 68 (15/25), 271– Prothero, D.R., 1994. The late eocene–oligocene extinctions. Ann. Rev. Earth Planet. 300. Sci. 22, 145–165. Hoorn, C., Wesselingh, F.P., terSteege, H., Stadler, T., Sanmartín, I., Sanchez- Pyron, R.A., Wiens, J.J., 2011. A large-scale phylogeny of Amphibia with over 2,800 Meseguer, A., Anderson, C.L., Jaramillo, C.M., Bermudez, C., Figueiredo, J.D., species, and a revised classification of extant frogs, salamanders, and caecilians. Riff, F.R., Negri, H., Hooghiemstra, J., Lundberg, T., Sarkinen, A.Mora., Antonelli, Mol. Phylogenet. Evol. 61, 543–583. A., 2010a. Amazonia through time: Andean uplift, climate change, landscape Rambaut, A., Drummond, A.J., 2007. Tracer. Version 1.5. (accessed 20.12.09). Hoorn, C., Wesselingh, F.P., Hovikoski, J., Guerrero, J., 2010b. The development of the Reece, J.S., Bowen, B.W., Smith, D.G., Larson, A., 2010. Molecular phylogenetics of Amazonian mega-wetland (Miocene; Brazil, Colombia, Peru, Bolivia. In: Hoorn, moray eels (Muraenidae) demonstrates multiple origins of a shell-crushing jaw C., Wesselingh, F.P. (Eds.), Amazonia, Landscape and Species Evolution: A Look (Gymnomuraena, Echidna) and multiple colonizations of the Atlantic ocean. into the Past. Wiley-Blackwell, Oxford, pp. 162–184. Mol. Phylogenet. Evol. 57, 829–835. Jansen, M., Bloch, R., Schulze, A., Pfenninger, M., 2011. Integrative inventory of Ribeiro, M.C., Metzger, J.P., Martensen, A.C., Ponzoni, F.J., Hirota, M.M., 2009. The Bolivia’s lowland amphibians reveals hidden diversity. Zool. Scr. 40 (6), 567– Brazilian Atlantic forest: how much is left, and how is the remaining forest 583. distributed? Implications for conservation. Biol. Conserv. 142 (6), 1141– Katoh, K., Misawa, K., Kuma, K., Miyata, T., 2002. MAFFT: a novel method for rapid 1153. multiple sequence alignment based on fast Fourier transform. Nucleic Acids Roig-Juñent, S., Domínguez, M.C., Flores, G.E., Mattoni, C., 2006. Biogeographic Res. 30, 3059–3066. history of South American arid lands: a view from its arthropods using TASS Kok, P.J.R., Kalamandeen, M., 2008. Introduction to the taxonomy of the amphibians analysis. J. Arid Environ. 66, 404–420. of Kaieteur National Park, Guyana. Abc Taxa 5, 278. Romero, E.J., 1986. Paleogene phytogeography and climatology of South-America. Lemmon, A.R., Brown, J.M., Stanger-Hall, K., Moriarty-Lemmon, E., 2009. The effect Ann. Missouri Bot. Gard. 73, 449–461. of ambiguous data on phylogenetic estimatesobtained by maximum likelihood Ronquist, F., Huelsenbeck, J.P., 2003. MRBAYES 3: Bayesian phylogenetic inference and Bayesian inference. Syst. Biol. 58, 130–145. under mixed models. Bioinformatics 19, 1572–1574. Liu, L., Yu, L., Pearl, D.K., Edwards, S.V., 2009. Estimating species phylogenies using Santos, J.C., Coloma, L.A., Summers, K., Caldwell, J.P., Ree, R., Cannatella, D.C., 2009. coalescence times among sequences. Syst. Biol. 58, 468–477. Amazonian amphibian diversity is primarily derived from late Miocene Andean Loebmann, D., Orrico, V.G.D., Haddad, C.F.B.H., 2011. First record of Adelophryne lineages. Plos Biol. 7 (3), 448–461. baturitensis Hoogmoed, Borges & Cascon, 1994 for the state of Pernambuco, Simmons, M.P., 2012. Radical instability and spurious branch support by likelihood northeastern Brazil (Anura, Eleutherodactylidae, Phyzelaphryninae). Herpetol. when applied to matrices with non-random distributions of missing data. Mol. Notes 4, 75–77. Phylogenet. Evol. 62, 472–484. Lynch, J.D., 2005. Discovery of the richest frog fauna in the world – an exploration of Stuart, S.N., Chanson, J.S., Cox, N.A., Young, B.E., Rodrigues, A.S.L., Fischman, D.L., the forests to the north of Leticia. Rev. Acad. Colomb. Cienc. 29 (113), 581–588. Waller, R.W., 2004. Status and trends of amphibian declines and extinctions Malhi, Y., Roberts, J.T., Betts, R.A., Killeen, T.J., Li, W.H., Nobre, C.A., 2008. Climate worldwide. Science 306 (5702), 1783–1786. change, deforestation, and the fate of the Amazon. Science 319, 169–172. Stuart, S.N., Hoffmann, M., Chanson, J.S., Cox, N.A., Berridge, R.J., Ramani, P., Young, Marshall, D.C., 2010. Cryptic failure of partitioned Bayesian phylogenetic analyses: B.E., 2008. Threatened Amphibians of the World. Lynx Edicions, Barcelona, lost in the land of long trees. Syst. Biol. 59 (1), 108–117. Spain, IUCN, Gland, Switzerland and Conservation International, Arlington, MacCulloch, R.D., Lathrop, A., Kok, P.J.R., Minter, L.R., Khan, S.Z., Barrio-Amorós, C.L., Virginia. 