UNIVERSIDADE ESTADUAL DE CAMPINAS INSTITUTO DE BIOLOGIA

Ana Flávia Alves Versiane

Phylogenetic Studies in Microlicia D.Don (, Microlicieae)

Estudos filogenéticos em Microlicia D.Don (Melastomataceae, Microlicieae)

Campinas 2019

Ana Flávia Alves Versiane

Phylogenetic Studies in Microlicia D.Don (Melastomataceae, Microlicieae))

Estudos filogenéticos em Microlicia D.Don (Melastomataceae, Microlicieae

Thesis presented to the Institute of Biology of the University of Campinas in partial fulfillment of the requirements for the degree of Doctor, in the area of Biology

Tese apresentada ao Instituto de Biologia da Universidade Estadual de Campinas como parte dos requisitos exigidos para a obtenção do título de Doutora em Biologia Vegetal

Orientado: Prof. Dr. Renato Goldenberg Coorientadora: Profa. Dra. Rosana Romero

ESTE TRABALHO CORRESPONDE À VERSÃO FINAL DA TESE DEFENDIDA PELA ALUNA ANA FLÁVIA ALVES VERSIANE, E ORIENTADA PELO PROF. DR. RENATO GOLDENBERG.

Campinas 2019

Campinas, 02 de dezembro de 2019

COMISSÃO EXAMINADORA

Prof. Dr. Renato Goldenberg Dra. Thais Nogales da Costa Vasconcelos Dra. Maria José Reis da Rocha Dra. Bárbara Simões Santos Leal Prof. Dr. André Olmos Simões

Os membros da Comissão Examinadora acima assinaram a Ata de defesa, que se encontra no processo de vida acadêmica do aluno.

A Ata da defesa com as respectivas assinaturas dos membros encontra-se no SIGA/Sistema de Fluxo de Dissertação/Tese e na Secretaria do Programa de Pós-Graduação em Biologia Vegetal do Instituto de Biologia.

AGRADECIMENTOS

Aqui deixo os meus sinceros agradecimentos àqueles que contribuíram direto ou indiretamente para a realização deste trabalho...

Ao Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq pela bolsa concedida (Processo: 142384/2018-6). O presente trabalho foi realizado com apoio da Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) Código de Financiamento 001. Ao meu orientador Renato Goldenberg pela oportunidade, paciência e ajuda durante todos estes quatro anos. À minha coorientadora Rosana Romero, por mais essa parceria e voto de confiança. Aos meus colaboradores, Dr. Fabian Michelangeli, Dr. Marcelo Reginato e Dr. Cassiano Welker pela ajuda com a escrita, discussões dos resultados e tomadas de decisões; Ao Fabian também agradeço pelo envio de algumas sequências de Microlicieae e ao Marcelo pela ajuda e sugestões das análises no R. Aos membros da pré-banca, Maria José Reis da Rocha, Duane Fernandes de Souza Lima e Gustavo Hiroaki Shimizu, pelas valiosas sugestões e correções. Aos membros da banca Maria José Reis da Rocha, Thais Vasconcelos, André Olmos Simão e Bárbara Leal que gentilmente aceitaram o convite de participar da banca, e aos suplentes André Vito Scatigna, Gustavo Hiroaki Shimizu e Lucas de Freitas Bacci. Ao Programa de Pós-Graduação em Biologia Vegetal da UNICAMP pelo apoio financeiro para a viagem de campo ao Parque Nacional da Serra da Gandarela. Aos professores do Programa de Pós-Graduação em Biologia Vegetal da UNICAMP pelos ensinamentos e incentivos. Ao Herbário UEC, especialmente à curadora Livia Cordi e a Kamila Pinheiro de Lima que sempre estavam à disposição. À Universidade Federal de Uberlândia, Instituto de Biologia por todo custeio das viagens de campo realizadas em Minas Gerais na Serra do Cipó, Diamantina e Serra do Cabral e em Goiás na Chapada dos Veadeiros, Minaçu e Niquelândia. Ao Herbário Uberlândense: à Dona Bia e Dona Cida pela preparação das exsicatas e à técnica Lilian Flávia pela logística de recebimento e envio de materiais, elaboração de etiquetas e por ajudar com qualquer problema que surgisse.

Aos curadores dos herbários ESA, BHCB, HUFU, MBM, SPF, UEC e UPCB pelo acolhimento nas visitas realizadas. Ao Flávio Carmo quem nos forneceu importantes informações para que a espécies tipo de Trembleya pudesse ser coletada na Serra do Gandarela. Ao Rodrigo Valentim e Gabriel Paranhos que, despretensiosamente, encontraram a espécie tipo de Microlicia em Diamantina. À Carol Dias e à Rafa Cabral, primeiras pessoas que me ensinaram as técnicas de extração de DNA tendo muita paciência e carinho para isso. Ao Vinicius Brito (Duartina) pela ajuda em meus primeiros passos em Barão Geraldo. Aos meus queridos amigos das repúblicas Sibipiruna (Samuel, Camila, Pietro, Amanda, Lucas, Thuane e Mariana), e Fazendinha (Anninha, Ana Paula e Carol Jansen) obrigada por todo carinho, conversas e amor! Aos meus queridos amigos Luiz Henrique e Leonardo por todo apoio e amizade. À Suzana, Carol Devides, Elisa, Andreza e Juliana Amaral pela companhia, colaboração e ajuda no laboratório de molecular. Ao Marcelinho Monge e ao Gu Shimizu pela amizade, conversas, conselhos e confortos. Aos amigos de Uberlândia, Carol Sbecker, Cassio, Daniela, Guilherme, Isabela, Jamile, Jean, João e Karla pela amizade e apoio de sempre. Ao Jimi, Laura e Noah que sempre com muito carinho me receberam em sua casa durante minhas idas à Uberlândia. À Duane e à Carol Harrison pelas vezes que estive em Curitiba e que me receberam e me hospedaram tão bem. À Ingrid e ao Zé por todo amor, carinho e cuidado. Aos meus pais, Ana Maria e Pedro, e minha irmã Lara pelo carinho, apoio e amor. E, por fim, agradeço à Carol, minha companheira de vida e de luta. Por todas vezes que precisei ela estava ao meu lado, sempre segurando a barra até mesmo quando eu não mais aguentava. Obrigada pela dedicação, paciência e amor.

RESUMO

Microlicieae é uma das maiores tribos de Melastomataceae, composta por 256 espécies distribuídas entre os gêneros Chaetostoma, Lavoisiera, Microlicia, Poteranthera, Rhynchanthera, Stenodon e Trembleya, os quais apresentam ocorrência quase endêmica aos campos rupestres no Brasil. Microlicia é o maior gênero, com 161 espécies, que frequentemente apresenta sobreposições morfológicas com os demais gêneros da tribo. Estudos anteriores para investigar o monofiletismo dos gêneros em Microlicieae foram limitados pela baixa amostragem dos táxons, principalmente em Microlicia. O principal objetivo deste estudo é propor uma nova hipótese filogenética para Microlicieae com ênfase em Microlicia a fim de ampliar o conhecimento sobre a história evolutiva do gênero, contribuindo para sua melhor delimitação morfológica e taxonômica. No capítulo 1, foram utilizadas sequências de DNA plastidial (atpF-atpH e trnS-trnG), nuclear ribossomal (nrITS, nrETS) e nuclear low-copy (waxy) de 124 espécies de todos os gêneros de Microlicieae, das quais 71 são Microlicia. Além disso, caracteres morfológicos tradicionalmente utilizados para delimitação de Microlicia em relação aos demais gêneros da tribo foram minuciosamente avaliados. As árvores filogenéticas foram estimadas usando análises de Verossimilhança e Inferência Bayesiana. Para identificar táxons de posição incerta nas árvores filogenéticas foi utilizado o método RogueNaRok e a abordagem treespace para visualizar incongruências entre as árvores de genes. Microlicia, como atualmente delimitado, é parafilético, e a maioria dos caracteres morfológicos utilizados para a circunscrição de cada gênero na tribo é homoplástica. Portanto, propõe-se a inclusão de Chaetostoma, Lavoisiera, Stenodon e Trembleya em Microlicia para que esta resulte em um gênero monofilético. Desta forma, após os dados obtidos neste trabalho, Microlicieae deve apresentar apenas três gêneros, Microlicia, Poteranthera e Rhynchanthera. No capítulo 2, são propostas as mudanças taxonômicas e nomenclaturais, com base nos resultados do Capítulo 1. Uma circunscrição ampla e detalhada para Microlicia é apresentada; 67 táxons são transferidos para o gênero, dos quais 48 são novas combinações, e 19 são novos nomes, e cinco nomes são reestabelecidos. Com isso, Microlicia possui 236 espécies, tornando-se um dos maiores gêneros em Melastomataceae.

