Programa De Pós-Graduação Em Genética, Conservação E Biologia Evolutiva

Instituto Nacional de Pesquisas da Amazônia – INPA Programa de Pós-Graduação em Genética, Conservação e Biologia Evolutiva

Filogenia e biogeografia de três famílias de aves do Neotrópico

Mateus Ferreira

Manaus, Amazonas Fevereiro, 2018 Mateus Ferreira

Filogenia e biogeografia de três famílias de aves do Neotrópico

Orientador: Dra. Camila Cherem Ribas

Tese apresentada ao Instituto Nacional de Pesquisas da Amazônia como requisito para obtenção do grau de doutor em Genética, Conservação e Biologia Evolutiva.

Manaus, Amazonas Fevereiro, 2018 1 Folha reservada para a banca julgadora da versão final da tese 2

3 Ficha Catalográfica 4

iii

5 Agradecimentos 6 Agradeço primeiramente a minha orientadora Camila Ribas, pela paciência e confiança 7 que depositou em mim durante esses anos de orientação. Sem sombra de dúvidas, esse trabalho 8 não seria possível sem essa amizade e parceria. 9 Ao meu co-orientador, Joel Cracraft, com quem tive a sorte de trabalhar durante o meu 10 doutorado sanduíche. Pelas excelentes conversas e orientações sobre biogeografia e sobre os 11 padrões e processos que moldaram a diversidade de aves no mundo. 12 À Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) e ao 13 programa de Pós-Graduação em Genética, Conservação e Biologia Evolutiva, do Instituto 14 Nacional de Pesquisas da Amazônia, pela concessão da bolsa de doutorado no país e bolsa 15 sanduíche (# 88881.133440/2016-01), que tornaram este projeto possível. 16 Aos curadores e responsáveis pelas coleções científicas que gentilmente cederam 17 material para que este trabalho fosse desenvolvido: Fátima Lima e Antonita Santana (MPEG); 18 Marlene Freitas (INPA); Nate Rice (ANSP), Cristina Miyaki (LGEMA), Donna Dittman e 19 Robb Brumfield (LSU), Paul Sweet e Tom Trombone (AMNH), Mark Robbins (KU), John 20 Bates e Ben Marks (FMNH), Brian K. Schmidt (USNM), Sharon Birks (UWBM). E, a todas as 21 pessoas envolvidas nas expedições de coleta dessas institutições. 22 Ao projeto “Dimensions US-Biota: Assembly and evolution of the Amazon biota and 23 its environment: an integrated approach”, um projeto financiado conjuntamente pela Fundação 24 de Amparo à Pesquisa de São Paulo (FAPESP #2012/50260-6) e pelo National Science 25 Fundation (NSF DEB 1241056). Cujo apoio e financiamento foram essenciais para a execução 26 das várias etapas desse doutorado. 27 A todos os colegas do EBBA, pela constante ajuda e pelas excelentes discussões e 28 incentivos, e pelo café, especialmente pelos cafés: Robs, Fernanda, Rafael, Claudinha, Érico, 29 Erik, Lídia, Renatinha, Jessica, Nelson, Carol, Waleskinha e todo mundo que passou por aqui. 30 Ao pessoal que me aguentou durante esse doutorado: Maricota, Leandro, Marina, Ana, 31 Marizita, Derek, Miquéias, Pedro, Cadu e Manu. Em especial à Romina, pela caminhada lado 32 a lado durante toda a execução desse projeto, pelos puxões de orelha quando eu precisei e por 33 ter me aguentado todo esse período. 34 Ao pessoal do LTBM, Giselle e Paula, pela excelente companhia, pelos cafés e ajudas 35 quando precisei.

36 To everyone who received me at the AMNH during my sandwich fellowship: Lydia, 37 Bill, Tom, Paul, Gabi, Brian, Luke, and Peter. A special thanks to Jessica and Laís for all the 38 support and friendship during my time in NY. 39 Também gostaria de agradecer ao Laboratório Nacional de Computação Científica 40 (LNCC/MCTI) por fornecer recursos de computação de alto desempenho através do 41 supercomputador SDumont, fundamentais para as análises realizadas neste estudo. 42 Por fim, um agradecimento especial para a minha família, que me apoiou 43 incondicionalmente em todo esse percurso, e cuja ajuda foi essencial para a finalização deste 44 doutorado. 45

46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

65 “Nothing in biology makes sense except in the light of evolution”

66 Theodosius Dobzhansky

68 “Life and Earth evolve together”

69 Leon Croizat

71 72 Resumo 73 O Neotrópico é uma das regiões com os maiores índices de biodiversidade do planeta e muito 74 tem se questionado sobre a origem de tamanha diversidade. Acredita-se que os padrões de 75 diversidade atual dentro da região sejam um resultado da complexa história geomorfológica e 76 climática da região. Entre os eventos geomorfológicos mais discutidos estão o soerguimento 77 dos Andes e consequente reestruturação da drenagem continental, e o fechamento do Istmo do 78 Panamá, que permitiu a troca intercontinental de biotas. Neste trabalho foram selecionadas três 79 famílias de aves do Neotrópico. A família Trogonidae tem uma distribuição Pantropical, 80 ocorrendo também nas regiões subtropicais e tropicais da África e Ásia, no entanto, a maior 81 diversidade encontra-se justamente na região Neotropical. As famílias Bucconidae e Galbulidae 82 são duas famílias irmãs endêmicas do Neotrópico. Foram selecionadas amostras de todas as 83 espécies e quase todas as subespécies descritas para os três grupos. Para as espécies amplamente 84 distribuídas foram selecionadas amostras ao longo de toda a distribuição e uma análise prévia 85 para verificar a estrutura filogeográfica de cada grupo, com base nesses resultados, foram 86 selecionadas amostras para o sequenciamento de milhares de loci de regiões Ultra Conservadas 87 (Ultraconserved Elements, UCE). Dessa forma, compilamos três estudos nessa tese. No 88 primeiro capítulo, foi estudado um complexo de aves da família Galbulidae associada aos 89 ambientes de areia branca na região Amazônica. Através da comparação entre marcadores 90 moleculares com diferentes métodos de herança, DNA mitocondrial e nuclear (UCE), pudemos 91 observar um conflito entre esses dois marcadores. Através deste conflito foi possível propor um 92 modelo de diversificação para os ambientes de areia branca na região. No segundo capítulo 93 analisamos a diversificação global da família Trogonidae, com o auxílio dos UCEs 94 reconstruímos a relação filogenética entre todas as espécies da família e estimamos uma árvore 95 datada da diversificação de Trogonidae. No terceiro e último capítulo, analisamos os padrões 96 de diversificação das famílias Galbulidae e Bucconidae através de uma abordagem 97 filogeográfica e filogenética. Neste trabalho pudemos observar que a diversidade do grupo se 98 encontra claramente subestimada. 99 100 101

vii

102 Abstract 103 The Neotropical region has one of the highest biodiversity index in the planet and several 104 hypotheses have been proposed to explain the origin of such diversity. Currently, landscape 105 and climatic evolution are credited to be the two main processes responsible for shaping the 106 patterns. Landscape evolution includes, for example, the Andean uplift and consequent 107 continental drainage reconfiguration, and the closure of the Isthmus of Panama, which 108 allowed the Great American Biotic Interchange. In the present study we selected three 109 Neotropical families of birds. Trogonidae has a Pantropical distribution, members of this 110 family inhabit tropical and subtropical regions of Africa, Asia, however, the highest diversity 111 is currently found in the Americas. Galbulidae and Bucconidae are sister families and 112 endemics to the Neotropics. WE sampled all species and almost all subspecies currently 113 recognized for this three families, and for widespread species we thoroughly sampled 114 throughout their distributions to uncover hidden phylogeographic patterns. Based on these 115 results, we selected the samples to sequence thousands of Ultraconserved Elements (UCE). 116 Thus, we compiled three studies for this thesis. In the first chapter, we studied one Galbulidae 117 species complex associated with the Amazonian White-sand environments. We compared 118 between molecular markers that have different heritage systems, the mtDNA and nuDNA 119 (UCE), where we recovered contrasting histories between markers, and based on these results 120 we proposed a diversification model for the White-sand environments. In the second chapter, 121 we analyzed the global diversification of Trogonidae, employing thousands of UCE loci to 122 propose a phylogenetic hypothesis between all species currently recognized, and we also 123 estimated a fossil calibrated time tree for Trogonidae diversification. At last, in the third 124 chapter, we analyzed the diversification patterns for Galbulidae and Bucconidae using a 125 phylogeographic/phylogenetic approach. In this chapter it was clear how these groups 126 diversity in underestimated by currently taxonomic approach.

127

viii

128 Sumário 129 130 Agradecimentos ...... iv 131 Resumo ...... vii 132 Abstract ...... viii 133 Introdução Geral ...... 1 134 Objetivos...... 7 135 Capítulo 1 ...... 8 136 Abstract ...... 10 137 1. Introduction ...... 11 138 2. Methods ...... 13 139 2.1. Taxon sampling ...... 13 140 2.2. DNA extraction, amplification and sequencing...... 14 141 3. Results ...... 17 142 3.1. Sanger sequencing and haplotype networks ...... 17 143 3.2. mtDNA genome and time tree ...... 18 144 3.3. UCE sequencing, RAxML and Species trees ...... 18 145 4. Discussion ...... 19 146 4.1. mtDNA and nuDNA incongruence ...... 19 147 4.2. Biogeography of WSE avifauna ...... 23 148 4.3. Evolution in the White-sand environments ...... 24 149 5. Conclusion ...... 26 150 Acknowledgements ...... 27 151 Funding ...... 27 152 References ...... 28 153 Capítulo 2 ...... 38 154 Abstract ...... 39 155 Introduction ...... 40 156 Results ...... 43 157 UCE sequencing...... 43 158 Phylogenetic inference ...... 43 159 Time-calibrated tree ...... 44 160 Discussion ...... 44 161 Phylogenomic contribution to the reconstruction of Trogonidae diversification ...... 44 162 Diversification and biogeography of Trogons ...... 46 163 Africa and Asia diversification ...... 48 164 Neotropical diversification ...... 50 ix

165 Conclusion ...... 52 166 Materials and Methods ...... 53 167 Taxon sampling and DNA extraction ...... 53 168 UCE and exons assembly ...... 53 169 Phylogenetic relationships and species tree analysis ...... 54 170 Dating analysis ...... 55 171 Acknowledgements ...... 55 172 References ...... 56 173 Capítulo 3 ...... 71 174 Abstract ...... 72 175 Introduction ...... 73 176 Material and Methods ...... 75 177 Sampling and DNA isolation ...... 75 178 Phylogeographic structure and UCE sampling ...... 76 179 UCE assembly ...... 76 180 Phylogenetic relationship ...... 77 181 Results ...... 77 182 Phylogeographic results ...... 77 183 UCE sequencing ...... 78 184 Phylogenetic results ...... 78 185 Discussion ...... 79 186 Phylogenetic results ...... 79 187 Galbulidae systematics ...... 80 188 Bucconidae systematics ...... 83 189 Conclusion ...... 88 190 Acknowledgements ...... 88 191 References ...... 89 192 Síntese Geral ...... 107 193 Referências Bibliográficas ...... 108 194 195 196 197

198

199

200 Introdução Geral 201 202 O Neotrópico é uma das regiões biogeográficas com uma das maiores biodiversidades 203 do mundo (Jetz et al., 2012; Holt et al., 2013), mesmo que uma grande parcela dessa diversidade 204 ainda seja desconhecida (Kier et al., 2005; Hopkins, 2007; Barrowclough et al., 2016). Na 205 região Neotropical, os biomas Mata Atlântica, Cerrado e Amazônia despontam como hotspots 206 de biodiversidade altamente ameaçados pela ação humana (Myers et al., 2000; Mittermeier et 207 al., 2003; Colombo e Joly, 2010). Em especial para a região Amazônica, que abrange mais de 208 40% da área total do Neotrópico, desde que Wallace (1852), fez suas primeiras observações 209 sobre a importância dos rios na delimitação da distribuição de diferentes espécies de primatas, 210 vários trabalhos foram realizados demonstrando a importância dos afluentes do rio Amazonas 211 na estruturação da diversidade alfa da região (Vanzolini e Willians, 1970; Cracraft, 1985; 212 Haffer, 1985; Ávila-Pires, 1995). A comparação e aparente congruência dos padrões de 213 distribuição geográfica permitiu a elaboração de algumas hipóteses sobre quais processos 214 poderiam ter dado origem a esses padrões (revisões em Haffer (1997) e Leite e Rogers (2013)), 215 incluindo as variações climáticas do Pleistoceno, em especial o Último Máximo Glacial (LGM 216 – Last Glacial Maximum, ca. 20.000 anos) (Haffer, 1969; Brown et al., 1974); a influência das 217 incursões marinhas (Nores, 1999; 2004); e a formação dos rios da bacia Amazônica (Bates et 218 al., 2004; Ribas et al., 2012). Contudo, essas hipóteses foram formuladas com base apenas na 219 distribuição geográfica dos táxons, com o advento da filogeografia (Avise et al., 1987; Avise, 220 2009) e técnicas de datação molecular (Bromham e Penny, 2003; Bromham et al., 2017) novas 221 teorias foram propostas e além da congruência entra a distribuição geográfica o tempo de 222 diversificação também passou a fazer parte da comparações (Donoghue e Moore, 2003). Como 223 consequência, a teoria dos refúgios associados aos eventos climáticos do LGM foi parcialmente 224 rejeitada, já que as espécies se mostraram mais antigas que os ciclos glaciais mais recentes 225 (Colinvaux et al., 2000; Bush e Oliveira, 2006). As incursões marinhas, por outro lado, seriam 226 muito antigas para explicar a origem das espécies (Hoorn, 1993), favorecendo o modelo da 227 hipótese dos rios como barreira. 228 Atualmente, no entanto, o que sabemos sobre a complexidade da diversidade Amazônica 229 sugere que mais de um processo está por trás de sua origem (Bush, 1995; Bates et al., 2008; 230 Smith et al., 2014). Todos os eventos que moldaram a paisagem do Neotrópico ao longo do 1

231 tempo podem ter influenciado a diversificação da biota, por exemplo: A) o fim do “isolamento 232 esplêndido” (Simpson, 1980; Dacosta e Klicka, 2008) após o estabelecimento do Istmo do 233 Panamá (Haug e Tiedeman, 1998; Coates e Stallard, 2013; Lessios, 2015; Odea et al., 2016). 234 B) O soerguimento da cadeia de montanha dos Andes (Garzione et al., 2008; Hoorn et al., 2010; 235 Horton, 2018), que influenciou drasticamente não só a drenagem da bacia Amazônica (Hoorn 236 e Wesselingh, 2010; Latrubesse et al., 2010; Shephard et al., 2010; Nogueira et al., 2013; Hoorn 237 et al., 2017), como também o clima de todo o continente (Hartley, 2003; Ehlers e Poulsen, 238 2009; Insel et al., 2009). C) A influência das flutuações climáticas do Pleistoceno também 239 voltou a fazer parte das discussões, especialmente com relação ao estabelecimento de diferentes 240 regimes de precipitação dentro do continente (Cheng et al., 2013; Wang et al., 2017). 241 Dessa forma, faz-se necessário investigar não somente a evolução do modelo através das 242 variáveis biológicas, mas também quais processos físicos podem ter influenciado a sua 243 diversificação (Baker et al., 2014). Por exemplo, o estabelecimento do atual curso 244 transcontinental do rio Amazonas, ainda bastante discutido na literatura, varia entre o final do 245 Mioceno (10 – 7 Ma) (Hoorn e Wesselingh, 2010; Hoorn et al., 2017), início do Plioceno (~5 246 Ma) (Latrubesse et al., 2010), ou ainda, durante o Pleistoceno (2,5 Ma) (Nogueira et al., 2013; 247 Rossetti et al., 2015). Nesse sentido, estudando um gênero de aves (Psophiidae: Psophia) que 248 é restrita aos ambientes de terra-firme, e dessa forma suscetível às mudanças na drenagem da 249 Amazônia, Ribas et al. (2012) propuseram um modelo de diversificação da fauna de terra firme 250 ao correlacionar os eventos de diversificação das espécies do gênero ao estabelecimento de 251 barreiras associadas aos principais afluentes da bacia, favorecendo o modelo do 252 estabelecimento do rio Amazonas durante o Pleistoceno. O modelo proposto por Ribas et al. 253 (2012) sugere que para compreender os fatores que influenciaram a evolução da paisagem, 254 como o efeito da formação de um determinado rio na diversificação de espécies de terra-firme, 255 deve-se buscar padrões congruentes, temporais e espaciais, de diversificação em grupos que 256 serão de fato afetados diretamente pela barreira (e.g. Polo, (2015)). Em contraponto, análises 257 que buscam explicar a diversificação na Amazônia através de um único processo, como por 258 exemplo, a importância dos rios como barreira utilizando uma ampla gama de táxons com 259 nichos variados (Oliveira et al., 2017; Santorelli et al., 2018; Smith et al., 2014) tendem a 260 refutar esta teoria, já que diferentes organismos respondem de diferentes maneiras aos 261 processos e eventos históricos. Dessa forma, aceitando que a diversificação na Amazônia é 262 inerentemente complexa, o teste de hipóteses deve ser feito de maneira dirigida, ou seja, deve- 263 se buscar grupos taxonômicos que tenham sido potencialmente influenciados pelas barreiras

264 em questão. Só assim será possível estabelecer a importância biológica de um determinado 265 evento e gerar dados importantes para o estabelecimento dos modelos de evolução 266 geomorfológica da região (Baker et al., 2014). 267 Essa iluminação recíproca entre os processos físicos e bióticos, no entanto, só é possível 268 levando em consideração o fato de que qualquer evento de diversificação só pode ser 269 correlacionado com um evento biogeográfico se duas condições forem respeitadas: 1) as 270 unidades biológicas utilizadas devem ser comparáveis entre si e devem representar linhagens 271 com uma história evolutiva única; 2) a relação filogenética entre essas linhagens deve 272 representar de fato a história de diversificação do grupo. 273 A primeira condição refere-se ao fato de que as unidades utilizadas no estudo devem 274 representar linhagens independentes. Geralmente, entende-se que espécies devem ser a unidade 275 básica para qualquer estudo de ecologia, evolução, ou biogeografia, no entanto, essa prática 276 pode ser particularmente problemática na Amazônia, uma vez que grande parte das espécies 277 amplamente distribuídas pela região na realidade representam um complexo de linhagens 278 evolutivas independentes (Ribas et al., 2012; D’horta et al., 2013; Fernandes et al., 2013; 279 Fernandes et al., 2014; Hrbek et al., 2014; Boubli et al., 2015; Thom e Aleixo, 2015; Byrne et 280 al., 2016; Carneiro et al., 2016; Boubli et al., 2017; Ferreira et al., 2017; Lima et al., 2017; 281 Ribas et al., 2018). Para aves, em particular, esse déficit entre a taxonomia atualmente 282 reconhecida e a real diversidade está diretamente relacionado ao fato de que a definição daquilo 283 que reconhecemos como espécie ainda é muito influenciado pelo tipo de conceito de espécie 284 utilizado, em especial o conceito biológico de espécie (Mayr, 1942), que implica no 285 reconhecimento de metapopulações isoladas reprodutivamente. No entanto, o reconhecimento 286 de isolamento reprodutivo em populações naturais é particularmente difícil, especialmente em 287 populações alopátricas, onde é impossível observar naturalmente esse contato. Mesmo em 288 populações parapátricas, o contato e estabelecimento de uma zona híbrida não necessariamente 289 ameaça o statu quo das espécies envolvidas (Weir et al., 2015). Especialmente, porque a 290 capacidade de hibridização entre espécies, mesmo distantes, parece ser uma característica 291 sinapomórfica para aves (Grant e Grant, 1992; Gill, 1998; Harrison e Larson, 2014). 292 O conceito de espécie, mesmo sendo um dos assunto centrais para os estudo de evolução 293 e ecologia, permanece ainda sem definição clara e é sem dúvida um dos pontos mais discutidos 294 dentro da biologia (Mayr, 1976; Donoghue, 1985; Isaac et al., 2004; De Queiroz, 2005; Aleixo, 295 2007; Joseph et al., 2008; Aleixo, 2009; De Queiroz, 2011; Cellinese et al., 2012; De Queiroz, 296 2012; Willis, 2017). Ressaltando o impacto dessa escolha entre um conceito ou outro e do nosso

297 atual conhecimento sobre a taxonomia de aves, um estudo recente demonstrou que a diversidade 298 das aves é, pelo menos, duas vezes maior do que a atualmente reconhecida (Barrowclough et 299 al., 2016). Por exemplo, dentro da Neotrópico, um dos padrões mais observados é a existência 300 de espécies amplamente distribuídas, compostas por diferentes subespécies morfologicamente 301 distintas e geograficamente estruturadas, as quais foram, no entanto, agrupadas dentro de uma 302 mesma espécie devido a existência da possibilidade dessas populações hibridizarem caso 303 entrem em contato. 304 A segunda condição está relacionada aos problemas de conflitos entre a história de um 305 único gene e a história da espécie (Degnan e Rosenberg, 2009; Knowles, 2009). Esse conflito 306 tem se tornado cada vez mais evidente em face do desenvolvimento de técnicas de 307 sequenciamento massivo em paralelo (Metzker, 2010). Apesar de estarem se tornando mais 308 acessíveis, o sequenciamento e análise do genoma completo para organismos não modelo ainda 309 é impraticável para trabalhos que requerem amostragem de muitos indivíduos. Dessa forma, 310 algumas técnicas de se utilizar representações reduzidas foram desenvolvidas. Duas abordagens 311 dominam o cenário atualmente, uma delas é a utilização de enzimas de restrição para sítios 312 específicos ao longo de todo o genoma, denominada RAD-seq (restriction-site-associated DNA 313 sequencing) (Davey et al., 2011); e a outra é a utilização de sondas de RNA desenvolvidas para 314 capturar regiões específicas do genoma (Grover et al., 2012; Lemmon et al., 2012; Lemmon e 315 Lemmon, 2013). Uma das abordagens de sequenciamento de captura é a técnica que utiliza 316 sondas específicas para regiões do genoma ultra conservadas (do inglês, Ultra Conserved 317 Elements, UCE) (Faircloth et al., 2012). Essas regiões ultra conservadas foram selecionadas 318 pois permitem a utilização de um mesmo conjunto de sondas para realizar estudos em diversos 319 níveis taxonômicos, pois apesar das regiões centrais serem altamente conservadas, as regiões 320 flanqueadoras possuem variação suficiente tanto para recuperar relações mais antigas 321 (Mccormack et al., 2012; Crawford et al., 2015; Faircloth et al., 2015), quanto mais recentes 322 (Bryson et al., 2016; Manthey et al., 2016), inclusive utilizadas em radiações adaptativas 323 (Meiklejohn et al., 2016), onde altos níveis de separação incompleta de linhagens (do inglês, 324 Incomplete Lineage Sorting, ILS) sejam esperados (Degnan e Rosenberg, 2006; Oliver, 2013). 325 De modo a tentar então lançar alguma luz sobre os possíveis eventos que moldaram a 326 diversificação da biota Neotropical, foram selecionadas três famílias de aves: Trogonidae, 327 Galbulidae e Bucconidae. As três famílias possuem representantes por toda a região 328 Neotropical, incluindo várias espécies, ou grupo de espécies, amplamente distribuídas. A 329 família Trogonidae tem distribuição Pantropical, estando ausente apenas da região Australiana.

330 Representantes dessa família, popularmente conhecidos como surucuás, são aves de médio 331 porte e sua dieta varia entre insetívora e onívora, apresentam plumagem com coloração bastante 332 chamativa, e são reconhecidas por serem más dispersoras, não sendo capazes de realizar voos 333 de longa distância (Collar, 2017). Apesar de apresentarem plumagem bastante distinta, a 334 morfologia interna da família é bastante conservada e a sua monofilia nunca foi questionada 335 (Livezey e Zusi, 2007; Collar, 2017). No entanto, a relação entre trogonídeos e outras aves não 336 passeriformes já foi bastante controversa (Cracraft, 1981; Maurer e Raikow, 1981; Monteros, 337 2000; Mayr, 2003; Livezey e Zusi, 2006). Atualmente, aceita-se que a família seja uma das 338 primeiras linhagens a diversificar dentro da radiação de Coracimorphae sendo grupo irmão de 339 todas as outras famílias do grupo Core Landbirds (Jarvis et al., 2015; Prum et al., 2015). 340 Atualmente são reconhecidas 45 espécies (Collar, 2017; Gill e Donsker, 2018; Ramsen et al., 341 2018) distribuídas em sete gêneros. A região Neotropical contém a maior diversidade da 342 família, com quatro gêneros e cerca de 30 espécies. A região Asiática contém dois gêneros e 12 343 espécies, e por último, a região Africana, possui um gênero com três espécies. Apesar da maior 344 diversidade da família ser encontrada no Neotrópico, a existência de fósseis na Europa (Mayr, 345 1999; Kristoffersen, 2002; Mayr, 2005) sugere uma origem no Paleártico e posterior dispersão 346 e colonização da distribuição atual. Diversos trabalhos já tentaram abordar a relação 347 filogenética entre os representantes da família (Monteros, 1998; Johansson e Ericson, 2005; 348 Moyle, 2005; Dacosta e Klicka, 2008; Ornelas et al., 2009; Hosner et al., 2010), entretanto, 349 nenhum foi capaz de resolver a relação entre os gêneros. O último trabalho publicado (Hosner 350 et al., 2010), e o único a incluir representantes de todos os gêneros, recuperou uma parafilia 351 entre regiões geográficas, sugerindo um cenário biogeográfico bem mais complexo, em que a 352 região Neotropical, por exemplo, tenha sido ocupada por pelo menos três linhagens distintas. 353 As famílias Galbulidae e Bucconidae formam um clado já bem estabelecido, tanto com 354 caracteres morfológicos (Livezey e Zusi, 2007), quanto dados moleculares (Hackett et al., 2008; 355 Prum et al., 2015). Dentro da ordem Piciformes, são as únicas famílias com representantes com 356 distribuição exclusivamente neotropical, formando o grupo irmão das outras famílias de 357 Piciformes (Prum et al., 2015). A família Galbulidae é composta por aves de pequeno a médio 358 porte, asas arredondadas e um bico longo e afilado utilizado para capturar insetos durante o 359 voo. Possui 19 espécies distribuídas em cinco gêneros diferentes (Tobias, 2017; Gill e Donsker, 360 2018; Ramsen et al., 2018). As espécies da família são geralmente agrupadas em oito grupos 361 zoogeográficos, seis desses grupos representam complexos de espécies com distribuições 362 alopátricas ou parapátricas, e dois são espécies amplamente distribuídas (Collar, 2017). A

363 família Bucconidae também inclui aves de pequeno a médio porte, asas curtas e arredondadas, 364 tendo como característica uma cabeça relativamente grande, atualmente são reconhecidas 35 365 espécies para a família distribuídas em nove gêneros (Gill e Donsker, 2018; Ramsen et al., 366 2018). Os trabalhos de filogeografia desenvolvidos com representantes da família Bucconidae 367 – Malacoptila (Ferreira et al., 2017), Monasa e Nonnula (Soares, 2016) e Nystalus (Duarte, 368 2015) – demonstraram que a diversidade reconhecida pela taxonomia tradicional para esses 369 grupos é subestimada, já que existem muito mais linhagens genéticas geograficamente isoladas 370 do que táxons reconhecidos, demonstrando a importância da condução dos estudos de 371 filogeografia para elucidar a delimitação taxônomica dessas espécies amplamente distribuídas. 372 Dessa forma, o presente trabalho tem por objetivo reconstruir a relação filogenética entre 373 todas as linhagens dessas três famílias de modo a reconstruir a história de diversificação desses 374 três grupos. Para tanto, foram amostrados indivíduos ao longo da distribuição de todas as 375 espécies amplamente distribuídas para uma análise prévia da estrutura genética de cada uma 376 dessas linhagens. Com base nos resultados obtidos previamente foram selecionadas amostras 377 representativas de cada um desses agrupamento, tentando incluir, sempre que possível, um 378 representante para cada um dos táxons reconhecidos. Para essas amostras foram sequenciados 379 mais de 2000 loci de UCE, e com base nessa representação reduzida do genoma foram 380 realizadas análises para a reconstrução filogenética dos grupos. 381 382

383

384

385 Objetivos 386 387 O objetivo geral foi investigar a história biogeográfica da região Neotropical com base 388 nas relações filogenéticas entre todos os táxons atualmente reconhecidos para as famílias 389 Trogonidae, Bucconidae e Galbulidae baseado em dados de sequenciamento genômico. Sendo 390 que para isso foi necessário: 391 Capítulo 1: revisar a taxonomia e compreender os processos de isolamento e fluxo 392 gênico em um contexto espacial; 393 Capítulo 2: compreender a origem de táxons Neotropicais em uma família amplamente 394 distribuída; 395 Capítulo 3: compreender a estrutura filogeográfica de espécies amplamente distribuídas 396 em duas famílias Neotropicais para com isso obter uma reconstrução filogenética representativa 397 da diversificação do grupo. 398 399

400 401 402 403 404 405 406 407 408 409

410 Capítulo 1

411 412 413 Ferreira, M.; Fernandes, A.M.; Aleixo, A.; Antonelli, 414 A.; Olsson, U.; Bates, J.M.; Cracraft, J.; Ribas, C.C. 415 Evidence for mtDNA capture in the jacamar Galbula 416 leucogastra / chalcothorax species-complex and 417 insights on the evolution of white-sand environments 418 in the Amazon basin. Molecular Phylogenetics and 419 Evolution (no prelo) 420 421 422

423 424 425 Manuscript submission to Molecular Phylogenetics and Evolution 426 Contribution type: Original article 427 428 Evidence for mtDNA capture in the jacamar Galbula leucogastra / chalcothorax species- 429 complex and insights on the evolution of white-sand environments in the Amazon basin. 430 431 Ferreira, Mateusa*; Fernandes, Alexandre M.b; Aleixo, Alexandrec; Antonelli, Alexandred,e,f; 432 Olsson, Urban d,f; Bates, Jonh M.g; Cracraft, Joelh; Ribas, Camila C.i 433 434 a Programa de Pós-Graduação em Genética, Conservação e Biologia Evolutiva, INPA, 435 Manaus, AM, Brazil 436 b Unidade Acadêmica de Serra Talhada, UFRPE, Serra Talhada, PE, Brazil 437 c Coordenação de Zoologia, MPEG, Belém, PA, Brazil 438 d Department of Biological and Environmental Sciences, University of Gothenburg, SE-413 439 19 Gothenburg, Sweden 440 e Gothenburg Botanical Garden, SE-413 19 Gothenburg, Sweden 441 f Gothenburg Global Biodiversity Centre, Box 461, SE-405 30 Gothenburg, Sweden 442 gDepartment of Ornithology, FMNH, Chicago, IL, USA 443 h Department of Ornithology, AMNH, New York, NY, USA 444 i Coordenação de Biodiversidade, INPA, Manaus, AM, Brazil 445 *Corresponding author 446 447 Correspondence: Mateus Ferreira, Coordenação de Biodiversidade, Instituto Nacional de 448 Pesquisas da Amazônia, CEP 69080-971, Manaus-AM, Brazil 449 E-mail: [email protected] 450

451 Abstract

452 Jacamar species are found throughout Amazonia, with several different species occupying

453 forested habitats, but one species-complex, Galbula leucogastra / chalcothorax, inhabits areas

454 of open vegetation, known as white-sand environments (WSE). Previous studies of WSE birds

455 recovered shallow genetic structure in mtDNA coupled with signs of gene flow among WSE

456 areas. Here we characterize diversification of the G. leucogastra/chalcothorax species-complex

457 with dense sampling across its distribution, using mitochondrial DNA and Ultraconserved

458 Elements (UCE) loci. We performed likelihood and Bayesian analysis to recover the

459 phylogenetic relationships among populations using a concatenated approach, as well as a

460 species-tree analysis using *BEAST. The mtDNA results recovered at least six geographically-

461 structured lineages in which G. chalcothorax was embedded within lineages of G. leucogastra.

462 In contrast, analysis of UCE data with both concatenated and species-tree approaches recovered

463 G. chalcothorax as sister to all G. leucogastra lineages. We hypothesize that the mitochondrial

464 genome of the Madeira population of G. leucogastra was captured by G. chalcothorax early in

465 their initial divergence, and we suggest how WSE evolution and the co-evolution of mtDNA

466 genes and nuclear genes might have played a role in this rare event.

467

468 Keywords: White-sand environments, Amazonia, Galbulidae, jacamars, mtDNA capture, UCE

469

470 1. Introduction

471 White-sand environments (WSE) represent a unique type of habitat within Amazonia.

472 Apart from the continuous forest habitats found all over the basin, the WSE consists of patches

473 of differentiated habitats scattered in the landscape and isolated by the forest matrix (Adeney

474 et al., 2016). WSE consist of a continuum from open non-forested habitats, such as campinas,

475 with a predominance of grass and shrubland, to denser vegetation, called campinaranas and

476 varillales, all associated with sandy soils. This insular characteristic of WSE continues to

477 intrigue researchers as to how the ecosystem and its specialized biota evolved, how it responded

478 to Pleistocene glacial cycles, and whether the specialized biota disperse through the forest

479 matrix among patches of WSE (Brown and Benson, 1977; Anderson, 1981; Capurucho et al.,

480 2013; Matos et al., 2016). Besides its characteristic fragmentation, WSE are more

481 physiologically stressful and challenging from an ecological and evolutionary perspective,

482 making them much more taxonomically selective, with overall diversity being smaller when

483 compared with adjacent forest areas (Borges, 2003; Fine et al., 2010; Laranjeiras et al., 2014;

484 Adeney et al., 2016), although several new species endemic to this habitat having recently been

485 described (Whitney and Alonso, 1998, 2005; Alonso and Whitney, 2001; Cohn-Haft and Bravo,

486 2013; Cohn-Haft et al., 2013). Some studies point to a recent and dynamic history for WSE

487 (Latrubesse and Franzinelli, 2002; Rossetti et al., 2012), yet this habitat is usually associated

488 with ancient soils (Adeney et al., 2016). Although some plant species have a loose association

489 with WSE (Fine and Baraloto, 2016), others are tightly associated with them, such as species

490 of Pagamea (Vicentini, 2016). The same can be observed for other organisms (Cohn-Haft,

491 2008; Vriesendorp et al., 2006), especially birds (Borges et al., 2016a; Borges et al., 2016b).

492 Therefore, the combination of persistent white-sand soils with recent climatic and landscape

493 changes must have had an important influence on the evolution and distribution of WSE and of

494 the biota that inhabits them.