2008. A new species of Adelophryne (Anura: Eleutherodactylidae) from Guyana, Sullivan, J., Joyce, P., 2005. Model selection in phylogenetics. Annu. Rev. Ecol. Evol. with additional description of A. gutturosa. Zootaxa 1884, 36–50. Syst. 36, 445–466. Mittermeier, R.A., Myers, N., Thomsen, J.B., da Fonseca, G.A.B., Olivieri, S., 1998. Swofford, D.L., 2002. PAUPÃ: Phylogenetic Analysis using Parsimony (Ãand Other Biodiversity hotspots and major tropical wilderness areas: approaches to Methods). Version 4.0b10. Sinauer Associates, Sunderland, Massachusetts. setting conservation priorities. Conserv. Biol. 12 (3), 516–520. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: Mori, S.A., Boom, B.A., Prance, G.T., 1981. Distribution patterns and conservation of molecular evolutionary genetics analysis using maximum likelihood, eastern Brazilian coastal forest tree species. Brittonia 33, 233–245. evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, Mueller, R.L., 2006. Evolutionary rates, divergence dates, and the performance of 2731–2739. mitochondrial genes in Bayesian phylogenetic analysis. Syst. Biol. 55 (2), 289– Tuomisto, H., Ruokolainen, K., Kalliola, R., Linna, A., Danjoy, W., Rodriguez, Z., 1995. 300. Dissecting Amazonian biodiversity. Science 269 (5220), 63–66. Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., Kent, J., 2000. Van Bocxlaer, I., Loader, S.P., Roelants, K., Biju, S.D., Menegon, M., Bossuyt, F., 2010. Biodiversity hotspots for conservation priorities. Nature 403 (6772), 853–858. Gradual adaptation toward a range-expansion phenotype initiated the global Oliveira-Filho, A.T., Fontes, M.A.L., 2000. Patterns of floristic differentiation among radiation of toads. Science 327, 679–682. Atlantic forests in southeastern Brazil and the influence of climate. Biotropica Vences, M., Thomas, M., Bonett, R.M., Vieites, D.R., 2005. Deciphering amphibian 32, 793–810. diversity through DNA barcoding: chances and challenges. Phil. Trans. Roy. Soc. Ortega-Andrade, H.M., 2009. Amphibia, Anura, Eleutherodactylidae, Adelophryne Lond. B 360, 1859–1868. adiastola Hoogmoed and Lescure, 1984: first countries records and distribution Vieites, D.R., Wollenberg, K.C., Andreone, F., Kohler, J., Glaw, F., Vences, M., 2009. extension from Ecuador and Brazil. Check List 5 (1), 139–143. Vast underestimation of Madagascar’s biodiversity evidenced by an integrative Ortiz-Jaureguizar, E., Cladera, G.A., 2006. Paleoenvironmental evolution of southern amphibian inventory. Proc. Natl. Acad. Sci. USA 106 (20), 8267–8272. South America during the Cenozoic. J. Arid Environ. 66, 498–532. Wake, D.B., Vredenburg, V.T., 2008. Are we in the midst of the sixth mass Padial, J.M., Miralles, A., de la Riva, I., Vences, M., 2010. The integrative future of extinction? A view from the world of amphibians. Proc. Natl. Acad. Sci. USA 105, taxonomy. Front. Zool. 7, 16. 11466–11473. Parra-Olea, G., Wake, D.B., 2001. Extreme morphological and ecological homoplasy Wiens, J.J., 1998. Does adding characters with missing data increase or decrease in tropical salamanders. Proc. Natl. Acad. Sci. USA 98, 7888–7891. phylogenetic accuracy? Syst. Biol. 47, 625–640. Pellegrino, K.C.M., Rodrigues, M.T., Waite, A.N., Morando, M., Yonenaga-Yassuda, Y., Wiens, J.J., 2003. Missing data, incomplete taxa, and phylogenetic accuracy. Syst. Sites Jr., J.W., 2005. Phylogeography and species limits in the Gymnodactylus Biol. 52, 528–538. darwinii complex (Gekkonidae, Squamata): genetic structure coincides with Wiens, J.J., Fetzner, J.W., Parkinson, C.L., Reeder, T.W., 2005. Hylid frog phylogeny river systems in the Brazilian Atlantic forest. Biol. J. Linn. Soc. 85, 13–26. and sampling strategies for speciose clades. Syst. Biol. 54 (5), 778– Pellegrino, K.C.M., Rodrigues, M.T., James Harris, D., Yonenaga-Yassuda, Y., Sites, 807. J.W., 2011. Molecular phylogeny, biogeography and insights into the origin of Wiens, J.J., Morrill, M.C., 2011. Missing data in phylogenetic analysis: reconciling parthenogenesis in the Neotropical genus Leposoma (Squamata: results from simulations and empirical data. Syst. Biol. 60, 719–731. Gymnophthalmidae): ancient links between the Atlantic forest and Amazonia. Will, K.P., Mishler, B.D., Wheeler, Q.D., 2005. The perils of DNA barcoding and the Mol. Phylogenet. Evol. 61 (2), 446–459. need for integrative taxonomy. Syst. Biol. 54, 844–851. Pfenninger, M., Schwenk, K., 2007. Cryptic animal species are homogeneously Wilson, E.O., 1992. The Diversity of Life. Belknap Press of Harvard University Press, distributed among taxa and biogeographical regions. BMC Evol. Biol. 7. Cambridge, Massachusetts. Pinto-da-Rocha, R., da Silva, M.B., Bragagnolo, C., 2005. Faunistic similarity and Zachos, J., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and historic biogeography of the harvestmen of southern and southeastern Atlantic aberrations in global climate 65 Ma to present. Science 292, 686–693. rain forest of Brazil. J. Arachnol. 33, 290–299.