PALAVRAS-CHAVE: Chaetostoma, Filogenia, Lavoisiera, Poteranthera, Reconstrução morfológica, Rhynchanthera, Stenodon, Trembleya

ABSTRACT

Microlicieae is one of the major tribes in Melastomataceae, with 256 species distributed in the genera Chaetostoma, Lavoisiera, Microlicia, Poteranthera, Rhynchanthera, Stenodon, and Trembleya, which have a near-endemic occurrence in the Brazilian campos rupestres. Microlicia is the largest genus with 161 species that often shows morphological overlap with the remaining genera in the tribe. Previous studies to investigate the monophyly of the genera in Microlicieae were limited due to the low taxon sampling, mainly in Microlicia. The main goal of this study was to propose a new phylogenetic hypothesis of Microlicieae with emphasis on Microlicia, in order to expand the knowledge about its evolutionary history and contributing to enhancing its morphological and taxonomic delimitation. In Chapter 1, were used plastidial (atpF-atpH e trnS-trnG), nuclear ribosomal (nrITS, nrETS), and nuclear low- copy (waxy) DNA sequences from 124 species from all genera in Microlicieae, of which 71 belong to Microlicia. Furthermore, the morphological characters traditionally used to circumscribe Microlicia and the other genera in the tribe were evaluated. Phylogenetic trees were estimated using Maximum Likelihood and Bayesian Inference analyses. To identify taxa with an uncertain position in the phylogenetic tree we used the RogueNaRok method and a treespace approach to visualize incongruences among the gene trees. Microlicia, as currently delimited, is paraphyletic and most of the characters used to circumscribe each genus in the tribe are homoplastic. Therefore, we propose merging Chaetostoma, Lavoisiera, Stenodon, and Trembleya in Microlicia in order to make it monophyletic. As result, our data show that Microlicieae should consist of three genera, Microlicia, Poteranthera, and Rhynchanthera. In Chapter 2, we proposed taxonomic and nomenclatural changes based on the results from Chapter 1. A broader and detailed circumscription for Microlicia is presented; 67 taxa were transferred to Microlicia, from which 48 are new combinations, 19 are new names, and five names were re-established in Microlicia. Thereby, Microlicia has 236 species and becomes one of the largest genera in Melastomataceae.

KEY WORDS: Chaetostoma, Lavoisiera, Morphological reconstruction, Phylogeny, Poteranthera, Rhynchanthera, Stenodon, Trembleya

Índice de Figuras

CAPÍTULO I. PHYLOGENETIC ANALYSIS OF MICROLICIEAE (MELASTOMATACEAE), WITH

EMPHASIS ON THE RE-CIRCUMSCRIPTION OF THE LARGE GENUS MICROLICIA D.DON

Figure 1. Species from the recognized genera in Microlicieae ...... 25 Figure 2. Morphological variation in the recognized genera in Microlicieae ...... 34

Figure 3. Visualization of the tree space occupied by 18 different data sets: bootstrap tree, best tree from ML analysis and the maximum clade credibility (MCC) tree from the BI analysis ...... 36

Figure 4. Majority consensus tree from Bayesian inference analysis of nuclear and plastidial concatenated sequences...... 39 Figure 5. Stochastic mapping of selected characters in the Microlicieae tribe. A, Habit; B, Leaf venation; C, Flower arrangement ...... 42 Figure 6. Stochastic mapping of selected characters in the Microlicieae tribe. A, Petal number; B, Ovary loci number ...... 43

CAPÍTULO II. AND NOMENCLATURAL NOTES IN MICROLICIEAE

(MELASTOMATACEAE): COMBINATIONS AND NEW NAMES IN MICROLICIA D.DON

Figure 1. Phylogenetic relationships within the Microlicieae tribe, based on majority consensus tree from Bayesian Inference of nuclear and plastidial concatenate sequences, showing Chaetostoma, Lavoisiera, Stenodon and Trembleya nested within Microlicia ...... 92

Índice de Tabelas

INTRODUÇÃO

Tabela 1. Histórico da circunscrição genérica em Microlicieae. Um asterisco (*) indica gêneros que são atualmente considerados sinônimos de Siphanthera e dois asterisco (**) de Cambessedesia. Números indicam gêneros que hoje me dia pertencem a outras tribos: ¹Tribo Melastomateae; ²Tribo Marcetieae; ³Tribo Cambessedesieae; 4Tribo Lithobieae (Penneys et al., dados não publicados); 5Tribo Eriocnemeae (Penneys et al., dados não publicados) ...... 14

CAPÍTULO I: PHYLOGENETIC ANALYSIS OF MICROLICIEAE (MELASTOMATACEAE), WITH

EMPHASIS ON THE RE-CIRCUMSCRIPTION OF THE LARGE GENUS MICROLICIA D.DON

Table 1. List of molecular markers and primer sequences used in this study ...... 28 Table 2. Datasets and summary statistical obtained from phylogenetic analyses...... 35

Table 3. Summary statistics for the posterior probabilities (PP) values in Bayesian trees ...... 37

Table 4. Summary of the morphological characters indicating the most recent common ancestor (MRCA), the total number of changes (mean over × stochastic maps), the consistency index (CI) and the retention index (RI) in each character and the most common change between the character state, considering only the Microlicieae ...... 41

Sumário Introdução Geral ...... 13 Referências ...... 16

Capítulo I. Phylogenetic analysis of Microlicieae (Melastomataceae), with emphasis on the re- circumscription of the large genus Microlicia D.Don ...... 20 Abstract ...... 21 Introduction ...... 22 Material and Methods ...... 26 Taxon Sampling ...... 26 DNA extraction ...... 26 PCR amplification, purification, and sequencing ...... 27 Sequence alignment and model selection ...... 28 Phylogenetic inference ...... 29 Treespace...... 29 Rogue taxa ...... 30 Morphology: ancestral state reconstruction ...... 30 Characters and coding...... 31 Results ...... 35 Molecular markers ...... 35 Treespace...... 35 Phylogenetic analyses ...... 36 Ancestral state reconstruction ...... 40 Discussion ...... 45 Major lineages in Microlicieae...... 46 Rhynchanthera lineage...... 46 Poteranthera lineage...... 47 Microlicia s.l. lineages ...... 47 Unresolved species ...... 47 Trembleya s.str. clade...... 48 Stenodon and allies...... 49 Pinheiroi clade ...... 50 Lavoisiera clade ...... 50 Chaetostoma clade ...... 51

Viminalis clade ...... 52 Ericoides clade...... 53 Morphology ...... 53 Habit ...... 53 Leaf venation ...... 54 Flower arrangement ...... 54 Petals number...... 55 Bristle on the hypanthium apex ...... 55 Androecium ...... 55 Ovary ...... 56 Fruit...... 57 Key to genera in Microlicieae ...... 58 Conclusions ...... 58 Acknowledgments ...... 59 References ...... 59 Supplementary Material ...... 72

Capítulo II. Taxonomy and Nomenclatural notes in Microlicieae (Melastomataceae): combinations and new names in Microlicia D.Don ...... 88 Abstract ...... 89 Resumo ...... 89 Intoduction ...... 90 Material and Methods ...... 92 Results ...... 93 New circumscription of Microlicia ...... 93 Combinations or new names in Microlicia ...... 94 Re-established names...... 102 Acknowledgments ...... 103 References ...... 103

Considerações Finais ...... 107 Referências Gerais ...... 109 Anexo 1 ...... 123 Anexo 2 ...... 128

13

INTRODUÇÃO GERAL

Melastomataceae s.l. [incl. Memecylaceae (APG, 2016)] constitui um elemento característico de ecossistemas tropicais e está entre as maiores famílias de plantas tropicais do mundo, com cerca de 5000 espécies e 160−170 gêneros (Renner, 1993; Clausing & Renner, 2001). São ervas, trepadeiras lenhosas, arbustos, ou árvores ocorrendo em montanhas, florestas, savanas e/ou vegetação perturbada (Clausing & Renner, 2001). Os níveis de diversidade, endemismo ou abundância de seus membros em diferentes habitats tornam a família um importante grupo ecológico, bem como um excelente modelo para estudos evolutivos (Reginato et al., 2016). Como exemplo, os campos rupestres, onde a família está entre os grupos de plantas vasculares com maior diversidade de espécies (Giulietti & Pirani, 1988; Rapini et al. 2008). A família está organizada em três subfamílias e 16 tribos [Melastomoideae Naudin: Astronieae Triana, Bertolonieae Triana, Blakeeae Bentham & Hooker, Cambessedesieae Bochorny et al., Dissochaeteae Naudin, Henrietteeae Penneys et al., Marcetieae M.J.R.Rocha et al., Melastomateae Bartl., Merianeae Triana, Miconieae DC., Microlicieae Naudin, Rhexieae DC., Sonerileae Triana, Trioleneae Bacci et al.; Kibessioideae Naudin: Kibessieae Krasser; e Olisbeoideae Cogn.] (Renner, 1993; Clausing & Renner, 2001; Fritsch et al., 2004; Penneys et al., 2010; Goldenberg et al., 2012; Rocha et al., 2018; Veranso-Libalah et al., 2017; Bacci et al., 2019; Bochorny et al., 2019). Microlicieae está entre as maiores tribos da família com 256 espécies. É uma tribo monofilética tanto por caracteres moleculares quanto morfológicos (Renner, 1993; Clausing & Renner 2001; Fritsch et al., 2004; Michelangeli et al., 2013). Seus membros compartilham a presença de sementes reniformes, elipsoides ou oblongas com testa foveolada ou reticulada, anteras com pedoconectivo prolongado na base e frutos capsulares (Cogniaux, 1891; Almeda & Martins, 2001; Fritsch et al., 2004). A tribo foi proposta por Naudin (1849) há cerca de 170 anos, e desde então vem sofrendo alterações significativas em relação ao número de gêneros que a compõe (Naudin, 1849 [9 gêneros]; Triana, 1871 [17]; Cogniaux, 1891 [13]; Renner, 1993 [11]; Fritsch et al., 2004 [6]; Rocha et al., 2016b [7] (Tabela 1). Atualmente, é composta pelos gêneros Chaetostoma DC. (12 espécies), Lavoisiera DC. (41), Microlicia D.Don (161), Poteranthera Bong. (5), Rhynchanthera DC. (15), Stenodon Naudin (2) e Trembleya DC. (20) (Renner, 1990; Fritsch et al., 2004; Rocha et al., 2016b; Martins & Almeda 2017; Silva et al., 2018a; Medonza-Cifuentes et al., 2019; Romero et al., 2019; Flora do Brasil 2020). A maioria dos