495 The few phylogeographic studies of WSE birds that have been undertaken, show little

496 genetic diversity with no geographic structure throughout Amazonia (Polytmus theresiae,

497 Matos et al., 2016); or shallow but geographically structured genetic diversity, with significant

498 migration rates between some populations (Tachyphonus phoenicius, Matos et al., 2016;

499 Xenopipo atronitrens, Capurucho et al., 2013). In general, results obtained so far for WSE birds

500 suggest that: (1) black-water flooded forest (igapó), due to similarities to WSE in vegetation

501 structure, may facilitate dispersal between isolated WSE patches; and, (2) Pleistocene glacial

502 periods, especially the Last Glacial Maximum, are temporally correlated with geographical

503 expansion of populations of species specialized in WSE.

504 These studies have been based on mtDNA markers (Capurucho et al., 2013), or on a

505 combination of mtDNA and a single nuclear marker (Matos et al., 2016). Until recently, most

506 phylogeographic studies have employed mtDNA. Its characteristic maternal inheritance,

507 comparatively small effective population size, rapid rate of mutation, and lack of

508 recombination, coupled with the fact that it is easy to amplify and sequence, have long made

509 mtDNA markers ideal for phylogeographic studies (Avise et al., 1987; Avise, 2009). However,

510 there are potential biases and limitations associated with these data (Zink and Barrowclough,

511 2008) and hybridization and introgression could be overlooked (Carling and Brumfield, 2008).

512 Thus, the inclusion of nuclear markers often yields different perspectives. That said, no markers

513 are without biases and the inclusion of autosomal markers entail other problems, such as

514 discordances between gene trees and species-trees (Knowles, 2009), as well as between the

515 history of mtDNA and nuclear markers, especially because the small number of nuclear loci

516 employed usually do not have enough information in recent divergences (Zink and

517 Barrowclough, 2008; Daly-Engel et al., 2012; Toews and Brelsford, 2012; Sloan et al., 2017).

518 These latter studies demonstrated the value of using several different markers to truly

519 understand species/lineage histories, yet until the advent of high throughput parallel sequencing

520 techniques, such multi-loci analyses were very time-consuming and uneconomical (Metzker,

521 2010). One of the new genomic markers made accessible by next-generation sequencing

522 technologies is Ultra-Conserved Elements – UCE (Faircloth et al., 2012; McCormack et al.,

523 2013). The use of UCEs provides access to large quantities of genomic data to assess

524 relationships at multiple time and taxonomic scales (Faircloth et al., 2012), from very old

525 radiations (Moyle et al., 2016), to more recent ones (Smith et al., 2014; Harvey et al., 2016;

526 Manthey et al., 2016).

527 Here, we investigate a rare pattern of evolutionary diversification in Amazonian WSE

528 avifauna by reconstructing the phylogeography of a jacamar species-complex using genomic

529 data. The jacamars (family Galbulidae) are exclusive to the Neotropics, with 19 species and 5

530 genera, mostly associated with wooded, lowland forest habitat (Stotz et al., 1996; Tobias,

531 2017). In Amazonia, most species are restricted to upland (terra firme) and flooded (varzea and

532 igapó) forests, with only two species (Galbula leucogastra and G. chalcothorax) known to

533 occur in WSE (Borges et al., 2016a). Galbula leucogastra and G. chalcothorax were previously

534 considered subspecies of a single species (Peters, 1948; Haffer, 1974), but were split by Parker

535 and Remsen (1987), based on diagnostic plumage and size differences. A phylogeny of the

536 family, based on multiple gene regions, indicates that G. leucogastra and G. chalcothorax are

537 sister-species with high support (Witt, 2004). Here we first investigate the distribution of

538 mtDNA diversity within these two species by sampling individuals from throughout their

539 distributions. Then, based on these results, we obtained sequences of thousands of genomic

540 markers (UCE) for a subset of samples to reconstruct their history of diversification and make

541 inferences about the evolution of WSE.

542

543 2. Methods

544 2.1. Taxon sampling

545 We sampled 35 individuals covering almost the entire distribution of the Galbula

546 leucogastra / chalcothorax (Table S1). As outgroups, we used one sample of G. albicollis (Witt,

547 2004). All tissues sequenced are represented by voucher specimens deposited in ornithological

548 collections in Brazil and the USA (Table S1).

549 2.2. DNA extraction, amplification and sequencing

550 DNA was extracted using a modified phenol-chloroform protocol (Sambrook and

551 Russel, 2001). We used published DNA primers (Sorenson et al., 1999) to amplify and

552 sequence two mitochondrial genes (Cytochrome b [cytb], and NADH subunit 2 [ND2]) for all

553 individuals following standard PCR protocols. For a subset of individuals (see below) we

554 extracted DNA using the DNeasy kit (Qiagen Inc.) following the manufacturer’s protocol, and

555 sent the extracts to RapidGenomics® (Gainsville, FL) for sequencing, using a probe set

556 targeting 2321 loci of Ultra Conserved Elements (UCE) plus 98 conserved exons from genes

557 that were previously used in phylogenetic analysis (Harvey et al., 2017). Some of the exons

558 used were flanked by introns, which are more variable, and were the focus of this capture. More

559 information about the capture and sequencing of UCE loci can be found in Faircloth et al.

560 (2012).

561 2.3. Phylogenetic analysis and haplotype networks

562 Phylogenetic analysis of the mtDNA genes using the complete dataset (cytb and ND2,

563 N=35) was performed using Bayesian Inference (BI) implemented in MrBayes 3.2.6 (Ronquist

564 et al., 2012). Both genes were concatenated and the best partition scheme and substitution

565 model were selected by PartitionFinder 2.1.1 using the Bayesian Information Criteria (BIC)

566 (Lanfear et al., 2016). We partitioned the genes by codon position, considering possible

567 saturation in the codon’s third position. Four parallel simultaneous runs were performed, for a

568 total of 4x107 generations, with trees sampled every 1000 generations. We discarded the first

569 10% of trees as burn-in after checking the ESS values of each run in Tracer 1.6 (Rambaut et

570 al., 2014). We used TCS v1.21 (Clement et al., 2000) to reconstruct haplotype networks.

571 2.3.1. UCE and exons assembly

572 Based on the results of the mtDNA, we selected eight samples for UCE sequencing

573 (Table 1). The raw data received from Rapid Genomics were processed using the Phyluce script

574 pack (Faircloth 2016). Sequences with adapter contamination, and those of low-quality, were

575 trimmed using illumiprocessor (Faircloth, 2013) and Trimmomatic (Bolger et al., 2014). After

576 the sequences were ‘cleaned’ we employed Trinity RNASeq assembler r201331110 (Grabherr

577 et al., 2011) to assemble the contigs using a de novo method. The contigs were then compared

578 with the UCE database to identify which UCE loci were sequenced. Since Trinity does not

579 recover information on heterozygote loci we performed a second round of assembly using the

580 contigs that were identified as a reference to map the clean reads back to it using the Bowtie2

581 (Langmead et al., 2009; Langmead and Salzberg 2012) plugin in Geneious R7.1 (Kearse et al.,

582 2012). The consensus sequence of each individual, derived from the reads, mapped back to each

583 reference, was called using a threshold of 75% with a depth of at least 5 reads. We then aligned

584 each locus using MAFFT (Katoh and Standley, 2013) with default options, and prepared the

585 input matrix for the subsequent analysis. To infer the phylogenetic relationship among all

586 samples we concatenated all the UCE loci and employed RAxML v8.2 (Stamatakis, 2014)

587 under a Maximum Likelihood analysis. Since we recovered almost all UCE loci for each

588 sample, we only used loci that were shared among all samples, with the final matrix having

589 2271 loci. This matrix was analyzed by running RAxML to search for the optimal tree, under

590 the fast hill climbing algorithm, and bootstraping was performed with the autoMRE algorithm

591 in the program.

592 The 98 exons targeted were from 47 different genes. Because some of the sequences

593 included intronic regions, which are prone to indels, de novo assembly was not an option.

594 Therefore, we mapped all the probes to the Galbula dea genome, identified the genes that were

595 targeted, and then used the whole gene-sequences to map the reads back following the same

596 approach that we used for UCE loci. Since recombination is not expected to happen inside one

597 gene, all exonic regions recovered belonging to the same gene were considered to be connected

598 in the species-tree (ST) analysis.

599 2.3.2. Mitochondrial genome assembly and time tree

600 As a byproduct of the UCE sequencing we also recovered the complete mtDNA genome.

601 We mapped all the Trinity contigs from each specimen to two reference mtDNA genomes from

602 representatives of close related families, the Downy Woodpecker, Dryobates pubescens (Aves,

603 Picidae; NC_027936.1), and the Ivory-billed Araçari, Pteroglossus azara (Aves,

604 Ramphastidae; DQ780882.1, Prum et al., 2015). After we identified the contigs from each

605 individual we used those contigs to map back the reads of that same specimen, again using

606 Bowtie2 to check for coverage depth. Incongruences found between reads and contigs were

607 checked manually. The complete mtDNA genomes were then aligned using MAFFT (Katoh

608 and Standley, 2013) under default options. The mtDNA genomes downloaded from GenBank

609 were used to import annotations. Coding regions were manually checked for codon translations,

610 and translated protein sequences were compared to check for frame shifts and stop codons. We

611 employed the concatenated coding regions in BEAST 1.8.2 (Drummond et al., 2012) to

612 estimate a time tree calibrated with the cytochrome b mutational rate of 0.0105 (normal

613 distribution, SD=0.0034) substitution.lineage-1.million years-1 (Weir and Schluter, 2008). The

614 best partition scheme and substitution model were selected by PartitionFinder 2.1.1 under the

615 Bayesian Information Criteria (BIC) (Lanfear et al., 2016). Two independent runs of 108

616 generations were performed sampling trees every 1000 generations. Convergence, posterior

617 distributions, and ESS values were checked in Tracer 1.6 (Rambaut et al., 2014).

618 2.3.3. Species-tree analysis

619 Considering the possibility that concatenation might result in highly supported but

620 inaccurate results (Kubatko and Degnan 2007; Weisrock et al., 2012, but see Gatesy and

621 Springer 2014), we performed a species-tree analysis, which infers the most likely species-tree

622 based on individual gene trees, using the StarBEAST2 (Ogilvie et al., 2017) template in the

623 BEAST v.2.4.6 package (Bouckaert et al., 2014). Even though StarBEAST2 was developed to

624 deal with huge amounts of data, we selected only the loci that had more than four parsimony

625 informative sites (PIS) among our samples. This latter step reduces the total time of analysis

626 and also avoids including loci lacking phylogenetic signal, which would create noise in the

627 analysis. We employed PartitionFinder2 (Lanfear et al., 2016) to check for the best partition

628 scheme and substitution model. Trees models were unlinked, except for exons from the same

629 gene, in which case we linked tree models across different partitions. We used a Yule model of

630 speciation, and ploidy was set to 2.0, unless genes were from the Z chromosome (in which case,

631 ploidy=1.5). We also included the complete mtDNA as a single locus, with ploidy=1.0.

632

633

634 3. Results

635 3.1. Sanger sequencing and haplotype networks

636 We sequenced 996 bp and 1013 bp, respectively, of the cytb and ND2 dataset. The best

637 partitioning scheme consisted of four partitions (cytb_pos1 = K80+I; ND2_pos2+cytb_pos2 =

638 HKY; ND2_pos3+cytb_pos3 = GTR+G; ND2_pos1 = HKY+I). The BI analysis, and the

639 haplotype network, recovered eight allopatric mtDNA lineages, six of them are well-supported

640 clades, while two of them are represented by a single individual each (Fig. 1). Although all

641 clades corresponding to the allopatric lineages had strong support, basal relationships among

642 them were poorly supported, the only exceptions being the sister relationships between Guiana

643 and Negro clades and between G. chalcothorax and the Madeira lineage of G. leucogastra.

644 Haplotypes networks were recovered using the concatenated matrix of cytb and ND2 in

645 which all missing data were discarded (1304 bp). Almost all networks were indicative of recent

646 population expansion, with little to no genetic diversity within lineages, except for the Madeira

647 lineage and for G. chalcothorax, for which we recovered a different haplotype for each

648 specimen. It is worth noting that samples from different banks of the Tapajós River are

649 separated by six mutational steps (Fig. 1: light and dark green), and that samples of G.

650 chalcothorax (Fig.1: light and dark brown) exhibit almost the same number of mutations among

651 them as they do in relation to the haplotypes from the Madeira lineage.

652 3.2. mtDNA genome and time tree

653 We recovered the complete mitochondrial genome from all samples sequenced for

654 UCEs. In contrast to our cytb+ND2 tree, the tree based on all the mtDNA coding genes was

655 highly supported (Fig. 2). Molecular dating indicates that diversification of the mtDNA lineages

656 started in the Middle Pleistocene, at about 1.5 million years ago (mya) (95%HPD = 2.4 - 0.75).

657 Although all nodes were recovered with high support, the first three splits occurred in a short

658 period of time, with short internodes, suggesting a rapid radiation among lineages from

659 southern, northern and western Amazonia (Fig. 2). The earliest divergence is suggested to have

660 been between populations separated by the Amazon River (Fig. 2). In both mtDNA analyses

661 (cytb+ND2 and mtDNA genome), G. chalcothorax was recovered as the sister-group to the G.

662 leucogastra lineage from the west bank of Madeira River, with their divergence dating of

663 around 0.74 mya (95%HPD = 1.21 – 0.38), therefore rendering G. leucogastra paraphyletic.

664 The lineages from the north bank of the Amazon River were also recovered as sister-groups,

665 and diverged roughly around the same time, 0.61 mya (95%HPD = 1 – 0.31). The most recent

666 divergence occurred between lineages separated by the Tapajós River at 0.28 mya (95%HPD =

667 0.47 – 0.13).

668 3.3. UCE sequencing, RAxML and Species trees

669 The complete UCE matrix, which included only those loci shared among all samples,

670 contained 2271 UCE loci, with mean locus length of 543.06 bp (see Table 1 for total number

671 of reads, Trinity contigs, UCE and exon loci recovered from each sample; for alignments, total

672 number of loci, and locus information, see Table 2). The concatenated RAxML tree recovered

673 G. chalcothorax as sister to all other samples of G. leucogastra with high bootstrap support

674 (p=100, Fig. 3). Thus, the earliest divergence is here suggested to have occurred between an

675 eastern and a western population, unlike the pattern suggested by the mitochondrial data. The

676 first split within G. leucogastra is between lineages north and south of the Amazon River,

677 followed by a split across the Madeira River (p=100), and then younger splits across the Tapajós

678 (p=96) and the Aripuanã (p=76).

679 For the StarBEAST species-tree we used 124 loci that had more than four parsimony

680 informative sites. The species-tree was identical in topology to the concatenated RAxML UCE

681 phylogeny, with some differences in statistical support, including two nodes without strong

682 support in the species-tree (p<0.95) (Fig. 3). In both the concatenated and the species-trees, we

683 found contrasting differences compared to the mtDNA genome tree. Besides the nature of the

684 earliest split in the complex, the most significant one is that the nuclear data recover G.

685 leucogastra as monophyletic and sister to G. chalcothorax with strong statistical support; the

686 mtDNA genome tree, in contrast, found G. leucogastra to be paraphyletic, and G. chalcothorax

687 as sister to the G. leucogastra Madeira lineage (Fig. 2). Furthermore, in the UCE trees the G.

688 leucogastra Aripuanã lineage (Fig. 3, dark pink) was strongly clustered with samples

689 distributed east of the Madeira River (Fig. 3).

690

691 4. Discussion

692 4.1. mtDNA and nuDNA incongruence

693 Historically the Purplish Jacamar (G. chalcothorax) was considered a subspecies of the

694 Bronzy Jacamar (G. leucogastra) (Peters, 1948; Haffer, 1974). Parker and Remsen (1987)

695 proposed that the two taxa be recognized as separate species based on their distinct phenotypes:

696 G. leucogastra is bronzy-green, with some suffused metallic blue, and a white belly, whereas

697 G. chalcothorax is tinged reddish-purple, and has a black belly with only the feathers tips being

698 white. Although these color characters seem to fluctuate across populations, G. chalcothorax is

699 distinctly larger than G. leucogastra (Haffer, 1974). Parker and Remsen (1987) also suggested

700 that Haffer (1974) did not recognize G. chalcothorax as a full species because of the supposition

701 they would interbreed if the two taxa came together, but they also noted (p. 98) that “the absence

702 of major river barriers between their ranges suggests that no interbreeding occurs or would

703 occur”.

704 The structure recovered by the mtDNA data within G. leucogastra, with five well

705 supported mtDNA clades, suggests that current taxonomic treatment misrepresents the diversity

706 within this species, which currently includes only two subspecies: G. l. leucogastra and G. l.

707 viridissima (Griscom and Greeway, 1941). Surprisingly, mtDNA data also revealed that all G.

708 leucogastra specimens from the Madeira clade, the geographically closest to G. chalcothorax

709 is sister to G. chalcothorax with high support, but with no shared haplotypes among species

710 (Fig. 1, 2). In contrast, the UCE concatenated RAxML tree as well as the UCE species-tree

711 recovered G. leucogastra and G. chalcothorax as monophyletic sister species, with the Madeira

712 lineage of G. leucogastra sister to G. leucogastra lineages from SE Amazonia (i.e. Aripuanã

713 and Tapajós lineages, Fig. 3). Multiple explanations have been proposed for conflict in

714 mitochondrial and nuclear histories (summarized in Table 1).

715 In the G. leucogastra / G. chalcothorax diversification mitochondrial capture may have

716 been influenced by the populational and ecological context of differentiation within WSE. After

717 the lineages in the south diverged east and west of the Madeira, the ancestral lineages of G.

718 chalcothorax and those of the Madeira met and hybridized. Although it is difficult to determine

719 when the contact started, it ended around 0.74 mya, as shown in our mtDNA time tree.

720 Moreover, even though the ranges of G. leucogastra and G. chalcothorax appear to be currently

721 allopatric (Tobias 2017), they approach each other between the Purus and Juruá rivers (Fig. 1).

722 Therefore, past gene flow may have been possible during drier climatic periods in SW

723 Amazonia (see below) (Mayle et al., 2004; Bush, 2017). mtDNA clades found within G.

724 leucogastra are more structured and differentiated than the clades found within the other WSE

725 birds, but all of them agree in recovering a well supported clade in northern Amazonia, and

726 with the Madeira being an important barrier in the south (Cracraft, 1985; Borges et al., 2012;

727 Ribas et al., 2012; Fernandes et al., 2013; Fernandes et al., 2014; Ferreira et al., 2017). The

728 maintenance of such structured mtDNA lineages may indicate that little or no gene flow is

729 present between the lineages, suggesting that the forest matrix is important for maintaining

730 allopatry.

731 Although mtDNA may reflect species boundaries (Hill, 2017), recent studies have

732 shown a number of cases in which apparent mtDNA paraphyly is not just derived from improper

733 taxonomy (McKay and Zink, 2010) but also from mtDNA introgression among adjacent

734 populations (see also Toews and Brelsford, 2012). For example, a mitochondrial sweep was

735 proposed in the certhiola complex in the Old World warbler genus Locustella, which is

736 comprised of three species (certhiola, ochotensis and pleskei). Phylogenetic studies using

737 mtDNA and nuDNA recovered conflicting results in that pleskei was paraphyletic relative to

738 certhiola and ochotensis on the mtDNA tree, whereas the nuDNA species-tree recovered

739 species monophyly (Drovetski et al., 2015). In addition, these authors found signs of

740 asymmetrical introgression, in which the species expanding its range (ochotensis) appears to

741 have invaded the species with a smaller ranges (pleskei), resulting in mtDNA introgression from

742 the species with large Ne to the one with smaller Ne. nuDNA introgression was in the opposite

743 direction, causing the paraphyly observed in the mtDNA tree (Drovetski et al., 2015). A similar

744 scenario was found in the eastern Australian rosellas (Platycercus, Shipham et al., 2017). The

745 three species of the subgenus Violania showed discordances between RADseq data and mtDNA

746 trees. Whereas the RAD trees recovered P. eximius as sister to the clade P. venustus and P.

747 adscitus, the mtDNA phylogeny recovered P. venustus as sister to P. adscitus and P. eximius.

748 Furthermore, when data for the isolated Tasmanian P. e. diemenensis were added, the same

749 relationship as those from the RAD data were recovered, suggesting that the subspecies of P.

750 eximius from the mainland (P. e. eximius) captured the mtDNA from P. adscitus.

751 Although these two documented cases represent examples of how mitochondrial sweeps

752 could occur in populations with known zones of hybridization, genetic and phenotypic data

753 suggest that there is no current hybrid zone corresponding to the conflict between UCE and

754 mtDNA reported here. Furthermore, isolation might lead to co-evolution of mitochondrial and

755 nuclear genes involved in cellular respiration, which could function as a post-zygotic barrier to

756 gene flow, due to Bateson-Dobzhansky-Muller Incompatibility (BDMI) (Orr, 1996). Given the

757 fragmented distribution of WSE in Amazonia, it is possible that the occupation of new patches,

758 or the fragmentation of previously continuous habitats into smaller patches due to landscape

759 evolution, followed by some time in allopatry, could lead to the mtDNA structure we observe

760 nowadays and consequent coevolution with nuclear background. All lineages we recovered for

761 this complex have deep structure in the mtDNA haplotypes, even between lineages with

762 adjacent distributions, such as Guiana and Negro lineages or Aripuanã and Tapajós lineages

763 (Fig. 3). Sex-biased traits, such as differential dispersal, hybrid fitness or mate choice are

764 commonly used to explain discordances between mtDNA and nuDNA (Excoffier, 2009; Toews

765 and Brelsford, 2012). However, in a recent review of this process, Bonnet et al. (2017)

766 simulated several scenarios and observed that the only way to have massive discordance in all

767 simulations, without detectable nuclear introgression, is when there is positive selection acting

768 on mitochondrial lineages. Surprisingly, Bonnet et al. (2017) were unable to detect

769 mitochondrial adaptive introgression using Tajima’s D and Fu’s Fs tests, reinforcing the

770 argument that these tests have low statistical power to detect adaptive introgression (Bonnet et

771 al., 2017). In addition, the mtDNA can accumulate deleterious mutations quickly, and in small

772 populations, drift could spread these deleterious mutations across the whole population in short

773 periods of time. Therefore, small populations may accumulate several deleterious mutations

774 and the “defective” mtDNA lineage can be supplanted by a foreign mtDNA lineage (Llopart et

775 al., 2014; Hulsey et al., 2016; Sloan et al., 2017). This hypothesis can be more plausible if

776 effects of the mtDNA sweep are more beneficial than the disadvantageous effects of

777 mitonuclear incompatibilities. This event, combined with the fact that both species occupy

778 different habitats inside the WSE, could explain why we observe the incongruences between

779 mtDNA and nuDNA.

780 4.2. Biogeography of WSE avifauna

781 In phylogeographic studies of the Black Manakin (Xenopipo atronitrens, Pipridae),

782 Capurucho et al. (2013) found the largest mtDNA divergences to correspond to populations

783 found across the Branco and Amazonas rivers. Similar results were observed for the Red-

784 shouldered Tanager (Tachyphonus phoenicius, Thraupidae, Matos et al., 2016), but with greater

785 isolation between opposite margins of the Amazon river. The divergence times estimated

786 between northern and southern lineages within X. atronitrens and T. phoenicius were 0.92 and

787 0.88 Ma, respectively, both slightly younger than the mean age estimate we obtained for the

788 first divergence on the mtDNA tree (~1.5 Ma, 95%HPD = 0.75 - 2.4) in G. leucogastra, but

789 with overlap of confidence intervals. Another WSE specialist studied, the Green-tailed

790 Goldenthroat (Polytmus theresiae, Trochilidae), showed no genetic structure, but exhibited

791 signs of recent population expansion (Matos et al., 2016). Signs of recent gene flow among

792 otherwise isolated populations of the aforementioned species contrast with the highly-

793 structured lineages recovered here. Although we found evidence for an ancient capture event

794 of mtDNA lineages, there is no evidence of current gene flow between G. leucogastra and G.

795 chalcothorax. Xenopipo atronitrens and G. leucogastra/chalcothorax are found in both WSE

796 and black-water flooded forest, T. phoenicius in WSE and savannas, and P. theresiae in WSE,

797 black-water flooded forest and savannas (Borges et al., 2016b). When compared to the other

798 WSE species, G. leucogastra and G. chalcothorax are the only exclusive insectivores, meaning

799 that they need not have as extensive foraging areas as do frugivores or nectarivores (Levey and

800 Stiles, 1992), and hence they are potentially more prone to isolation and differentiation (Burney

801 and Brumfield, 2009).

802 4.3. Evolution in the White-sand environments

803 White-sand environments cover an area of approximately 5% of the Amazon basin

804 (Adeney et al., 2016). They can be covered by different kinds of vegetations, from open

805 grasslands to different types of forest. In general, these communities grow on nutrient poor and

806 highly acidic soils, usually associated with quarzitic sand, even though some clay and silt can

807 also be found with varying amounts of organic matter (Adeney et al., 2016). This complex

808 environment, however, does not share a single history, since different patches of WSE may

809 have different geological origins (Prance and Schubart 1978; Frasier et al., 2008).

810 Podzolization, a natural process in which all nutrients are leached away from the top layers of

811 soil, leaving only sand (Sauer et al., 2007), appears to be a principal cause of in loco formation

812 of the white sand soils, especially in northeastern Amazonia (Nascimento et al., 2004). In

813 central, northwestern, and southern Amazon, white sand soils can be found as fluvial deposits

814 of ancient rivers (Roddaz et al., 2005), or abandoned ancient paleochannels (Latrubesse, 2002;

815 Cordeiro et al., 2016). In western Amazonia, white sand formations date to before the Andean

816 uplift, and are probably a result of westward rivers flowing from the Guiana and Brazilian

817 shields to the Pacific Ocean, during the Early Miocene (Hoorn, 1993). These sandy sediments

818 were reorganized and recycled multiple times within the basin during the Andean uplift, giving

819 patches of WSE in western Amazonia a very insular and scattered characteristic, especially

820 because most of these sediments are now covered by more recent clay-rich sediments derived

821 from the Andes. This mosaic of sediments is reflected in soils with distinct edaphic conditions,

822 which influence floristic composition that ultimately influences local bird communities

823 (Pomara et al., 2012).

824 Phylogeographic studies of WSE specialized birds suggest that they have recently

825 occupied the Amazonian WSE from east to west (Whitney and Alonso, 1998; Capurucho et al.,

826 2013; Matos et al., 2016). Also, most of WSE birds have sister groups inhabiting other open

827 vegetation habitats and not the adjacent Amazonian humid forest formations, such as terra-

828 firme or varzea (Rheindt et al., 2008; Capurucho et al., 2013; McGuire et al., 2014; Matos et

829 al., 2016). This suggests the colonization of Amazonian WSE by lineages that had already

830 evolved in open habitats, instead of ancestral lineages from neighboring humid forest. In this

831 sense, Galbula leucogastra and Galbula chalcothorax are unlike other WSE taxa since all other

832 Galbula species are found in forest habitats (Witt, 2004; Tobias, 2017).

833 The WSE were probably more widespread throughout the continent before Andean

834 uplift, thus extant WSE lineages of birds should be the ones resilient enough to endure the

835 reconfiguration of the Amazon basin (Campbell et al., 2006; Hoorn et al., 2010; Nogueira et

836 al., 2013). The pattern of more genetic diversity in the east we observe today should be then

837 related to the fact that during the Pleistocene climatic cycles, eastern Amazonia experienced

838 greater fluctuations in precipitation (Wang et al., 2017). Although these cyclical oscillations

839 were not enough to replace forest with savannas (Bush, 2017; Wang et al., 2017), they may

840 have affected forest structure (Barthe et al., 2017; Cowling et al., 2001). This could have

841 facilitated contact between different patches of WSE in the East, especially for birds that can

842 use black-water flooded forest, allowing them to expand their distribution and colonize

843 previously unoccupied patches of WSE. The paleoclimatic record (Cheng et al., 2013; Wang et

844 al., 2017) suggests that western Amazonia remained as humid as it is today throughout the

845 Pleistocene climatic oscillations, while eastern Amazonia experienced about 42% less rainfall

846 when compared with modern values (Wang et al., 2017). Even though eastern Amazonia

847 experienced drier climate, there is no evidence suggesting replacement of forest by savanna

848 (Bush et al., 2017). So, the existence of WSE in western Amazonia would occur only in

849 scattered patches in recycled quartzite soils reminiscent of ancient fluvial deposits (Hoorn,

850 1993), or as fluvial deposits of ancient rivers (Latrubesse, 2002). This scenario of contracting

851 WSE areas in the west, because of recycling soils during Andean uplift, and WSE expansion in

852 the east, especially during Pleistocene climatic cycles, would probably explain the smaller Ne

853 of G. chalcothorax and the mtDNA capture from G. leucogastra as it expanded its distribution

854 during dry cycles.

855

856 5. Conclusion

857 Here we shown an instance of clear discordance between phylogenetic relationships

858 recovered using mtDNA and nuclear data. Interestingly, nuclear data agrees with current

859 taxonomy, which is based on phenotypic patterns, while the mtDNA relationships seem to be

860 related to an old event of mtDNA capture. The capture event relates to what is currently known

861 about the distinct biogeographical histories of WSE in Eastern and Western Amazonia. While

862 these results raise important issues about apparent mtDNA paraphyly of taxa and the

863 straightforward use of mtDNA relationships in taxonomy, they also show that interesting

864 biogeographic histories can be uncovered when enough data is available, allowing for a

865 comparison with mtDNA. This will be an important contribution of NGS for studies for recent

866 speciation and taxonomy.

867

868

869 Acknowledgements

870 We thank the curator and curatorial assistants of the Academy of Natural Science of Drexel

871 University, Philadelphia, USA (ANSP); Field Museum of Natural History, Chicago, USA

872 (FMNH); Instituto Nacional de Pesquisas da Amazônia, Manaus, Brazil (INPA); Lousiana State

873 University Museum of Natural Science, Baton Rouge, USA (LSUMZ); and Museu Paraense

874 Emílio Goeldi, Belém, Brazil (MPEG), for borrowing tissue samples under their care. We thank

875 S. W. Cardiff and N. Rice for helping us with LSUMZ and ANSP specimens, respectively. We

876 are also grateful for all collectors involved in the fieldwork throughout Amazon that make this

877 paper possible. We thank J. M. G. Capurucho and S. H. Borges for early inputs on this paper.

878

879 Funding

880 Support to M.F.’s graduate research was provided by CAPES PhD fellowship, and CAPES

881 PDSE fellowship (# 88881.133440/2016-01), support also from the AMNH Frank M. Chapman

882 Memorial Fund. Support to A.M.F. during his post-doc studies was provided by CNPq

883 (#500488/2012-6). Laboratory and Sanger sequencing costs were partly covered by grants to

884 A. Aleixo (CNPq # 471342/2011-4 and FAPESPA # ICAAF 023/2011) and A.Antonelli from

885 the European Research Council under the European Union’s Seventh Framework Programme

886 (FP/2007-2013, ERC Grant Agreement n. 331024), the Knut and Alice Wallenberg Foundation

887 through a Wallenberg Academy Fellowship, the Swedish Research Council (2015-04857), and

888 the Swedish Foundation for Strategic research. A.Aleixo, C.C.R., J.M.B., J.C. and M.F. also

889 thanks the grant Dimensions US-Biota-São Paulo: Assembly and evolution of the Amazon biota

890 and its environment: an integrated approach, co-funded by the US National Science Fundation

891 (NSF DEB 1241056) to J.C. and the Fundação de Amparo à Pesquisa do Estado de São Paulo

892 (FAPESP grant #2012/50260-6) to Lucia Lohmann. A. Aleixo and C.C.R. are supported by 27

893 CNPq research productivity fellowships. The authors acknowledge the National Laboratory for

894 Scientific Computing (LNCC/MCTI, Brazil) for providing HPC resources of the SDumont

895 supercomputer, which have contributed to the research results reported within this paper.

896

897 References 898 899 Adeney, J.M., Christensen, N.L., Vicentini, A., Cohn-Haft, M., 2016. White-sand Ecosystems 900 in Amazonia. Biotropica 48, 7-23. 901 Alonso, J.A., Whitney, B.M., 2001. A new Zimmerius tyrannulet (Aves: Tyrannidae) from 902 white sand forests of northern Amazonian Peru. Wilson Bulletin 113, 1-9. 903 Anderson, A.B., 1981. White-sand vegetation of Brazilian Amazonia. Biotropica 13, 199-210. 904 Avise, J.C., 2009. Phylogeography: retrospect and prospect. J Biogeogr 36, 3-15. 905 Avise, J.C., Arnold, J., Ball, R.M., Bermingham, E., Lamb, T., Neigel, J.E., Reeb, C.A., 906 Saunders, N.C., 1987. Intraspecific phylogeography: The mitochondrial DNA bridge 907 between population genetics and systematics. Ann Rev Ecol Syst 18, 489-522. 908 Barthe, S., Binelli, G., Hérault, B., Scotti-Saintagne, C., Sabatier, D., Scotti, I., 2017. Tropical 909 rainforest that persisted: inferences from the Quaternary demographic history of eight tree 910 species in the Guiana Shield. Mol Ecol 26, 1161-1174. 911 Bolger, A.M., Lohse, M., Usadel, B., 2014. Trimmomatic: a flexible trimmer for Illumina 912 sequence data. Bioinformatics 30, 2114-2120. 913 Bonnet, T., Leblois, R., Rousset, F., Crochet, P.-A., 2017. A reassessment of explanations for 914 discordant introgressions of mitochondrial and nuclear genomes. Evolution 71, 2140- 915 2158. 916 Borges, S.H., 2003. Species poor but distinct: bird assemblages in white sand vegetation in Jaú 917 National Park, Brazilian Amazon. Ibis 146, 114-124. 918 Borges, S.H., Da Silva, J.M.C., 2012. A New Area of Endemism for Amazonian Birds in the 919 Rio Negro Basin. Wilson J Ornithol 124, 15-23. 920 Borges, S.H., Cornelius, C., Moreira, M., Ribas, C.C., Cohn-Haft, M., Capurucho, J.M.G., 921 Vargas, C., Almeida, R., 2016a. Bird communities in Amazonian white-sand vegetation 922 patches: effects of landscape configuration and biogeographic context. Biotropica 48, 923 121-131. 924 Borges, S.H., Cornelius, C., Ribas, C., Almeida, R., Guilherme, E., Aleixo, A., Dantas, S., 925 Santos, M.P.D., Moreira, M., 2016b. What is the avifauna of Amazonian white-sand 926 vegetation. Bird Conserv Int 26, 192-204. 927 Bouckaert, R., Heled, J., Kuhnert, D., Vaughan, T., Wu, C.H., Xie, D., Suchard, M.A., 928 Rambaut, A., Drummond, A.J., 2014. BEAST 2: a software platform for Bayesian 929 evolutionary analysis. PLoS computational biology 10, e1003537. 930 Brown, K.S., Benson, W.W., 1977. Evolution in Modern Amazonian Non-Forest Islands: 931 Heliconius hermathena. Biotropica 9, 95-117. 932 Burney, C.W., Brumfield, R.T., 2009. Ecology predicts levels of genetic differentiation in 933 neotropical birds. Am Nat 174, 358-368. 934 Bush, M.B., 2017. Climate science: The resilience of Amazonian forests. Nature 541, 167-168. 935 Campbell, K.E., Frailey, C.D., Romero-Pittman, L., 2006. The Pan-Amazonian Ucayali 936 Peneplain, late Neogene sedimentation in Amazonia, and the birth of the modern Amazon 937 River system. Palaeogeogr Palaeoclimatol Palaeoecol 239, 166-219.