14

gêneros apresentam uma distribuição exclusiva à América do Sul (Fritsch et al., 2004), com exceção de Rhynchanthera que também ocorre no sul do México e América Central (Renner, 1990). Ocorre em áreas de cerrado, veredas, campo limpo e campo sujo, porém apresenta-se quase exclusiva aos campos rupestres, com mais de 90% de suas espécies endêmicas a esta formação vegetacional. Até o momento, o único trabalho filogenético para Microlicieae proposto por Fritsch et al. (2004), indicou Rhynchanthera como a primeira linhagem a se divergir na tribo; Lavoisiera e Trembleya foram reconhecidos como clados bem suportados, porém os autores sugeriram um aumento da amostragem de espécies de Trembleya a fim de confirmar seu monofiletismo; os autores apontaram ainda que Chaetostoma, Microlicia e Stenodon apresentavam ramos com baixo suporte; enquanto que Poteranthera não foi amostrado neste trabalho, uma vez que havia sido excluído de Microlicieae por Renner (1993) e Almeda & Martins (2001). Contudo, recentemente Rocha et al. (2016b) restabeleceram Poteranthera à tribo Microlicieae com base em dados moleculares e morfologia da semente e antera.

Tabela 1. Histórico da circunscrição genérica em Microlicieae. Um asterisco (*) indica gêneros que são atualmente considerados sinônimos de Siphanthera e dois asterisco (**) de Cambessedesia. Números indicam gêneros que hoje me dia pertencem a outras tribos: ¹Tribo Melastomateae; ²Tribo Marcetieae; ³Tribo Cambessedesieae; 4Tribo Lithobieae (Penneys et al., dados não publicados); 5Tribo Eriocnemeae (Penneys et al., dados não publicados).

Naudin (1849) Triana (1871) Cogniaux (1891) Renner (1993) Fritsch et al., (2004) Rocha et al., (2016b) Centradenia¹ Bucquetia¹ Bucquetia¹ Bucquetia¹ Chaetostoma Chaetostoma Chaetostoma Cambessedesia³ Cambessedesia³ Cambessedesia³ Lavoisiera Lavoisiera Lavoisiera Castratella¹ Castratella¹ Castratella¹ Microlicia Microlicia Meisneria*² Centradenia¹ Centradenia¹ Chaetostoma Rhynchanthera Poteranthera Microlicia Chaetostoma Chaetostoma Eriocnema4 Stenodon Rhynchanthera Rhynchanthera Eriocnema5 Eriocnema5 Lavoisiera Trembleya Stenodon Siphanthera² Lavoisiera Lavoisiera Lithobium4 Trembleya Stenodon Lithobium4 Lithobium4 Microlicia Trembleya Meisneria*² Microlicia Rhynchanthera Microlicia Poteranthera Stenodon Poteranthera Pyramia**³ Trembleya Pyramia**³ Rhynchanthera Rhynchanthera Siphanthera² Stenodon Stenodon Svitramia¹ Trembleya Trembleya Tulasnea*²

15

Segundo a classificação tradicional de Microlicieae, Rhynchanthera distingue-se morfologicamente dos demais gêneros de Microlicieae pelas flores haplostêmones com presença de estaminódios (Renner, 1990). Poteranthera são ervas anuais pequenas com tricomas glandulares na margem das folhas (Kriebel, 2012; Rocha et al., 2016b; Almeda & Pacífico, 2018), podendo apresentar estaminódios ou não. Chaetostoma apresenta uma coroa de tricomas no ápice externo do hipanto (Koschnitzke & Martins, 2006; Silva et al., 2018a). Stenodon apresenta ramos com casca grossa e decorticantes e estames com apêndice ventral inconspícuo (Fritsch et al., 2004). Lavoisiera distingue-se dos demais gêneros pelo ovário parcialmente ínfero a ínfero e cápsula com deiscência predominantemente acrópeta (deiscente da base para o ápice), com columela persistente (Fritsch et al., 2004; Martins & Almeda, 2017). Trembleya é caracterizado pelas nervuras reticuladas visíveis na face abaxial das folhas e inflorescências em dicásios reduzidos ou não, com flores sempre associadas a um par de brácteas basais (Martins, 1997). Ainda de acordo com a classificação tradicional, Microlicia seria o maior gênero da tribo, e a maioria de suas espécies é endêmica do Brasil, as quais apresentam três principais centros de diversidade localizados nos estados da Bahia, Goiás e Minas Gerais (Romero, 2003). Até o momento seis espécies podem ser encontradas fora do território brasileiro na Bolívia [Microlicia arenariifolia DC. e M. weddellii Naudin], Colômbia [M. colombiana HumbertoMend. & R.Romero], Guyana e Venezuela [M. benthamiana Triana and M. guanayana Wurdack] e Peru [M. sphagnicola Gleason] (Romero, 2003; Mendoza-Cifuentes et al., 2019). O gênero, de acordo com a circunscrição utilizada até o momento, é reconhecido pelas folhas desprovidas de nervação reticulada, flores pentâmeras, raro hexâmeras ou octâmeras, estames com apêndice ventral bem desenvolvido, ovário trilocular, com ápice glabro, fruto com deiscência basípeta (deiscente do ápice para base) e columela decídua (Almeda & Martins, 2001; Romero, 2003). Embora Microlicieae seja uma tribo bem delimitada (Fritsch et al., 2004; Michelangeli et al., 2013; Rocha et al., 2016b), à medida que estudos vem sendo feitos, observa-se que os caracteres tradicionalmente utilizados na delimitação dos gêneros apresentam alto grau de sobreposições, tornando difícil a distinção genérica (e.g. Almeda & Martins, 2001; Fidanza et al., 2013; Romero et al., 2015, 2017, 2019; Romero & Versiane, 2014, 2016; Pacifico & Fidanza, 2017; Pacifico et al., 2017; Diniz-Neres & Silva, 2017; Martins & Almeda 2017). Mesmo com esses avanços sobre o conhecimento da sistemática de Microlicieae (e.g., Fritsch et al., 2004; Michelangeli et al., 2013; Rocha et al., 2016a), estes trabalhos ou tiveram uma

16

amostragem limitada (Fritsch et al., 2004) ou não tiveram a tribo como foco principal de estudo (Michelangeli et al., 2013, Rocha et al., 2016a). Diante do exposto, o presente estudo apresenta a análise filogenética mais completa para Microlicieae até o momento, com 44% dos seus táxons amostrados e inclui quatro marcadores moleculares diferentes em relação ao último estudo filogenético (i.e., Fritsch et al., 2004). O principal objetivo deste estudo é propor uma nova hipótese filogenética para Microlicieae com ênfase em Microlicia a fim de ampliar o conhecimento sobre a história evolutiva do gênero, contribuindo para sua melhor delimitação morfológica e taxonômica. Desta forma, no capítulo 1, é investigada a relação filogenética de Microlicia com os gêneros Chaetostoma, Lavoisiera, Poteranthera, Rhynchanthera, Stenodon e Trembleya, e os caracteres morfológicos tradicionalmente utilizados na delimitação de cada gênero em relação à Microlicia são minunciosamente avaliados. No capítulo 2, é apresentada uma nova circunscrição para Microlicia; e são propostas as alterações nomenclaturais com base nos resultados do capítulo 1.