938 Capurucho, J.M.G., Cornelius, C., Borges, S.H., Cohn-Haft, M., Aleixo, A., Metzger, J.P., 939 Ribas, C.C., 2013. Combining phylogeography and landscape genetics of Xenopipo 940 atronitens (Aves: Pipridae), a white sand campina specialist, to understand Pleistocene 941 landscape evolution in Amazonia. Biol J Linn Soc 110, 60-76. 942 Carling, M.D., Brumfield, R.T., 2008. Haldane's rule in an avian system: using cline theory and 943 divergence population genetics to test for differential introgression of mitochondrial, 944 autosomal, and sex-linked loci across the Passerina bunting hybrid zone. Evolution 62, 945 2600-2615. 946 Cheng, H., Sinha, A., Cruz, F.W., Wang, X., Edwards, R.L., d'Horta, F.M., Ribas, C.C., Vuille, 947 M., Stott, L.D., Auler, A.S., 2013. Climate change patterns in Amazonia and biodiversity. 948 Nature Commun 4, 1411. 949 Clement, M., Posada, D., Crandall, K.A., 2000. TCS: a computer program to estimate gene 950 genealogies. Mol Ecol 9, 1657-1659. 951 Cohn-Haft, M., 2008. Estudos temático de impacto da BR-319: Aves. Internal Report. INPA, 952 Manaus, Brazil. 953 Cohn-Haft, M., Bravo, G.A., 2013. A new species of Herpsilochmus antwren from west of the 954 Rio Madeira in Amazonian Brazil. In: del Hoyo, J., Elliot, A., Sargatal, J., Christie, D. 955 (Eds.), Handbook of the birds of the world. Special volume. New species and global 956 index. Lynx Ediciones, Barcelona, Spain, pp. 272-276. 957 Cohn-Haft, M., Santos, M.A., Fernandes, A.M., Ribas, C., 2013. A new species of Cyanocorax 958 jay from savannas of the central Amazon. In: del Hoyo, J., Elliot, A., Sargatal, J., Christie, 959 D. (Eds.), Handbook of the birds of the world. Special volume. New species and global 960 index. Lynx Ediciones, Barcelona, Spain, pp. 306-310. 961 Cordeiro, C.L.O., Rossetti, D.F., Gribel, R., Tuomisto, H., Zani, H., Ferreira, C.A.C., Coelho, 962 L., 2016. Impact of sedimentary processes on white-sand vegetation in an Amazonian 963 megafan. J Trop Ecol 32, 498-509. 964 Cowling, S.A., Maslin, M.A., Sykes, M.T., 2001. Paleovegetation Simulations of Lowland 965 Amazonia and Implications for Neotropical Allopatry and Speciation. Quat Res 55, 140- 966 149. 967 Cracraft, J., 1985. Historical biogeography and patterns of differentiation within the South 968 American Avifauna: Areas of Endemism. Ornithol Monogr 36, 49-84. 969 Daly-Engel, T.S., Seraphin, K.D., Holland, K.N., Coffey, J.P., Nance, H.A., Toonen, R.J., 970 Bowen, B.W., 2012. Global phylogeography with mixed-marker analysis reveals male- 971 mediated dispersal in the endangered scalloped hammerhead shark (Sphyrna lewini). PloS 972 one 7, e29986. 973 Drovetski, S.V., Semenov, G., Red'kin, Y.A., Sotnikov, V.N., Fadeev, I.V., Koblik, E.A., 2015. 974 Effects of asymmetric nuclear introgression, introgressive mitochondrial sweep, and 975 purifying selection on phylogenetic reconstruction and divergence estimates in the Pacific 976 clade of Locustella warblers. PloS one 10, e0122590. 977 Drummond, A.J., Suchard, M.A., Xie, D., Rambaut, A., 2012. Bayesian phylogenetics with 978 BEAUti and the BEAST 1.7. Mol Biol Evol 29, 1969-1973. 979 Excoffier, L., Foll, M., Petit, R.J., 2009. Genetic Consequences of Range Expansions. Ann Rev 980 Ecol Evol Syst 40, 481-501. 981 Faircloth, B.C., 2013. illumiprocessor: a trimmomatic wrapper for parallel adapater and quality 982 trimming. http://dx.doi.org/10.6079/J9ILL. 983 Faircloth, B.C., 2016. PHYLUCE is a software package for the analysis of conserved genomic 984 loci. Bioinformatics 32, 786-788. 985 Faircloth, B.C., McCormack, J.E., Crawford, N.G., Harvey, M.G., Brumfield, R.T., Glenn, 986 T.C., 2012. Ultraconserved elements anchor thousands of genetic markers spanning 987 multiple evolutionary timescales. Syst Biol 61, 717-726. 29

988 Fernandes, A.M., 2013. Fine-scale endemism of Amazonian birds in a threatened landscape. 989 Biodivers Conserv 22, 2683-2694. 990 Fernandes, A.M., Cohn-Haft, M., Hrbek, T., Farias, I.P., 2014. Rivers acting as barriers for bird 991 dispersal in the Amazon. Rev Bras Ornitol 22, 363-373. 992 Ferreira, M., Aleixo, A., Ribas, C.C., Santos, M.P.D., 2017. Biogeography of the Neotropical 993 genus Malacoptila (Aves: Bucconidae): the influence of the Andean orogeny, Amazonian 994 drainage evolution and palaeoclimate. J Biogeogr 44, 748-759. 995 Fine, P.V.A., Baraloto, C., 2016. Habitat endemism in white-sand forests: Insights into the 996 mechanisms of lineage diversification and community assembly of the Neotropical Flora. 997 Biotropica 48, 24-33. 998 Fine, P.V.A., García-Villacorta, R., Pitman, N.C.A., Mesones, I., Kembel, S.W., 2010. A 999 Floristic Study of the White-Sand Forests of Peru. Ann Missouri Bot Gard 97, 283-305. 1000 Frasier, C.L., Albert, V.A., Struwe, L., 2008. Amazonian lowland, white sand areas as ancestral 1001 regions for South American biodiversity: Biogeographic and phylogenetic patterns in 1002 Potalia (Angiospermae: Gentianaceae). Org Divers Evol 8, 44-57. 1003 Funk, D.J., Omland, K.E., 2003. Species-level paraphyly and polyphyly: Frequency, causes, 1004 and consequences, with insights from animal mitochondrial DNA. Ann Rev Ecol Evol 1005 Syst 34, 397-423. 1006 Gatesy, J., Springer, M.S., 2014. Phylogenetic analysis at deep timescales: unreliable gene 1007 trees, bypassed hidden support, and the coalescence/concatalescence conundrum. Mol 1008 Phylogenet Evol 80, 231-266. 1009 Grabherr, M.G., Haas, B.J., Yassour, M., Levin, J.Z., Thompson, D.A., Amit, I., Adiconis, X., 1010 Fan, L., Raychowdhury, R., Zeng, Q., Chen, Z., Mauceli, E., Hacohen, N., Gnirke, A., 1011 Rhind, N., di Palma, F., Birren, B.W., Nusbaum, C., Lindblad-Toh, K., Friedman, N., 1012 Regev, A., 2011. Full-length transcriptome assembly from RNA-Seq data without a 1013 reference genome. Nature Biotechnol 29, 644-652. 1014 Griscom, L., Greenway, J.C., 1941. Birds of Lower Amazonia. Bull Mus Comp Zool 88, 83- 1015 344. 1016 Haffer, J., 1974. Avian speciation in Tropical South America. Nuttall Ornithological Club, 1017 Harvard University, Cambridge. 1018 Harvey, M.G., Smith, B.T., Glenn, T.C., Faircloth, B.C., Brumfield, R.T., 2016. Sequence 1019 Capture versus Restriction Site Associated DNA Sequencing for Shallow Systematics. 1020 Syst Biol 65, 910-924. 1021 Harvey, M.G., Aleixo, A., Ribas, C., Brumfield, R.T., 2017. Habitat association predicts genetic 1022 diversity and population divergence in Amazonian birds. American Naturalist Early 1023 View. 1024 Hill, G.E., 2017. The mitonuclear compatibility species concept. Auk 134, 393-409. 1025 Hoorn, C., 1993. Marine incursions and the influence of Andean tectonics on the Miocene 1026 depositional history of northwestern Amazonia: results of a palynostratigraphic study. 1027 Palaeogeogr Palaeoclimatol Palaeoecol 105, 267-309. 1028 Hulsey, C.D., Bell, K.L., García-de-León, F.J., Nice, C.C., Meyer, A., 2016. Do relaxed 1029 selection and habitat temperature facilitate biased mitogenomic introgression in a 1030 narrowly endemic fish? Ecol Evol 6, 3684-3698. 1031 Katoh, K., Standley, D.M., 2013. MAFFT Multiple Sequence Alignment Software Version 7: 1032 Improvements in Performance and Usability. Mol Biol Evol 30, 772-780. 1033 Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., 1034 Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., Drummond, 1035 A., 2012. Geneious Basic: An integrated and extendable desktop software platform for 1036 the organization and analysis of sequence data. Bioinformatics 28, 1647-1649. 1037 Knowles, L.L., 2009. Statistical Phylogeography. Ann Rev Ecol Evol Syst 40, 593-612. 30

1038 Kubatko, L.S., Degnan, J.H., 2007. Inconsistency of phylogenetic estimates from concatenated 1039 data under coalescence. Syst Biol 56, 17-24. 1040 Lanfear, R., Frandsen, P.B., Wright, A.M., Senfeld, T., Calcott, B., 2017. PartitionFinder 2: 1041 New methods for selecting partitioned models of evolution for molecular and 1042 morphological phylogenetic analyses. Mol Biol Evol 34, 772-773. 1043 Langmead, B., Salzberg, S.L., 2012. Fast gapped-read alignment with Bowtie 2. Nature 1044 Methods 9, 357-359. 1045 Langmead, B., Trapnell, C., Pop, M., Salzberg, S.L., 2009. Ultrafast and memory-efficient 1046 alignment of short DNA sequences to the human genome. Genome Biol 10, R25. 1047 Laranjeiras, T.O., Naka, L.N., Bechtold, C.L., Costa, T.V.V., Andretti, C.B., Cerqueira, M.C., 1048 Torres, M.d.F., Rodrigues, G.L., Santos, M.P.D., Vargas, C.F., Pacheco, A.M.F., Sardelli, 1049 C.H., Mazar-Barnett, J., Cohn-Haft, M., 2014. The avifauna of Viruá National Park, 1050 Roraima, reveals megadiversity in northern Amazonia. Rev Bras Ornitol 22, 138-171. 1051 Latrubesse, E.M., 2002. Evidence of Quaternary palaeohydrological changes in middle 1052 Amazonia: The Aripuanã-Roosevelt and Jiparaná "fans". Z Geomorph N F 129, 61-72. 1053 Latrubesse, E.M., Franzinelli, E., 2002. The Holocene alluvial plain of the middle Amazon 1054 River, Brazil. Geomorphology 44, 241-257. 1055 Levey, D.J., Stiles, F.G., 1992. Evolutionary precursors of long-distance migration: resource 1056 availability and movement patterns in Neotropical landbirds. Am Nat 140, 447-476. 1057 Llopart, A., Herrig, D., Brud, E., Stecklein, Z., 2014. Sequential adaptive introgression of the 1058 mitochondrial genome in Drosophila yakuba and Drosophila santomea. Mol Ecol 23, 1059 1124-1136. 1060 Manthey, J.D., Campillo, L.C., Burns, K.J., Moyle, R.G., 2016. Comparison of Target-Capture 1061 and Restriction-Site Associated DNA Sequencing for Phylogenomics: A Test in 1062 Cardinalid Tanagers (Aves, Genus: Piranga). Syst Biol 65, 640-650. 1063 Matos, M.V., Borges, S.H., d'Horta, F.M., Cornelius, C., Latrubesse, E., Cohn-Haft, M., Ribas, 1064 C.C., 2016. Comparative Phylogeography of two bird species, Tachyphonus phoenicius 1065 (Thraupidae) and Polytmus theresiae (Trochilidae), specialized in Amazonian White- 1066 sand vegetation. Biotropica 48, 110-120. 1067 Mayle, F.E., Beerling, D.J., Gosling, W.D., Bush, M.B., 2004. Responses of Amazonian 1068 ecosystems to climatic and atmospheric carbon dioxide changes since the last glacial 1069 maximum. Philos Trans R Soc Lond B Biol Sci 359, 499-514. 1070 McCormack, J.E., Hird, S.M., Zellmer, A.J., Carstens, B.C., Brumfield, R.T., 2013. 1071 Applications of next-generation sequencing to phylogeography and phylogenetics. 1072 Molecular phylogenetics and evolution 66, 526-538. 1073 McGuire, J.A., Witt, C.C., Remsen, J.V., Jr., Corl, A., Rabosky, D.L., Altshuler, D.L., Dudley, 1074 R., 2014. Molecular phylogenetics and the diversification of hummingbirds. Curr Biol 1075 24, 910-916. 1076 McKay, B.D., Zink, R.M., 2010. The causes of mitochondrial DNA gene tree paraphyly in 1077 birds. Mol Phylogenet Evol 54, 647-650. 1078 Metzker, M.L., 2010. Sequencing technologies - the next generation. Nat Rev Genet 11, 31-46. 1079 Moyle, R.G., Oliveros, C.H., Andersen, M.J., Hosner, P.A., Benz, B.W., Manthey, J.D., 1080 Travers, S.L., Brown, R.M., Faircloth, B.C., 2016. Tectonic collision and uplift of 1081 Wallacea triggered the global songbird radiation. Nature Commun 7, 12709. 1082 Nascimento, N.R., Bueno, G.T., Fritsch, E., Herbillon, A.J., Allard, T., Melfi, A.J., Astolfo, R., 1083 Boucher, H., Li, Y., 2004. Podzolization as a deferralitization process: a study of an 1084 Acrisol-Podzol sequence derived from Palaeozoic sandstones in the northern upper 1085 Amazon Basin. Eur J Soil Sci 55, 523-538.

1086 Nogueira, A.C.R., Silveira, R., Guimarães, J.T.F., 2013. Neogene–Quaternary sedimentary and 1087 paleovegetation history of the eastern Solimões Basin, central Amazon region. J South 1088 Am Earth Sci 46, 89-99. 1089 Ogilvie, H.A., Bouckaert, R.R., Drummond, A.J., 2017. StarBEAST2 brings faster species tree 1090 inference and accurate estimates of substitution rates. Mol Biol Evol 34, 2101-2114. 1091 Orr, H.A., 1996. Dobzhansky, Bateson, and the genetics of speciation. Genetics 144, 1331- 1092 1335. 1093 Parker, T.A., Remsen, J.V., 1987. Fifty-two Amazonian bird species new to Bolivia. Bull Brit 1094 Orn Cl 107, 94-107. 1095 Peters, J.L., 1948. Check-list of birds of the world. Harvard University Press, Cambridge, UK. 1096 Pomara, L.Y., Ruokolainen, K., Tuomisto, H., Young, K.R., 2012. Avian composition co-varies 1097 with floristic composition and soil nutrient concentration in Amazonian upland forests. 1098 Biotropica 44, 545-553. 1099 Prance, G.T., Schubart, H.O.R., 1978. Notes on the vegetation of Amazonia I. A preliminary 1100 note on the origin of the open white sand campinas of the lower Rio Negro. Brittonia 30, 1101 60-63. 1102 Prum, R.O., Berv, J.S., Dornburg, A., Field, D.J., Townsend, J.P., Lemmon, E.M., Lemmon, 1103 A.R., 2015. A comprehensive phylogeny of birds (Aves) using targeted next-generation 1104 DNA sequencing. Nature 526, 569-573. 1105 Rambaut, A., Suchard, M.A., Xie, D., Drummond, A.J., 2014. Tracer v1.6. 1106 http://beast.bio.ed.ac.uk/Tracer 1107 Rheindt, F.E., Christidis, L., Norman, J.A., 2008. Habitat shifts in the evolutionary history of a 1108 Neotropical flycatcher lineage from forest and open landscapes. BMC Evol Biol 8. 1109 Rheindt, F.E., Edwards, S.V., 2011. Genetic Introgression: An Integral but neglected 1110 component of speciation in birds. Auk 128, 620-632. 1111 Ribas, C.C., Aleixo, A., Nogueira, A.C., Miyaki, C.Y., Cracraft, J., 2012. A 1112 palaeobiogeographic model for biotic diversification within Amazonia over the past three 1113 million years. Proc Biol Sci 279, 681-689. 1114 Roddaz, M., Baby, P., Brusset, S., Hermoza, W., Maria Darrozes, J., 2005. Forebulge dynamics 1115 and environmental control in Western Amazonia: The case study of the Arch of Iquitos 1116 (Peru). Tectonophysics 399, 87-108. 1117 Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Hohna, S., Larget, B., 1118 Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes 3.2: efficient Bayesian 1119 phylogenetic inference and model choice across a large model space. Syst Biol 61, 539- 1120 542. 1121 Rossetti, D.F., Bertani, T.C., Zani, H., Cremon, E.H., Hayakawa, E.H., 2012. Late Quaternary 1122 sedimentary dynamics in Western Amazonia: Implications for the origin of open 1123 vegetation/forest contrasts. Geomorphology 177-178, 74-92. 1124 Sambrook, J., Russel, D.W., 2001. Molecular Cloning: A laboratory manual. Cold Spring 1125 Harbor Laboratory Press, Cold Spring Harbor, New York. 1126 Sauer, D., Sponagel, H., Sommer, M., Giani, L., Jahn, R., Stahr, K., 2007. Podzol: Soil of the 1127 year 2007. A review on its genesis, occurrence, and functions. J Plant Nutr Soil Sci 170, 1128 581-597. 1129 Shipham, A., Schmidt, D.J., Joseph, L., Hughes, J.M., 2015. Phylogenetic analysis of the 1130 Australian rosella parrots (Platycercus) reveals discordance among molecules and 1131 plumage. Mol Phylogenet Evol 91, 150-159. 1132 Shipham, A., Schmidt, D.J., Joseph, L., Hughes, J.M., 2017. A genomic approach reinforces a 1133 hypothesis of mitochondrial capture in eastern Australian rosellas. Auk 134, 181-192. 1134 Sloan, D.B., Havird, J.C., Sharbrough, J., 2017. The on-again, off-again relationship between 1135 mitochondrial genomes and species boundaries. Mol Ecol 26, 2212-2236 32

1136 Smith, B.T., Harvey, M.G., Faircloth, B.C., Glenn, T.C., Brumfield, R.T., 2014. Target capture 1137 and massively parallel sequencing of ultraconserved elements for comparative studies at 1138 shallow evolutionary time scales. Syst Biol 63, 83-95. 1139 Sorenson, M.D., Ast, J.C., Dimcheff, D.E., Yuri, T., Mindell, D.P., 1999. Primers for a PCR- 1140 based approach to mitochondrial genome sequencing in birds and other vertebrates. Mol 1141 Phylogenet Evol 12, 105-114. 1142 Stamatakis, A., 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of 1143 large phylogenies. Bioinformatics 30, 1312-1313. 1144 Stotz, D.F., Fitzpatrick, J.W., Parker, T.A., Moskovitz, D.K., 1996. Neotropical birds: Ecology 1145 and conservation. University of Chicago Press, Chicago, IL. 1146 Tobias, J.A., 2017. Jacamars (Galbulidae). In: del Hoyo, J., Elliot, A., Sargatal, J. (Eds.), 1147 Handbook of the birds of the world. Lynx Ediciones, Barcelona. (retrieved from 1148 http://www.hbw.com/node/52281 on 15 August 2017) 1149 Toews, D.P., Brelsford, A., 2012. The biogeography of mitochondrial and nuclear discordance 1150 in animals. Mol Ecol 21, 3907-3930. 1151 Vicentini, A., 2016. The Evolutionary History of Pagamea (Rubiaceae), a White-sand 1152 Specialist Lineage in Tropical South America. Biotropica 48, 58-69. 1153 Vriesendorp, C., Pitman, N.C., Rojas Moscoso, J.I., Pawlak, B.A., Chavéz, L.R., Méndez, L.C., 1154 Collantes, M.V., Rimachi, E.P.F., 2006. Perú: Matsés. Rapid biological inventories report 1155 16. The Field Museum, Chicago. 1156 Wang, N., Kimball, R.T., Braun, E.L., Liang, B., Zhang, Z., 2017. Ancestral range 1157 reconstruction of Galliformes: the effects of topology and taxon sampling. J Biogeogr 44, 1158 122-135. 1159 Weir, J.T., Schluter, D., 2008. Calibrating the avian molecular clock. Mol Ecol 17, 2321-2328. 1160 Weisrock, D.W., Smith, S.D., Chan, L.M., Biebouw, K., Kappeler, P.M., Yoder, A.D., 2012. 1161 Concatenation and concordance in the reconstruction of mouse lemur phylogeny: an 1162 empirical demonstration of the effect of allele sampling in phylogenetics. Mol Biol Evol 1163 29, 1615-1630. 1164 Whitney, B.M., Alonso, J.A., 1998. A new Herpsilochmus antwren (Aves: Thamnophilidae) 1165 from Northern Amazonian Peru and adjacent Ecuador: The role of edaphic heterogeneity 1166 of terra firme forest. Auk 115, 559-576. 1167 Whitney, B.M., Alonso, J.A., 2005. A new species of Gnatcatcher from white-sand forests of 1168 northern Amazonian Peru with revision of the Polioptila guianensis complex. Wilson 1169 Bull 117, 113-127. 1170 Witt, C.C., 2004. Rates of molecular evolution and their application to Neotropical Avian 1171 Biogeography. Department of Biological Sciences. Louisiana State University, p. 132 p. 1172 Zink, R.M., Barrowclough, G.F., 2008. Mitochondrial DNA under siege in avian 1173 phylogeography. Mol Ecol 17, 2107-2121. 1174

1175 Author contibutions 1176 M.F. and A.M.F. developed the sampling plan, extracted DNA and sequenced all samples. M.F. 1177 performed all analysis. A.A.P., A.A., U.O., J.M.B., J.C. and C.C.R. were involved in intellectual merit, 1178 funding, and writing. All authors participated in writing the manuscript. 1179 Supporting information 1180 Additional supporting information may be found in the online version of this article. 1181 Table S1 Supplementary details of individuals. 1182

1183 Table 1 - Samples used for UCE sequencing, their voucher numbers, general locality, number of clean reads 1184 after Illumiprocessor, number of contigs assembled by Trinity, and total UCE loci recovered from Trinity. 1185 Species Museum voucher Locality Clean reads Trinity contigs UCE loci G. chalcothorax LSUMZ B2803 N of Napo River, Iquitos, Peru 1,524,126 6,537 2,230 G. leucogastra INPA A4182 145 Km WWS of Apuí, AM, Brazil 2,540,148 12,163 2,269 G. leucogastra INPA A4672 Right bank of Jatapú River, AM, Brazil 2,209,895 9,491 2,263 G. leucogastra LSUMZ B35619 Arapiuns River, PA, Brazil 4,394,658 10,853 2,246 G. leucogastra LSUMZ B9608 Nicolás Suarez, Pando, Bolívia 1,677,988 5,074 1,928 G. leucogastra MPEG 59360 Novo Airão, AM, Brazil 2,372,950 6,026 1,957 G. leucogastra MPEG 75618 Right bank of Tapajós River, PA, Brazil 1,346,149 6,117 2,263 G. leucogastra MPEG 73685 Novo Aripuanã, AM, Brazil 1,466,240 6,896 2,227 G. albirostris INPA A064 Amazonas, Brazil 2,809,416 16,718 2,256 1186 1187 Table 2 – Summary of each method, including number of loci, total length, mean length size of each loci, 1188 minimum and maximum length, number of Parsimony Informative sites. 1189

Method Complete Exons Species Tree†

Number of loci 2271 47 124 Total lenght (bp) 1,233,287 47,580 80,085 Mean lenght size (bp) 543.06 849.64 645.85 Min - Max lenght (bp) 118 – 1,305 182 - 3093 347 – 3093 Number of PI sites (mean) 2003 (0.88) 190 (3.39) 744 (6) 1190 †without the mtDNA 1191 1192 Table 3 – Possible causes of conflict in mitochondrial and nuclear DNA histories. 1193 Inferred process Reference Funk and Omland, 2003; McKay and Zink, 2010; Incomplete lineage sorting Zink and Barrowclough, 2008 Incomplete sampling Shipham et al., 2015, 2017 Improper taxonomy McKay and Zink, 2010 Adaptive introgression Bock et al., 2014; Dobler et al., 2014 Bonnet et al., 2017; Daly-Engel et al., 2012; Demography or Sex-biased traits Rheindt and Edwards, 2011; Sloan et al., 2017 1194

1195 Figure 1 - Map of sequenced individuals, phylogenetic Bayesian tree recovered, and haplotype networks. The 1196 colors in the tree, map and networks are correspondent, and the tree and networks are based on two mtDNA genes 1197 (2009 bp, cytb and ND2). Posterior probabilities obtained at each node are indicated on the tree, red circles 1198 represent pp=1. The brown labeled points are G. chalcothorax, all other lineages are G. leucogastra. Terminal 1199 names in red are samples used in the UCE analysis. 1200

1201 1202

1203 Figure 2 – Chronogram recovered by BEAST using all mtDNA coding genes with a calibration derived from the 1204 mutational rate of the cytb gene (Weir and Schluter 2008). Posterior probabilities obtained at each node are 1205 indicated in the tree, red circles represents pp>98, associated confidence interval (95% HPD) for diversification 1206 time (blue bar), and the median time of divergence. Colors are correspondent with Figure 1. 1207

1208 1209

1210 Figure 3 - Comparison between the concatenated UCE RAxML tree (left) and the StarBEAST2 species tree (right). 1211 Bootstrap support for the RAxML tree, and the posterior probability for the StarBEAST species tree, is show near 1212 the nodes. Colors are correspondent with Figure 1. 1213

1214 1215 1216 1217

1218 1219 1220 1221 1222 1223 1224 1225 1226 1227

1228 Capítulo 2

1229 1230 1231 Ferreira, M.; Aleixo, A.; Bates, J. M.; Cracraft, J.; 1232 Ribas, C. C. Phylogenomics of trogons (Aves: 1233 Trogonidae) shed light on the Quaternary 1234 biogeography of tropical forests and the connections 1235 between Asia, North and South America. Manuscrito 1236 formatado para Molecular Biology and Evolution 1237

1238 1239 1240 Manuscript submission to Molecular Biology and Evolution 1241 Contribution type: Article 1242 1243 Phylogenomics of trogons (Aves: Trogonidae) shed light on the Quaternary 1244 biogeography of tropical forests and the connections between Asia, North 1245 and South America 1246 1247 Ferreira, Mateus1*; Aleixo, Alexandre2; Bates, John M.3; Cracraft, Joel4; Ribas, Camila C.5 1248 1249 1 Programa de Pós-Graduação em Genética, Conservação e Biologia Evolutiva, INPA, 1250 Manaus, AM, Brazil 1251 2 Coordenação de Zoologia, MPEG, Belém, PA, Brazil 1252 3 Department of Ornithology, FMNH, Chicago, IL, USA 1253 4 Department of Ornithology, AMNH, New York, NY, USA 1254 5 Coordenação de Biodiversidade, INPA, Manaus, AM, Brazil 1255 *Corresponding author 1256 1257 Correspondence: Mateus Ferreira, Coordenação de Biodiversidade, Instituto Nacional de 1258 Pesquisas da Amazônia, CEP 69080-971, Manaus-AM, Brazil 1259 E-mail: [email protected] 1260 1261 1262 Abstract

1263 The pantropical distribution of trogons always drew attention of biogeographers, with

1264 species distributed all over the forests regions of subtropical and tropical Africa, Asia and

1265 America, several studies tried to reconstruct the phylogenetic relationships without, however,

1266 being able to achieve conclusive results. For the first time, all genera and almost all currently

1267 recognized species, 43 out of 45, were sampled and sequenced for thousands of ultraconserved

1268 elements (UCE) to reconstruct the family phylogenetic hypothesis. We analysed the

1269 concatenated dataset using different treatments for missing data with RAxML and ExaBayes,

1270 we also estimated a species tree using SVDquartets. We also estimated a fossil calibrated time

1271 tree for trogons diversification sampling 177 individuals of the Core Landbirds for RAG1 and

1272 RAG2 genes. Our results were congruent among all methods with high nodal support,

1273 disagreement between treatments (Species Tree x concatenated) were observed only at the basal

1274 nodes. In general, our results support the monophyly of the different biogeographical regions,

1275 with Apaloderma species being sister to the Asian (Harpactes and Apalharpactes) and the

1276 Neotropical trogons (Euptilotis, Pharomachrus, Priotelus, and Trogon). Trogonidae initial

1277 diversifications occurred around 20 Ma, and continued till the Pleistocene, where most of the

1278 Neotropical species appeared. Based on these results, we proposed how the climate changes

1279 since the Late Oligocene influenced forest distributions and how the establishment of land

1280 bridges between continents helped shape the family diversification.

1281

1282 Introduction

1283 The Trogonidae have some of the most colourful and exquisite plumages among birds.

1284 Representatives of this family, usually known as trogons or quetzals, can be found in forested

1285 tropical and subtropical regions of Africa, Asia and America (Collar 2017). The monophyly of

1286 the family was never questioned due to the morphological homogeneity among species

1287 (Livezey and Zusi 2007; Collar 2017), the most iconic feature that differentiate trogons and

1288 quetzals from other birds is the heterodactyl foot, in which digits 1 and 2 are directed backwards

1289 and the basal half of digits 3 and 4 are fused and directed forward (Maurer and Raikow 1981;

1290 Mayr 2009). However, it is precisely this unique feature that makes trogons so difficult to relate

1291 with extant birds. Despite several attempts to reconstruct the relationship between trogons and

1292 other birds, most of the morphological (Cracraft 1981; Maurer and Raikow 1981; Mayr 2003;

1293 Livezey and Zusi 2007) and the first molecular analyses (Monteros 2000; Hackett, et al. 2008;

1294 McCormack, et al. 2013) were unable to recover conclusive results about their phylogenetic

1295 relationships. Only recently, employing genomic representations, trogons were shown to be a

1296 sister group to a clade containing mousebirds (Coliiformes), cuckoo rollers (Leptosomiformes)

1297 and other Core Landbirds (Jarvis, et al. 2014; Prum, et al. 2015).

1298 Although the relationship with other birds is partially resolved, the relationships within

1299 the family are still pending conclusive results. Historically, the genera and species within each

1300 biogeographic region were considered monophyletic. The highest diversity is found in the

1301 Neotropical region, with four genera, Euptilotis, Pharomachrus, Priotelus and Trogon, and ~30

1302 species ranging from southwestern USA to northern Argentina. The Indo-Malaysian region

1303 comprises 2 genera, Apalharpactes and Harpactes, and 12 species, ranging from southern India,

1304 Southeast Asia, Philippines, the Malay Peninsula, Borneo, Philippines, Sumatra and Java, while

1305 the African region includes only one genus, Apaloderma and tree species. Although trogons are

1306 currently found only in tropical and subtropical regions, fossil records indicate that they had a

1307 wider distribution in the past. Two fossils from Europe, Primotrogon wintersteini (Mayr 1999)

1308 from the Middle Oligocene, and ?P. pumilio (Mayr 2005), from the Middle Eocene, are credited

1309 to be sister group to all other extant species (Mayr 2009). Whereas Septentrogon madseni

1310 (Kristoffersen 2002), from the transitional Paleocene-Eocene Fur Formation in north-western

1311 Denmark shares morphological characteristics that put him inside the Trogonidae. The presence

1312 of these fossils in Europe suggests a widespread lineage occurring in regions that are currently

1313 unsuitable for them. The similarity between fossils and extant trogons also indicates that this

1314 lineage suffered little morphological changes through time. This apparent conservatism of

1315 morphological characteristics also makes the inferences of phylogenetic relationships among

1316 extant species difficult.

1317 The first molecular phylogenetic hypothesis for trogons was based on two mitochondrial

1318 genes and included 20 out of the ca. 45 species (Monteros 1998). This study supported the

1319 hypothesis of monophyly of the biogeographic regions, recovering the Neotropical genera sister

1320 to the Asian, with the African clade sister to these two (Monteros 1998). Following studies that

1321 increased the number of genes and/or samples, however, couldn’t recover the monophyly of the

1322 Neotropical genera, nor the relationship among the different regions (Johansson and Ericson

1323 2005; Moyle 2005; DaCosta and Klicka 2008; Hosner, et al. 2010). The most recent paper

1324 (Hosner, et al. 2010), and the first one to sample the genus Apalharpactes, recognized six clades

1325 (Apaloderma, Apalharpactes, Harpactes, Pharomachrus/Euptilotis, Priotelus, and Trogon)

1326 with uncertain relationships among them, but showing evidences of Apalharpactes being more

1327 closely related with the African Apaloderma, than to the other Asian genus, Harpactes,

1328 implying a very complex biogeographical pattern, with two independent colonizations of Asia.

1329 A similar pattern suggested for the Neotropical genera, which group three distinct clades

1330 (Hosner, et al. 2010).

1331 This uncertainty regarding phylogenetic relationships so far was probably related to the

1332 scarcity of signal due to a low number of loci employed in previous studies. Genomic analyses

1333 using a reduced representation of the genome can increase phylogenetic information and avoid

1334 confounding the histories of single genes with the species relationships (Degnan and Rosenberg

1335 2009; Knowles 2009). Also, since the correct interpretation of biotic evolution can shed light

1336 on the landscape evolution (Baker, et al. 2014), a robust and well supported phylogenetic

1337 hypothesis is of extreme importance for defining hypothesis in biogeography (Donoghue and

1338 Moore 2003; Lexer, et al. 2013). In this sense, a prominent approach to study systematics using

1339 genomic markers is the use of probes for Ultraconserved Elements (UCE)(Faircloth, et al. 2012;

1340 McCormack, et al. 2012; McCormack and Faircloth 2013; McCormack, et al. 2013; Faircloth,

1341 et al. 2015). These probes, have been employed to reconstruct deep (Faircloth, et al. 2015;

1342 Moyle, et al. 2016; Branstetter, et al. 2017; Esselstyn, et al. 2017) and shallow (Bryson, et al.

1343 2016; Manthey, et al. 2016) phylogenetic relationships, even where high incomplete lineage

1344 sorting is expected, such as in cases of rapid evolutionary radiation (Meiklejohn, et al. 2016).