REFERÊNCIAS BIBLIOGRÁFICAS

Almeda F, Martins AB. 2001. New combinations and new names in some Brazilian Microlicieae (Melastomataceae), with notes on the delimitation of Lavoisiera, Microlicia and Trembleya. Novon 11: 1–7. Almeda F, Pacífico R. 2018. Neotropical Poteranthera (Melastomataceae: Microlicieae) Revisited. Systematic Botany 43(2): 552−556. DOI 10.1600/036364418X697274 APG - The Angiosperm Phylogeny Group 2016. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering : APG IV. Botanical Journal of the Linnean Society 181: 1–20. Bacci LF, Michelangeli FA, Goldenberg R. 2019. Revisiting the classification of Melastomataceae: implications for habit and fruit evolution. Botanical Journal of the Linnean Society 190: 1–24. doi.org/10.1093/botlinnean/boz006 Bochorny T, Michelangeli FA, Almeda F, Goldenberg R. 2019. Phylogenetics, morphology and circumscription of Cambessedesieae: a new Neotropical tribe of Melastomataceae. Botanical Journal of the Linnean Society 190(3): 281–302. doi.org/10.1093/botlinnean/boz018

17

Clausing G, Renner SS. 2001. Molecular phylogenetics of Melastomataceae and Memecylaceae: implications for character evolution. American Journal of Botany 88(3): 486–498. Cogniaux, CA. 1891. Melastomataceae. In De Candolle ALPP, De Candolle, ACP (eds.). Monographiae Phanerogamarum. Paris: G. Masson 7: 1–1256. Diniz-Neres DO, Silva MJ. 2017. A remarkable new species of Microlicia (Melastomataceae) including an identification key to species of the genus from the Chapada dos Veadeiros National Park, Goiás, Brazil. Systematic Botany 42(3): 554–561. DOI 10.1600/036364417X696014 Fidanza K, Martins AB, Almeda F. 2013. Four new species of Trembleya (Melastomataceae: Microlicieae) from Serra do Cabral, Minas Gerais, Brazil. Brittonia. DOI 10.1007/s12228-012-9281-x. Flora do Brasil 2020 em construção. Jardim Botânico do Rio de Janeiro.Disponível em: . Acesso em: 15 Jan 2020. Fritsch PW, Almeda F, Renner SS, Martins AB, Cruz BC. 2004. Phylogeny and circumscription of the near-endemic Brazilian tribe Microlicieae (Melastomataceae). American Journal of Botany 91(7): 1105–1114. Giulietti AM & Pirani JR. 1988. Patterns of geographic distribution of some plant species from the Espinhaço Range, Minas Gerais and Bahia, Brazil. In: Vanzolini, P.E., & Heyer, W.R. (eds) Proceedings of a workshop on Neotropical distribution patterns. Academia Brasileira de Ciências, Rio de Janeiro, 39–69 Goldenberg R, Fraga CN, Fontana AP, Nicolas AN, Michelangeli FA. 2012. Taxonomy and phylogeny of Merianthera (Melastomataceae). Taxon 61(5): 1040−1056. Koschnitzke C, Martins AB. 2006. Revisão taxonômica de Chaetostoma DC. (Melastomataceae, Microlicieae). Arquivos do Museu Nacional 64(2): 95–119. Kriebel R. 2012. A synopsis of the genus Poteranthera (Melastomeae: Melastomataceae) with the description of a new, apparently pollinator deceiving species. Brittonia 62: 6–14. Martins E. 1997. Revisão taxonômica do gênero Trembleya DC. (Melastomataceae). Tese não publicada. Universidade Estadual de Campinas. Martins AB, Almeda F. 2017. A monograph of the Brazilian endemic genus Lavoisiera (Melastomataceae: Microlicieae). Phytotaxa 135(1): 1–194. Mendoza-Cifuentes H, Ariza W, Granados DE, Romero R. 2019. A new species of Microlicia (Melastomataceae): first record of the genus for Colombia. PhytoKeys 122: 87– 96./doi.org/10.3897/phytokeys.122.34171

18

Michelangeli FA, Guimarães PJF, Penneys DS, Almeda F, Kriebel R. 2013. Phylogenetic relationships and distribution of New World Melastomeae (Melastomataceae). Botanical Journal of the Linnean Society 171: 38–60. Naudin C. 1849. Additions à la flore du Brésil méridional. Description de genres nouveaux, et rectification de quelques anciens genres appartenant à la famille des Mélastomacées. Annales des Sciences Naturelles, Botanique, série 3 12: 126–159. Pacifico RB, Fidanza K. 2017. Microlicia sciophylla (Melastomataceae: Microlicieae), a new micro-endemic species from the rocky fields of Minas Gerais, Brazil. Kew Bulletin 72: 5. DOI 10.1007/S12225-017-9688-3 Pacifico RB, Fidanza K, Almeda F. 2017. Two new species of Microlicia (Melastomataceae) from the Rupestrian Grasslands of Serra do Cipó, Minas Gerais, Brazil. Phytotaxa 316(1): 39–50. doi.org/10.11646/phytotaxa.316.1.3 Pacifico R, Almeda F, Do Carmo AP, Fidanza K. 2019. A new species of Trembleya (Melastomataceae: Microlicieae) with notes on leaf anatomy and generic circumscription Phytotaxa 391(5): 289–300. doi.org/10.11646/phytotaxa.391.5.2 Penneys DS, Michelangeli FA, Judd WS, Almeda F. 2010. Henrietteeae (Melastomataceae): A new Neotropical berry-fruited tribe. Systematic Botany 35(4): 783– 800. DOI 10.1600/036364410X539862 Rapini A, Ribeiro PL, Lambert S. & Pirani JR. 2008. A flora dos campos rupestres da Cadeia do Espinhaço. Megadiversidade 4: 15–23. Reginato M, Neubig KM, Majure LC, Michelangeli FA. 2016. The first complete plastid genomes of Melastomataceae are highly structurally conserved. PeerJ 4: e2715. DOI 10.7717/peerj.2715 Renner SS. 1990. A revision of Rhynchanthera (Melastomataceae). Nordic Journal of Botany 9: 601–630. Renner SS. 1993. Phylogeny and classification of the Melastomataceae and Memecylaceae. Nordic Journal of Botany 13: 519–540. Rocha MJR, Batista JAN, Guimarães PJF, Michelangeli FA. 2016a. Phylogenetic relationships in the Marcetia alliance (Melastomeae, Melastomataceae) and implications for generic circumscription. Botanical Journal of the Linnean Society 181: 585–609. Rocha MJR, Batista JAN, Guimarães PJF, Michelangeli FA. 2016b. Phylogenetic placement and a new circumscription of Poteranthera (Microlicieae; Melastomataceae). Phytotaxa 263 (3): 219–232. http://dx.doi.org/10.11646/phytotaxa.263.3.3

19

Rocha MJR, Guimarães PJF, Michelangeli FA, Batista JAN. 2018. Taxonomy of Marcetieae: A new Neotropical tribe of Melastomataceae. International Journal of Plant Sciences 179(1): 50−74. DOI: 10.1086/694932 Romero R. 2003. Revisão taxonômica de Microlicia sect. Chaetostomoides (Melastomataceae). Revista Brasileira de Botânica 26(4): 429–435. Romero R, Caetano APS, Santos KF, De-Paula OC, Versiane AFA. 2019. Two new species of Microlicia D.Don (Melastomataceae) from campos rupestres in eastern Brazil. Nordic Journal of Botany 37(10): 1–11. https://doi.org/10.1111/njb.02350 Romero R, Silva KR, Simão DG. 2015. Microlicia cogniauxiana and Microlicia naudiniana (Melastomataceae), two new species from the Espinhaço Range, Brazil. Systematic Botany 40(4). DOI 10.1600/036364415X690058. Romero R, Versiane AFA. 2014. A new species of Microlicia and a checklist of Melastomataceae from the mountains of Capitólio municipality, Minas Gerais, Brazil. Phytotaxa 170(2): 118–124. Romero R, Versiane AFA. 2016. Microlicia candolleana (Melastomataceae): a new endemic species to the Espinhaço range, Minas Gerais, Brazil. Phytotaxa 261(3): 275–281. http://dx.doi.org/10.11646/phytotaxa.261.3.7. Romero R, Versiane AFA, Fontelas JC, Diniz-Neres DO, de Paula, CO. 2017. A new species of Microlicia (Melastomataceae) with bright golden cuticle from Chapada dos Veadeiros, Goiás (Central Brazil). Systematic Botany 42(3): 562–568. DOI 10.1600/036364417X696069 Silva DN, Diniz DO, Koschnitzke C, Guimarães PJF, Silva MJ, Maia, VH. 2018a.

Chaetostoma hexapetalum (Microlicieae, Melastomataceae): A new species from the

Chapada dos Veadeiros, Goiás, Brazil. Systematic Botany 43(4): 986–992.

Triana J. 1871. Les Melastomacées. Transactions of the Linnean Society of London 28: 1–

188.