1345 Therefore, trogons represent a great study model on how genomic representation may

1346 elucidate uncertain phylogenetic relationships, and to understand how the landscape evolution

1347 shaped the family diversification, due to its pantropical geographic distribution and preference

1348 for forested habitats. Here, we aim (1) to generate and unprecedent and robust analyses of

1349 phylogenetic relationships within the Trogonidae family, using nearly complete sampling of all

1350 recognized speces and a genomic representation of more than 2,000 UCE loci, (2) to investigate

1351 the monophyly of main biogeographical regions, and (3) to reconstruct a calibrated tree to infer

1352 the timing of diversification, and how it was influenced by the global events on geography and

1353 climate.

1354

1355 Results

1356 UCE sequencing

1357 The reference sequences we extracted from the Apaloderma vittatum genome (Gilbert,

1358 Jarvis, Li, Consortium, et al. 2014) included 2,228 loci. The mean number of sequences for

1359 each individual was 2,080,592, and a mean number of UCE loci was 2,222, with only one toe

1360 pad sample (AMNH 322898) recovering less than 2000 loci (Table 1). The complete matrix

1361 contained 1421 loci, with mean locus length of 510.27 base pairs, and a total of 37,880

1362 parsimony informative (PI) sites, mean of 26.6 per locus (Table 2). The incomplete matrices

1363 with 95% and 75% completeness have 2,210 and 2,217 loci, with mean locus length of 499.77

1364 and 495.95 base pairs, and 55,060 and 57,259 PI sites, with mean of 24.91 and 25.83 sites per

1365 locus (Table 2).

1366 Phylogenetic inference

1367 The tree topologies were congruent among all methods and with high node support, apart

1368 from the SVDq analyses, in which the basal nodes presented low support. The concatenated

1369 RAxML and ExaBayes phylogenies recovered the Asian trogons sister to the Neotropical, and

1370 these two sisters to the African clade with high support (Fig. 1). All the ExaBayes analyses,

1371 including the complete and the two incomplete datasets, recovered the same topology with all

1372 nodes with the maximum posterior probability (Fig. 1). Although the topologies recovered by

1373 RAxML trees were congruent with ExaBayes, some of the basal nodes received low support.

1374 The same was observed with SVDq.

1375 Within the Asian group, Apalharpactes was sister to Harpactes, but with low support in

1376 the RAxML (Table 3) analyses. Within Harpactes we recovered three groups: (1) the distinct

1377 H. oreskios; (2) the two small-bodied species H. duvaucelli and H. orrhophaeus; and (3) the

1378 large-bodied species, containing the other species, with clearly defined and high support

1379 supported relationships (Fig. 1). The Neotropical clade was recovered with high nodal support

1380 (Table 3), showing the quetzals, Euptilotis and Pharomachrus, as sister to Priotelus and Trogon

1381 (Fig. 1). Pharomachrus moccino, the only Central America species, is sister to all other

1382 Pharomachrus species. The two Andean species, P. antisianus and P. auriceps, are not closely

1383 related (Fig. 1). Within Trogon, the most diverse genus in the family, we recovered 5 clades,

1384 all of which include species at both sides of the Andes (Fig. 1).

1385 Time-calibrated tree

1386 The concatenated matrix of RAG1 and RAG2 sequences includes 4757 base pairs for 177

1387 representatives of the Core Land birds (Claramunt and Cracraft 2015; Prum, et al. 2015)

1388 (Supplementary Table 1). Phylogenetic analysis of this matrix recovered a well-supported tree.

1389 Trogonidae diversification started in the Early Miocene, the first of four divergence events are

1390 close to each other, around 20 Ma (Fig. 2). While the Asian species originated during the Late

1391 Miocene/ Early Pliocene, most Neotropical species originated during the Late

1392 Pliocene/Pleistocene (Fig. 2).

1393

1394 Discussion

1395 Phylogenomic contribution to the reconstruction of Trogonidae diversification

1396 Recovering basal relationships in the Trogonidae phylogeny has proven to be challenging,

1397 and previous studies have failed to resolve the relationships among genera (Monteros 1998; 44

1398 Mayr 2003; Johansson and Ericson 2005; Moyle 2005), either because of incomplete taxon

1399 sampling or inadequate number of markers. Monteros (1998) using only two mtDNA genes

1400 recovered a tree topology similar to the one we recovered, in which taxa from different

1401 biogeographical regions were monophyletic. However, the relationships among genera were

1402 not well supported, and Apalharpactes was not sampled. Johansson and Ericson (2005), and

1403 then Moyle (2005), increased the sampling and added a few nuclear introns, yet there were few

1404 improvements in phylogenetic resolution. Moyle (2005) recovered a paraphyletic Neotropical

1405 group, with the quetzals being sister to all other genera, and the Asian and African group sister

1406 to each other embedded within Trogon and Priotelus. Johansson and Ericson (2005) based on

1407 a combined analysis of mtDNA and three nuclear introns recovered a topology similar to ours,

1408 however, node support for the Neotropical group, and the node grouping Asia and the

1409 Neotropics, received low to moderate support. Hosner, et al. (2010) were the first to include an

1410 Apalharpactes sample, but their results were also inconclusive, as relationships among genera

1411 were poorly supported and biogeographical groups, except for Africa, were not monophyletic.

1412 Our phylogenetic results were the first to recover with moderate to high support the

1413 relationship of almost all currently recognized species, as our analyses recovered most of the

1414 nodes with high statistical support (Fig. 1). The nodes that did not receive full support at the

1415 base of the tree (Table 3) are connected by short internodes, probably as a result of an ancient

1416 rapid radiation (Whitfield and Lockhart 2007). Recurrent issues arising from rapid radiations

1417 usually include incomplete lineage sorting (ILS), represented by conflict among gene trees due

1418 to successive events of speciation in short periods of time, which can be accentuated by large

1419 population sizes (Oliver 2013; Suh, et al. 2015). ILS probably was also the main cause of low

1420 support in previous studies that employed few genetic markers, as they could have conflicting

1421 histories (Knowles 2009; Oliver 2013) and probably lacked strong phylogenetic signal to

1422 recover the deep phylogenetic relationships (Salichos and Rokas 2013). Evidence of gene tree

1423 incongruence was strongly observed in the whole-genome analysis of bird diversification,

1424 where there was no single gene tree that fully corroborated the combined topology (Jarvis, et

1425 al. 2014). However, counterintuitive, increasing the number of markers does not necessarily

1426 means an improvement in poorly supported nodes. Instead, expanding the number of markers

1427 increases the probability of discordance among them (Oliver 2013), and thus, notably in events

1428 of rapid radiation, some divergences are expected not to behave as a fully bifurcating tree, but

1429 more like a network (Bapteste, et al. 2013; Suh, et al. 2015) because most genes will have

1430 discordant histories due to ILS (Degnan and Rosenberg 2006). Therefore, concatenation may

1431 be the best approach when the number of possible sites supporting a relationship is concentrated

1432 in a few loci diluted in a high number of loci affected by ISL (Gatesy and Springer, 2014;

1433 Springer and Gatesy, 2016). Nonetheless, based on our results, after the first events of

1434 diversification, most of nodes were recovered with high statistical support for all analysis,

1435 including the Neotropical node, which means that, even though we probably do not have enough

1436 confidence to allege the correct order of events that trogons went through their initial

1437 diversification, we may still infer some hypothesis based on current distribution and ecology.

1438 Diversification and biogeography of Trogons

1439 Trogons are still-hunting predators feeding on insects or small vertebrates, but most of

1440 Asian and Neotropical species also feed on fruits, with quetzals being mostly frugivores. They

1441 inhabit the midstory and canopy of tropical and subtropical forest, with some species occurring

1442 in forested patches of open habitats (e.g. Trogon curucui). Most species are territorialists, with

1443 small territories, and lack the capacity to fly over long distances, usually flying from perch to

1444 perch in short sallies (Collar 2017). The morphological conservatism of fossils compared to

1445 extant species suggests that trogons have not underwent large ecological shifts (Mayr 1999,

1446 2003; Mayr 2005), hence their historical distribution probably was affected by the distribution

1447 of suitable habitats through time. Although nowadays there is no continuous patch of suitable

1448 habitats, i.e. forested habitat, between Africa, Asia and America, during the Early Miocene, due

1449 to a warmer climate, most of the dry land was covered by forest habitats, such as the broad-leaf

1450 deciduous (Mixed Mesophytic) forest that covered most of the Northern Hemisphere (Baskin

1451 and Baskin 2016), and forests dominated by deciduous conifers that extended even over the

1452 Article Circle (Jahren 2007; Jahren and Sternberg 2008).

1453 The abundance of forests during the Tertiary is due to both warmer temperatures and

1454 twice the current amount of CO2 concentrations (Zachos, et al. 2001). However, after the

1455 Eocene Climatic Optimum (52 to 50 Ma), in which global mean temperatures were 8-10°C

1456 higher, the world temperature started to cool down with two climatic aberrations, where the

1457 amount of ice in polar regions increased drastically. The first one, known as Oi-1, happened

1458 just above the limits between Eocene and Oligocene (34 Ma) (Zachos, et al. 2001), this

1459 glaciation event caused rapid expansions of Antarctic continental ice-sheets and global

1460 temperatures remained low until a warming trend at the end of Oligocene (Zachos, et al. 2001).

1461 This warm phase that followed extended from the Late Oligocene until middle Miocene (~15

1462 Ma) with the Mid-Miocene Climatic Optimum (17 to 15 Ma) and it was followed by a gradual

1463 cooling, with the culmination in the Glacial cycles throughout the Plio/Pleistocene (Zachos, et

1464 al. 2001). The second aberration, Mi-1, happened during this warm period at the end of the

1465 Oligocene (~23 Ma), and was followed by a series of glaciation events (Zachos, et al. 2001),

1466 period well within the confidence interval for the initial diversification events we recovered in

1467 our time calibrated phylogeny. Both aberrations probably influenced the distribution and rates

1468 of diversification in some groups that have similar distributions as trogons, such as ferns

1469 (Bauret, et al. 2017; Hennequin, et al. 2017), and flowering plants (Li, et al. 2017). Interestingly,

1470 other groups of birds that have similar distributions present different patterns of diversification

1471 than trogons; woodpeckers (Aves: Picidae) and kingfishers (Aves: Alcedinidae) apparently

1472 have dispersed to the New World from the Old World more than once, however these events

1473 seem to be younger than those we recovered for trogons, around 15 to 5 Ma for woodpeckers

1474 (Shakya, et al. 2017), and 10 to 5 Ma for kingfishers (Andersen, et al. 2017). This pattern

1475 suggests that dispersal between Asia and America was possible during a long period of time,

1476 probably experiencing cycles of connection and disconnection due to climatic variations

1477 (Zachos, et al. 2001). Therefore, our temporal framework supports an ancestral lineage

1478 distributed over the Palearctic region (Claramunt and Cracraft 2015), with dispersal to Asia,

1479 Africa and America during a short period of time, causing the poorly supported nodes we

1480 observed in our analysis.

1481 Africa and Asia diversification

1482 Even though African and Asian linages are as old as the Neotropical, only 6% and 31%

1483 of species diversity are found in these areas, respectively. Although contentious, there are

1484 probably many reason for the uneven diversity among areas. Monteros (1998) suggests that

1485 competitive exclusion might play a role in this pattern, as African and Asian trogons need to

1486 compete with other groups of frugivores birds, such as mousebirds (Colliformes), hornbills

1487 (Bucerotidae), barbets (Megalaimidae and Lybiidae), turacos (Musophagidae), and several

1488 families of passerines (Irenidae, Pycnonotidae, etc). While the Neotropical trogons are, along

1489 cotingas (Cotingidae) and toucans (Ramphastidae), one of the most important family for seed

1490 dispersal in this region (Collar, et al. 2017).

1491 Inside Africa, except for Apaloderma narina which has six recognized subspecies, the

1492 other two, A. vittatum and A. aequatoriale are monotypic (Collar 2017). However, no

1493 phylogeographic study was conducted to evaluate genetic structure within these species, with

1494 recent studies using other organism as models showing shallow genetic structure probably

1495 originated by aridification of the continent as a response of Plio/Pleistocene climatic

1496 fluctuations (Bowie, et al. 2004; Bowie, et al. 2006; Voelker, et al. 2010). The diversification

1497 event we recovered between A. vittatum and A. narina happened around 7.4 Ma (Fig. 2) and

1498 precedes the beginning of the most drastic climatic fluctuations of the Pliocene, making any

1499 assumption of what may have caused this very hard, in particular considering that Africa has

1500 been geomorphologically stable for the last 40 Ma (Potts and Behrensmeyer 1992). Also, A.

1501 vittatum inhabits the montane forests, while A. narina and A. aequatoriale, inhabits the

1502 lowlands, and although we could not sample A. aequatoriale, previous work recovered it as

1503 sister species to A. narina (Hosner, et al. 2010). Suggesting that other mechanisms may be

1504 responsible for Apaloderma species diversification (Moritz, et al. 2000).

1505 In contrast with previous studies (Hosner, et al. 2010), our analyses recovered the

1506 monophyly of Asian trogons. Although the bootstrap support was moderate for this node in the

1507 likelihood analysis, it was recovered with high statistical support in the Bayesian analysis

1508 (Table 3). This suggest that after the initial diversification of the family, at least two Paleartic

1509 lineages (Claramunt and Cracraft 2015) colonized the Sundaland, the continental shelf that

1510 extended from SE Asia and comprises the Malay Peninsula, and the islands of Borneo, Java,

1511 and Sumatra. The time of diversification we found for Apalharpactes and Harpactes is

1512 consistent with the Hymalayan uplift acceleration, derived from India-Asia continental collision

1513 (Hall 2012; Hu, et al. 2017), and with the intermittent glaciations that followed the Mi-1

1514 glaciation at the Oligocene-Miocene boundary (Zachos, et al. 2001). These two events

1515 combined may have shaped Asian trogons diversification, however, making assumptions about

1516 Haparctes diversification involves a very complex history, and it is difficult based on extant

1517 species distribution to make any assumption about possible biogeographic barriers. Current

1518 geography of SE Asia and the Sunda islands can be misleading, the Sunda shelf was once

1519 exposed and covered by forest (Hall 2012; Bruyn, et al. 2014), and sea-level fluctuations were

1520 responsible for islands “formation” and connectivity, especially during the climatic fluctuations

1521 of the Pleistocene (Woodruff 2010). This mechanism is suggested as a possible explanation for

1522 Southeast Asia bird diversification (Lim, Rahman, et al. 2010; Lim, Zou, et al. 2010; Lim, et

1523 al. 2017). However, most of the Harpactes diversification events precede the Pleistocene, and

1524 occurred between the Mid-Miocene Climatic Optimum (17-15 Ma) (Zachos, et al. 2001) and

1525 the Early Pliocene, much older than the diversification events of the Neotropical clade, for

1526 example. The only phylogeographic study conducted so far, with the Philippine Trogon

1527 (Harpactes ardens), demonstrated geographical structure among different island matching

1528 subspecies distribution (Hosner, et al. 2014), whereas H. kasumba, H. diardii and H.

1529 erythrocephalus showed little to no genetic variation in the mtDNA for the few samples used

1530 (Hosner, et al. 2010). Therefore, further studies, with broad sampling are necessary to

1531 understand how the Pleistocene climate, and sea level fluctuation, influenced population

1532 structure, which in turn may shed some light on the initial diversification of this genus.

1533 Neotropical diversification

1534 For the first time, Neotropical trogons were recovered as a monophyletic group with high

1535 statistical support (Monteros 1998; Johansson and Ericson 2005; Moyle 2005; Hosner, et al.

1536 2010). Although most of extant diversity is currently found in Central and South America,

1537 trogons arrived first in the Americas through the Beringia Bridge, northwest North America,

1538 and colonized the whole west coast, during a period when there were vast forests covering

1539 North America (Baskin and Baskin 2016). Therefore, tracing back the events related with the

1540 initial divergences would require extensive palaeontological investigation. The overall trend we

1541 observe in this clade diversification is that Central American lineages occupied South America

1542 through the Panamanian Isthmus, and most of divergence events postdate the Mid-Miocene

1543 Climatic Optimum (17-15 Ma), which marks the beginning of the cooling trend that escalated

1544 to the Plio-Pleistocene glaciations. Also during this period, there was extensive orogenic

1545 activity in Mexico, including the uplift of Sierra Madre Occidental (34 – 15 Ma) (Ferrari, et al.

1546 2007) and the formation of the Trans-Mexican Volcanic Belt (35 – 2.5 Ma) (Ferrari, et al. 2000).

1547 Both events triggered climatic changes, which in turn influenced the establishment of major

1548 biomes in Mexico (Ferrari, et al. 1999), that have been shown to have influenced diversification

1549 in Amazillia hummingbirds (Ornelas, et al. 2014), and some plants (Lavin, et al. 2004; Becerra

1550 2005; Arakaki, et al. 2011).

1551 Another major event that shaped Neotropical trogons diversification was the

1552 establishment of the connection between North and South America, through the uplift of the

1553 Isthmus of Panama. The Great American Biotic Interchange allowed inter-continental exchange

1554 of biotas that were previously isolated in both continents and is of great importance for shaping

1555 bird assemblages and diversification (Weir, et al. 2009; Smith and Klicka 2010). Early studies

1556 suggested that the connection was only fully established at 3 Ma (Haug and Tiedeman 1998;

1557 Coates and Stallard 2013; Odea, et al. 2016), however, even though contentious in the literature

1558 (Farris, et al. 2011; Montes, et al. 2012; Bacon, et al. 2013; Bacon, et al. 2015a, b; Hoorn and

1559 Flantua 2015; Lessios 2015; Montes, et al. 2015; Odea, et al. 2016), this date was broadly used

1560 as a calibration point in phylogenetic studies attempting to integrate and synthesize patterns of

1561 dispersion across the Isthmus (review in Bacon, et al. (2015a)). Our results suggest that trogon

1562 dispersion across the Isthmus started as early as 6.5 Ma, with the split of Pharomachrus

1563 moccino from the other Pharomachrus species, and happened at least six additional times

1564 within Trogon diversification, all of them after 4 Ma. These results are also supported by a

1565 former study using only one mitochondrial marker for Trogon (DaCosta and Klicka 2008).

1566 Finally, the most notorious accomplishment of Neotropical trogons was to colonize the

1567 Greater Antilles. The genus Priotelus, which includes species endemic to the islands of Cuba,

1568 P. temnurus, and Hispaniola, P. roseigaster, split from Trogon around 17 Ma (Fig. 2). Trogons

1569 are well known for being weak fliers, so the chances of the ancestor of Priotelus to have

1570 dispersed through the ocean to colonize not just one, but two Caribbean islands are low. One

1571 possible explanation is the land bridge that once connected Central America to South America,

1572 known as GAARlandia (Greater Antilles + Aves Ridge) land bridge (Iturralde-Vinent 1994,

1573 2006). Although this land connection is credited to be much older (35 – 33 Ma) (Alonso, et al.

1574 2011; Rícan, et al. 2013; Nieto-Blázquez, et al. 2017) than the split of Priotelus and Trogon,

1575 during the Middle-Late Miocene, the emerged islands that were part of the land bridge were

1576 still connected by shallow seas (Iturralde-Vinent 2006), and sea levels fluctuations may have

1577 facilitated the dispersal to these islands. Fabre, et al. (2014) studying Caribbean rodents found

1578 a similar age (16.5 Ma) for the subfamily of rodents that occupy the Greater Antilles. However,

1579 the sister group is from South America, and the authors suggested that the ancestor of this group

1580 colonized the Caribbean Islands via rafting. Our results imply in a more complex scenario for

1581 the Greater Antilles colonization, and further studies are required to evaluate this late

1582 connection.

1583

1584 Conclusion

1585 In this study we recovered the phylogenetic relationships among almost Trogonidae taxa

1586 using a genomic approach. Coupled with our fossil calibrated time tree, we were able to propose

1587 a model of diversification that related not only how the climate change since the Late Oligocene,

1588 but also the connections between continents, shaped the family diversification. The monophyly

1589 of the different biogeographical regions was recovered, and even though some nodes at the base

1590 of the tree received low support, the pattern of rapid radiation is clear at the initial stages of

1591 trogons diversification. Also, even though trogons are currently restricted to subtropical and

1592 tropical regions, they were widespread lineages in the past, and their diversification was

1593 influenced by forest distribution through time. Our results also identified some interestingly

1594 new questions to be pursued: Are Neotropical trogons species really younger than African and

1595 Asian, or is it just a sampling artifact? What was the influence of past sea level fluctuations in

1596 the diversification of Harpactes? Is competition preveting diversification in Apaloderma?

1597

1598 Materials and Methods

1599 Taxon sampling and DNA extraction

1600 We sampled 48 individuals comprising all genera and currently recognized species of the

1601 Trogonidae family, except for the African Bare-cheeked Trogon (Apaloderma aequatoriale),

1602 and the narrow endemic Javan Trogon (Apalharpactes reinwardtii) (Collar 2017; Gill, et al.

1603 2018; Remsen, et al. 2018). All samples are represented by voucher specimens deposited in

1604 ornithological collections at the American Museum of Natural History (AMNH), Academy of

1605 Natural Sciences of Drexel University (ANSP), Field Museum of Natural History (FMNH),

1606 Instituto Nacional de Pesquisas da Amazônia (INPA), Kansas University (KU), Laboratório de

1607 Genética e Evolução Molecular de Aves - USP (LGEMA), Louisiana Museum of Natural

1608 History (LSUMZ), Museu Paraense Emílio Goeldi (MPEG), Smithsonian Institution National

1609 Museum of Natural History (USNM) and Burke Museum (UWBM) (Appendix S1).

1610 DNA from fresh tissue was extracted with the DNeasy kit (Qiagen Inc.), following the

1611 manufacture’s protocol. For taxa lacking fresh tissues we cut toepad clips from museum

1612 specimens with a sterile surgical blade and processed in a dedicated room for ancient DNA

1613 (aDNA Lab, AMNH). Toepads were rinsed with 100% ethanol, and ultra-pure water prior to

1614 digestion to remove any inhibitor that could cause problems in downstream procedures. We

1615 then extracted DNA with the DNeasy kit (Qiagen Inc.), replacing the regular silica columns

1616 with the QIAquick columns, to ensure maximum DNA yield. All extracts were sent to Rapid

1617 Genomics (Gainsville, FL) for library prep and target-capture sequence 2321 loci of

1618 Ultraconserved Elements (UCE) plus 98 conserved exons from 46 genes that were previously

1619 employed in phylogenetic analyses (Hackett, et al. 2008; Kimball, et al. 2009; Harvey, et al.

1620 2017).

1621 UCE and exons assembly

1622 The raw sequence data were processed with the Phyluce script pack (Faircloth 2016). We

1623 employed illumiprocessor (Faircloth 2013) and Trimmomatic (Bolger, et al. 2014) to remove

1624 adapter contamination and low-quality reads. To assemble a reference genome, we mapped the

1625 UCE and exons probes back to the Apaloderma vittatum genome (Gilbert, Jarvis, Li,

1626 Consortium, et al. 2014) using the script phyluce_probe_run_multiple_lastzs_sqlite, and then,

1627 phyluce_probe_slice_sequence_from_genomes to extract the probe region plus 500 base pairs

1628 from each flanking region. Apaloderma exonic regions were identified based on the Gallus

1629 gallus genes, and annotations of CDS and exons were copied to the reference sequences inside

1630 Geneious version R10.2.3 (Kearse, et al. 2012). With these sequences as a reference we mapped

1631 back the clean reads of each individual employing Bowtie2 (Langmead and Salzberg 2012)

1632 plugin 7.2.1 inside Geneious. The consensus sequences were called with the highest quality

1633 threshold and a depth of at least 4 reads. Each locus was aligned with MAFFT (Katoh and

1634 Standley 2013) under default parameters.

1635 Phylogenetic relationships and species tree analysis

1636 Since the intergeneric relationship among trogons are still mostly unresolved (Monteros

1637 1998; Johansson and Ericson 2005; Moyle 2005; Hosner, et al. 2010), we first performed a

1638 maximum likelihood analyses in RAxML v8.2 (Stamatakis 2014), and a Bayesian Inference

1639 analyses in ExaBayes v.1.4 (Aberer, et al. 2014), using the concatenated matrix with three

1640 treatments for missing data: a complete matrix, where no missing data was allowed, and two

1641 where the missing data was allowed, a 95% and 75% completeness matrix, in which each locus

1642 should have at least 95% or 75%, respectively, of all individuals in the matrix. As outgroups

1643 we selected one mousebird (Colius striatus, (Gilbert, Jarvis, et al. 2014b)), and a roller

1644 (Leptosomus discolor, (Gilbert, Jarvis, et al. 2014a)), suggested by recent studies as the closest

1645 relatives to the Trogonidae family (Jarvis, et al. 2014; Prum, et al. 2015). We also estimated a

1646 species tree using the SVDquartets (Chifman and Kubatko 2014) implemented in PAUP*

1647 v4a(build157) (Swofford 2002), that samples quartets of individuals for each gene tree and infer

1648 an unrooted phylogeny, performing a species tree using a coalescent approach. We

1649 exhaustively sampled all quartets and performed a 100 bootstrap to quantify the support for

1650 each node.

1651 Dating analysis

1652 To date the Trogonidae phylogeny we employed the slow evolving recombination-

1653 activating genes (RAG-1 and RAG-2) and a dense sampling for the Core Landbirds group

1654 (Telluraves), with the same calibration points used by Claramunt and Cracraft (2015). The

1655 concatenated matrix was partitioned by codon and the best partition and substitution model

1656 schemes were selected by PartitionFinder2 (Lanfear, et al. 2017).

1657

1658 Acknowledgements

1659 The authors thankfully acknowledge all the curators and curatorial assistants of the

1660 American Museum of Natural History, New York, USA (AMNH), Academy Academy of

1661 Natural Science of Drexel University, Philadelphia, USA (ANSP); Field Museum of Natural

1662 History, Chicago, USA (FMNH); Instituto Nacional de Pesquisas da Amazônia, Manaus, Brazil

1663 (INPA); Kansas University (KU), Laboratório de Genética e Evolução Molecular de Aves –

1664 USP (LGEMA), Lousiana State University Museum of Natural Science, Baton Rouge, USA

1665 (LSUMZ); and Museu Paraense Emílio Goeldi, Belém, Brazil (MPEG), Smithsonian Institution

1666 National Museum of Natural History (USNM), for borrowing tissue samples under their care.

1667 We are also grateful for all collectors involved in the fieldwork that make this paper possible.

1668 We thank L. Moraes for early input on this paper. MF acknowledge CAPES for his PhD

1669 fellowship, and CAPES PDSE fellowship (# 88881.133440/2016-01) and the support from the

1670 AMNH Frank M. Chapman Memorial Fund. The authors also thanks the grant Dimensions US-

1671 Biota-São Paulo: Assembly and evolution of the Amazon biota and its environment: an

1672 integrated approach, co-funded by the US National Science Fundation (NSF DEB 1241056) to

1673 J.C. and the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP grant

1674 #2012/50260-6) to Lucia Lohmann. AA and CCR are supported by CNPq research productivity

1675 fellowships. The authors acknowledge the National Laboratory for Scientific Computing

1676 (LNCC/MCTI, Brazil) for providing HPC resources of the SDumont supercomputer, which

1677 have contributed to the research results reported within this paper.

1678

1679 References

1680 1681 Aberer AJ, Kobert K, Stamatakis A. 2014. ExaBayes: massively parallel bayesian tree inference 1682 for the whole-genome era. Mol Biol Evol 31:2553-2556. 1683 Alonso R, Crawford AJ, Bermingham E. 2011. Molecular phylogeny of an endemic radiation 1684 of Cuban toads (Bufonidae: Peltophryne) based on mitochondrial and nuclear genes. J 1685 Biogeogr 39:434-451. 1686 Andersen MJ, McCullough JM, Mauck WM, Smith BT, Moyle RG. 2017. A phylogeny of 1687 kingfishers reveals an Indomalayan origin and elevated rates of diversificatio on oceanic 1688 islands. J Biogeogr early view. 1689 Arakaki M, Christin P-A, Nyffeler R, Lendel A, Eggli U, Ogbum RM, Spriggs E, Moore MJ, 1690 Edwards EJ. 2011. Contemporaneous and recent radiations of the world's major succulent 1691 plant lineages. PNAS 108:8379-8384. 1692 Bacon CD, Mora A, Wagner WL, Jaramillo CA. 2013. Testing geological models of evolution 1693 of the Isthmus of Panama in a phylogenetic framework. Botanical Journal of the Linnean 1694 Society 171:287-300. 1695 Bacon CD, Silvestro D, Jaramillo C, Smith BT, Chakrabarty P, Antonelli A. 2015a. Biological 1696 evidence supports an early and complex emergence of the Isthmus of Panama. PNAS 1697 112:6110-6115. 1698 Bacon CD, Silvestro D, Jaramillo C, Smith BT, Chakrabarty P, Antonelli A. 2015b. Reply to 1699 Lessios and Marko et al.: Early and progressive migration across the Isthmus of Panama 1700 is robust to missing data and biases: Fig. 1. PNAS 112:E5767-E5768. 1701 Baker PA, Fritz SC, Dick CW, Eckert AJ, Horton BK, Manzoni S, Ribas CC, Garzione CN, 1702 Battisti DS. 2014. The emerging field of geogenomics: Constraining geological problems 1703 with genetic data. Earth-Sci Rev 135:38-47. 1704 Bapteste E, Iersel Lv, Janke A, Kelchner S, Kelk S, McInerney JO, Morrison DA, Nakhleh L, 1705 Steel M, Stougie L, et al. 2013. Networks: expanding evolutionary thinking. Trends Genet 1706 29:439-441. 1707 Baskin JM, Baskin CC. 2016. Origins and relationships of the Mixed Mesophytic Forest of 1708 Oregon–Idaho, China, and Kentucky: Review and Synthesis. Ann MO Bot Gard 101:525- 1709 552. 1710 Bauret L, Gaudeul M, Sundue MA, Parris BS, Ranker TA, Rakotondrainibe F, Hennequin S, 1711 Ranaivo J, Selosse M-A, Rouhan G. 2017. Madagascar sheds new light on the molecular

1712 systematics and biogeography of grammitid ferns: New unexpected lineages and 1713 numerous long-distance dispersal events. Mol Phylogenet Evol 111:1-17. 1714 Becerra JX. 2005. Timing the origin and expansion of the Mexican tropical dry forest. PNAS 1715 102:10919-10923. 1716 Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for Illumina sequence 1717 data. Bioinformatics 30:2114-2120. 1718 Bouckaert R, Heled J, Kuhnert D, Vaughan T, Wu CH, Xie D, Suchard MA, Rambaut A, 1719 Drummond AJ. 2014. BEAST 2: a software platform for Bayesian evolutionary analysis. 1720 PLoS Comput Biol 10:e1003537. 1721 Bowie RC, Fjeldså J, Hackett SJ, Crowe TM. 2004. Molecular evolution in space and through 1722 time: mtDNA phylogeography of the Olive Sunbird (Nectarinia olivacea/obscura) 1723 throughout continental Africa. Mol Phylogenet Evol 33:56-74. 1724 Bowie RCK, Fjeldså J, Hackett SJ, Bates JM, Crowe TM. 2006. Coalescent mdels reveal the 1725 relative roles of ancestral polymorphism, vicariance, and dispersal in shaping 1726 phylogeographical structure of an African montane forest robin. Mol Phylogenet Evol 1727 38:171-188. 1728 Branstetter MG, Longino JT, Ward PS, Faircloth BC. 2017. Enriching the ant tree of life: 1729 enhanced UCE bait set for genome-scale phylogenetics of ants and other Hymenoptera. 1730 Methods in Ecology and Evolution. 1731 Bruyn M, Stelbrink B, Morley RJ, Hall R, Carvalho GR, Cannon CH, Bergh Gvd, Meijaard E, 1732 Metcalfe I, Boitani L, et al. 2014. Borneo and Indochina are major evolutionary hotspots 1733 for Southeast Asian biodiversity. Syst Biol 63:879-901. 1734 Bryson RW, Jr., Faircloth BC, Tsai WLE, McCormack J, Klicka J. 2016. Target enrichment of 1735 thousands of ultraconserved elements sheds new light on early relationships within New 1736 World sparrows (Aves: Passarelidae). Auk 133:451-458. 1737 Chifman J, Kubatko LS. 2014. Quartet inference from SNP data under the coalescent model. 1738 Bioinformatics 30:3317-3324. 1739 Claramunt S, Cracraft J. 2015. A new time tree reveals Earth history's imprint on the evolution 1740 of modern birds. Sci Adv 1:e1501005. 1741 Coates AG, Stallard RF. 2013. How old is the Isthmus of Panama? Bull Mar Sci 89:801-813. 1742 Collar N. 2017. Trogons (Trogonidae). Barcelona: Lynx Ediciones. 1743 Cracraft J. 1981. Toward a phylogenetic classification of the recent birds of the world (Class 1744 Aves). Auk 98:681-714. 1745 DaCosta JM, Klicka J. 2008. The Great American Interchange in birds: a phylogenetic 1746 perspective with the genus Trogon. Mol Ecol 17:1328-1343. 1747 Degnan JH, Rosenberg NA. 2006. Discordance of species trees with their most likely gene 1748 trees. PLoS Genet 2:e68. 1749 Degnan JH, Rosenberg NA. 2009. Gene tree discordance, phylogenetic inference and the 1750 multispecies coalescent. Trends Ecol Evol 24:332-340. 1751 Handbook of the birds of the world [Internet]. Barcelona: Lynx Ediciones; 2017 [cited 2017 1752 12/19]. Available from: http://www.hbw.com/ 1753 Donoghue PC, Moore BR. 2003. Toward an integrative historical biogeography. Integr Comp 1754 Biol 43:261-270. 1755 Esselstyn JA, Oliveros CH, Swanson MT, Faircloth BC. 2017. Investigating difficult nodes in 1756 the placental mammal tree with expanded taxon sampling and thousands of 1757 Ultraconserved Elements. Syst Biol early view. 1758 Fabre P-H, Vilstrup JT, Raghavan M, Sarkissian CD, Willerslev E, Douzery EJP, Orlando L. 1759 2014. Rodents of the Caribbean: origin and diversification of hutias unravelled by next- 1760 generation museomics. Biol Lett 10:201440266.