20

CAPÍTULO I

Phylogenetic analysis of Microlicieae (Melastomataceae), with emphasis on the re-circumscription of the large genus

1 Microlicia D.Don

Ana Flávia Alves Versiane, Rosana Romero, Marcelo Reginato, Cassiano Aimberê

Dornelles Welker, Fabián Armando Michelangeli and Renato Goldenberg

1Manuscrito será submetido à revista Botanical Journal of the Linnean Society

21

Microlicieae is a monophyletic tribe compound by seven genera: Chaetostoma,

Lavoisiera, Poteranthera, Rhynchanthera, Stenodon, and Trembleya. Microlicia is a diverse genus comprising 161 species with predominant distribution in the campos rupestres from Brazil. Its delimitation is not well-established mainly in regarding the generic boundaries with Lavoisiera and Trembleya. This study presents a phylogenetic hypothesis for Microlicieae focusing on Microlicia, with the specific goals of investigating the monophyly of Microlicia; providing an appropriate classification of

Microlicia and related genera; and investigating morphological characters to circumscribe clades and/or genera in the tribe. Plastidial (atpF-atpH and trnS-trnG), nuclear ribosomal (nrITS, nrETS) and nuclear low copy (waxy) DNA sequences were used. The phylogenetic trees were estimated using Bayesian Inference and Maximum

Likelihood analyses. The history of 12 morphological characters was estimated based on ancestral state reconstruction analyses. According to the results, Microlicia is paraphyletic with Chaetostoma, Lavoisiera, Stenodon, and Trembleya nested within it and most characters traditionally used to diagnose the genera in Microlicieae are homoplastic. Thus, here is proposed that all four previously segregated genera

(Chaetostoma, Lavoisiera, Stenodon, and Trembleya) should be included in a broadly circumscribed Microlicia to achieve its monophyly. Therefore, Microlicieae has now consisted of Rhynchanthera, Poteranthera, and Microlicia genera.

ADDITIONAL KEYWORDS: ancestral reconstruction – campos rupestres –

Chaetostoma – Lavoisiera – paraphyletic – phylogenetic – Poteranthera – Stenodon – treespace – Trembleya

22

INTRODUCTION

0Melastomataceae are among the largest tropical plant families worldwide with about

170 genera and 5000 species (Renner 1993; Clausing & Renner, 2001; Veranso-Libalah et al., 2017). Phylogenetic analyses in the family have been recovering the tribes

Melastomateae Bartl., Marcetieae M.J.R. Rocha, P.J.F. Guim. & Michelang.,

Microlicieae Naudin, and Rhexiae DC. in a clade with species that share anthers with pedoconnectives and capsular fruits (Clausing & Renner, 2001; Michelangeli et al.,

2004; Michelangeli et al., 2013; Rocha et al., 2016a; Veranso-Libalah et al., 2017;

Guimarães et al., 2019).

Microlicieae is a monophyletic tribe based on both morphological and molecular evidence (Clausing & Renner, 2001; Fritsch et al., 2004; Michelangeli et al., 2013;

Rocha et al., 2016a), forming a group with recent diversification and a nearly-endemic distribution in Brazil (Fritsch et al., 2004; Simon et al., 2009). The tribe is characterized by stamens with a prolonged pedoconnective, anthers with rostrate apex, capsular fruits, and reniform to ellipsoid seeds with foveolate or lacunate-reticulated testa (Naudin,

1849; Triana, 1871; Cogniaux, 1891; Almeda & Martins, 2001; Fritsch et al., 2004;

Rocha et al., 2016a; Martins & Almeda, 2017). Currently, is compound by seven genera: Chaetostoma DC. (12 accepted species), Lavoisiera DC. (41), Microlicia D.Don

(161), Poteranthera Bong. (five), Rhynchanthera DC. (15), Stenodon Naudin (two), and

Trembleya DC. (20) (Fig. 1; Renner, 1990; Rocha et al., 2016b; Martins & Almeda

2017; Silva et al., 2018a; Flora do Brasil 2020)

In a previous phylogenetic study for Microlicieae, Fritsch et al. (2004) indicated

Rhynchanthera as the first-diverging lineage and recognized Lavoisiera and Trembleya as well-supported clades. However, this study suggested increasing taxon sampling in

Trembleya to confirm its monophyly. Stenodon showed low supported branches. Also, according to the authors, more samples and more phylogenetically informative data

23 were required to adequately assess the monophyly and phylogenetic relationships of both Chaetostoma and Microlicia. Poteranthera was not sampled by Fritsch et al.

(2004) since it was previously excluded from Microlicieae by Renner (1993) and

Almeda & Martins (2001). Nevertheless, Rocha et al. (2016b) recently reestablished

Poteranthera in the tribe based on molecular data and morphological features of seeds and anthers.

Microlicia is the largest genus in the Microlicieae and has predominant distribution in the campos rupestres and cerrado from Brazil with over 140 species occurring in this area (Flora do Brasil 2020). Until now, six species can be found outside the Brazilian territory in Bolivia [Microlicia arenariifolia DC. and M. weddellii

Naudin], Colombia [M. colombiana HumbertoMend. & R.Romero], Guyana and

Venezuela [M. benthamiana Triana and M. guanayana Wurdack], and Peru [M. sphagnicola Gleason] (Romero, 2003; Mendoza-Cifuentes et al., 2019).

Due to its high diversity, Microlicia presents non-exclusive and/or polymorphic characters that usually are not informative for its circumscription which remains controversial (see Almeda & Martins, 2001). Nowadays, Microlicia can be recognized by its isolated flowers, usually pentamerous, rarely hexamerous or octamerous, bristles absent on the hypanthium apex, isomorphic to dimorphic stamens, tetrasporangiate or polysporangiate anthers, three locular ovary with a glabrous apex, and fruits with basipetal dehiscence and deciduous columella (Almeda & Martins, 2001; Romero 2003;

Versiane et al., this paper). However, most of these characters used to identify the species in Microlicia are also used to place species in the other genera within

Microlicieae, mainly in Lavoisiera and Trembleya.

As morphological studies advance with these genera (e.g. Romero, 2000, 2003,

2005, 2010, 2013a; Romero & Woodgyer, 2010, 2014, 2018; Fidanza et al., 2013;

Romero et al., 2014, 2015, 2017, 2019a, 2019b; Romero & Versiane, 2014, 2016;

24

Pacifico & Fidanza, 2015, 2017; Pacifico et al., 2017, 2019; Diniz-Neres & Silva, 2017;

Diniz & Silva, 2018), is possible to realize that some characters traditionally used to segregate Microlicia, Lavoisiera, and Trembleya are not reliable. The characters include absent or presence of secondaries and tertiaries nerves on the leaves, flowers solitary or disposed in inflorescence, number of petals, number of loci in the ovary, capsule with basipetal or acropetal dehiscence, and deciduous or persistent columella (see Almeda &

Martins, 2001; Fritsch et al., 2004; Fidanza et al., 2013; Martins & Almeda, 2017).

The morphological delimitation among Microlicia and the remaining genera,

Rhynchanthera, Chaetostoma, Stenodon, and Poteranthera, has not been subject to the same issues as the genera above. Rhynchanthera can be easily distinguished by haplostemonous flowers with staminodes (Renner, 1990); Chaetostoma can be recognized by the crown of bristles on the hypanthium apex (Koschnitzke & Martins,

2006; Silva et al., 2018a); Stenodon has thick, woody and decorticating branches, and also stamens with an inconspicuous ventral appendage (Fritsch et al., 2004); finally,

Poteranthera are small, annual herbs and has glandular hairs on the leaf margin

(Kriebel, 2012; Rocha et al., 2016b; Almeda & Pacifico, 2018).

In view of clarifying the relationship and the morphological delimitation of

Microlicia within Microlicieae, this study aimed to provide a phylogenetic hypothesis for the tribe focusing on Microlicia. Here were included both an increase in the number of taxa and molecular markers regarding Fritsch et al. (2004). Moreover, morphological characters traditionally used in genera delimitation were evaluated to improve the classification of Microlicieae. The specific goals were (1) to investigate the monophyly of Microlicia; (2) to provide the most appropriate classification of Microlicia and related genera; and (3) to investigate morphological characters that may circumscribe clades and/or genera in the tribe

25

Figure 1. Species from the recognized genera in Microlicieae: A, Chaetostoma armatum (Spreng.) Cogn.; B, Lavoisiera cordata Cogn.; C, Lavoisiera imbricata (Thunb.) DC.*; D, Microlicia amplexicaulis Cogn.; E, Microlicia ericoides D.Don*; F, Microlicia longipedicellata (Cogn.) Almeda & A.B. Martins ; G, Poteranthera pusilla Bong.*; H, Rhynchanthera grandiflora (Aubl.) DC.*; I, Stenodon suberosus Naudin*; J, Trembleya rosmarinoides DC.*; K, Trembleya pradosiana Netto; L, Trembleya chamissoana Naudin ex Cogn. *Type species of the genus (B−F, I−L: Versiane; A, H: Goldenberg; G: Fernandes)

26

MATERIAL AND METHODS

TAXON SAMPLING

A total of 130 taxa in 10 genera were sampled to produce a phylogenetic hypothesis for

Microlicieae focusing on Microlicia. Of these, 113 accepted species from the seven genera in Microlicieae, representing about 44% of 256 species in this group; eleven are new unpublished species of Lavoisiera, Microlicia, and Trembleya. The number of species sampled per genus were: Chaetostoma (5/41% of the species), Lavoisiera

(17/40%), Microlicia (71/45%), Rhynchanthera (4/36%), Poteranthera (1/20%),

Stenodon (1/50%), Trembleya (14/70%). Sampling included the type species of all genera in Microlicieae. Most samples were collected in the field and determined by the authors. Whenever possible, the morphological variation and geographical distribution for each genus were sampled. Only five samples were obtained from herbarium specimens (Chaetostoma stenocladon (Naudin) Kosch. & A.B. Martins, Microlicia flava R. Romero, M. hatschbachii Wurdack, M. minima Markgr., and M. parvula

(Markgr.) Kosch. & A.B. Martins). Vouchers were deposited at HUFU, NY, UEC and

UPCB (acronyms according to Thiers, 2020). As outgroup, six accessions representing three genera from lineages that are phylogenetically related to Microlicieae were chosen: L., (2 spp.) (Rhexieae), Marcetia DC., (2 spp.) and Siphanthera Pohl ex

DC., (2 spp) (Marcetieae). The trees were rooted in Marcetieae following the results of

Rocha et al. (2016a). The specimens sampled for this study are listed in Supplementary

Material (Table S1).

DNA EXTRACTION

Total genomic DNA from silica gel dried leaves was extracted using the CTAB method following Doyle & Doyle (1987) with adjustments. Specimens with a high concentration of secondary compounds had 1.5 ml of sorbitol and 30 μL of sarkosyl

27 additioned to the protocol to improve the DNA extraction. DNA from herbarium specimens (2–6 mg) was extracted using NucleoSpin 96 Plant II extraction Kit

(Macherey-Nagel) following the manufacturer’s instructions, however, the protocol was slightly modified by adding 2 μL of proteinase K and incubating on thermoblock for a period of three hours.

PCR AMPLIFICATION, PURIFICATION, AND SEQUENCING

Were amplified and sequenced two nuclear ribosomal loci, the internal transcribed spacer (nrITS) and external transcribed spacer (nrETS), one low-copy nuclear locus, the

Granule-bound starch synthase 1 (waxy) and two plastid non-coding loci, the intergenic spacers regions atpF-atpH and trnS-trnG, which are frequently used in phylogenetic studies in Melastomataceae (Michelangeli et al., 2004, 2013; Reginato & Michelangeli

2016a, 2016b; Rocha et al., 2016a; Bacci et al., 2018; Bochorny et al., 2019). Due to its large size (over 900 bp), the nrITS region was amplified in two fragments that overlap

(nrITS1+5.8S and nrITS2). Sequence data from Marcetieae, Rhexieae (both except for the marker waxy), and for Chaetostoma cupressinum (D.Don) Koschnitzke & A.B.

Martins, Lavoisiera crassifolia Mart. & Schrank ex DC., L. macrocarpa Naudin, L. subulate Triana, Microlicia amblysepala Ule, Poteranthera pusilla, Rhynchanthera bracteata Triana, R. grandiflora, and R. serrulata (L.C.Rich.) DC. were obtained from

GenBank (https://www.ncbi.nlm.nih.gov/genbank/). PCR-primers used in this study are detailed in Table 1. The amplification of all DNA regions was performed in 25 μL reactions containing 1.5 μL of genomic DNA, 2.5 μL of each primer and betaine, 12.5

μL of Green Master® Taq DNA polymerase, 2 μL of DMSO 5% or 10% and bovine serum albumin (BSA) [] 10 mg/mL. The PCR profile reactions were similar for all marlers, with small adjustments in the annealing temperature and extension time, as follows: initial denaturation at 94°C for 2 min; 40 cycles of denaturation at 94°C for 30s

28

(except for waxy, 94°C for 45s); annealing at 52°C for 30s (nrITS), 55°C for 37s

(nrETS), 54°C for 45s (waxy), 57°C for 60s (atpH-atpF), 55°C for 45s (trnS−trnG);

extension at 72°C for 40s (except for waxy which was 72°C for 75s); and a final

extension at 72°C for 7 min for all markers. PCR products were purified with

Polyethylene glycol (PEG 20%) by precipitation of the primers and dNTPs following

the protocol suggested by Dunn & Blattner (1987). Cycle sequencing reactions were

carried out with the same amplification primers using the sequencing service at

Macrogen, Inc. All sequences generated in this study were deposited in GenBank

(https://www.ncbi.nlm.nih.gov/genbank/) (see Supplementary Material, Table S1).

Table 1. List of molecular markers and primers sequences used in this study. Region Primer Sequence (5'−3') Reference nrITS1+5.8S NY183 CCTTATCATTTAGAGGAAGGAG Michelangeli et al. (2004) NY887 ATTGATGGTTCGCGGGATTCTGC nrITS2 NY017 GCATCGATGAAGAACGCAGC Michelangeli et al. (2004) NY207 CAGTGCCTCCTGCGACA nrETS NY320 AGACAAGCATATGACTACTGGCAGG Kriebel et al. (2015) NY1428 ACGTGTCGCGTCTAGCAGGCT waxy NYF1 GRGGTCTTGGGGACGTGCTC Reginato & Michelangeli (2016a) NYR AGCAGTGTGCCARTCGTTGG atpF-atpH NY822 ACTCGCACACACTCCCTTTCC Reginato et al. (2010) NY823 GCTTTTATGGAAGCTTTAACAAT trnS-trnG NY904 GAACGAATCACACTTTTACCAC Shaw et al. (2005) NY905 GCCGCTTTAGTCCACTCAGC

SEQUENCE ALIGNMENT AND MODEL SELECTION

The sequence fragments generated for each sample from bidirectional reads were

assembled and edited using Geneious v.9.1.2 (Biomatters Ltd.). DNA sequence

alignment was performed using the MAFFT v.7 algorithm with the strategy E-INS-i

(Katoh, 2013). The best evolutionary DNA model for each marker was determined

under the Bayesian Information Criteria (BIC) using PartitionFinder v.2.1.1 (Lanfear et

al., 2012) as implemented on the CIPRES Science Gateway platform

(http://www.phylo.org/; Miller et al., 2010) using the 3-model scheme (GTR, HKY,

29

K80) with or without four discrete rate categories approximating a gamma distribution

(+G) and/or invariant sites (+I).

PHYLOGENETIC INFERENCE

Phylogenetic analyses were performed individually for each marker, and the congruence among the topologies was visibly evaluated. Once no significant conflicts among the topologies were found, the same analyses were done with the concatenated alignments.

The phylogenetic trees for each marker and the concatenated data set were estimated using maximum likelihood (ML) and Bayesian inference (BI).

Maximum likelihood analyses were performed with RAxML using default parameters (Stamatakis, 2006) and run through the CIPRES Science Gateway

(http://www.phylo.org/; Miller et al., 2010). Bootstrap support values (hereafter indicated as BS) were estimated based on 1,000 replicates. The BS within 50–70% were considered as weak, 71–85% as moderate, and >85% as strong (Kress et al., 2002).

Bayesian analyses were performed using MrBayes v.3.1.2 run through the

CIPRES Science Gateway (http://www.phylo.org/; Miller et al., 2010). The analyses were run for 100,000,000 generations with four Markov chain Monte Carlo (MCMC), of two independent runs, sampling one tree every 2,000 generations. The convergence of the MCMC runs, effective sampling size (ESS) values and likelihood scores were assessed in TRACER v.1.6 (Rambaut & Drummond, 2013). The first 25% of samples from each run were discarded as “burn-in”. Groups with posterior probabilities (PP) <

0.90 were considered as weakly supported, PP within 0.91–0.95 as moderately supported, and PP 0.95 > as strongly supported.

TREESPACE

30

Ideally, a single phylogenetic tree could be used to visualize the evolutionary history of a set of sequences (Jombart et al., 2017). However, several biological and statistical factors may result in phylogenetic incongruence between gene trees (Jeffroy et al.,

2006; Gautier & Daubin, 2008; Kumar et al., 2012), and the existence of incongruence impedes the achievement of the primary goals of evolutionary research (Som, 2014).

Here, we used a treespace approach (Hillis et al., 2005) to visualize putative incongruence across our gene trees. This method provides a simple framework for exploring landscapes of phylogenetic trees and investigating phylogenetic incongruence using tree distances (Jombart et al., 2017). The analyses were performed using the R package treespace v.1.1.3 (Jombart et al., 2017), with default parameters. In our treespace analysis, we included the bootstrap tree and the best tree set from ML analysis and the Maximum Clade Credibility (MCC) tree from the BI analysis for each data set.

ROGUE TAXA

The resolution in a consensus tree and the branch support on the best-known tree can be substantially deteriorated by rogues (Aberer et al., 2013). Rogues are one or few taxa whose position is unstable due to missing data, an elevated substitution rate causing homoplasy, or extremely low rates inside and outside the clade, all of which can cause low BS (Sanderson & Shaffer, 2002). Here, we used the RogueNaRok method (Aberer et al., 2013) to identify potential rogue taxa. RogueNaRok takes a bootstrap tree set from ML analysis as input and the analysis was performed in its web server

(http://rnr.h-its.org/). RogueNaRok parameters were set as: threshold = extended majority-rule consensus, optimize = support, maximum dropset size = 1.