1761 Faircloth BC. 2013. illumiprocessor: a trimmomatic wrapper for parallel adapater and quality 1762 trimming. doi: 10.6079/J9ILL 1763 Faircloth BC. 2016. PHYLUCE is a software package for the analysis of conserved genomic 1764 loci. Bioinformatics 32:786-788. 1765 Faircloth BC, Branstetter MG, White ND, Brady SG. 2015. Target enrichment of 1766 ultraconserved elements from arthropods provides a genomic perspective on relationships 1767 among Hymenoptera. Mol Ecol Resour 15:489-501. 1768 Faircloth BC, McCormack JE, Crawford NG, Harvey MG, Brumfield RT, Glenn TC. 2012. 1769 Ultraconserved elements anchor thousands of genetic markers spanning multiple 1770 evolutionary timescales. Syst Biol 61:717-726. 1771 Farris DW, Jaramillo C, Bayona G, Restrepo-Moreno SA, Montes C, Cardona A, Mora A, 1772 Speakman RJ, Glascock MD, Valencia V. 2011. Fracturing of the Panamanian Isthmus 1773 during initial collision with South America. Geology 39:1007-1010. 1774 Ferrari L, Conticelli S, Vaggelli G, Petrone CM, Manetti P. 2000. Late Miocene volcanism and 1775 intra-arc tectonics during the early development of the Trans-Mexican Volcanic Belt. 1776 Tectonophysics 318:161-185. 1777 Ferrari L, Lopez-Martínez M, Aguirre-Díaz G, Carrasco-Núñez G. 1999. Space-time patterns 1778 of Cenozoic arc volcanism in central Mexico: From the Sierra Madre Occidental to the 1779 Mexican Volcanic Belt. Geology 27:303-306. 1780 Ferrari L, Valencia-Moreno M, Bryan S. 2007. Magmatism and tectonics of the Sierra Madre 1781 Occidental and its relation with the evolution of the western margin of North America. 1782 Geol Soc Am Spec Pap 422:1-39. 1783 Gatesy J, Springer MS. 2014. Phylogenetic analysis at deep timescales: unreliable gene trees, 1784 bypassed hidden support, and the coalescence/concatalescence conundrum. Mol 1785 Phylogenet Evol 80:231-266. 1786 Gilbert M, Jarvis ED, Li B, Consortium TAG, Wang J, Zhang G. 2014. Genomic data of the 1787 Bar-tailed trogon (Apaloderma vittatum) GigaScience Database. 1788 Gilbert MP, Jarvis ED, Li B, Li C, Consortium TAG, Wang J, Zhang G. 2014a. Genomic data 1789 of the Cucko roller (Leptosomus discolor). GigaScience Database. 1790 Gilbert MP, Jarvis ED, Li B, Li C, Consortium TAG, Wang J, Zhang G. 2014b. Genomic data 1791 of the Speckled mousebird (Colius striatus). GigaScience Database. 1792 Gill F, Donsker D. 2018. IOC World Bird List (v8.1). doi: 10.14344/IOC.ML.8.1 1793 Hackett SJ, Kimball RT, Reddy S, Bowie RC, Braun EL, Braun MJ, Chojnowski JL, Cox WA, 1794 Han KL, Harshman J, et al. 2008. A phylogenomic study of birds reveals their 1795 evolutionary history. Science 320:1763-1768. 1796 Hall R. 2012. Late Jurassic-Cenozoic reconstructions of the Indonesian region and the Indian 1797 Ocean. Tectonophysics 570-571:1-41. 1798 Harvey MG, Aleixo A, Ribas C, Brumfield RT. 2017. Habitat association predicts genetic 1799 diversity and population divergence in Amazonian birds. Am Nat Early View. 1800 Haug GH, Tiedeman R. 1998. Effect of the formation of the Isthmus of Panama on Atlantic 1801 Ocean thermohaline circulation. Nature 393:673-676. 1802 Hennequin S, Rouhan G, Salino A, Duan Y-F, Lepeigneux M-C, Guillou M, Ansell S, Almeida 1803 TE, Zhang L-B, Schneider H. 2017. Global phylogeny and biogeography of the fern genus 1804 Ctenitis (Dryopteridaceae), with a focus on the Indian Ocean region. Mol Phylogenet 1805 Evol 112:277-289. 1806 Hoorn C, Flantua S. 2015. An early start for the Panama land bridge. Science 348:186-187. 1807 Hosner PA, Sánchez-González LA, Peterson AT, Moyle RG. 2014. Climate-driven 1808 diversification and Pleistocene Refugia in Philippine birds: Evidence from 1809 phylogeographic structure and palaeoenvironmental niche modeling. Evolution 68:2658- 1810 2674. 58

1811 Hosner PA, Sheldon FH, Lim HC, Moyle RG. 2010. Phylogeny and biogeography of the Asian 1812 trogons (Aves: Trogoniformes) inferred from nuclear and mitochondrial DNA sequences. 1813 Mol Phylogenet Evol 57:1219-1225. 1814 Hu X, Wang J, An W, Garzanti E, Li J. 2017. Constraining the timing of the India-Asia 1815 continental collision by the sedimentary record. Sci China Earth Sci 60:603. 1816 Iturralde-Vinent MA. 1994. Cuba geology: a new plate-tectonic synthesis. J Petrol Geol 17:39- 1817 70. 1818 Iturralde-Vinent MA. 2006. Meso-Cenozoic Caribbean paleogeography: implications for the 1819 historical biogeography of the region. Int Geol Rev 48:791-827. 1820 Jahren AH. 2007. The Artic Forest of the Middle Eocene. Annu Rev Earth Planet Sci 35:509- 1821 540. 1822 Jahren AH, Sternberg LSL. 2008. Annual patterns within tree rings of the Artict middle Eocene 1823 (ca. 45 Ma): Isotopic signatures of precipitation, relative humidity, and deciduousness. 1824 Geology 36:99-102. 1825 Jarvis ED, Mirarab S, Aberer AJ, Li B, Houde P, Li C, Ho SY, Faircloth BC, Nabholz B, 1826 Howard JT, et al. 2014. Whole-genome analyses resolve early branches in the tree of life 1827 of modern birds. Science 346:1320-1331. 1828 Johansson US, Ericson PGP. 2005. A re-evaluation of basal phylogenetic relationships within 1829 trogons (Aves: Trogonidae) based on nuclear DNA sequences. J Zool Syst Evol Res 1830 43:166-173. 1831 Katoh K, Standley DM. 2013. MAFFT Multiple Sequence Alignment Software Version 7: 1832 Improvements in Performance and Usability. Mol Biol Evol 30:772-780. 1833 Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, 1834 Markowitz S, Duran C, et al. 2012. Geneious Basic: An integrated and extendable desktop 1835 software platform for the organization and analysis of sequence data. Bioinformatics 1836 28:1647-1649. 1837 Kimball RT, Braun EL, Barker FK, Bowie RC, Braun MJ, Chojnowski JL, Hackett SJ, Han 1838 KL, Harshman J, Heimer-Torres V, et al. 2009. A well-tested set of primers to amplify 1839 regions spread across the avian genome. Mol Phylogenet Evol 50:654-660. 1840 Knowles LL. 2009. Statistical Phylogeography. Annu Rev Ecol Evol Syst 40:593-612. 1841 Kristoffersen AV. 2002. An early Paleogene trogon (Aves: Trogoniformes) from the Fur 1842 Formation, Denmark. J Vert Paleontol 22:661-666. 1843 Lanfear R, Frandsen PB, Wright AM, Senfeld T, Calcott B. 2017. PartitionFinder 2: New 1844 methods for selecting partitioned models of evolution for molecular and morphological 1845 phylogenetic analyses. Mol Biol Evol 34:772-773. 1846 Langmead B, Salzberg SL. 2012. Fast gapped-read alignment with Bowtie 2. Nat Methods 1847 9:357-359. 1848 Lavin M, Schrire BP, Lewis GP, Pennington RT, Delgado-Salinas A, Thulin M, Hughes CE, 1849 Matos AB, Wojciechowski MF. 2004. Metacommunity process rather than continental 1850 tectonic history better explains geographically structured phylogenies in legumes. Phil 1851 Trans R Soc B 359:1509-1522. 1852 Lessios HA. 2015. Appearance of an early closure of the Isthmus of Panama is the product of 1853 biased inclusion of data in the metaanalysis. PNAS 112:E5765. 1854 Lexer C, Mangili S, Bossolini E, Forest F, Stölting KN, Pearman PB, Zimmermann NE, 1855 Salamin N, Carine M. 2013. ‘Next generation’ biogeography: towards understanding the 1856 drivers of species diversification and persistence. J Biogeogr 40:1013-1022. 1857 Li M, Ohi-Toma T, Gao Y-D, Xu B, Zhu Z-M, Ju W-B, Gao X-F. 2017. Molecular 1858 phylogenetics and historical biogeography of Sorbus senso stricto (Rosaceae). Mol 1859 Phylogenet Evol 111:76-86.

1860 Lim HC, Gawin DF, Shakya SB, Harvey MG, Rahman MA, Sheldon FH. 2017. Sundaland's 1861 east-west rain forest population structure: variable manifestation in four polytypic bird 1862 species examined using RAD-Seq and plumage analyses. J Biogeogr 44:2259-2271. 1863 Lim HC, Rahman MA, Lim SLH, Moyle RG, Sheldon F. 2010. Revisiting Wallace's haunt: 1864 Coalescent simulations and comparative niche modeling reveal historical mechanisms 1865 that promoted avian population divergence in the Malay Archipelago. Evolution 65:321- 1866 334. 1867 Lim HC, Zou F, Taylor SS, Marks BD, Moyle RG, Voelker G, Sheldon F. 2010. Phylogeny of 1868 magpie-robins and shamas (Aves: Turdidae: Copsychus and Trichixos): implications for 1869 island biogeography in Southeast Asia. J Biogeogr 37:1894-1906. 1870 Livezey BC, Zusi RL. 2007. Higher-order phylogeny of modern birds (Theropoda, Aves: 1871 Neornithes) based on comparative anatomy. II. Analysis and discussion. Zoological 1872 Journal of the Linnean Society 149:1-95. 1873 Manthey JD, Campillo LC, Burns KJ, Moyle RG. 2016. Comparison of Target-Capture and 1874 Restriction-Site Associated DNA Sequencing for Phylogenomics: A Test in Cardinalid 1875 Tanagers (Aves, Genus: Piranga). Syst Biol 65:640-650. 1876 Maurer DR, Raikow RJ. 1981. Appendicular myology, phylogeny and classificaiton of the 1877 avian order Coraciiformes (inluding Trogoniformes). Ann Carnegie Mus 50:417-434. 1878 Mayr G. 1999. A new trogon from the Middle Oligocene of Céreste, France. Auk 116:427-434. 1879 Mayr G. 2005. New trogons from the early Tertiary of Germany. Ibis 147:512-518. 1880 Mayr G. 2003. On the phylogenetic relationships of trogons (Aves, Trogonidae). Journal of 1881 Avian Biology 34:81-88. 1882 Mayr G. 2009. Paleogene Fossil Birds. Germany: Springer. 1883 McCormack JE, Faircloth BC. 2013. Next-generation phylogenetics takes root. Mol Ecol 1884 22:19-21. 1885 McCormack JE, Faircloth BC, Crawford NG, Gowaty PA, Brumfield RT, Glenn TC. 2012. 1886 Ultraconserved elements are novel phylogenomic markers that resolve placental mammal 1887 phylogeny when combined with species-tree analysis. Genome Res 22:746-754. 1888 McCormack JE, Harvey MG, Faircloth BC, Crawford NG, Glenn TC, Brumfield RT. 2013. A 1889 phylogeny of birds based on over 1,500 loci collected by target enrichment and high- 1890 throughput sequencing. PLoS One 8:e54848. 1891 Meiklejohn KA, Faircloth BC, Glenn TC, Kimball RT, Braun EL. 2016. Analysis of a rapid 1892 evolutionary radiation using Ultraconserved Elements: Evidence for a bias in some 1893 multispecies coalescent methods. Syst Biol 65:612-627. 1894 Metzker ML. 2010. Sequencing technologies - the next generation. Nature Reviews Genetics 1895 11:31-46. 1896 Monteros AE. 2000. Higher-level phylogeny of Trogoniformes. Mol Phylogenet Evol 14:20- 1897 34. 1898 Monteros AE. 1998. Phylogenetic relationships among the Trogons. Auk 115:937-954. 1899 Montes C, Cardona A, Jaramillo C, Pardo A, Silva JC, Valencia V, Ayala C, Perez-Angel LC, 1900 Rodriguez-Parra LA, Ramirez V, et al. 2015. Middle Miocene closure of the Central 1901 American Seaway. Science 348:226-229. 1902 Montes C, Cardona A, McFadden R, Moron SE, Silva CA, Restrepo-Moreno S, Ramirez DA, 1903 Hoyos N, Wilson J, Farris D, et al. 2012. Evidence for middle Eocene and younger land 1904 emergence in central Panama: Implications for Isthmus closure. Geol Soc Am Bull 1905 124:780-799. 1906 Moritz C, Patton JL, Schneider CJ, Smith TB. 2000. Diversification of rainforest fauna: An 1907 integrated molecular approach. Annu Rev Ecol Syst 31:533-563. 1908 Moyle RG. 2005. Phylogeny and biogeographical history of Trogoniformes, a pantropical bird 1909 order. Biol J Linn Soc 84:725-738. 60

1910 Moyle RG, Oliveros CH, Andersen MJ, Hosner PA, Benz BW, Manthey JD, Travers SL, Brown 1911 RM, Faircloth BC. 2016. Tectonic collision and uplift of Wallacea triggered the global 1912 songbird radiation. Nat Commun 7:12709. 1913 Nieto-Blázquez M, Antonelli A, Roncal J. 2017. Historical biogeography of endemic seed plant 1914 genera in the Caribbean: Did GAARlandia play a role? Ecol Evol Early View. 1915 Odea A, Lessios HA, Coates AG, Eytan RI, Restrepo-Moreno SA, Cione AL, Collins LS, de 1916 Queiroz A, Farris DW, Norris RD, et al. 2016. Formation of the Isthmus of Panama. Sci 1917 Adv 2:e1600883-e1600883. 1918 Oliver J. 2013. Microevolutionary processes generate phylogenomic discordance at ancient 1919 divergences. Evolution 67:1823-1830. 1920 Ornelas JF, González C, de los Monteros AE, Rodríguez-Gómez F, García-Feria LM, Riddle 1921 B. 2014. In and out of Mesoamerica: temporal divergence of Amazilia hummingbirds pre- 1922 dates the orthodox account of the completion of the Isthmus of Panama. Journal of 1923 Biogeography 41:168-181. 1924 Potts R, Behrensmeyer AK. 1992. Late Cenozoic terrestrial ecosystems. In: Beherensmeyer 1925 AK, Damuth JD, DiMichele WA, Potts R, Sues HD, WIng SL, editors. Terrestrial 1926 ecosystems through time. Evolutionary paleoecology of terrestrial plants and animals. 1927 Chicago: University of Chicago Press. 1928 Prum RO, Berv JS, Dornburg A, Field DJ, Townsend JP, Lemmon EM, Lemmon AR. 2015. A 1929 comprehensive phylogeny of birds (Aves) using targeted next-generation DNA 1930 sequencing. Nature 526:569-573. 1931 Tracer v1.6 [Internet]. 2014. Available from: http://beast.bio.ed.ac.uk/Tracer 1932 Remsen JV, Jr., Areta JI, Cadena CD, Claramunt S, Jaramillo C, Pacheco JF, Pérez-Emen J, 1933 Robbins MB, Stiles FG, Stotz DF, et al. 2018. A classification of the bird species of South 1934 America. American Ornithologists' Union. 1935 http://www.museum.lsu.edu/~Remsen/SACCBaseline.htm 1936 Rícan O, Piálek L, Zardoya R, Doadrio I, Zrzavý J. 2013. Biogeography of the Mesoamerican 1937 Cichlidae (Teleostei: Heroini): colonization through the GAARlandia land bridge and 1938 early diversification. J Biogeogr 40:579-593. 1939 Salichos L, Rokas A. 2013. Inferring ancient divergences requires genes with strong 1940 phylogenetic signals. Nature 497:327-331. 1941 Shakya SB, Fuchs J, Pons JM, Sheldon F. 2017. Tapping the woodpecker tree for evolutionary 1942 insight. Mol Phylogenet Evol in press. 1943 Smith BT, Klicka J. 2010. The profound influence of the Late Pliocene Panamanian uplift on 1944 the exchange, diversification, and distribution of New World birds. Ecography. 1945 Springer MS, Gatesy J. 2015. The gene tree delusion. Mol Phylogenet Evol 94:1-33. 1946 Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of 1947 large phylogenies. Bioinformatics 30:1312-1313. 1948 Suh A, Smeds L, Ellegren H. 2015. The Dynamics of Incomplete Lineage Sorting across the 1949 Ancient Adaptive Radiation of Neoavian Birds. PLoS Biol 13:e1002224. 1950 Swofford R. 2002. PAUP*. Phylogenetic analysis using parsimony (* and other methods). 1951 Version 4. Version 4. Sunderland, Massachusetts: Sinauer Associates. 1952 Voelker G, Outlaw RK, Bowie RCK. 2010. Pliocene forest dynamics as a primary driver of 1953 African bird speciation. Global Ecol Biogeogr 19:111-121. 1954 Weir JT, Bermingham E, Schluter D. 2009. The Great American Biotic Interchange in birds. 1955 PNAS 106:21737-21742. 1956 Weir JT, Schluter D. 2008. Calibrating the avian molecular clock. Mol Ecol 17:2321-2328. 1957 Whitfield JB, Lockhart PJ. 2007. Deciphering ancient rapid radiations. Trends Ecol Evol 1958 22:258-265.

1959 Woodruff DS. 2010. Biogeography and conservation in Southeast Asia: how 2.7 million years 1960 of repeated environmental fluctuations affect today's patterns and the future of the 1961 remaining refugial-phase biodiversity. Biodivers Conserv 19:919-941. 1962 Zachos J, Pagani M, Sloan L, Thomas E, Billups K. 2001. Trens, rhytms, and aberrations in 1963 global climate 65 Ma to present. Science 27:686-693. 1964 1965

1966 Table 1 – Samples used in this study, the museum voucher numbers, locality and geographical coordinates 1967 (when available), number of UCE reads, and loci recovered for each sample. Clean UCE Species Museum voucher Locality reads loci Apaloderma vittatum SRP028834 Tanzania: Udzungwa Mts. - 2,228*

Apaloderma narina AMNH DOT-12430 Liberia: Lofa, Ziggida (08°02'15.5"N 9°31'49.5"W) 3,607,056 2,228

Apalharpactes mackloti LSUMZ B-49104 Indonesia: Sumatra 1,664,511 2,220

Apalharpactes mackloti AMNH 633881 Indonesia: Sumatra, Bandar-Baroe (03°15'57.6''N 98°30'49.9''E) 2,758,684 2,080 Philippines: Barrio Via, Sitio Hot Springs, Baggao Mun. (17°50'N, Harpactes ardens USNM 607340 1,193,041 2,208 122°01'E) Malaysia: Sabah, Klias Forest Reserve (05°19’34’’N Harpactes diardii AMNH DOT-563 3,601,173 2,226 115°40’25’’E) Harpactes oreskios ANSP 16308 Malaysia: Sabah, Mendolong (04°54'27.6"N 115°47'04.5"E) 5,208,017 2,228

Harpactes orrhophaeus AMNH DOT-15159 Malaysia: Sabah, Mt. Lucia (04°27’37.8’’N 117°55’20.4’’E) 4,250,801 2,228 Malaysia: Sabah, Imbak Valley, ca 60 km S Telupid (5°06’N Harpactes duvaucelli LSUMZ B-38592 887,312 2,222 117°01’51’’E) Harpactes fasciatus AMNH 778649 India: Dangs, Bhawandagad 5,386,424 2,218 Vietnam: Quang Nam, Ngoc Linh Range (15°11’00’’N Harpactes erythrocephalus AMNH DOT-12240 2,126,329 2,224 108°02’00’’E) Harpactes wardii AMNH 307761 Myanmar: Laukkaing 5,151,969 2,198 Malaysia: Sabah, Tambuman, Mt. Trus Madi (05°35’09’’N Harpactes whiteheadi LSUMZ B-52627 11,299,280 2,228 116°29’26’’E) Malaysia: Sabah, Ulu Tungud Forest Reserve, Melian Range Harpactes kasumba AMNH DOT-15326 4,264,359 2,228 (05°50’48’’N 117°10’57’’E) USA: Arizona, Ramsey Canyon Preserve (31°26'50.2"N Euptilotis neoxenus AMNH DOT-11080 1,955,116 2,186 110°18'25.8"W) Brazil: Amazonas, Parque Nacional do Jaú (01°49’50’’S Pharomachrus pavoninus INPA A-1993 2,080,592 2,215 61°35’45’’W) Pharomachrus auriceps AMNH 175988 Ecuador: Baeza, Arriba (0°27’54’’S 77°53’44.9’’W) 6,034,956 2,210 hargitti Pharomachrus auriceps FMNH 473723 Peru: Rodriguez de Mendoza (06°S 77°W) 2,620,376 2,221 auriceps Venezuela: Near village of Junquito on Colonia Tovar Rd Pharomachrus fulgidus AMNH 322895 4,665,318 1,864 (10°27’23’’N 67°04’31’’W) Pharomachrus moccino AMNH 326512 Honduras: Mt Pucca, Gracias (14°34’43’’N 88°38’30’’W) 5,630,314 2,215

Pharomachrus antisianus ANSP 19429 Ecuador: Napo, 12 km NNE El Chaco; Mirador 5,651,764 2,228

Priotelus temnurus ANSP 20257 Cuba 1,644,934 2,220 Dominican Republic: Parque Nacional Sierra Baoruco, Pueblo Priotelus roseigaster KU 8098 1,431,709 2,221 Viejo (18°12’N 71°32’W) Trogon clathratus USNM 613996 Panama: Bocas del Toro, Los Planes (08°35’43’’N 82°14’16’’W) 3,200,785 2,158 Ecuador: Esmeraldas, 20 km ENE Muisne (0°38’51’’N Trogon mesurus ANSP 19305 7,341,190 2,142 79°59’59’’W) Trogon massena KU 2073 Mexico: Campeche, Silvituc (18°13’48’’N 90°12’W) 1,689,867 2,224

Trogon comptus LSUMZ B-11829 Ecuador: Esmeraldas, El Placer (0°52’N 78°33’W) 2,072,859 2,228 Brazil: Amazonas, Parque Nacional do Jaú (01°49’50’’S Trogon melanurus INPA A-1955 2,451,461 2,225 61°35’45’’W) Brazil: Pará, Aveiro, left bank Tapajós River (03°42.3’S Trogon viridis INPA A-5240 1,893,902 2,226 55°35.5’W) Trogon chionurus LSUMZ B-28571 Panama: Colón, Achiote Road (09°13’32’’N 80°0’56’’W) 1,879,103 2,225 El Salvador: La Paz, Aeropuerto Internacional El Salvador Trogon melanocephalus USNM 646857 1,521,530 2,224 (13°25’57’’N 89°03’50’’W) Mexico: Michoacán, Lazaro Cardenas, La Mira (18°05.71’N Trogon citreolus UWBM 101087 1,311,613 2,224 102°23.71’W) Trogon bardii LSUMZ B-71992 Costa Rica: Osa, Los Charces (08°40’19’’N 83°30’19’’W) 2,036,944 2,226

Trogon violaceus MPEG CN437 Brazil: Pará, Alenquer, ESEC Grão-Pará (0°09’S 55°11’W) 1,251,316 2,222 Peru: Tumbes, El caucho Biological Station (3°49’25’’S Trogon caligatus LSUMZ B-66270 4,878,667 2,150 80°15’37’’W) Trogon ramonianus INPA A-5449 Brazil: Pará, Santarém, Rio Arapiuns (3°19’S 55°20’W) 2,665,900 2,228 Brazil: Pará, Aveiro, left bank Tapajós River (3°42.3’S Trogon curucui INPA A-5286 1,157,694 2,221 55°35.5’W) Brazil: Minas Gerais, RPPN Serra do Caraça (20°07’01’’S Trogon aurantius LGEMA 15782 1,162,924 2,213 43°29’16’’W) Brazil: Santa Catarina, Blumenau, Vila Itoupava (26°39’59’’S Trogon surrucura MPEG SC015 2,005,634 2,224 49°05’41’’W) Trogon rufus tenellus LSUMZ B-26564 Panama: Colón, Gamboa (9°09’25’’N 79°45’36’’W) 4,118,529 2,228 Brazil: Pará, Aveiro, left bank Tapajós River (3°42.3’S Trogon rufus amazonicus INPA A-5284 3,892,857 2,228 55°35.5’W) Trogon rufus chrysochlorus LGEMA 9557 Brazil: São Paulo, Ubatuba (23°23’24’’S 45°05’24’’W) 1,161,086 2,225

El Salvador: Sonsonate: Izalco, Canton Las Laja (13°45’35’’N Trogon elegans FMNH 434014 475,853 2,201 89°40’21’’W) Mexico: Jalisco, Puerto los Mazos, Sierra de Manantlan Trogon mexicanus FMNH 343220 1,322,925 2,222 (19°28’09’’N 103°56’51’’W) Trogon aurantiiventris LSUMZ B-41625 Panama: Bocas del Toro, Chiriqui (8°47’29’’N 82°12’35’’W) 6,441,454 2,228 Mexico: Oaxaca, San Gabriel Mixtepec, Sierra de Miahuatlan Trogon collaris puella FMNH 394272 292,340 2,114 (16°09’56’’N 97°01’29’’W) Trogon collaris collaris MPEG CN450 Brazil: Pará, Alenquer, ESEC Grão-Pará (0°09’S 55°11’W) 1,361,638 2,221 Guyana: Potaro-Siparuni, Kopinang Mountain (4°57’54’’N Trogon personatus LSUMZ B-48503 1,826,664 2,228 59°54’49’’W) 1968 1969 Table 2 – Summary information of each method, including number of loci, total length of the concatenated 1970 alignment, mean length size per locus, minimum and maximum length, and the total number of the Parsimony 1971 Informative (PI) sites. Complete 75% 95% Number of loci 1421 2110 2217 Total lenght 725090 1054512 1099526 Mean length size 510.27 499.77 495.95 Min-max length 259-1145 162-1145 162-1145 Number of PI sites 37,880 55,060 57,259 1972 1973 Table 3 – Node support for recalcitrant nodes in the Trogonidae phylogeny. RAxML ExaBayes SVDq 75% 95% complete 75% 95% complete 95% Asian + Neotropical 70 62 84 1.0 1.0 1.0 - Apalharpactes + Harpactes 60 46 52 1.0 1.0 1.0 - Neotropical 100 100 100 1.0 1.0 1.0 100 1974

1975

1976 Figure 1 – Phylogeny of Trogonidae inferred with ExaBayes summarizing the results from other analyses. The 1977 circle at each node represent the statistical support for the RAxML analyses and the species tree reconstruction 1978 inferred by SVDq. Green lines represent distribution shifts from Central America to South America. Trogon 1979 species were group in five species groups highlighted with grey boxes: “rufus”, “collaris”, “melanurus”, “viridis”, 1980 and “violaceus”. 1981

1982

1983

1984 Figure 2 – Time-calibrated phylogeny of Trogonidae inferred from the concatenated dataset of RAG1 and RAG2 1985 genes using BEAST. This tree represents part of the tree calibrated using (Claramunt and Cracraft 2015) 1986 calibrations, complete taxon data in Supplementary Table 1. The basal nodes were constrained to match the UCE 1987 topology, all other nodes have a red circle, if the posterior probability is 1.0, or the posterior is written next to the 1988 node. Timings of major splits are shown next to each node. Blue bars represent the 95% HPD estimates of node 1989 height. Green lines represent distribution shifts from Central America to South America. The top-right figure 1990 represents the whole tree with calibration points as red circles. 1991

1992

1993

1994 Supplementary Table S1 – Table containing taxonomic information on all specimens employed in the RAG time 1995 tree. The RAG1 and RAG2 column refers to GenBank accession numbers for these two genes. Taxonomy follows 1996 del Hoyo, et al. (2017). 1997 Order Family Species RAG1 RAG2 Passeriformes Thraupidae Thraupis cyanocephala AY057035 AY443236 Passeriformes Emberizidae Emberiza schoeniclus AY056992 AY443143 Passeriformes Passeridae Passer montanus AF143738 AY443198 Passeriformes Prunellidae Prunella collaris AY057024 AY443213 Passeriformes Dicaeidae Dicaeum aeneum AY443282 AY443139 Passeriformes Regulidae Regulus calendula AY057028 AY443220 Passeriformes Irenidae Irena cyanogaster AY056999 AY443158 Passeriformes Nectariniidae Nectarinia olivacea AY057009 AY443180 Passeriformes Turdidae Catharus ustulatus AY443265 AY443114 Passeriformes Cinclidae Cinclus cinclus AY056985 AY443119 Passeriformes Mimidae Mimus patagonicus AY057005 AY443173 Passeriformes Sturnidae Sturnus vulgaris AY057032 AY443232 Passeriformes Troglodytidae Troglodytes aedon AY057038 AY443241 Passeriformes Certhiidae Certhia familiaris AY056983 AY443115 Passeriformes Sittidae Sitta carolinensis AY443332 AY443227 Passeriformes Sylviidae Sylvia nanna AY057033 AY443233 Passeriformes Pycnonotidae Pycnonotus barbatus AY057027 AY443219 Passeriformes Hirundinidae Hirundo rustica AY443290 AY443154 Passeriformes Aegithalidae Aegithalos iouschensis AY056976 AY443103 Passeriformes Locustellidae Megalurus palustris AY319988 AY799840 Passeriformes Remizidae Remiz pendulinus AY443328 AY443222 Passeriformes Promeropidae Promerops cafer AY443323 AY443212 Passeriformes Monarchidae Monarcha axillaris AY057006 AY443176 Passeriformes Laniidae Lanius excubitor AY443293 AY443160 Passeriformes Artamidae Artamus leucorhynchus AY056980 AY443109 Passeriformes Artamidae Artamus cyanopterus AY443262 AY443108 Passeriformes Artamidae Cracticus quoyi AY443278 AY443135 Passeriformes Vangidae Vanga curvirostris AY057040 AY443244 Passeriformes Platysteiridae Batis mixta AY443263 AY443110 Passeriformes Vireonidae Vireo philadelphia AY057041 AY443245 Passeriformes Melanocharitidae Melanocharis nigra AY057002 AY443167 Passeriformes Melanocharitidae Melanocharis vesteri AY443299 AY443168 Passeriformes Orthonychidae Orthonyx teminckii AY057012 AY443309 Passeriformes Climacteridae Climacteris erythrops AY443268 AY443121 Passeriformes Menuridae Menura novaehollandiae AY057004 AY443171 Passeriformes Furnariidae Furnarius rufus AY056995 AY443149 Passeriformes Rhinocryptidae Scytalopus magellanicus AY443331 AY443226 Passeriformes Thamonophilidae Terenura sharpei JX213518 JX213481 Passeriformes Pipridae Piprites chloris FJ501717 FJ501897 Passeriformes Pipridae Piprites pileata JF970177 KC157559 Passeriformes Pipridae Lepidothrix coronata FJ501655 FJ501835 Passeriformes Pipridae Antilophia galeata FJ501600 FJ501780 Passeriformes Oxyrunchidae Oxyruncus cristatus FJ501689 FJ501878 Passeriformes Cotingidae Cotinga cayana FJ501623 FJ501803

Passeriformes Cotingidae Laniisoma elegans FJ501651 FJ501831 Passeriformes Cotingidae Phoenicircus nigricollis FJ501705 FJ501885 Passeriformes Tyrannidae Tyrannus tyrannus AF143739 AY443243 Passeriformes Sapayoidae Sapaoya aenigma DQ320606 DQ320573 Passeriformes Dendrocolaptidae Dendrocolaptes certhia FJ461166 FJ460982 Passeriformes Pittidae Pitta sordida AY443219 AY443206 Passeriformes Acanthisittidae Acanthisitta chloris AY056975 AY443102 Psittaciformes Psittacidae Psittacus erithacus EF517674 EF517687 Psittaciformes Psittacidae Alisterus scapularis KT954426 EF517677 Psittaciformes Psittacidae Melopsittacus undulatus XM_005150647.1 XM_005150646.1 Psittaciformes Psittacidae Micropsitta brujinii EF517673 EF517681 Psittaciformes Psittacidae Amazona aestiva LMAW01003202 LMAW01003202 Psittaciformes Psittacidae Myopsitta monachus DQ143328 - Psittaciformes Psittacidae Agapornis personata EF517672 EF517679 Psittaciformes Cacatuidae Calyptorhynchus funereus KT954425 EF517680 Psittaciformes Strigopidae Nestor notabilis XM_010020228.1 XM_010020229.1 Falconiformes Falconidae Falco peregrinus AY461399 KT954538 Falconiformes Falconidae Falco cherrug XM_005441067.1 XM_005441068.2 Falconiformes Falconidae Daptrius ater AY461397 KT954537 Falconiformes Falconidae Micrastur gilvicollis AY461403 KT954536 Cariamiformes Cariamidae Cariama cristata XM_009699718.1 XM_009699720.1 Piciformes Ramphastidae Pteroglossus aracari KT954416 KT954525 Piciformes Capitonidae Capito niger KT954414 KT954523 Piciformes Semnornidae Semnornis frantzii KT954415 KT954524 Piciformes Lybiidae Trachyphonus erythrocephalus KT954413 KT954522 Piciformes Lybiidae Lybius hirsutus KT954412 KT954521 Piciformes Megalaimidae Megalaima oorti KT954411 KT954520 Piciformes Picidae Melanerpes carolinus KT954418 KT954527 Piciformes Picidae Picoides pubescens XM_009905561.1 XM_009905562.1 Piciformes Picidae Picumnus cirratus AF295195 - Piciformes Indicatoridae Indicator variegatus KT954417 KT954526 Piciformes Bucconidae Bucco capensis MPEG_ARA018 Piciformes Bucconidae Nystalus maculatus MPEG_MARJ045 Piciformes Bucconidae Nonnula rubecula INPA_A4705 Piciformes Bucconidae Monasa atra INPA_A8299 Piciformes Bucconidae Chelidoptera tenebrosa MPEG_JTW1160 Piciformes Bucconidae Hapaloptila castanea LSU_12059 Piciformes Bucconidae Micromonacha lanceolata LSU_4489 Piciformes Bucconidae Cyphos macrodactylus MPEG_AMA354 Piciformes Bucconidae Notharchus tectus LSU_28765 Piciformes Bucconidae Hypnellus bicinctus FMNH_339641 Piciformes Bucconidae Nystactes tamatia MPEG_JRT134 Piciformes Bucconidae Notharchus ordii LSU_25460 Piciformes Bucconidae Notharchus hyperrhynchus MPEG_GAPTO296 Piciformes Bucconidae Malacoptila fulvogularis FMNH_321031 Piciformes Bucconidae Malacoptila rufa LSU_103572 Piciformes Galbulidae Jacamalcyon tridactyla MPEG_800 Piciformes Galbulidae Brachygalba lugubris MPEG_293 Piciformes Galbulidae Jacamerops aureus MPEG_JAP375 68

Piciformes Galbulidae Galbacyrhynchus purusianus INPA_A1429 Piciformes Galbulidae Galbula dea INPA_A2288 Piciformes Galbulidae Galbula leucogastra MPEG_AMZ190 Piciformes Galbulidae Galbula ruficauda MPEG_MARJ109 Piciformes Galbulidae Galbula cyanescens MPEG_PUC159 Piciformes Galbulidae Galbula albirostris MPEG_JAP616 Piciformes Galbulidae Galbula cyanicollis MPEG_FLJA056 Coraciformes Alcedinidae Chloroceryle americana KT954422 KT954533 Coraciformes Alcedinidae Halcyon malimbica DQ111819 KT954532 Coraciformes Alcedinidae Alcedo leucogaster DQ111794 KT954531 Coraciformes Momotidae Momotus momota KT954421 KT954530 Coraciformes Todidae Todus angustirostris KT954420 KT954529 Coraciformes Coraciidae Coracias caudata AF143737 AY443126 Coraciformes Brachypteracidae Brachypteracias leptosomus KT954423 KT954534 Coraciformes Meropidae Merops pusillus KT954419 KT954528 Coraciformes Meropidae Merops nubicus XM_008938323.1 XM_008938322.1 Bucerotiformes Upupidae Upupa epops KT954409 KT954517 Bucerotiformes Phoeniculidae Phoeniculus purpureus KT954408 KT954516 Bucerotiformes Bucerotidae Buceros rhinoceros XM_010145185.1 XM_010145184.1 Bucerotiformes Bucerotidae Buceros bicornis KT954407 KT954515 Bucerotiformes Bucerotidae Tockus camurus KT954406 KT954514 Leptosomatiformes Leptosomidae Leptosomus discolor XM_009958543.1 XM_009958545.1 Colliformes Coliidae Colius colius KT954404 KT954512 Colliformes Coliidae Colius striatus XM_010201405.1 XM_010209029.1 Strigiformes Strigidae Strix occidentalis DQ482641 KT954508 Strigiformes Strigidae Ninox novaeseelandiae KT954400 KT954507 Strigiformes Tytonidae Tyto alba XM_009975325.1 XM_009975324.1 Strigiformes Tytonidae Phodilus badius KT954402 KT954510 Accipitrifromes Accipitridae Buteo jamaicensis EF078718 KT954506 Accipitrifromes Accipitridae Elanus caeruleus EF078724 KT954505 Accipitrifromes Pandionidae Pandion haliaetus EF078706 KT954504 Accipitrifromes Sagittaridae Sagittarius serpentarius KT954399 KT954503 Accipitrifromes Cathartidae Cathartes aura EF078766 KT954502 Accipitrifromes Accipitridae Aquila chrysateos XM_011594630.1 XM_011594629.1 Accipitrifromes Accipitridae Haliaeetus albicilla XM_009928640.1 XM_009928639.1 Accipitrifromes Accipitridae Haliaeetus leucocephalus XM_010586008.1 XM_010586006.1 Trogoniformes Trogonidae Apaloderma vittatum XM_009874816.1 XM_009869619.1 Trogoniformes Trogonidae Apaloderma narina AMNH_DOT12430 Trogoniformes Trogonidae Apalharpactes mackloti LSU_49104 Trogoniformes Trogonidae Apalharpactes mackloti AMNH_633881 Trogoniformes Trogonidae Harpactes ardens AY625239 - Trogoniformes Trogonidae Harpactes ardens USNM_607340 Trogoniformes Trogonidae Harpactes diardii AMNH_DOT563 Trogoniformes Trogonidae Harpactes oreskios AY625238 - Trogoniformes Trogonidae Harpactes oreskios ANSP_16308 Trogoniformes Trogonidae Harpactes orrhopheus AY625241 - Trogoniformes Trogonidae Harpactes orrhopheus AMNH_DOT15159 Trogoniformes Trogonidae Harpactes duvaucelli LSU_38592 Trogoniformes Trogonidae Harpactes fasciatus AMNH_778649 69

Trogoniformes Trogonidae Harpactes erythrocephalus AMNH_DOT12240 Trogoniformes Trogonidae Harpactes wardii AMNH_307761 Trogoniformes Trogonidae Harpactes whiteheadii LSU_52627 Trogoniformes Trogonidae Harpactes kasumba AMNH_DOT15326 Trogoniformes Trogonidae Euptilotis neoxenus AMNH_DOT11080 Trogoniformes Trogonidae Pharomachrus pavoninus LSU_4986 Trogoniformes Trogonidae Pharomachrus auriceps hargitti AMNH_175988 Trogoniformes Trogonidae Pharomachrus auriceps auriceps FMNH_473723 Trogoniformes Trogonidae Pharomachrus fulgidus AMNH_322895 Trogoniformes Trogonidae Pharomachrus moccino AMNH_326512 Trogoniformes Trogonidae Pharomachrus antisianus ANSP_19429 Trogoniformes Trogonidae Priotelus temnurus ANSP_20257 Trogoniformes Trogonidae Priotelus roseigaster KU_8098 Trogoniformes Trogonidae Trogon clathratus USNM_613996 Trogoniformes Trogonidae Trogon mesurus ANSP_19305 Trogoniformes Trogonidae Trogon massena KU_2073 Trogoniformes Trogonidae Trogon comptus LSU_11829 Trogoniformes Trogonidae Trogon melanurus INPA_A1995 Trogoniformes Trogonidae Trogon viridis INPA_A5240 Trogoniformes Trogonidae Trogon chionurus LSU_28571 Trogoniformes Trogonidae Trogon melanocephalus USNM_646857 Trogoniformes Trogonidae Trogon citreolus UWBM_101087 Trogoniformes Trogonidae Trogon bardii LSU_71992 Trogoniformes Trogonidae Trogon violaceus MPEG_CN437 Trogoniformes Trogonidae Trogon caligatus LSU_66270 Trogoniformes Trogonidae Trogon ramonianus INPA_A5449 Trogoniformes Trogonidae Trogon curucui INPA_A5286 Trogoniformes Trogonidae Trogon aurantius LGEMA_15782 Trogoniformes Trogonidae Trogon surrucura MPEG_SC015 Trogoniformes Trogonidae Trogon elegans FMNH_434014 Trogoniformes Trogonidae Trogon rufus amazonicus INPA_A5284 Trogoniformes Trogonidae Trogon rufus tenellus LSU_26564 Trogoniformes Trogonidae Trogon rufus chrysochlorus LGEMA_9557 Trogoniformes Trogonidae Trogon mexicanus FMNH_343220 Trogoniformes Trogonidae Trogon aurantiiventris LSU_41625 Trogoniformes Trogonidae Trogon collaris puella FMNH_394272 Trogoniformes Trogonidae Trogon collaris collaris MPEG_CN450 Trogoniformes Trogonidae Trogon personatus LSU_48503 1998

1999

2000

2001

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

2012 Capítulo 3

2013 2014 2015 Ferreira, M.; Aleixo, A.; Bates, J. M.; Cracraft, J.; 2016 Ribas, C. C. Phylogeography and phylogenomics of 2017 jacamars (Aves: Galbulidae) and puffbirds (Aves: 2018 Bucconidae) reveal underestimation of species 2019 diversity and recurrent biogeographic patterns in the 2020 Neotropics. Manuscrito formatado para Zoological 2021 Journal of Linnean Society 2022

2023 2024 2025 Manuscript submission to Zoological Journal of Linnean Society 2026 Contribution type: Article 2027 2028 Phylogeography and phylogenomics of jacamars (Aves: Galbulidae) and 2029 puffbirds (Aves: Bucconidae) reveal underestimation of species diversity 2030 and recurrent biogeographic patterns in the Neotropics 2031 2032 Ferreira, Mateus1*; Aleixo, Alexandre2; Bates, John M.3; Cracraft, Joel4; Ribas, Camila C.5 2033 2034 1 Programa de Pós-Graduação em Genética, Conservação e Biologia Evolutiva, INPA, 2035 Manaus, AM, Brazil 2036 2 Coordenação de Zoologia, MPEG, Belém, PA, Brazil 2037 3 Department of Ornithology, FMNH, Chicago, IL, USA 2038 4 Department of Ornithology, AMNH, New York, NY, USA 2039 5 Coordenação de Biodiversidade, INPA, Manaus, AM, Brazil 2040 *Corresponding author 2041 2042 Correspondence: Mateus Ferreira, Coordenação de Biodiversidade, Instituto Nacional de 2043 Pesquisas da Amazônia, CEP 69080-971, Manaus-AM, Brazil 2044 E-mail: [email protected] 2045 2046 Short running title: Galbuliformes phylogenomic 2047 2048 2049 2050 Abstract 2051 Galbulidae (jacamars) and Bucconidae (puffbirds) are sister families endemic to the 2052 Neotropical region. Together they comprise 57 species and more than a 100 described 2053 subspecies. Both families have their highest diversity in Amazonia. Within Galbulidae, most 2054 species have restricted and parapatric / allopatric distributions in relation to other closely related 2055 species, while within Buccondiae, species are widespread and polytypic. In this study, we 2056 obtained DNA sequence data for over 400 samples, and used previous published results, of all 2057 widespread species to uncover phylogeographic patterns. Then, based on these results, we 2058 selected and sequenced thousands of Ultraconserved Elements to reconstruct the phylogenetic 2059 relationships among these phylogeographic groups and propose the first phylogenetic 2060 hypothesis for these two families with dense taxon sampling. Our phylogeographic results 2061 recovered phylogeographic breaks in almost all studied groups, most of them associated with 2062 the main tributaries of the Amazon River, and many corresponding to already described 2063 subspecies. We then reconstructed phylogenetic relationships based on over 2,000 UCE loci 2064 using a concatenated approach in a Bayesian Inference framework. Overall, most nodes had 72

2065 high support, and the relationships among genera, species and instraspecific diversity were 2066 discussed. We propose the recognition of all subspecies that received support from the 2067 phylogeographic and phylogenomic approaches as distinct species. We found evidence of 2068 paraphyly of several species and proposed taxonomic changes to deal with that. Also, we 2069 propose a new genus of puffbirds, Cryptobucco gen. nov., to accommodate the paraphyly of 2070 Notharchus species.

2071

2072 Introduction 2073 Species usually are the basic unit of any study in evolutionary biology. Considering they 2074 should represent the lowest and only non-arbitrary rank above individuals, species are the basic 2075 operational unit for comparing any intrinsic evolutionary aspect, such as physiology, behaviour, 2076 morphology, etc. However, we still lack a broad and comprehensive concept for species 2077 recognition (Cellinese, Baum & Mishler, 2012; de Queiroz, 2007; de Queiroz, 2012). In birds, 2078 taxonomy has been historically influenced by the Biological Species Concept (Mayr, 1942; 2079 Mayr, 1976), based on reproductive isolation as the main criterion for species delimitation. 2080 Therefore, since this concept was adopted several distinct allopatric populations were lumped 2081 as subspecies due to morphological similarities pending further investigation to prove the 2082 absence of gene flow (Peters, 1945; Peters, 1948). This implies that allopatric and parapatric 2083 populations, even if diagnosably distinct, should only be recognized as full species if there is 2084 evidence of reproductive isolation (Gill, 2014). 2085 In the Neotropical region, and especially in Amazonia, one of the main issues that 2086 obscures the recognition of diversity patterns is the fact that most widespread species are in fact 2087 complexes of taxa, usually diagnosable and geographically structured, that are lumped under 2088 the same species name due to their morphological similarities and physical isolation. Many of 2089 these polytypic species, when thoroughly sampled, prove to include distinct lineages, 2090 sometimes not even closely related to each other (Bravo, Chesser & Brumfield, 2012; Bravo, 2091 Remsen, Whitney & Brumfield, 2012; Fernandes, Wink, Sardelli & Aleixo, 2014; Isler, Bravo 2092 & Brumfield, 2013; Lopes, Chaves, Aquino, Silveira & Santos, 2017; Lutz, Weckstein, Patane, 2093 Bates & Aleixo, 2013; Ribas, Aleixo, Nogueira, Miyaki & Cracraft, 2012; Ribas, Aleixo, 2094 Gubili, d'Horta, Brumfield & Cracraft, 2018; Tobias, Bates, Hackett & Seddon, 2008). The 2095 recognition of these hidden lineages is critical for appropriate hypothesis formulation in 2096 macroevolution and biogeography (Donoghue & Moore, 2003; Lexer, Mangili, Bossolini, 2097 Forest, Stölting, Pearman, Zimmermann, Salamin & Carine, 2013). For example, Amazonian

2098 areas of endemism were recognized based on congruent distribution patterns of bird species 2099 (Borges & Da Silva, 2012; Cracraft, 1985), and have been used as a basis to formulate 2100 hypothesis of biotic diversification in Amazonia (Haffer, 1969; Haffer, 1974; Haffer, 1997). 2101 Considering that any biogeographic study should be based on a taxonomy that correctly 2102 recognizes the evolutionary units included in the studied groups, for the present study we 2103 densely sampled all recognized taxa within two sister families of birds restricted to the 2104 Neotropical region. Galbulidae and Bucconidae form a clade, sometimes recognized in its own 2105 order Galbuliformes, that diverged from all the other Piciformes during the early Eocene and 2106 diverged from each other in the Late Eocene (Prum, Berv, Dornburg, Field, Townsend, 2107 Lemmon & Lemmon, 2015). Although the ancestor was from the Afrotropical region the two 2108 families’ entire diversification happened inside the Neotropical region (Claramunt & Cracraft, 2109 2015). Hence, making these two families excellent models to understand how landscape 2110 evolution of the Neotropical region influenced diversification. However, there are no 2111 phylogenetic hypotheses about relationships within these two families, and the few 2112 phylogeographic studies conducted so far with Bucconidae species showed that the diversity is 2113 highly underestimated by current species limits (Almeida, 2013; Duarte, 2015; Ferreira, Aleixo, 2114 Ribas & Santos, 2017; Soares, 2016). Although Galbulidae species were never subjected to 2115 phylogeographic studies, with 19 species distributed in 5 genera, jacamar distributions were 2116 used as models by Haffer (1974), together with other families, when he proposed his theory for 2117 Amazonian diversification (Haffer, 1974). Haffer recognized eight zoogeographic groups, five 2118 were composed of species complexes, and two were widespread polytypic species. Bucconidae, 2119 in turn, are composed of 38 species distributed in 12 genera. However, half of those species 2120 consist of polytypic groups lumped as subspecies due to morphological similarities. Groups 2121 such as the White-fronted Nunbird, Monasa morphoeus, and the Rusty-breasted Nunlet, 2122 Nonnnula rubecula, are composed of several subspecies, which in fact still underestimate the 2123 phylogeographic structure recovered for them (Soares, 2016). On the other hand, Malacoptila 2124 species are widespread species for which only a few subspecies were described, however, 2125 phylogeographic patterns indicated a great underestimation of taxonomic diversity. For 2126 example, for a single species, the Rufous-necked Puffbird (M. rufa), that only includes two 2127 subspecies described, ten distinct genetic lineages were recovered (Ferreira et al., 2017). Due 2128 to these first results, the present study focused on sampling all named taxa described for these 2129 two families, and sampling all widespread species throughout their distribution to uncover 2130 phylogeographic patterns. Based on these results, we selected samples representing all

2131 phylogeographic groups and sequenced thousands of Ultraconserved Elements (UCE) 2132 (Faircloth, McCormack, Crawford, Harvey, Brumfield & Glenn, 2012; McCormack & 2133 Faircloth, 2013; McCormack, Harvey, Faircloth, Crawford, Glenn & Brumfield, 2013) to 2134 recover their phylogenetic relationships. Our aims are (1) to characterize the phylogeographic 2135 patterns and population structure within widespread species, recognizing the cryptic diversity 2136 within them, when present; (2) propose a densely sampled phylogenetic hypothesis for these 2137 two families; and (3) discuss patterns of diversification in the entire clade. 2138

2139 Material and Methods 2140 Sampling and DNA isolation 2141 We sampled 436 individuals from almost all named taxa currently recognized within 2142 Galbuliformes (Gill & Donsker, 2018; Peters, 1948; Piacentini, Aleixo, Agne, Mauricio, 2143 Pacheco, Bravo, Brito, Naka, Olmos, Posso, Silveira, Betini, Carrano, Franz, Lees, Lima, Pioli, 2144 Schunck, do Amaral, Bencke, Cohn-Haft, Figueiredo, Straube & Cesari, 2015; Rassmussen & 2145 Collar, 2002; Remsen, Areta, Cadena, Claramunt, Jaramillo, Pacheco, Pérez-Emen, Robbins, 2146 Stiles, Stotz & Zimmer, 2018 Tobias, 2017), and when available, we used published sequences 2147 to select samples for UCE sequencing. All samples are represented by voucher specimens 2148 deposited at the ornithological collections of the American Museum of Natural History 2149 (AMNH), Academy of Natural Sciences of Drexel University (ANSP), Field Museum of 2150 Natural History (FMNH), Instituto Nacional de Pesquisas da Amazônia (INPA), Kansas 2151 University (KU), Laboratório de Genética e Evolução Molecular de Aves - USP (LGEMA), 2152 Louisiana Museum of Natural History (LSUMZ), Museu Paraense Emílio Goeldi (MPEG), 2153 Smithsonian Institution National Museum of Natural History (USNM) and Burke Museum 2154 (UWBM) (Table S1). 2155 DNA from fresh tissue was extracted with the DNeasy kit (Qiagen Inc.), following the 2156 manufacturer’s protocol. For taxa lacking fresh tissues we sampled toe pad clips from museum 2157 specimens at the American Museum of Natural History (AMNH). Toe pads were cut from 2158 specimens with a sterile surgical blade and processed in a dedicated room for ancient DNA 2159 (aDNA Lab, AMNH). They were rinsed with 100% ethanol, and twice with ultra-pure water 2160 prior to digestion to remove any inhibitor that could cause problems in downstream procedures. 2161 We then extracted DNA with the DNeasy kit (Qiagen Inc.), replacing the regular silica columns 2162 with the QIAquick (Qiagen Inc.) columns, to ensure maximum DNA yield. 2163

2164 Phylogeographic structure and UCE sampling 2165 Widespread species that lacked previous studies were sampled throughout their 2166 distributions to uncover phylogeographic structure. We amplified one mitochondrial gene 2167 (NADH subunit 2 – ND2) following conventional PCR protocols and sequenced both strands 2168 with BigDye® Terminator v3.1 in an ABI 3130/3130XL automated capillary sequencer 2169 (Applied Biosystems®) following manufacturer’s protocols. The sequences were edited on 2170 Geneious version 10.2.3 (Kearse, Moir, Wilson, Stones-Havas, Cheung, Sturrock, Buxton, 2171 Cooper, Markowitz, Duran, Thierer, Ashton, Meintjes & Drummond, 2012) and aligned with 2172 MAFFT (Katoh & Standley, 2013) under default parameters. We analysed each species 2173 complex independently. Within Galbulidae we analysed five species complexes: 1) 2174 Brachygalba and Jacamaralcyon; 2) Jacamerops; 3) Galbula dea; 4) Galbula cyanicollis, G. 2175 chalcocephala, and G. albirostris; and 5) G. ruficauda, G. pastazae, G. cyanescens, G. 2176 tombacea, and G. galbula. We used a previous study to select samples for G. leucogastra and 2177 G. chalcothorax (Ferreira et al., submitted). For Bucconidae, we gathered data in this study for 2178 five polytypic species or species complexes: 1) Bucco capensis; 2) Cyphos macrodatylus; 3) 2179 Notharchus tectus; 4) Notharchus ordii, N. hyperrhynchus, N. macrorhynchus, N. swainsoni, 2180 and N. pectorales; and 5) Chelidoptera tenebrosa. Sample selection for the genera Monasa, 2181 Nonnula, Malacoptila, and Nystalus was based on previous studies (Almeida, 2013; Duarte, 2182 2015; Ferreira et al., 2017; Soares, 2016). The best evolutionary model for each matrix was 2183 selected by jModelTest 2.1.10 (Darriba, Taboada, Doallo & Posada, 2012). We performed a 2184 Bayesian inference analysis (BI) implemented in MrBayes 3.2.6 (Ronquist, Teslenko, van der 2185 Mark, Ayres, Darling, Hohna, Larget, Liu, Suchard & Huelsenbeck, 2012) with four parallel 2186 simultaneous runs consisting of a total of 4x107 generations, sampling trees every 1000 2187 generations. ESS values, stationarity, and convergence among runs were checked in Tracer 1.6 2188 (Rambaut, Suchard, Xie & Drummond, 2014). Based on these results we selected our samples 2189 for UCE sequencing. All extracts were sent to Rapid Genomics (Gainsville, FL) for library prep 2190 and target-capture sequencing of 2321 loci of Ultraconserved Elements (UCE) (Faircloth et al., 2191 2012; McCormack et al., 2013). 2192 UCE assembly 2193 The raw sequence data were processed with the Phyluce script pack (Faircloth, 2016). 2194 We employed illumiprocessor (Faircloth, 2013) and Trimmomatic (Bolger, Lohse & Usadel, 2195 2014) to remove adapter contamination and low-quality reads. We assembled our targeted 2196 regions using a reference genome for each family. For Bucconidae, we used the Collared

2197 puffbird (Bucco capensis), and for Galbulidae, the Paradise jacamar (Galbula dea) genomes. 2198 We mapped the UCE probes back to each genome using the script 2199 phyluce_probe_run_multiple_lastzs_sqlite, and then, phyluce_probe_slice_sequence_from_g- 2200 enomes to extract the probe region plus 500 base pairs from each flanking region (Faircloth, 2201 2016). With these sequences as a reference we mapped back the clean reads of each individual 2202 employing Bowtie2 (Langmead & Salzberg, 2012) plugin 7.2.1 inside Geneious version 10.2.3 2203 (Kearse et al., 2012). The consensus sequences were called with the highest quality threshold 2204 and a depth of at least 4 reads. Each locus was aligned with MAFFT (Katoh & Standley, 2013) 2205 under default parameters. 2206 Phylogenetic relationship 2207 Even though the sister relationship between Galbulidae and Bucconidae is well 2208 established (Hackett et al., 2008; Livezey & Zusi, 2007; Prum et al., 2015), we used the 2209 Rhinoceros hornbill (Buceros rhinoceros, Bucerotidae)(Gilbert, Jarvis, Li, Li, Avian Genome 2210 Consortium, Wang & Zhang, 2014b), the Northern Carmine bee-eater (Merops nubicus, 2211 Meropidae)(Gilbert, Jarvis, Li, Li, Avian Genome Consortium, Wang & Zhang, 2014c), and 2212 the Downy woodpecker (Picoides pubescens, Picidae)(Gilbert, Jarvis, Li, Li, Avian Genome 2213 Consortium, Wang & Zhang, 2014a) as outgroups. To recover the phylogenetic relationships, 2214 we performed a Bayesian Inference analysis in ExaBayes v1.4 (Aberer, Kobert & Stamatakis, 2215 2014) employing the concatenated matrix of all UCE loci with 75% completeness, where only 2216 loci that had at least 75% of all individuals were selected. Four parallel chains consisting of 2217 4x107 generations were performed. 2218

2219 Results 2220 Phylogeographic results 2221 With a few exceptions, we obtained the whole ND2 sequence for all samples. 2222 Phylogenetic trees and maps of samples and lineages’ distributions can be found in the 2223 Supplementary Material (Figures S1-S10). Overall, most species complexes contained 2224 phylogeographic structure in the mtDNA that matches known areas of endemism for birds. The 2225 only two widespread species that apparently lacked phylogeographic structure were Cyphos 2226 macrodactylus and Chelidoptera tenebrosa. The phylogeographic breaks were more 2227 conspicuous in birds with stronger association with terra-firme forests [Fig. S2-S4, S6, 2228 Malacoptila spp. (Ferreira et al., 2017), Monasa morphoeus and Nonnula rubecula (Soares, 2229 2016)]. However, species associated with other habitats, such as várzeas, open habitats (i.e.

2230 non-forested) or white-sand environments also showed structure [Fig. S1, S5, S8-S9, also 2231 Galbula leucogastra/chalcothorax, Nystactes (Almeida, 2013), Nystalus spp. (Duarte, 2015), 2232 and Nonnula ruficapilla (Soares, 2016)]. Nonetheless, some lineages are represented by a single 2233 individual and additional samples should be collected and analysed to make further 2234 assumptions. It is also worth to note that some species were paraphyletic in the mtDNA. The 2235 most remarkable are the complex Brachygalba lugubris, B. albogularis (Fig. S1), G. albirostris, 2236 G. cyanicolis, and G. chalcocephala (Fig. S4), G. ruficauda, G. cyanescens (Fig. S5); 2237 Notharchus tectus, N. subtectus (Fig. S8); N. hypperhynchus, N. swainsoni, N. macrorhynchus 2238 (Fig. S9). 2239 UCE sequencing 2240 The reference sequences we assembled from the Collared puffbird (Bucco capensis) and 2241 the Paradise jacamar (Galbula dea) genomes included 2226 and 2279 sequences, respectively. 2242 The mean number of sequences was 2,240,885 reads; and a mean number of 2191 UCE loci per 2243 sample (Table S1). The matrix for Galbulidae contained 2165 loci, while for Bucconidae, the 2244 matrix had 2158 loci. 2245 Phylogenetic results 2246 In general, the ExaBayes tree is well supported, with most of the nodes with lower support 2247 found near the tips (Fig. 2, 3). Galbulidae consisted of two clades, the first comprises 2248 Jacamaralcyon and Brachygalba, and the other, Jacamerops, Galbacyrhynchus, and Galbula 2249 (Fig. 1). Within Bucconidae, some genera were paraphyletic. Bucco, that previously included 2250 four species (Gill & Donsker, 2018; Peters, 1948; Piacentini et al., 2015; Remsen et al., 2018), 2251 comprises three distinct genera as previously suggested by morphological characters 2252 (Rassmussen & Collar, 2018): B. capensis Linneus, 1766 is the family and genus type and more 2253 closely related to Nystalus; Cyphos macrodactylus von Spix, 1824, is sister to the clade that 2254 comprises Notharchus, Hypnelus, and Nystactes; and finally, Nystactes tamatia (J. F. Gemelin, 2255 1788), and N. noanamae (Hellmayer, 1909), more closely related with Hypnelus species (Fig. 2256 3). Notharchus was also paraphyletic, with Hypnelus and Nystactes embedded within it. N. 2257 tectus and N. subtectus were sister to Hypnelus, Nystactes, and the remaining Notharchus 2258 species (Fig. 1, 3). 2259 The relationships within genera in the UCE trees (Fig. 2, 3) mostly agreed with the 2260 mtDNA phylogeographic structure. Most notably is the paraphyly of Brachygalba lugubris in 2261 relation to B. albogularis (Fig. 2), and the polyphyletic status of Galbula ruficauda, in which 2262 the lineages from Central America (G. melanogenia), and northern South America (G. pallens,

2263 G. ruficauda), including G. pastazae, are sister group to the clade comprising the species group 2264 of G. albicollis (albicollis, chalcocephala, cyanicollis) and G. galbula (galbula, pastazae, 2265 cyanescens, rufoviridis). Also, in contrast with the mtDNA, the two samples of G. cyanescens 2266 are sister to G. heterogyna and G. rufoviridis from the Brazilian Shield, instead of being 2267 embedded between them (Fig. S5). For puffbirds, the UCE tree also recovered the paraphyly of 2268 N. tectus subspecies (Fig. 3), and for the hyperrhynchus group (Fig. S9), we recovered N. 2269 macrorhynchus sister to N. swainsoni and N. hyperrhynchus, rendering the Amazonian group 2270 paraphyletic. 2271

2272 Discussion 2273 Phylogenetic results 2274 Our dense sampling coupled with the use of UCE loci provided good insights about 2275 genera and species relationships. We sampled all species, and almost all subspecies, for the two 2276 families, and characterized the spatial distribution of mtDNA lineages for all widespread 2277 species. Predominantly, our results indicate a severe disparity between currently recognized 2278 species and the potential number of independent evolutionary units within these clades. 2279 Avian taxonomy has historically been greatly influenced by the Biological Species 2280 Concept (BSC), which assumes that reproductive isolation is required for recognition of species 2281 status (de Queiroz, 2005). This condition, can be easily detected in sympatric taxa, however, 2282 for parapatric and allopatric populations, natural observations are very hard to detect. 2283 Consequently, many morphologically distinct taxa have been lumped into species complexes, 2284 pending further analysis to prove them different. Thus, the null hypothesis for species 2285 recognition has been of peer-reviewed publications proving that essential reproductive isolation 2286 is true among allopatric populations. It implies that we should be looking for reasons that 2287 differentiate allopatric populations, either through genetic evidence or some other characteristic 2288 that would lead to reproductive isolation, rather than assuming that they already are 2289 reproductive isolated, because they are not in contact, and looking for evidence proving the 2290 contrary (Gill, 2014). Albeit avian taxonomy and systematics is probably the best known among 2291 vertebrates, there are still many taxa to be described (Barrowclough, Cracraft, Klicka & Zink, 2292 2016), and although species concept, or criteria, are amid one of the most controversial topics 2293 in biology (Aleixo, 2007; Dayrat, Cantino, Clarke & de Queiroz, 2008; de Queiroz, 2012), the 2294 appropriate understanding of a lineage’s evolutionary history is essential to several fields, 2295 including conservation and biogeography (Avendaño, Arbeláez-Cortés & Cadena, 2017; Ribas,

2296 Gaban-Lima, Miyaki & Cracraft, 2005; Tobias, Bates, Hackett & Seddon, 2008), especially in 2297 the emergent field of geogenomics (Baker, Fritz, Dick, Eckert, Horton, Manzoni, Ribas, 2298 Garzione & Battisti, 2014). Therefore, we are confident our results provide great insight about 2299 Galbulidae and Bucconidae systematics and will enable future biogeography studies to uncover 2300 how the landscape evolution of South America shaped this group’s diversity. 2301 Galbulidae systematics 2302 Galbulidae currently recognized diversity includes 19 species distributed in 5 genera 2303 (Tobias, 2017). Our results, however, show that this diversity is severely underestimated. In 2304 addition to the fact that most widespread species have genetic lineages structured 2305 geographically, we also found evidence that, at least four species are para- (Brachygalba spp.) 2306 or polyphyletic (Galbula ruficauda complex). Conceding that we recognize all subspecies that 2307 were monophyletic in our analyses and elevate them to species status, the species diversity of 2308 Galbulidae practically doubles, from 19 to 37 species, including at least six new taxa that need 2309 to be formally described. Biogeographically, there is also some noteworthy patterns that arouse 2310 from the mtDNA data. All widespread species presented some degree of genetic structure in 2311 the known areas of endemism in Amazonia (Borges & Da Silva, 2012; Cracraft, 1985). Most 2312 of the larger Amazonian tributaries, including rivers such as the Negro, Madeira, Solimões, and 2313 Amazonas delimit lineages in opposite margins, however, if they were responsible for causing 2314 these divergences still need to be investigated. 2315 According to our phylogenetic hypothesis for Galbulidae, there are now eight main 2316 groups of species: 2317 1. Brachygalba and Jacamaralcyon 2318 Brachygalba and Jacamaralcyon species were recovered as sisters to all other jacamars. 2319 The monotypic Jacamaralcyon species, Jacaramaralcyon trydactyla (Viellot, 1817), is 2320 endemic to the Atlantic Forest, inhabiting semi-deciduous or gallery forest. This species was 2321 recovered as sister to all other Brachygalba species (Fig. 1), which prefer forest edges and open 2322 habitats throughout the Amazon basin and north South America. B. goeringii Sclatter, PL & 2323 Salvin, 1869 and B. salmoni Sclatter, PL & Salvin, 1879 represent two distinct lineages within 2324 Brachygalba radiation (Fig. 2), with very distinct plumages and restricted distributions in 2325 northern South America. B. goeringii was recovered as sister to all other Brachygalba species, 2326 and B. salmoni, as sister group to the species group of B. lugubris (naumburgae, obscuriceps, 2327 lugubris, and melanosterna) and B. albogularis (von Spix, 1824), from the Amazon basin (Fig. 2328 2). Because B. albogularis was embedded within B. lugubris lineages, we recommend that the

2329 current subspecies of B. lugubris should now be elevated to species status. This way, we resolve 2330 the paraphyly of B. lugubris, and fully recognize all its diversity. Further studies are necessary 2331 to completely understand B. obscuriceps Zimmer, JT & Phelps, 1947 and B. naumburgae 2332 Chapman, 1931 distributions, especially regarding the relationship between B. lugubris, and B. 2333 l. fulviventris Sclater, PL, 1891 and B. l. caquetae Chapman, 1917. 2334 2. Jacamerops 2335 Jacamerops individuals are so distinct from the other jacamars that were once considered 2336 to belong to a separate subfamily Jacameropinae. Although this treatment is no longer followed, 2337 Jacamerops are by far the bulkiest jacamars, inhabiting the midstory and canopy of continuous 2338 forest in the Amazon basin. Among the four subspecies recognized, J. a. ridgway Todd, 1943 2339 formed a well supported clade in both analyses. (Fig. 2, S2), while J. a. aureus (Statius Müller, 2340 PL, 1776) was monophyletic in the mtDNA analysis (Fig. S2) but paraphyletic in the UCE 2341 analysis, with the two individuals from the Guiana Shield as sister to all other J. aureus 2342 individuals (Fig. 2). Since the type from J. a. aureus is British Guiana (Peters, 1948), we 2343 consider that only this group should be recognized as J. aureus, while the second lineage should 2344 receive a new name (Fig. 2). An interesting biogeographic pattern that arouse from Jacamerops 2345 data was the sister relationship between J. penardi Bangs & Barbour, 1922, from Central 2346 America, and J. isidori Deville 1849, from the Madeira-Solimões interfluve. A similar pattern 2347 was found in the Hylophylax species complex (Fernandes et al., 2014). Finally, J. ridgwayi 2348 Todd, 1943 requires further study to fully evaluate all diversity present in this group, our results 2349 suggest the presence of at least 4 mtDNA lineages, each separated by the main rivers of the 2350 Brazilia Shield. 2351 3. Galbalcyrhynchus 2352 Galbacyrhynchus species are endemic to floodplain forests from Western Amazon. 2353 Galbalcyrhynchus purusianus Goeldi, 1904 was considered conspecific with G. leucotis Des 2354 Murs, 1845, and they were actually considered male and female forms of the same species. 2355 Nonetheless, the parapatric distribution and the apparently lack of intermediate forms (Haffer, 2356 1974) render these two taxa the status of distinct species (Fig. 2). 2357 4. Galbula dea complex 2358 Previously allocated in the genus Urogalba, Galbula dea individuals are the most 2359 morphologically distinct among Galbula species. Our results recovered six distinct mtDNA 2360 lineages (Fig. S3) that matches with the UCE results (Fig. 2), in which four already have 2361 associated names. G. dea (Linnaeus, 1758) from the Guiana Shield; G. brunneiceps (Todd,