MORPHOLOGY: ANCESTRAL STATE RECONSTRUCTION

31

A morphological matrix was compiled to investigate the evolution of selected morphological characters and to identify putative diagnostic characters for each formed clade. These characters were used to circumscribe the genera in Microlicieae by Don

(1823), Candolle (1828), Naudin (1849), Triana (1871), Cogniaux (1883, 1891), Renner

(1993), Almeda & Martins (2001), Martins & Almeda (2017). The characters were coded by observations in herbarium specimens (BHCB, CEN, ESA, HUFU, IBGE,

MBM, NY, RB, SPF, UB, UEC, UEG, UPCB, US) (acronyms according to Thiers,

2020). and bibliographical survey (Renner, 1990; Martins, 1997; Romero, 2003;

Woodgyer, 2005; Koschnitzke & Martins, 2006; Romero & Woodgyer, 2010; Romero

2013b; Rocha et al., 2016b; Martins & Almeda, 2017, Romero et al., 2017). The morphological matrix was edited using Mesquite v.3.04 (Maddison & Maddison, 2001).

All characters were mapped on the MCC tree, in which the stable posterior distributions were combined using LogCombiner v.1.7.5. and summarized on TreeAnnotator v.1.7.5

(Bouckaert et al., 2014).

Three models of morphological character evolution (“ER”, Equal Rates; “SYM”,

Symmetric; and “ARD”, All Rates Different) were first evaluated under the corrected

Akaike information criterion (AICc) using the fitDiscrete function of the R package geiger v.2.0.6 (Harmon et al., 2008). Ancestral state reconstruction was performed through stochastic mapping implemented in the R package phytools v.0.6 (Revell,

2012), where for each character 1000 stochastic maps were generated and summarized using the functions make.simmap and describe.simmap (Revell, 2012). Taxa with polymorphic data were treated as having the same probability for each possible state.

The results were plotted over the phylogenetic tree using the basic functions of the R package ape (Paradis et al., 2004).

Characters and Coding

32

Twelve characters including habit, leaves, perianth, gynoecium, androecium, fruit and trichomes were coded (see Supplementary Material, Table S2).

Habit: herbaceous (0); woody (1). Subshrub, shrub, small tree and tree have woody tissue while in herbs it is lacking.

Leaf secondary veins: absent (0); present (1). The secondary and tertiary veins may form a reticulated pattern visible on the adaxial and/or abaxial surface, or they may be absent and only primary veins are visible (Fig. 2A).

Flower arrangement: solitary (0); dichasium (1); paired (2); glomerule (3). The inflorescence terminology followed Martins & Almeda (2017). Solitary flowers do not have bracteoles at the flower base (pedicel), while dichasia, paired and glomerular arrangements have these bracteoles.

Petal number: less than five (0); five (1); more than five (2). The number of petals is extremely important to recognize some groups in Melastomataceae, such as

Acisanthera (Rocha et al., 2016a), Marcetia (Rocha et al., 2016a), Pterolepis (Renner,

1994), and Siphanthera (Almeda & Robinson, 2011). In Microlicieae the variation ranges from five to nine, within and between each genus (Almeda & Martins, 2001;

Fritsch et al., 2004).

Bristles on the hypanthium apex: absent (0); present (1). The bristles are a series of rigid set and erect structures that form a crown on the hypanthium apex, on the outer surface (Fig. 2B1−B2).

Stamens dimorphism: isomorphic (0); subisomorphic (1); dimorphic (2).

Isomorphic stamens occur when the stamens from both whorls have the same length and form, and the ventral appendage is usually inconspicuous or only articulated to the filament (Fig. 2C1−C2); in subisomorphic stamens the antepetalous and the antesepalous are almost the same size, and the shape can be the same or not in both whorls, with a conspicuous ventral appendage or only articulated to the filament (Fig.

33

2E1−E2); and dimorphic occurs when the stamens from the antepetalous whorl are ½ or less as long as the antesepalous and with a distinct shape and a conspicuous ventral appendage in the antesepalous whorl (Fig. 2D1−D2).

Stamens number: androecium with one whorl of fertile stamens (0); androecium with two whorls of fertile stamens (1). Most genera of Melastomataceae have androecia with two whorls, both with fertile stamens (Rocha et al., 2016a). However, some species or even genera have androecia with one whorl of fertile stamens; the second whorl may be reduced to staminodes or lacking.

Anthers, number of sporangia: tetrasporangiate (0); polysporangiate (1). In

Melastomataceae, the anthers are usually described as tetrasporangiate and bilocular at maturity (Baumgratz et al., 1996; Almeda & Martins, 2001). Nevertheless, multilocular anthers (polysporangiate) may also occur in Microlicia, i.e., when the thecae are internally divided into numerous small locules along the dorso-lateral face (Baumgratz et al., 1996; Lima et al., 2018). In general, tetra and polysporangiate anthers can be distinguished without a stereomicroscope, with tetrasporangiate anthers having a smooth surface (Fig. 2D1−D2) while polysporangiate anthers have a bullate surface

(Fig. 2C1, 2E1−E2).

Ovary locule number: two (0); three (1); four (2); five (3); more than five (4). This character is highly variable between and within each genus of Microlicieae.

Ovary position: (0) superior; (1) partly inferior; (2) inferior. Although the ovary position is considered a stable character in angiosperms (Endress, 2010, 2011),

Melastomataceae species have perigynous flowers in which the ovary varies from superior to inferior (Basso-Alves et al., 2017). The ovary is superior when there is no adnation between ovary wall and hypanthium (Fig. 2F), partly inferior when the adnation reaches half the length of the ovary (Fig. 2G) and inferior when the ovary is fully adnate to the hypanthium (Fig. 2H).

34

Figure 2. Morphological variation in the recognized genera in Microlicieae. A, leaf blade showing the secondary and tertiary veins in Trembleya (M. Monge et al., 2595, UEC); B1, bristles on the hypanthium apex in Chaetostoma, B2, detail of the same bristles (F. Almeda et al., 9434, UEC); C1, isomorphic stamen and tetrasporangiate anther in Microlicia (W. Ganev, 2803, UEC), C2, flower with isomorphic stamens (W. Ganev, 2803, UEC); D1−D2, dimorphic stamens and polysporangiate anthers in Microlicia

(F.S. Meyer, 2053, UEC); E1−E2, subisomorphic stamens and tetrasporangiate anthers in Microlicia (F.S.

Meyer, 1185, UEC); F, superior ovary in Microlicia (F.S. Meyer, 2053, UEC); G, semi-inferior ovary in

Lavoisiera (F. Almeda et al., 8562, UEC); H, inferior ovary in Lavoisiera (I.M. Araújo et al., 233, UEC);

I, fruit showing the basipetal dehiscence in Microlicia (F. Almeda et al., 8305, UEC); J, fruit showing the acropetal dehiscence in Lavoisiera (F. Almeda et al., 8525, UEC); K, persistent columella in Lavoisiera

(F. Almeda et al., 8525, UEC).

35

RESULTS

MOLECULAR MARKERS

A total of 501 new sequences from five DNA regions were generated during this study,

and the final alignment of the concatenated sequences included 4,475 base pairs (bp),

from which 1746 (39%) were variable and 1113 (24.9%) parsimony informative within

the ingroup. From these, the nrETS was the most variable marker (65.3%) and the atpF-

atpH the less variable (26.6%). The characteristics of each aligned locus and the best fit

nucleotide substitution model are provided in Table 2.

Table 2. Datasets and summary statistical obtained from phylogenetic analyses.

nrITS nrETS Waxy atpF-atpH trnS-trnG Concatenated Concatenated Total

nuclear plastidial

Number of taxa sequenced 127 119 95 107 86 129 114 129 Alignment length (bp) 1004 674 843 927 1027 2521 1954 4475 Variable sites 326 (32.5%) 440 (65.3%) 381 (45.2%) 247 (26.6%) 352 (34.3%) 1147 (45.5%) 599 (30.7%) 1746 (39%) Parsimony informative sites (bp) 246 (24.5%) 361 (53.6%) 166 (19.7%) 115 (12.4%) 225 (21.9%) 773 (30.7%) 340 (17.4%) 1113 (24.9%) Retention index (RI) 0.85 0.79 0.6 0.69 0.75 0.74 0.68 0.70 DNA model (BIC criterion) GTR+I+G GTR+G K80+G GTR+I+G GTR+I+G - - -

TREESPACE

A treespace depicting topological distances between the gene tree bootstrap sets, the

concatenated bootstrap set, the best tree, and the MCC trees of all data set is presented

in Figure 3. In the first axis, there is not a clear separation of the bootstrap tree sets

across all data sets, while in the second axis the plastidial and waxy trees are closer, the

nrITS appears on an intermediary position overlapping with the Nuclear and Total

concatenated trees, and the nrETS on the other extreme. The Nuclear and Total

concatenated bootstrap sets show the highest cohesion in the treespace, indicating less

phylogenetic uncertainty. Also, these two data sets are highly overlapping in the tree

space, indicating great similarity. All data sets, to some extent and with varied intensity,

present some degree of overlap among them, indicating that incongruences among the

36 data sets could at least be partially explained by uncertainty of the nuclear and plastidial concatenated trees (available at the Supplementary Material - Figures S1−S2).

Figure 3. Visualization of the tree space occupied by 18 different data sets: bootstrap tree, best tree from ML analysis and the maximum clade credibility (MCC) tree from the BI analysis.