2362 1943) from the Negro-Solimões interfluve; G. phainopepla (Todd, 1943) from the Solimões- 2363 Madeira interfluve; and G. amazonum (Sclater, PL, 1855). The last lineage, from the Madeira- 2364 Tapajós interfluve was considered to be part of G. d. brunneiceps (Peters, 1948), however, since 2365 the type locality from G. brunneiceps is Manacapurú, Rio Solimões, Brazil, we suggest that a 2366 new name should be given to this lineage. 2367 5. G. leucogastra/chalcothorax 2368 This complex includes the only jacamars that inhabit white-sand environments (Adeney, 2369 Christensen, Vicentini & Cohn-Haft, 2016) in the Amazon basin. Although highly structured 2370 throughout its distribution (Ferreira et al., submitted) this group lacks morphological 2371 distinctiveness among genetic lineages, thus further systematic and taxonomic work is required 2372 before the proposition of any change in nomenclature. 2373 6. Galbula albirostris, G. chalcocephala, and G. cyanicollis 2374 These tree species were formerly considered conspecifics in G. albirostris Latham, 1790 2375 (Peters, 1948), later Haffer (1974) recognized G. cyanicollis Cassin, 1851, based on the lack of 2376 interbreeding between these two forms. Our results support the recognition of all three species, 2377 with G. albirostris restricted to the Guiana Shield, east of Negro River; G. chalcocephala 2378 Deville, 1849 in between the west bank of lower Negro river, west of Branco River, and north 2379 of Solimões all the way down to the west bank of the upper Ucayali River (Harvey, Seeholzer, 2380 Cáceres A, Winger, Tello, Camacho, Aponte Justiniano, Judy, Ramírez, Terrill, Brown, León, 2381 Bravo, Combe, Custodio, Zumaeta, Tello, Bravo, Savit, Ruiz, Mauck & Barden, 2014); and at 2382 last, G. cyanicollis, along the south bank of Amazon River. This group of species, in contrast 2383 with other jacamars, only inhabits the interior of forests, mainly in terra-firme habitats. Not 2384 surprisingly, the mtDNA showed lineages separated by the main Amazonian tributaries (Fig. 2385 S4). However, some lineages presented some interesting biogeographic patterns, such as the 2386 distinct lineage at the lower portion of Madeira-Tapajós interfluve, that is also found in other 2387 groups of birds, such as Rhegmatorhina berlespchi (Ribas et al., 2018), Malacoptila rufa 2388 (Ferreira et al., 2017), and Glyphorhynchus spirurus (Fernandes, Gonzalez, Wink & Aleixo, 2389 2013). Another pattern, that has not been reported before for birds, is the distinct lineage 2390 between the Purus and Tapajós Rivers (Fig. S4). This is the first evidence of a lineage of and 2391 understory terra-firme bird that has n structure related to the Madeira River. 2392 7. G. melanogenia, G. pastazae, G. pallens, and G. ruficauda. 2393 Although G. melanogenia Sclater, PL, 1852, was first described as a full species, it was 2394 later lumped together with G. rufoviridis Cabanis, 1851 in G. ruficauda Cuvier, 1816 due to

2395 morphological similarity (Peters, 1948). Our results, however, recovered this group as sister to 2396 the clade containing G. cyanicolis and G. galbula complex (Fig. 2). Also, G. pastazae 2397 Taczanowski and Berlepsch, 1885, probably the only jacamar to live in high altitudes, is 2398 embedded between G. melanogenia and G. ruficauda (Fig. S5, 2). Therefore, our 2399 recommendation is that G. melanogenia, from Central America and the Pacific coast of 2400 Colombia and Ecuador, along with G. pallens Bangs, 1898 and G. ruficauda Cuvier, 1816 2401 should be recognized as species. Further studies are required to check the validity of G. r. 2402 brevirostris Cory, 1913. 2403 8. Galbula galbula, G. tombacea, G. cyanescens and G. rufoviridis 2404 This group is often regarded as G. galbula (Linnaeus, 1766) species group due to 2405 morphological and ecology similarity. Usually associated with forest edges and floodplains 2406 forest, while G. albirostris species group, its sister clade (Fig. 2), is usually associated with the 2407 interior of terra-firme forests. Despite been associated with floodplain forests, and therefore, 2408 not “bounded” by rivers, there are no previous reports of hybridization among these taxa. We 2409 found, however, that the individual INPA A019 is phenotypically G. tombacea (checked by 2410 M.F.), however, the mtDNA clustered with G. cyanescens Deville, 1849 (Fig. S5). This is the 2411 only reported case of hybrids among this group, the other individual that could be a hybrid - G. 2412 cyanescens, voucher MPEG MAD305 - is phenotypically G. cyanescens (checked by Fátima 2413 Lima), even though the individual was collected in the right bank of Madeira River, supposedly 2414 the limit between distributions of G. cyanescens and G. heterogyna Todd, 1932. Another 2415 important pattern that we can observe in this group is the apparently discordance between the 2416 mtDNA and UCE trees (Fig. S5, 3). Our mtDNA tree recovered G. cyanescens as one lineage 2417 embedded within lineages of G. rufoviridis and G. heterogyna. It also recovered G. rufoviridis 2418 as paraphyletic (Fig. S5). The UCE tree instead, recovered G. cyanescens as sister to G. 2419 heterogyna and G. rufoviridis (Fig. 2). In addition, all samples we sequenced for G. rufoviridis 2420 were recovered as monophyletic and sister to G. heterogyna. Thus, this might be an evidence 2421 of mtDNA capture (Sloan, Havird & Sharbrough, 2017), in which probably G. cyanescens 2422 captured the mtDNA lineage of G. heterogyna. However, further studies are required to 2423 understand the direction and timing of this event. 2424 Bucconidae systematics 2425 Our results showed that, similar to the situation with Galbulidae, Bucconidae diversity is 2426 underestimated. In addition, we found evidence of genera paraphyly. Phylogeographic patterns 2427 recovered for widespread puffbird species varied from little to no genetic structure, as in

2428 Chelidoptera and Cyphos, to highly structured, as in Malacoptila (Ferreira et al., 2017), 2429 Nonnula rubecula and N. ruficapilla, and Monasa morphoeus (Soares, 2016). Historically, 2430 apart from the morphologically explicit genera Hapaloptila, Chelidoptera, Malacoptila, 2431 Micromonacha, Monasa, and Nonnula, all other species were lumped in Bucco Brisson, and 2432 later split in Notharchus and Hypnelus. Currently, although some authors recognize different 2433 genera for former Bucco species (i.e. Cyphos, and Nystactes) (Rassmussen & Collar, 2018), 2434 many others still keep several species within the genus Bucco (Gill & Donsker, 2018; Piacentini 2435 et al., 2015; Remsen et al., 2018). Our results however recovered Bucco as polyphyletic, and 2436 thus, we favor the recognition of Cyphos Spix, 1824 (which has priority over Argicus Cabanis 2437 & Heine, 1863) and Nystactes Gloger 1827. Also, we recovered Notharchus specie as 2438 paraphyletic, with the species group of N. tectus (Boddaert, 1783) as sister to the clade 2439 containing Hypnelus, Nystactes and the other species of Notharchus. One way to resolve this 2440 paraphyly would be to include Hypnelus and Nystactes in the genus Notharchus, however, both 2441 Nystactes and Hypnelus species are morphologically distinct from any of Notharchus species. 2442 Therefore, since the type species of Notharchus is N. hyperrhynchus Sclater, PL, 1856, we 2443 propose a new generic name for this group: 2444 Cryptobucco, gen. nov. Ferreira, Aleixo, Bates, Cracraft & Ribas 2445 Type species: Bucco tectus Boddaert, 1783 2446 Included taxa: Cryptobucco tectus (Boddaert, 1783), comb. nov.; Cryptobucco picatus 2447 (Sclater, PL, 1856), comb. nov.; Cryptobucco subtectus (Sclater, PL, 1860), comb. nov. 2448 Etymology: The genus name Cryptobucco alludes to the fact that this group, first 2449 described as Bucco and then placed in Notharchus, represents a hidden diversity inside 2450 Bucconidae that was until now concealed due to morphological similarity among species of 2451 Notharchus and the new genus. The name is of masculine gender. 2452 1. Bucco capensis and Nystalus 2453 Bucco capensis Linnaeus, 1766 and Nystalus species were recovered as sister to all other 2454 puffbirds. The sister relationship we recovered between B. capensis and Nystalus is validated 2455 by the bill-tip morphology that was previously used to separate former Bucco species in the 2456 genera Cyphos and Nystactes (Rassmussen & Collar, 2018). Our results for B. capensis samples 2457 recovered three clades in the UCE tree (Fig. 3) in contrast to the four clades found in the mtDNA 2458 analysis (Fig. S7). Our UCE analysis also favor the recognition of B. dugandi Gilliard, 1949 2459 and suggest the presence of a new taxon yet undescribed. Nystalus relationships found here 2460 were similar to a previous study that used only one mtDNA marker (Duarte, 2015), which

2461 recovered N. maculatus (Gmelin, JF, 1788) and N. striatipectus (Sclater, PL, 1854) as sister to 2462 all remaining species, and N. chacuru (Vieillot, 1816) as sister to N. radiatus (Sclater, PL, 1854) 2463 and the N. striolatus species complex: N. obamai Whitney et al., 2013; N. striolatus (Pelzeln, 2464 1856), and N. torridus Bond & Meyer de Schauensee, 1940. 2465 2. Chelidoptera 2466 The Swallow-winged puffbird, Chelidoptera tenebrosa Pallas, 1782, is by far the most 2467 distinct puffbird, aberrant both in morphology and in ecology. With swallow-like morphology, 2468 they are highly specialized in aerial activity, and its flying proficiency is probably the cause for 2469 the lack of genetic structure we found in the mtDNA (Fig. S10). However, we were unable to 2470 sample UCE from the two toe pad samples, from the subspecies C. t. pallida Cory, 1913, from 2471 Northwest Venezuela; and C. t. brasiliensis Sclater, PL, 1862, from the east coast in Brazil. 2472 3. Monasa and Nonnula 2473 Monasa and Nonnula were the focus of a recent phylogeographic study (Soares, 2016). 2474 Species from both genera presented high levels of genetic structure in the mtDNA, and we 2475 sampled one individual per mtDNA lineage that were uncovered previously. We recovered 2476 Monasa as sister to Chelidoptera, and these two sisters to Nonnula (Fig. 1). Relationships inside 2477 each genus (Fig. 3) were also congruent to Soares (2016). In addition to this previous study, we 2478 were able to sample three toe pads representing three subspecies of M. morphoeus (Hahn & 2479 Küster, 1823): M. m. morphoeus (Hahn & Küster, 1823) from the east coast of Brazil; M. m. 2480 pallescens Cassin, 1860; and M. m. grandior Sclater, PL & Salvin, 1868, both from Central 2481 America. However, their phylogenetic relationship with other subspecies of M. morphoeus was 2482 uncertain (Fig. S11) and further studies are required to fully understand if the phylogeographic 2483 structure found in the mtDNA matches the UCE. For Nonnula, our results support the paraphyly 2484 of N. ruficapilla (Tschudi, 1844), with N. amaurocephala Chapman, 1921 is embedded within 2485 it. Both genera are being studied using broader sampling of individuals and molecular markers. 2486 4. Malacoptila 2487 Malacoptila UCE topology was congruent with the concatenated dataset topology from 2488 Ferreira et al. (2017), placing M. fulvogularis Sclater, PL, 1854 as sister to all other species. 2489 This result changes the previous biogeographic interpretations, and a more detailed study 2490 focusing on this genus is necessary, to fully understand the relationship of Malacoptila species, 2491 including the position of M. mystacalis (Lafresnaye, 1850), that in the concatenated UCE tree 2492 was recovered as sister to all other species (Fig. 3). Since, M. mystacalis UCE contigs were 2493 shorter due to DNA degradation common in toe pad samples (McCormack, Tsai & Faircloth,

2494 2016), we assembled a small subset of Malacoptila samples to minimize the effects of missing 2495 data, and yet, M. mystacalis was again, recovered as sister to all other species of Malacoptila 2496 (Fig. S12). Further sampling of this narrow endemic species is required to confirm if this pattern 2497 is true, or an artefact of toe pad sequencing error. 2498 5. Hapaloptila 2499 The monotypic Hapaloptila castanea (Verreaux, J, 1866) was recovered as sister group 2500 to Micromonacha, Cyphos, Cryptobucco, Hypnelus, Nystactes, and Notharchus (Fig. 1, 3). 2501 Very distinct in morphology, this species is specialized in cloud forests, usually above 1,500 2502 m, and even though it can be found in both sides of the Andes, no subspecies was ever described. 2503 The two specimens we samples are from opposite sides of Andes, however a more focused 2504 work on this species is required to understand the relationships among these apparently disjunct 2505 populations. 2506 6. Micromonacha 2507 Micromonacha lanceolata (Deville, 1849) occurs in the middle and upper stories of 2508 forests in both sides of the Andes, usually below 1,500 m. With populations also found in 2509 Panama and Costa Rica. Although no subspecies is currently recognized (Rassmussen & Collar, 2510 2018), populations from Central America were historically recognized in a distinct subspecies 2511 M. l. austinsmithi Dwight and Griscom, 1942. Our results recovered the sample from Panama 2512 as sister to the other two samples from Peru and Brazil, however, we refrain from making any 2513 nomenclatural change pending better sampling of this group to fully understand its diversity. 2514 7.Cyphos 2515 Cyphos macrodactylus Spix, 1824 can only be found east of the Andes, mostly near water 2516 inside terra-firme and varzea forests in Western Amazon. Our phylogeographic sampling 2517 showed almost no genetic structure, only the westernmost sample showed some difference. If 2518 this is, in fact, a phylogeographic structure, or just an artifact in sampling, still needs to be 2519 investigated. The described subspecies C. m. caurensis (Cherrie, 1916) from the Caura River 2520 region, Venezuela, is currently considered undifferentiated from the nominal form (Rassmussen 2521 & Collar, 2018), and probably does not correspond to this phylogeographic break, additional 2522 sampling is required for further assumptions. 2523 8. Cryptobucco 2524 The three species included in the newly described genus Cryptobucco, were first 2525 described as full species, and later lumped and considered conspecific as C. tectus (Boddaert, 2526 1783) (Peters, 1948). Recently, C. subtectus regained its status as full species (Rassmussen &

2527 Collar, 2018), but C. tectus and C. picatus are still considered subspecies (Gill & Donsker, 2528 2018; Remsen et al., 2018). Our results recovered C. picatus as sister to a clade containing C. 2529 tectus and C. subtectus, both in the mtDNA and the UCE tree. Biogeographically, implying that 2530 the two forms found in the Amazon, C. picatus and C. tectus, are not sister. Therefore, we 2531 propose the recognition of these taxa as full species and a more extensive work should be carried 2532 out to understand the limits of distribution of both Amazonian species, and if there is any 2533 contact, what are the implications of it. 2534 9. Hypnelus and Nystactes 2535 The sister relationship of Hypnelus and Nystactes is supported by the autapomorphic bifid 2536 bill tip in both genera, that is also present in Notharchus, although less pronounced in the later 2537 (Rassmussen & Collar, 2018). Hypnelus species are restricted to northern South America, with 2538 H. ruficollis (Wagler, 1829) having three subspecies, and H. bicinctus (Gould, 1837), two. Their 2539 specific status has been questioned based on hybridization in part of their distribution (Donegan, 2540 Quevedo, Verhelst, Cortés-Herrera, Ellery & Salaman, 2015), however without a thorough 2541 genetic and geographic sampling, this decision remains questionable. Nystactes noanamae 2542 (Hellmayr, 1909) and the species group of N. tamatia (Gmelin, JF, 1788), form the sister group 2543 of Hypnelus (Fig. 1, 3). Nystactes noanamae, is a restricted-range species, present only in a 2544 small portion of northwest Colombia, and currently considered Near-threatened by IUCN 2545 (Rassmussen & Collar, 2018). Its sister species, N. tamatia, is associated with the flooded 2546 forests in Amazonia, rarely found far from the water, even when in terra firme. Previous 2547 phylogeographic study found six genetic lineages for N. tamatia, one lineage was composed by 2548 only one sample though (Almeida, 2013). Nevertheless, our results corroborate the 2549 relationships previously found, and further studies are being conducted to understand the 2550 relationships and distribution of each lineage (Almeida, 2013). 2551 10. Notharchus 2552 Notharchus species can be grouped into three distinct groups based on distribution and 2553 morphology. Notharchus ordii (Cassin, 1851), as sister to all other species, is restricted to 2554 Amazonia, and unusually uncommon in collections. Its habitat preference and current 2555 distribution is virtually unknown. The sampling we gathered for the mtDNA sequencing 2556 actually represents all tissue samples available, and the apparent phylogeographic structure we 2557 found (Fig. S9) may only represent an artifact of sampling. Notharchus pectorales (Gray, GR, 2558 1846) is restricted to Northwest Colombia and East Panama. The last groups of species, is the 2559 group centered in N. macrorhynchus (Gmelin, JF, 1788). The ND2 analyses recovered a

2560 polytomy between the N. swainsoni (Gray, GR, 1846) N. macrorhynchus, and several lineages 2561 of N. hyperrhynchus, including one lineage from Central America (Fig. S9). Our UCE tree, in 2562 contrast, recovered N. macrorhynchus sister to N. swainsoni and N. hyperrhynchus. This result 2563 corroborates the recognition of N. hyperrhynchus and N. swainsoni as full species and renders 2564 the two Amazonian groups as non-sister lineages. Although the two subspecies of N. 2565 hyperrhynchus seem to be paraphyletic in the UCE topology, the geographical relationship 2566 seem to be reasonable, and a reappraisal of this subspecies distribution is desirable in further 2567 studies. 2568

2569 Conclusion 2570 The results presented here corroborate most of the diversity historically described in these 2571 two families, but also hidden patterns that need further investigation. With our thorough 2572 sampling of practically all widespread species and species complexes we were able to recover 2573 the phylogeographic patterns for the entire diversification of jacamars and puffbirds. This study 2574 is the first one to present a phylogenetic hypothesis for this two families employing a genomic 2575 dataset. Based on this tree we resolved some relationships that were obscured by morphological 2576 similarities among taxa, such as the recognition of the different species previously lumped into 2577 Galbula ruficauda, and even described a new puffbird genus to allocate the paraphyletic 2578 Notharchus species. Overall, the results presented here are another instance reinforcing the fact 2579 that Neotropical bird diversity still is underestimated, and that we still need exploratory research 2580 to fully comprehend diversity patterns, especially in the super complex Amazonian Basin, 2581 which will be of extreme importance for future biogeographical interpretations and better 2582 conservation planning. 2583

2584 Acknowledgements

2585 The authors thankfully acknowledge all the curators and curatorial assistants of the 2586 American Museum of Natural History, New York, USA (AMNH), Academy Academy of 2587 Natural Science of Drexel University, Philadelphia, USA (ANSP); Field Museum of Natural 2588 History, Chicago, USA (FMNH); Instituto Nacional de Pesquisas da Amazônia, Manaus, Brazil 2589 (INPA); Kansas University (KU), Laboratório de Genética e Evolução Molecular de Aves – 2590 USP (LGEMA), Lousiana State University Museum of Natural Science, Baton Rouge, USA 2591 (LSUMZ); and Museu Paraense Emílio Goeldi, Belém, Brazil (MPEG), Smithsonian Institution 2592 National Museum of Natural History (USNM), for borrowing tissue samples under their care. 88

2593 We are also grateful for all collectors involved in the fieldwork that make this paper possible. 2594 To J. McKay for helping with some laboratory procedures at the AMNH. MF acknowledge 2595 CAPES for his PhD fellowship, and CAPES PDSE fellowship (# 88881.133440/2016-01) and 2596 the support from the AMNH Frank M. Chapman Memorial Fund. The authors also thank the 2597 grant Dimensions US-Biota-São Paulo: Assembly and evolution of the Amazon biota and its 2598 environment: an integrated approach, co-funded by the US National Science Fundation (NSF 2599 DEB 1241056) to J.C. and the Fundação de Amparo à Pesquisa do Estado de São Paulo 2600 (FAPESP grant #2012/50260-6) to Lucia Lohmann; PEER-USAID Cycle 5 to CCR. AA and 2601 CCR are supported by CNPq research productivity fellowships (#310880/2012-2 and 2602 #308927/2016-8, respectively). The authors acknowledge the National Laboratory for 2603 Scientific Computing (LNCC/MCTI, Brazil) for providing HPC resources of the SDumont 2604 supercomputer, which have contributed to the research results reported within this paper. 2605 2606

2607 References 2608 Aberer AJ, Kobert K, Stamatakis A. 2014. ExaBayes: massively parallel bayesian tree 2609 inference for the whole-genome era. Mol Biol Evol 31: 2553-2556. 2610 Adeney JM, Christensen NL, Vicentini A, Cohn-Haft M. 2016. White-sand Ecosystems in 2611 Amazonia. Biotropica 48: 7-23. 2612 Aleixo A. 2007. Conceitos de espécie e o eterno conflito entre continuidade e operacionalidade: 2613 uma proposta de normatização de critérios para o reconhecimento de espécies pelo 2614 Comite Brasileiro de Registros Ornitológicos. Rev Bras Ornitol 15: 297-310. 2615 Almeida B. 2013. Filogeografia de Bucco tamatia (Aves: Bucconidae): uma linhagem 2616 associada a florestas alagadas na Amazônia. Mestrado em Zoologia, UFPA. 2617 Avendaño JE, Arbeláez-Cortés E, Cadena CD. 2017. On the importance of geographic and 2618 taxonomic sampling in phylogeography: A reevaluation of diversification and species 2619 limits in a Neotropical thrush (Aves, Turdidae). Mol Phylogenet Evol 87-97. 2620 Baker PA, Fritz SC, Dick CW, Eckert AJ, Horton BK, Manzoni S, Ribas CC, Garzione 2621 CN, Battisti DS. 2014. The emerging field of geogenomics: Constraining geological 2622 problems with genetic data. Earth-Sci Rev 135: 38-47. 2623 Barrowclough GF, Cracraft J, Klicka J, Zink RM. 2016. How Many Kinds of Birds Are 2624 There and Why Does It Matter? PLoS One 11: e0166307. 2625 Bolger AM, Lohse M, Usadel B. 2014. Trimmomatic: a flexible trimmer for Illumina sequence 2626 data. Bioinformatics 30: 2114-2120. 2627 Borges SH, Da Silva JMC. 2012. A New Area of Endemism for Amazonian Birds in the Rio 2628 Negro Basin. The Wilson J Ornithol 124: 15-23. 2629 Bravo GA, Chesser RT, Brumfield RT. 2012. Isleria, a new genus of antwren (Aves: 2630 Passeriformes: Thamnophilidae). Zootaxa 3195: 61-67. 2631 Bravo GA, Remsen JV, Jr., Whitney BM, Brumfield RT. 2012. DNA sequence data reveal 2632 a subfamily-level divergence within Thamnophilidae (Aves: Passeriformes). Mol 2633 Phylogenet Evol 65: 287-293.

2634 Cellinese N, Baum DA, Mishler BD. 2012. Species and phylogenetic nomenclature. Syst Biol 2635 61: 885-891. 2636 Cherrie GK. 1916. Two new birds from Venezuela. Bull Am Mus Nat Hist 35: 389. 2637 Claramunt S, Cracraft J. 2015. A new time tree reveals Earth history's imprint on the 2638 evolution of modern birds. Sci Adv 1: e1501005. 2639 Cracraft J. 1985. Historical biogeography and patterns of differentiation within the South 2640 American Avifauna: Areas of Endemism. Ornithol Monogr 36: 49-84. 2641 Darriba D, Taboada GL, Doallo R, Posada D. 2012. jModelTest 2: more models, new 2642 heuristics and parallel computing. Nat Methods 9: 772. 2643 Dayrat B, Cantino PD, Clarke JA, de Queiroz K. 2008. Species Names in the PhyloCode: 2644 the approach adopted by the International Society for Phylogenetic Nomenclature. Syst 2645 Biol 57: 507-514. 2646 de Queiroz K. 2005. Ernst Mayr and the modern concept of species. Proc Natl Acad Sci U S A 2647 102 Suppl 1: 6600-6607. 2648 de Queiroz K. 2007. Species concepts and species delimitation. Syst Biol 56: 879-886. 2649 de Queiroz K. 2012. Biological nomenclature from Linnaeus to the PhyloCode. In: Bell CJ, 2650 ed. The herpetological legacy of Linnaeus: A celebration of the Linnaean Tercentenary: 2651 Bibliotheca Herpetologica 9. 135-145. 2652 Donegan T, Quevedo A, Verhelst JC, Cortés-Herrera O, Ellery T, Salaman P. 2015. 2653 Revision of the status of bird species occurring or reported in Colombia 2015, with 2654 discussion of BirdLife International's new taxonomy. Conservacion Colombiana 23. 2655 Donoghue PC, Moore BR. 2003. Toward an integrative historical biogeography. Integr Comp 2656 Biol 43: 261-270. 2657 Duarte SR. 2015. Filogenia molecular do gênero Nystalus (Bucconidae, Aves): enfoque na 2658 estruturação populacional em N. maculatus e N. chacuru. Mestrado em Biologia 2659 Animal, UNB. 2660 Faircloth BC. 2013. illumiprocessor: a trimmomatic wrapper for parallel adapater and quality 2661 trimming. http://dx.doi.org/10.6079/J9ILL. 2662 Faircloth BC. 2016. PHYLUCE is a software package for the analysis of conserved genomic 2663 loci. Bioinformatics 32: 786-788. 2664 Faircloth BC, McCormack JE, Crawford NG, Harvey MG, Brumfield RT, Glenn TC. 2665 2012. Ultraconserved elements anchor thousands of genetic markers spanning multiple 2666 evolutionary timescales. Syst Biol 61: 717-726. 2667 Fernandes AM, Cohn-Haft M, Hrbek T, Farias IP. 2014. Rivers acting as barriers for bird 2668 dispersal in the Amazon. Rev Bras Ornitol 22: 363-373. 2669 Fernandes AM, Gonzalez J, Wink M, Aleixo A. 2013. Multilocus phylogeography of the 2670 Wedge-billed Woodcreeper Glyphorynchus spirurus (Aves, Furnariidae) in lowland 2671 Amazonia: widespread cryptic diversity and paraphyly reveal a complex diversification 2672 pattern. Mol Phylogenet Evol 66: 270-282. 2673 Fernandes AM, Wink M, Sardelli CH, Aleixo A. 2014. Multiple speciation across the Andes 2674 and throughout Amazonia: the case of the spot-backed antbird species complex 2675 (Hylophylax naevius/Hylophylax naevioides). J Biogeogr: 41: 1094-1104 2676 Ferreira M, Aleixo A, Ribas CC, Santos MPD. 2017. Biogeography of the Neotropical genus 2677 Malacoptila (Aves: Bucconidae): the influence of the Andean orogeny, Amazonian 2678 drainage evolution and palaeoclimate. Journal of Biogeography 44: 748-759. 2679 Gilbert MP, Jarvis ED, Li B, Li C, Avian Genome C, Wang J, Zhang G. 2014a. Genomic 2680 data of the Downy Woodpecker (Picoides pubescens). GigaScience Database. 2681 Gilbert MP, Jarvis ED, Li B, Li C, Avian Genome C, Wang J, Zhang G. 2014b. Genomic 2682 data of the Javan rhinoceros hornbill (Buceros rhinoceros silvestris). GigaScience 2683 Database. 90

2684 Gilbert MP, Jarvis ED, Li B, Li C, Avian Genome C, wang J, Zhang J. 2014c. Genomic 2685 data of the Northern Carmine bee-eater (Merops nubicus). GigaScience Database. 2686 Gill F, Donsker D. 2018. IOC World Bird List (v8.1). doi: 10.14344/IOC.ML.8.1 2687 Gill FB. 2014. Species taxonomy of birds: Which null hypothesis? Auk 131: 150-161. 2688 Haffer J. 1969. Speciation in amazonian forest birds. Science 165: 131-137. 2689 Haffer J. 1974. Avian speciation in Tropical South America. Nuttall Ornithological Club: 2690 Harvard University, Cambridge. 2691 Haffer J. 1997. Alternative models of vertebrate speciation in Amazonia: an overview. 2692 Biodivers Conserv 6: 451-476. 2693 Harvey MG, Seeholzer GF, Cáceres A D, Winger BM, Tello JG, Camacho FH, Aponte 2694 Justiniano MA, Judy CD, Ramírez SF, Terrill RS, Brown CE, León LAA, Bravo 2695 G, Combe M, Custodio O, Zumaeta AQ, Tello AU, Bravo WAG, Savit AZ, Ruiz 2696 FWP, Mauck WM, Barden O. 2014. The avian biogeography of an Amazonian 2697 headwater: the Upper Ucayali River, Peru. Wilson J Ornithol 126: 179-191. 2698 Isler ML, Bravo GA, Brumfield RT. 2013. Taxonomic revision of Myrmeciza (Aves: 2699 Passeriformes: Thamnophilidae) into 12 genera based on phylogenetic, morphological, 2700 behavioral, and ecological data. Zootaxa 3717: 469. 2701 Katoh K, Standley DM. 2013. MAFFT Multiple Sequence Alignment Software Version 7: 2702 Improvements in Performance and Usability. Mol Biol Evol 30: 772-780. 2703 Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper 2704 A, Markowitz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A. 2012. 2705 Geneious Basic: An integrated and extendable desktop software platform for the 2706 organization and analysis of sequence data. Bioinformatics 28: 1647-1649. 2707 Langmead B, Salzberg SL. 2012. Fast gapped-read alignment with Bowtie 2. Nat Methods 9: 2708 357-359. 2709 Lexer C, Mangili S, Bossolini E, Forest F, Stölting KN, Pearman PB, Zimmermann NE, 2710 Salamin N, Carine M. 2013. ‘Next generation’ biogeography: towards understanding 2711 the drivers of species diversification and persistence. J Biogeogr 40: 1013-1022. 2712 Livezey BC, Zusi RL. 2007. Higher-order phylogeny of modern birds (Theropoda, Aves: 2713 Neornithes) based on comparative anatomy. II. Analysis and discussion. Zool J Linn 2714 Soc 149: 1-95. 2715 Lopes LE, Chaves AV, Aquino MM, Silveira LF, Santos FR. 2017. The striking polyphyly 2716 of Suiriri: Convergent evolution and social mimicry in two cryptic Neotropical birds. J 2717 Zool Syst Evol Res Early view. 2718 Lutz HL, Weckstein JD, Patane JS, Bates JM, Aleixo A. 2013. Biogeography and spatio- 2719 temporal diversification of Selenidera and Andigena Toucans (Aves: Ramphastidae). 2720 Mol Phylogenet Evol 69: 873-883. 2721 Mayr E. 1942. Systematics and the Origin of Species. Columbia Univ. Press: New York. 2722 Mayr E. 1976. Species concept and definitions Topics in the Phylosophy of Biology. 2723 Netherlands: Springer. 353-371. 2724 McCormack JE, Faircloth BC. 2013. Next-generation phylogenetics takes root. Mol Ecol 22: 2725 19-21. 2726 McCormack JE, Harvey MG, Faircloth BC, Crawford NG, Glenn TC, Brumfield RT. 2727 2013. A phylogeny of birds based on over 1,500 loci collected by target enrichment and 2728 high-throughput sequencing. PLoS One 8: e54848. 2729 McCormack JE, Tsai WLE, Faircloth BC. 2016. Sequence capture of ultraconserved 2730 elements from bird museum specimens. Mol Ecol Resour 16: 1189-1203. 2731 Peters JL. 1945. Check-list of birds of the world. Harvard University Press: Cambrigde. 2732 Peters JL. 1948. Check-list of birds of the world. Harvard University Press: Cambridge, UK.

2733 Piacentini VD, Aleixo A, Agne CE, Mauricio GN, Pacheco JF, Bravo GA, Brito GRR, 2734 Naka LN, Olmos F, Posso S, Silveira LF, Betini GS, Carrano E, Franz I, Lees AC, 2735 Lima LM, Pioli D, Schunck F, do Amaral FR, Bencke GA, Cohn-Haft M, 2736 Figueiredo LFA, Straube FC, Cesari E. 2015. Annotated checklist of the birds of 2737 Brazil by the Brazilian Ornithological Records Committee. Revista Brasileira de 2738 Ornitologia 23: 91-298. 2739 Prum RO, Berv JS, Dornburg A, Field DJ, Townsend JP, Lemmon EM, Lemmon AR. 2740 2015. A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA 2741 sequencing. Nature 526: 569-573. 2742 Rambaut A, Suchard MA, Xie D, Drummond AJ. 2014. Tracer v1.6. Availabe at: 2743 http://tree.bio.ed.ac.uk/software/tracer/ 2744 Rassmussen P, Collar N. 2002. Family Bucconidae (Puffbirds). In: del Hoyo J, Elliot A and 2745 Sargatal J, eds. Handbook of the birds of the world: Lynx Edicions. 2746 Rassmussen P, Collar N. 2018. Puffbirds (Bucconidae). In: del Hoyo J, Elliot A, Sargatal J, 2747 Christie DA and Juana E, eds. Handbook of the birds of the world Alive. Barcelona: 2748 Lynx Ediciones. 2749 Remsen JV, Jr., Areta JI, Cadena CD, Claramunt S, Jaramillo C, Pacheco JF, Pérez- 2750 Emen J, Robbins MB, Stiles FG, Stotz DF, Zimmer KJ. 2018. A classification of the 2751 bird species of South America. American Ornithologists' Union. 2752 http://www.museum.lsu.edu/~Remsen/SACCBaseline.htm 2753 Ribas CC, Aleixo A, Gubili C, d'Horta FM, Brumfield RT, Cracraft J. 2018. Biogeography 2754 and diversification of Rhegmatorhina (Aves: Thamnophilidae): Implications for the 2755 evolution of Amazonian landscapes during the Quaternary. J Biogeogr Early View. 2756 https://doi.org/10.1111/jbi.13169 2757 Ribas CC, Aleixo A, Nogueira AC, Miyaki CY, Cracraft J. 2012. A palaeobiogeographic 2758 model for biotic diversification within Amazonia over the past three million years. Proc 2759 Biol Sci 279: 681-689. 2760 Ribas CC, Gaban-Lima R, Miyaki CY, Cracraft J. 2005. Historical biogeography and 2761 diversification within the Neotropical parrot genus Pionopsitta (Aves: Psittacidae). J 2762 Biogeogr 32: 1409-1427. 2763 Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Hohna S, Larget B, Liu 2764 L, Suchard MA, Huelsenbeck JP. 2012. MrBayes 3.2: efficient Bayesian phylogenetic 2765 inference and model choice across a large model space. Syst Biol 61: 539-542. 2766 Sloan DB, Havird JC, Sharbrough J. 2017. The on-again, off-again relationship between 2767 mitochondrial genomes and species boundaries. Mol Ecol 26: 2212-2236. 2768 Soares LMS. 2016. Sistemática molecular e diversificação dos gêneros Nonnula e Monasa 2769 (Aves: Bucconidae). Doutorado em Zoologia, UFPA. 2770 Tobias JA. 2017. Jacamars (Galbulidae). In: del Hoyo J, Elliot A and Sargatal J, eds. Handbook 2771 of the birds of the world. Barcelona: Lynx Ediciones. 2772 Tobias JA, Bates JM, Hackett SJ, Seddon N. 2008. Comment on "The latitudinal gradient in 2773 recent speciation and extinction rates of birds and mammals". Science 319: 901; author 2774 reply 901. 2775

2776

2777

2778 Figure 1 – Phylogeny of the Galbulidae and Bucconidae families inferred with ExaBayes. All nodes in this tree 2779 receive the maximum posterior probability. The two genomes used as reference sequence were included in this 2780 analysis. 2781

2782 2783 2784

2785 Figure 2 – Phylogeny of the Galbulidae inferred by ExaBayes with the 75% completeness matrix. Node support 2786 is indicated near it, if no support is indicated posterior probability is 1.0. 2787

2788 2789

2790

2791 Figure 3 – Phylogeny of the Bucconidae inferred by ExaBayes with the 75% completeness matrix. Node support 2792 is indicated near it, if no support is indicated posterior probability is 1.0. 2793

2794 2795 95

2796 Figure S1 – Phylogenetic relationship and map with sample distribution of Brachygalba and Jacamaralcyon 2797 species. Phylogenetic tree was recovered by MrBayes using the mtDNA gene ND2 (1041bp). Red circles mean 2798 posterior probabilities of 1.0, values that differs are indicated near the node. Samples highlighted in red were the 2799 samples selected for UCE analysis. The maps contain sample localities and approximate lineage distribution. 2800 Colours are correspondent between the tree and the map. 2801

2802

2803

2804

2805 Figure S2 – Phylogenetic relationship and map with sample distribution of Jacamerops aureus complex. 2806 Phylogenetic tree was recovered by MrBayes using the mtDNA gene ND2 (1041bp). Red circles mean posterior 2807 probabilities of 1.0, values that differs are indicated near the node. Samples highlighted in red were the samples 2808 selected for UCE analysis. The maps contain sample localities and approximate lineage distribution. Colours are 2809 correspondent between the tree and the map. 2810

2811

2812

2813

2814 Figure S3 – Phylogenetic relationship and map with sample distribution of Galbula dea complex. Phylogenetic 2815 tree was recovered by MrBayes using the mtDNA gene ND2 (1041bp). Red circles mean posterior probabilities 2816 of 1.0, values that differs are indicated near the node. Samples highlighted in red were the samples selected for 2817 UCE analysis. The maps contain sample localities and approximate lineage distribution. Colours are 2818 correspondent between the tree and the map. 2819

2820

2821

2822 Figure S4 – Phylogenetic relationship and map with sample distribution of the species complex of G. albirostris, 2823 G. chalcocephala and G. albirostris. Phylogenetic tree was recovered by MrBayes using the mtDNA gene ND2 2824 (1041bp). Red circles mean posterior probabilities of 1.0, values that differs are indicated near the node. Samples 2825 highlighted in red were the samples selected for UCE analysis. The maps contain sample localities and 2826 approximate lineage distribution. Colours are correspondent between the tree and the map. 2827

2828

2829

2830

2831 Figure S5 – Phylogenetic relationship and map with sample distribution of the species complex of G. galbula, G. 2832 tombacea, G. cyanescens, G. pastazae, and G. ruficauda. Phylogenetic tree was recovered by MrBayes using the 2833 mtDNA gene ND2 (1041bp). Red circles mean posterior probabilities of 1.0, values that differs are indicated 2834 near the node. Samples highlighted in red were the samples selected for UCE analysis. The maps contain sample 2835 localities and approximate lineage distribution. Colours are correspondent between the tree and the map. 2836

2837

2838

2839

100

2840 Figure S6 – Phylogenetic relationship and map with sample distribution of the species Bucco capensis. 2841 Phylogenetic tree was recovered by MrBayes using the mtDNA gene ND2 (1041bp). Red circles mean posterior 2842 probabilities of 1.0, values that differs are indicated near the node. Samples highlighted in red were the samples 2843 selected for UCE analysis. The maps contain sample localities and approximate lineage distribution. Colours are 2844 correspondent between the tree and the map. 2845

2846

2847

2848

101

2849 Figure S7 – Phylogenetic relationship and map with sample distribution of the species Cyphos macrodactylus. 2850 Phylogenetic tree was recovered by MrBayes using the mtDNA gene ND2 (1041bp). Red circles mean posterior 2851 probabilities of 1.0, values that differs are indicated near the node. Samples highlighted in red were the samples 2852 selected for UCE analysis. The maps contain sample localities and approximate lineage distribution. Colours are 2853 correspondent between the tree and the map. 2854

2855

2856

2857

102

2858 Figure S8 – Phylogenetic relationship and map with sample distribution of the species complex of Notharchus 2859 tectus, N. subtectus, and N. picatus. Phylogenetic tree was recovered by MrBayes using the mtDNA gene ND2 2860 (1041bp). Red circles mean posterior probabilities of 1.0, values that differs are indicated near the node. Samples 2861 highlighted in red were the samples selected for UCE analysis. The maps contain sample localities and 2862 approximate lineage distribution. Colours are correspondent between the tree and the map. 2863

2864

2865

2866

2867

103

2868 Figure S9 – Phylogenetic relationship and map with sample distribution of the species complex of Notharchus 2869 ordii, N. pectorales, N. swainsoni, N. macrorhynchus, and N. hyperrhynchus. Phylogenetic tree was recovered by 2870 MrBayes using the mtDNA gene ND2 (1041bp). Red circles mean posterior probabilities of 1.0, values that 2871 differs are indicated near the node. Samples highlighted in red were the samples selected for UCE analysis. The 2872 maps contain sample localities and approximate lineage distribution. Colours are correspondent between the tree 2873 and the map. 2874

2875

2876

2877

2878

104

2879 Figure S10 – Phylogenetic relationship and map with sample distribution of the species Chelidoptera tenebrosa. 2880 Phylogenetic tree was recovered by MrBayes using the mtDNA gene ND2 (1041bp). Red circles mean posterior 2881 probabilities of 1.0, values that differs are indicated near the node. Samples highlighted in red were the samples 2882 selected for UCE analysis. The maps contain sample localities and approximate lineage distribution. Colours are 2883 correspondent between the tree and the map. 2884

2885 2886 2887 2888 2889

105

2890 Figure S11 – Phylogenetic tree recovered for Monasa using a subset of samples to check for M. mystacalis 2891 phylogenetic affinity. The same tree topology was recovered by RAxML and ExaBayes, the RAxML bootstrap 2892 support were low overall. 2893

2894 2895

2896 Figure S12 – Phylogenetic tree recovered for Malacoptila using a subset of samples to check for M. mystacalis 2897 phylogenetic affinity. The same tree topology was recovered by RAxML and ExaBayes with high support, with 2898 only node receiving bootstrap support different from 100. 2899

2900 2901

2902

2903

2904

106

2905 Síntese Geral

2906 Neste trabalho coletamos dados que nos ajudaram a compreender a relação filogenética 2907 de três famílias de aves do Neotrópico. A utilização de dados de sequenciamento genômico e a 2908 inclusão de amostras representando quase todas as linhagens basais em cada família permitiu 2909 realizar inferências sobre a importância de uma amostragem ampla, tanto num sentido de 2910 amostras, quanto marcadores. No primeiro capítulo pudemos observar o impacto do conflito 2911 entre marcadores moleculares com diferentes padrões de herança, e quais as implicações 2912 biológicas deste conflito. Além disso, através da análise combinada da história dos dois 2913 marcadores foi possível propor um modelo de evolução das áreas de vegetação aberta 2914 relacionadas aos solos de areia branca dentro da bacia Amazônia. No segundo capítulo, 2915 recuperamos a relação filogenética da família Trogonidae utilizando quase todas as espécies 2916 descritas com base em uma matriz com mais de 2.000 marcadores moleculares. Com base 2917 nesses resultados traçamos um modelo de como a evolução do clima desde o final do Oligoceno 2918 e as conexões entre os continentes influenciaram a história de diversificação do grupo. Por fim, 2919 no terceiro capítulo, analisamos a diversidade intraespecífica de duas famílias endêmicas do 2920 Neotrópico e reconstruímos a primeira hipótese de relação filogenética para Galbulidae e 2921 Bucconidae utilizando dados genômicos. Neste capítulo pudemos observar como a percepção 2922 da diversidade nesses grupos é subestimada e influenciada pela taxonomia vigente, e que a 2923 amostragem densa ao longo da distribuição de espécies amplamente distribuídas pode revelar 2924 táxons e padrões ainda desconhecidos. 2925 De modo geral, este trabalho reforça a complexidade dos padrões de diversidade da biota 2926 Neotropical, e que ainda se faz necessário estudos para desvendar esses padrões, em especial 2927 na Amazônia. Além disso, fica claro que a diversidade real da região ainda está mascarada pela 2928 taxonomia vigente e revisões sistemáticas e taxonômicas são necessárias. Só através do 2929 reconhecimento dessa diversidade escondida é que será possível, não só traçar hipóteses sobre 2930 os processos que deram origem a tamanha diversidade, mas também traçar planos de 2931 conservação que reconheçam a história evolutiva de cada um desses grupos. 2932

107

2933 Referências Bibliográficas 2934 2935 ALEIXO, A. Conceitos de espécie e o eterno conflito entre continuidade e operacionalidade: uma 2936 proposta de normatização de critérios para o reconhecimento de espécies pelo Comite Brasileiro 2937 de Registros Ornitológicos. Revista Brasileira de Ornitologia, v. 15, n. 2, p. 297-310, 2007. 2938 2939 ______. Conceitos de espécie e suas implicações para a conservação. Megadiversidade, v. 5, n. 1-2, p. 2940 87-95, 2009. 2941 2942 ÁVILA-PIRES, T. C. S. Lizards of Brazilian Amazon (Reptilia: Squamata). Zool Verh Natuur Mus, 2943 v. 299, p. 1-637, 1995. 2944 2945 AVISE, J. C. Phylogeography: retrospect and prospect. J Biogeogr, v. 36, n. 1, p. 3-15, 2009. 2946 2947 AVISE, J. C. et al. Intraspecific phylogeography: The mitochondrial DNA bridge between population 2948 genetics and systematics. Annu Rev Ecol Syst, v. 18, p. 489-522, 1987. 2949 2950 BAKER, P. A. et al. The emerging field of geogenomics: Constraining geological problems with genetic 2951 data. Earth-Sci Rev, v. 135, p. 38-47, 2014. 2952 2953 BARROWCLOUGH, G. F. et al. How Many Kinds of Birds Are There and Why Does It Matter? Plos 2954 One, v. 11, n. 11, p. e0166307, 2016. 2955 2956 BATES, J. M. et al. Diversification in the Neotropics: Phylogenetic patterns and historical processes. 2957 Ornitologia Neotropical, v. 19, n. (Suppl.), p. 427-432, 2008. 2958 2959 BATES, J. M.; HAFFER, J.; GRISMER, E. Avian mitochondrial DNA sequence divergence across a 2960 headwater stream of the Rio Tapajós, a major Amazonian river. Journal of Ornithology, v. 2961 145, n. 3, p. 199-205, 2004. 2962 2963 BOUBLI, J. P. et al. Spatial and temporal patterns of diversification on the Amazon: A test of the 2964 riverine hypothesis for all diurnal primates of Rio Negro and Rio Branco in Brazil. Mol 2965 Phylogenet Evol, v. 82 Pt B, p. 400-12, 2015. 2966 2967 BOUBLI, J. P. et al. How many Pygmy Marmoset (Cebuella Gray, 1870) species are there? A 2968 taxonomic re-appraisal based on new molecular evidence. Mol Phylogenet Evol, v. 120, p. 170- 2969 82, 2017. 2970 2971 BROMHAM, L. et al. Bayesian molecular dating: opening up the black box. Biol Rev v. Early view, 2972 2017. doi: 10.1111/brv.12390 2973 2974 BROMHAM, L.; PENNY, D. The modern molecular clock. Nat Rev Genet, v. 4, n. 3, p. 216-24, Mar 2975 2003. 2976 2977 BROWN, K. S.; SHEPPARD, P. M.; TURNER, J. R. G. Quaternary refugia in tropical America: 2978 evidence from race formation in Heliconius butterflies. Proc R Soc Lond B, v. 187, p. 369-378, 2979 1974. 2980 2981 BRYSON, R. W., JR. et al. Target enrichment of thousands of ultraconserved elements sheds new light 2982 on early relationships within New World sparrows (Aves: Passarelidae). Auk, v. 133, n. 3, p. 2983 451-458, 2016. 2984

108

2985 BUSH, M. B. Amazonian speciation: A necessarily complex model. Journal of Biogeography, v. 21, 2986 p. 5-17, 1995. 2987 2988 BUSH, M. B.; OLIVEIRA, P. E. D. The rise and fall of the Refugial Hypothesis of Amazonian 2989 speciation: a paleoecological perspective. Biota Neotrop, v. 6, n. 1, 2006. 2990 2991 BYRNE, H. et al. Phylogenetic relationships of the New World titi monkeys (Callicebus): first appraisal 2992 of taxonomy based on molecular evidence. Front Zool, v. 13, n. 1, 2016. 2993 2994 CARNEIRO, J. C. et al. Phylogeny of the Titi monkeys of the Callicebus moloch group (Pitheciidae, 2995 Primates). Am J Primatol, v. 78, p. 904-913, 2016. 2996 2997 CELLINESE, N.; BAUM, D. A.; MISHLER, B. D. Species and phylogenetic nomenclature. Syst Biol, 2998 v. 61, n. 5, p. 885-891, 2012. 2999 3000 CHENG, H. et al. Climate change patterns in Amazonia and biodiversity. Nat Commun, v. 4, p. 1411, 3001 2013. 3002 3003 COATES, A. G.; STALLARD, R. F. How old is the Isthmus of Panama? Bull Mar Sci, v. 89, n. 4, p. 3004 801-813, 2013. 3005 3006 COLINVAUX, P. A.; DE OLIVEIRA, P. E.; BUSH, M. B. Amazonian and neotropical plant 3007 communities on glacial time-scales: The failure of the aridity and refuge hypothesis. 3008 Quaternary Science Reviews, v. 19, p. 141-169, 2000. 3009 3010 COLLAR, N. Trogons (Trogonidae). In: DEL HOYO, J.;ELLIOT, A., et al (Ed.). Handbook of the 3011 birds of the world Alive. Barcelona: Lynx Ediciones, 2018. 3012 3013 COLOMBO, A. F.; JOLY, C. A. Brazilian Atlantic Forest lato sensu: the most ancient Brazilian forest, 3014 and a biodiversity hotspot, is highly threatene by climate change. Braz J Biol, v. 70, p. 697-708, 3015 2010. 3016 3017 CRACRAFT, J. Toward a phylogenetic classification of the recent birds of the world (Class Aves). Auk, 3018 v. 98, n. 4, p. 681-714, 1981. 3019 3020 ______. Historical biogeography and patterns of differentiation within the South American Avifauna: 3021 Areas of Endemism. Ornitholigical Monographs, v. 36, p. 49-84, 1985. 3022 3023 CRAWFORD, N. G. et al. A phylogenomic analysis of turtles. Molecular Phylogenetics and 3024 Evolution, v. 83, p. 250-257, 2015. 3025 3026 D’HORTA, F. M. et al. Phylogeny and comparative phylogeography of Sclerurus (Aves: Furnariidae) 3027 reveal constant and cryptic diversification in an old radiation of rain forest understorey 3028 specialists. Journal of Biogeography, v. 40, n. 1, p. 37-49, 2013. 3029 3030 DACOSTA, J. M.; KLICKA, J. The Great American Interchange in birds: a phylogenetic perspective 3031 with the genus Trogon. Mol Ecol, v. 17, n. 5, p. 1328-43, Mar 2008. 3032 3033 DAVEY, J. W. et al. Genome-wide genetic marker discovery and genotyping using next-generation 3034 sequencing. Nat Rev Genet, v. 12, n. 7, p. 499-510, 2011 3035 3036 DE QUEIROZ, K. Ernst Mayr and the modern concept of species. Proc Natl Acad Sci U S A, v. 102 3037 Suppl 1, p. 6600-7, 3 2005. 3038 109

3039 ______. Branches in the lines of descent: Charles Darwin and the evolution of the species concept. Biol 3040 J Linn Soc, v. 103, p. 19-35, 2011. 3041 3042 ______. Biological nomenclature from Linnaeus to the PhyloCode. In: BELL, C. J. (Ed.). The 3043 herpetological legacy of Linnaeus: A celebration of the Linnaean Tercentenary: Bibliotheca 3044 Herpetologica 9, 2012. p.135-145. 3045 3046 DEGNAN, J. H.; ROSENBERG, N. A. Discordance of species trees with their most likely gene trees. 3047 PLoS Genet, v. 2, n. 5, p. e68, 2006. 3048 3049 DEGNAN, J. H.; ROSENBERG, N. A. Gene tree discordance, phylogenetic inference and the 3050 multispecies coalescent. Trends Ecol Evol, v. 24, n. 6, p. 332-340, 2009. 3051 3052 DONOGHUE, M. J. A critique of the Biological Species Concept and recommendations for a 3053 phylogenetic alternative. The Bryologist, v. 88, n. 3, p. 172-181, 1985. 3054 3055 DONOGHUE, P. C.; MOORE, B. R. Toward an integrative historical biogeography. Integr Comp Biol, 3056 v. 43, p. 261-270, 2003. 3057 3058 DUARTE, S. R. Filogenia molecular do gênero Nystalus (Bucconidae, Aves): enfoque na 3059 estruturação populacional em N. maculatus e N. chacuru. 2015. (Mestrado em Biologia 3060 Animal). Instituto de Biologia, UNB, Brasília. 3061 3062 EHLERS, T. A.; POULSEN, C. J. Influence of Andean uplift on climate and paleoaltimetry estimates. 3063 Earth Planet Sci Lett, v. 281, p. 238-248, 2009. 3064 3065 FAIRCLOTH, B. C. et al. Target enrichment of ultraconserved elements from arthropods provides a 3066 genomic perspective on relationships among Hymenoptera. Mol Ecol Resour, v. 15, n. 3, p. 3067 489-501, 2015. 3068 3069 FAIRCLOTH, B. C. et al. Ultraconserved elements anchor thousands of genetic markers spanning 3070 multiple evolutionary timescales. Syst Biol, v. 61, n. 5, p. 717-26, Oct 2012. 3071 3072 FERNANDES, A. M. et al. Multilocus phylogeography of the Wedge-billed Woodcreeper 3073 Glyphorynchus spirurus (Aves, Furnariidae) in lowland Amazonia: widespread cryptic diversity 3074 and paraphyly reveal a complex diversification pattern. Mol Phylogenet Evol, v. 66, n. 1, p. 3075 270-82, 2013. 3076 3077 FERNANDES, A. M. et al. Multiple speciation across the Andes and throughout Amazonia: the case 3078 of the spot-backed antbird species complex (Hylophylax naevius/Hylophylax naevioides). 3079 Journal of Biogeography, p. n/a-n/a, 2014. ISSN 03050270. 3080 3081 FERREIRA, M. et al. Biogeography of the Neotropical genus Malacoptila (Aves: Bucconidae): the 3082 influence of the Andean orogeny, Amazonian drainage evolution and palaeoclimate. Journal of 3083 Biogeography, v. 44, n. 4, p. 748-759, 2017. 3084 3085 GARZIONE, C. N. et al. Rise of the Andes. Science, v. 320, n. 5881, p. 1304-7, Jun 6 2008. 3086 3087 GILL, F.; DONSKER, D. IOC World Bird List (v8.1). 2018. 3088 3089 GILL, F. B. Hybrization in birds. Journal of Ornithology, v. 115, n. 2, p. 281-283, 1998. 3090 3091 GRANT, P. R.; GRANT, B. R. Hybridization of bird species. Science, v. 256, n. 5054, p. 193-197, 3092 1992.

110

3093 3094 GROVER, C. E.; SALMON, A.; WENDEL, J. F. Targeted sequence capture as a powerful tool for 3095 evolutionary analysis. Am J Bot, v. 99, n. 2, p. 312-319, 2012. 3096 3097 HACKETT, S. J. et al. A phylogenomic study of birds reveals their evolutionary history. Science, v. 3098 320, n. 5884, p. 1763-8, 2008. 3099 3100 HAFFER, J. Speciation in amazonian forest birds. Science, v. 165, n. 3889, p. 131-7, 1969. 3101 3102 ______. Avian zoogeography in the Neotropical lowlands. Ornitholigical Monographs, v. 36, p. 113- 3103 146, 1985. 3104 3105 ______. Alternative models of vertebrate speciation in Amazonia: an overview. Biodivers Conserv, v. 3106 6, p. 451-476, 1997. 3107 3108 HARRISON, R. G.; LARSON, E. L. Hybridization, introgression, and the nature of species boundaries. 3109 J Hered, v. 105 Suppl 1, p. 795-809, 2014. 3110 3111 HARTLEY, A. Andean uplift and climate change. Journal of the Geological Society, London, v. 160, 3112 p. 7-10, 2003. 3113 3114 HAUG, G. H.; TIEDEMAN, R. Effect of the formation of the Isthmus of Panama on Atlantic Ocean 3115 thermohaline circulation. Nature, v. 393, p. 673-676, 1998. 3116 3117 HOLT, B. G. et al. An update of Wallace's zoogeographic regions of the world. Science, v. 339, n. 3118 6115, p. 74-8, 2013. 3119 3120 HOORN, C. Marine incursions and the influence of Andean tectonics on the Miocene depositional 3121 history of northwestern Amazonia: results of a palynostratigraphic study. Palaeogeography, 3122 Palaeoclimatology, Palaeoecology, v. 105, n. 3-4, p. 267-309, 1993. 3123 3124 HOORN, C. et al. The Amazon at sea: Onset and stages of the Amazon River from a marine record, 3125 with special reference to Neogene plant turnover in the drainage basin. Global Plant Change, 3126 2017. 3127 3128 HOORN, C.; WESSELINGH, F. P. Amazonia: Landscape and species evolution: A look into past. 3129 Oxford: Wiley-Blackwell Publishing Ltd., 2010. 3130 3131 HOORN, C. et al. Amazonia through time: Andean uplift, climate change, landscape evolution, and 3132 biodiversity. Science, v. 330, n. 6006, p. 927-31, 2010. 3133 3134 HOPKINS, M. J. G. Modelling the known and unkown plant biodiversity of the Amazon Basin. J 3135 Biogeogr, v. 34, p. 1400-1411, 2007. 3136 3137 HORTON, B. K. Tectonic regimes of the Central and Southern Andes: Responses to variations in plate 3138 coupling during subduction. Tectonics, v. Early View, 2018. doi: 10.1002/2017TC004624 3139 3140 HOSNER, P. A. et al. Phylogeny and biogeography of the Asian trogons (Aves: Trogoniformes) 3141 inferred from nuclear and mitochondrial DNA sequences. Mol Phylogenet Evol, v. 57, n. 3, p. 3142 1219-25, 2010. 3143 3144 HRBEK, T. et al. A new species of river dolphin from Brazil or: how little do we know our biodiversity. 3145 PLoS One, v. 9, n. 1, p. e83623, 2014. 3146

111

3147 INSEL, N.; POULSEN, C. J.; EHLERS, T. A. Influence of the Andes Mountains on South American 3148 moisture transport, convection, and precipitation. Clim Dyn, v. 35, n. 7-8, p. 1477-1492, 2009. 3149 3150 ISAAC, N. J. B.; MALLET, J.; MACE, G. M. Taxonomic inflation: its influence on macroecology and 3151 conservation. Trends Ecol Evol, v. 19, n. 9, p. 464-469, 2004. 3152 3153 JARVIS, E. D. et al. Phylogenomic analyses data of the avian phylogenomics project. Gigascience, v. 3154 4, p. 4, 2015. 3155 3156 JETZ, W. et al. The global diversity of birds in space and time. Nature, v. 491, n. 7424, p. 444-8, 2012. 3157 3158 JOHANSSON, U. S.; ERICSON, P. G. P. A re-evaluation of basal phylogenetic relationships within 3159 trogons (Aves: Trogonidae) based on nuclear DNA sequences. J Zool Syst Evol Res, v. 43, n. 3160 2, p. 166-173, 2005. 3161 3162 JOSEPH, L. et al. Where and when does a ring start and end? Testing the ring-species hypothesis in a 3163 species complex of Australian parrots. Proc Biol Sci, v. 275, n. 1650, p. 2431-40, 2008. 3164 3165 KIER, G. et al. Global patterns of plant diversity and floristic knowledge. J Biogeogr, v. 32, p. 1107- 3166 1116, 2005. 3167 3168 KNOWLES, L. L. Statistical Phylogeography. Annu Rev Ecol Evol Syst, v. 40, n. 1, p. 593-612, 2009. 3169 3170 KRISTOFFERSEN, A. V. An early Paleogene trogon (Aves: Trogoniformes) from the Fur Formation, 3171 Denmark. J Vert Paleontol, v. 22, n. 3, p. 661-666, 2002. 3172 3173 LATRUBESSE, E. M. et al. The Late Miocene paleogeography of the Amazon Basin and the evolution 3174 of the Amazon River system. Earth-Sci Rev, v. 99, n. 3-4, p. 99-124, 2010. 3175 3176 LEITE, R. N.; ROGERS, D. S. Revisiting Amazonian phylogeography: insights into diversification 3177 hypotheses and novel perspectives. Organisms Diversity & Evolution, 2013. 3178 3179 LEMMON, A. R.; EMME, S. A.; LEMMON, E. M. Anchored hybrid enrichment for massively high- 3180 throughput phylogenomics. Syst Biol, v. 61, n. 5, p. 727-44, Oct 2012. 3181 3182 LEMMON, E. M.; LEMMON, A. R. High-Throughput Genomic Data in Systematics and 3183 Phylogenetics. Annu Rev Ecol Evol Syst, v. 44, n. 1, p. 99-121, 2013. 3184 3185 LESSIOS, H. A. Appearance of an early closure of the Isthmus of Panama is the product of biased 3186 inclusion of data in the metaanalysis. PNAS, v. 112, n. 43, p. E5765, 2015. 3187 3188 LIMA, M. G. M. et al. Capuchin monkey biogeography: understanding Sapajus Pleistocene range 3189 expansion and the current sympatry between Cebus and Sapajus. Journal of Biogeography, v. 3190 44, n. 4, p. 810-820, 2017. 3191 3192 LIVEZEY, B. C.; ZUSI, R. L. Phylogeny of Neornithes. Bull Carnegie Mus Nat Hist, v. 37, p. 1-544, 3193 2006. 3194 3195 ______. Higher-order phylogeny of modern birds (Theropoda, Aves: Neornithes) based on comparative 3196 anatomy. II. Analysis and discussion. Zoological Journal of the Linnean Society, v. 149, p. 1- 3197 95, 2007. 3198

112

3199 MANTHEY, J. D. et al. Comparison of Target-Capture and Restriction-Site Associated DNA 3200 Sequencing for Phylogenomics: A Test in Cardinalid Tanagers (Aves, Genus: Piranga). Syst 3201 Biol, v. 65, n. 4, p. 640-50, 2016. 3202 3203 MAURER, D. R.; RAIKOW, R. J. Appendicular myology, phylogeny and classificaiton of the avian 3204 order Coraciiformes (inluding Trogoniformes). Ann Carnegie Mus, v. 50, p. 417-434, 1981. 3205 3206 MAYR, E. Systematics and the Origin of Species. New York: Columbia Univ. Press, 1942. 3207 3208 ______. Species concept and definitions. In: (Ed.). Topics in the Phylosophy of Biology. Netherlands: 3209 Springer, 1976. p.353-371. 3210 3211 MAYR, G. A new trogon from the Middle Oligocene of Céreste, France. Auk, v. 116, n. 2, p. 427-434, 3212 1999. 3213 3214 ______. On the phylogenetic relationships of trogons (Aves, Trogonidae). Journal of Avian Biology, 3215 v. 34, p. 81-88, 2003. 3216 3217 MAYR, G. New trogons from the early Tertiary of Germany. Ibis, v. 147, p. 512-518, 2005. 3218 3219 MCCORMACK, J. E. et al. Ultraconserved elements are novel phylogenomic markers that resolve 3220 placental mammal phylogeny when combined with species-tree analysis. Genome Res, v. 22, 3221 n. 4, p. 746-54, 2012. 3222 3223 MEIKLEJOHN, K. A. et al. Analysis of a rapid evolutionary radiation using Ultraconserved Elements: 3224 Evidence for a bias in some multispecies coalescent methods. Systematic Biology, v. 65, n. 4, 3225 p. 612-627, 2016. 3226 3227 METZKER, M. L. Sequencing technologies - the next generation. Nature Reviews Genetics, v. 11, p. 3228 31-46, 2010. 3229 3230 MITTERMEIER, R. A. et al. Wilderness and biodiversity conservation. Proc Natl Acad Sci U S A, v. 3231 100, n. 18, p. 10309-13, 2003. 3232 3233 MONTEROS, A. E. Phylogenetic relationships among the Trogons. Auk, v. 115, n. 4, p. 937-954, 1998. 3234 3235 MONTEROS, A. E. Higher-level phylogeny of Trogoniformes. Mol Phylogenet Evol, v. 14, n. 1, p. 3236 20-34, 2000. 3237 3238 MOYLE, R. G. Phylogeny and biogeographical history of Trogoniformes, a pantropical bird order. Biol 3239 J Linn Soc, v. 84, p. 725-738, 2005. 3240 3241 MYERS, N. et al. Biodiversity hotspots for conservation priorities. Nature, v. 403, n. 6772, p. 853-8, 3242 2000. 3243 3244 NOGUEIRA, A. C. R.; SILVEIRA, R.; GUIMARÃES, J. T. F. Neogene–Quaternary sedimentary and 3245 paleovegetation history of the eastern Solimões Basin, central Amazon region. Journal of South 3246 American Earth Sciences, v. 46, p. 89-99, 2013. 3247 3248 NORES, M. An alternative hypothesis for the origin of Amazonian bird diversity. Journal of 3249 Biogeography, v. 26, n. 3, p. 475-485, 1999. 3250 3251 ______. The implications of Tertiary and Quaternary sea level rise events for avian distribution patterns 3252 in the lowlands of northern South America. Global Ecol Biogeogr, v. 13, p. 149-161, 2004.

113

3253 3254 ODEA, A. et al. Formation of the Isthmus of Panama. Science Advances, v. 2, n. 8, p. e1600883, 2016. 3255 3256 OLIVEIRA, U.; VASCONCELOS, M. F.; SANTOS, A. J. Biogeography of Amazon birds: rivers limit 3257 species composition, but not areas of endemism. Scientific Reports, v. 7, p. 2992, 2017. 3258 3259 OLIVER, J. Microevolutionary processes generate phylogenomic discordance at ancient divergences. 3260 Evolution, v. 67, n. 6, p. 1823-1830, 2013. 3261 3262 ORNELAS, J. F.; GONZALEZ, C.; ESPINOSA DE LOS MONTEROS, A. Uncorrelated evolution 3263 between vocal and plumage coloration traits in the trogons: a comparative study. J Evol Biol, v. 3264 22, n. 3, p. 471-84 3265 3266 3267 POLO, E. M. Influência da formação do curso atual do rio Negro na origem da diversificação 3268 regional em alguns grupos de aves. 2015. (Mestrado em Genética). GCBEv, INPA, Manaus, 3269 AM. 3270 3271 PRUM, R. O. et al. A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA 3272 sequencing. Nature, v. 526, n. 7574, p. 569-73, 2015. 3273 3274 RAMSEN, J. V., JR. et al. A classification of the bird species of South America. American 3275 Ornithologists' Union. 2018. 3276 3277 RIBAS, C. C. et al. Biogeography and diversification of Rhegmatorhina (Aves: Thamnophilidae): 3278 Implications for the evolution of Amazonian landscapes during the Quaternary. J Biogeogr, v. 3279 Early View, 2018. doi: 10.1111/jbi.13169 3280 3281 RIBAS, C. C. et al. A palaeobiogeographic model for biotic diversification within Amazonia over the 3282 past three million years. Proc Biol Sci, v. 279, n. 1729, p. 681-9, 2012. 3283 3284 ROSSETTI, D. F. et al. Mid-Late Pleistocene OSL chronology in western Amazonia and implications 3285 for the transcontinental Amazon pathway. Sedimentary Geology, v. 330, p. 1-15, 2015. 3286 3287 SANTORELLI JR., S.; MAGNUSSON, W. E.; DEUS, C. P. Most species are not limited by an 3288 Amazonian river postulated to be a border between endemism areas. Sci Rep, v. 8, p. 2294, 3289 2018. 3290 3291 SHEPHARD, G. E. et al. Miocene drainage reversal of the Amazon River driven by plate–mantle 3292 interaction. Nature Geoscience, v. 3, n. 12, p. 870-875, 2010. 3293 3294 SIMPSON, G. G. Splendid isolation: the curious history of South American Mammals. Yale 3295 University Press, 1980. 3296 3297 SMITH, B. T. et al. The drivers of tropical speciation. Nature, v. 515, n. 7527, p. 406-9, 2014. 3298 3299 SOARES, L. M. S. Sistemática molecular e diversificação dos gêneros Nonnula e Monasa (Aves: 3300 Bucconidae). 2016. (Doutorado em Zoologia). Museu Paraense Emílio Goeldi, UFPA, Belém, 3301 PA. 3302 3303 THOM, G.; ALEIXO, A. Cryptic speciation in the white-shouldered antshrike (Thamnophilus aethiops, 3304 Aves - Thamnophilidae): the tale of a transcontinental radiation across rivers in lowland 3305 Amazonia and the northeastern Atlantic Forest. Mol Phylogenet Evol, v. 82 Pt A, p. 95-110, 3306 2015

114

3307 3308 TOBIAS, J. A. Jacamars (Galbulidae). Handbook of the birds of the world, Barcelona, 2017. Acesso 3309 em: 26 February 2018. 3310 3311 VANZOLINI, P. E.; WILLIANS, E. E. South american anoles: the geographic differentiation and 3312 evolution of the anolis Chrysolepis species group (Sauria, Iguanidae). Arquivos de Zoologia, 3313 v. 19, n. 3-4, p. 125-298, 1970. 3314 3315 WALLACE, A. R. On the monkeys of the Amazon. Alfred Russel Wallace Writings, v. 3, 1852. 3316 Disponível em: < http://digitalcommons.wku.edu/dlps_fac_arw/3 >. 3317 3318 WANG, X. et al. Hydroclimate changes across the Amazon lowlands over the past 45,000 years. 3319 Nature, v. 541, n. 7636, p. 204-207, 2017. 3320 3321 WEIR, J. T. et al. Hybridization in headwater regions, and the role of rivers as drivers of speciation in 3322 Amazonian birds. Evolution, v. 69, n. 7, p. 1823-34, 2015 3323 3324 WILLIS, S. C. One species or four? Yes!...and, no. Or, arbitrary assignment of lineages to species 3325 obscures the diversification processes of Neotropical fishes. Plos One, v. 12, n. 2, p. e0172349, 3326 2017. 3327

115