PHYLOGENETIC ANALYSES

In the data set, no individual was flagged as rogue. Among the gene trees, the median

PP across nodes ranged from 0.05 in the atpF-atpH tree to 0.59 in the nrETS tree. The nrETS tree also had the higher amount of nodes with PP ≥ 0.95 (25% of the nodes), followed by nrITS (20%), waxy (19%); the plastid gene trees had lower PP values

(atpF-atpH, 6% of the nodes, trnS-trnG, 19%). The median PP of the total concatenated tree was 0.76 and 38% of the nodes had support ≥ 95%. All splits in the backbone have low PP values and are usually supported by only one of the five gene trees. A summary of the PP values of the trees is listed in Table 3.

37

Table 3. Summary statistics for the posterior probabilities (PP) values in Bayesian trees.

Median % < 0.9 0.9 ≤ % < 0.95 % ≥0.95 nrITS 0.08 78 2 20 nrETS 0.59 70 5 25 Single trees Waxy 0.33 76 5 19 atpF-atpH 0.05 91 3 6 trnS-trnG 0.18 80 1 19 Nuclear 0.6 66 6 28 Concatenated Plastidial 0.04 87 1 12 trees Total 0.76 57 5 38

The phylogenetic results showed three main lineages in Microlicieae (Fig. 4):

Rhynchanthera (clade A), Poteranthera (clade B) and the remaining five genera in the tribe, hereafter referred to as Microlicia s.l. Within Microlicia s.l. (Fig. 4), seven well- supported clades were found: Trembleya s.s. (clade C), Stenodon and allies (clade D),

Pinheiroi (clade E), Lavoisiera (clade F), Chaetostoma (Clade G), Viminalis (clade H); and Ericoides (clade I). Nevertheless, the resolution of their relationships is still impossible to infer, since they are clustered together into a polytomy. As well as,

Microlicia flava, Trembleya hatschbachii Wurdack & Martins, T. phlogiformis DC., and

T. pradosiana which are unresolved species with no placement in clades along the phylogenetic tree (Fig. 4).

Clades A and B. Rhynchanthera (clade A: PP = 1; BS = 100) is the sister group of all other lineages in the tribe (PP = 1; BS = 100) and, together with Poteranthera

(clade B), are the lineages which probably first diverged in the tribe Microlicieae (Fig.

4). Poteranthera is the sister group of Microlicia s.l. (PP =1; BS = 78).

Clade C. Trembleya s.s. is a strongly supported clade (PP = 1; BS = 98) in this molecular phylogenetic hypothesis. It contains eight of 14 species of Trembleya sampled here, including the type species of the genus, T. rosmarinoides. (Fig. 1J).

38

Figure 4. Majority consensus tree from Bayesian inference analysis of nuclear and plastidial concatenated sequences. Numbers on the left are posterior probabilities (PP) from Bayesian inference analysis and on the right are bootstrap support values (BS) from maximum likelihood analysis (only PP ≥ 0.90 and BS ≥ 71 are shown). The type species of each genus is marked with an asterisk (*). The colours and form of the symbols represent distinct genera. The letters indicate clades discussed in the text.

39

Figure 4. continuation

40

Clade D. Stenodon and allies includes the type species of the genus Stenodon (S. suberosus, Fig. 1I) and 15 of 71 species of the genus Microlicia sampled here.

However, it had a strongly supported clade only in the BI analysis (PP = 0.96).

Clade E. Pinheiroi is a well-supported clade (PP = 0.98; BP = 94) with only two species from Bahia states nested within it: Microlicia pinheiroi Wurdack and M. giulietiana A.B. Martins & Almeda.

Clade F. Lavoisiera clade has all 18 species of the genus Lavoisiera sampled in this study. It forms a strongly supported clade (PP = 1; BS = 89%), however, the relationships among its species are not resolved.

Clade G. Chaetostoma clade is fomed by all five species of the genus

Chaestostoma sampled here. The clade is highly supported in the BI and moderately supported in the ML analysis (PP = 1; BS = 83%).

Clade H. Viminalis clade has five of 71 species of Microlicia sampled in this study. It has support only in the BI analysis (clade K: PP = 1). Microlicia viminalis is the sister group of the remaining species in this clade (PP =1; BS = 80).

Clade I. Ericoides clade includes the type species of the genus Microlicia (M. ericoides), most of the Microlicia species sampled here (52 out of the 70), including all undescribed species, and some species of Trembleya (4 out of the 14). It is a strongly supported clade in the BI analysis but without support in the ML (clade L: PP = 0.96).

ANCESTRAL STATE RECONSTRUCTION

The ancestral state reconstructions indicated that most characters have some degree of homoplasy in Microlicieae. The ovary locule number and flower arrangement are the characters with more change in the tribe (Table 4). However, the bristles on the hypanthium apex, can be considered a diagnostic character to Chaetostoma clade and

41

the partly inferior and/or inferior ovary and the fruit acropetal dehiscence only for

Lavoisiera clade (Fig. S4A and B). A summary of the ancestral state reconstructions

indicating the probable state at the most recent common ancestor (MRCA) of

Microlicieae, the total number of changes, the consistency index (CI), retention index

(RI), and the most common change between the character states is available in Table 5.

All changes between all character states are in Supplementary Material (Table S5).

Following, for the ancestral state reconstruction, are also presented the characters

considered most homoplastic in Microlicieae (Figs. 5−6). The remaining characters are

in Supplementary Material (Figs. S3−S4).

Table 4. Summary of the morphological characters indicating the most recent common ancestor (MRCA), the total number of changes (mean over × stochastic maps), the consistency index (CI) and the retention index (RI) in each character and the most common change between the character state, considering only the Microlicieae. Both CI and RI measure the degree of tree likeness; if the CI and RI values are equal to one, this means that there is no homoplasy.

Total Character MRCA of Microlicieae CI RI Most common change changes Habit Woody 2 0.5 0.5 woody to herbaceous Leaf secondary veins Uncertain 18 0.08 0.43 absent to presente Flower arrangement Solitary 25 0.17 0.55 solitary to dichasium Petal number Five 9 0.25 0.57 five to more than five Bristles on the hypanthium apex Absent 1 1 1 absent to presente Stamens dimorphism Dimorphic 12 0.18 0.31 subisomorphic to dimorphic Stamen number Androecium of two whorls 4 0.33 0.6 androecium of two whorls to one whorl Anthers, number of sporangia Tetrasporangiate 13 0.08 0.45 tetrasporangiate to polysporangiate Ovary locule number Three 30 0.2 0.57 three to four Ovary position Superior 1 1 1 superior to partly inferior Fruit dehiscence Basipetal 3 0.33 0.87 basipetal to acropetal Columella Caducous 1 1 1 caducous to persistente

The morphological reconstruction analysis shows that the woody is the

hypothetical MRCA of Microlicieae and Poteranthera is the unique lineage with

herbaceous species, while the Rhynchanthera and Microlicia s.l. lineages are woody

species (Fig. 5A). The most commom change within Microlicieae is from woody to

herbaceous habit and this character had only two changes (Table 4).

42

Figure 5. Stochastic mapping of selected characters in the Microlicieae tribe. A, Habit; B, Leaf venation; C, Flower arrangement.

43

Figure 6. Stochastic mapping of selected characters in the Microlicieae tribe. A, Petal number; B, Ovary loci number.

According to the morphological analyses, the MRCA for the leaf venation is

uncertain, however this character has changed 18 times within the tribe (Table 4). The

leaf venation probably evolving independently in some species along the phylogentic

tree and the presence of secondary and tertiaries veins is shared by all species in

Trembleya s.s. clade, absent for Chaetostoma clade, and in the remaing clade both

conditions are presented (Fig. 5B).

44

In the flower arrangement, solitary flowers are the MRCA in Microlicieae

(Table 4, Fig. 5C). The most common change in from solitary flowers to dichasia, and the less frequent is from glomerular to paired or reduced to one flower (Tables 4 and

S3). Pentamerous flowers are the MRCA state in Microlicieae (Fig. 6A), and five petals to more than five is the most common change (Table 4). Lavoisiera clade shows the most considerable variation in this character (Fig. 6A, clade I). Hypanthium lacking bristles is the MRCA in Microlicieae, and the presence of bristles is an autopomorphie to the Chaetostoma clade (Fig. S3A, clade H).

The reconstruction of stamen dimorphism (Fig. S3B) in the Microlicieae suggests that dimorphic ones is the MRCA (Table 4). The stamen form changed at least

12 times in Microlicieae, and subisomorphic to dimorphic is the most common change

(Table 4). Androecia with diplostemonous flowers is MRCA in Microlicieae and the most common change is from the androecia with two whorls to one whorl (Table 4).

Here, the Rhynchanthera, Poteranthera, and Lavoisiera clades have species with haplostemonous flowers (Fig. S3C; clades A, B, and F, respectively), in the remaining clades the flowers are diplostemonous. In Microlicieae, tetrasporagiate anthers are the

MRCA (Table 4, Fig. S3D) and the unique change is from tetrasporangiate to polysporangiate that probably happened 13 times (Table 4).

Three locules is the MRCA in Microlicieae. The number of loci is the most homoplastic character with 30 change (Table 4, Fig. 6B; clade F); the superior ovary is the MRCA (Fig. S4A; clade F); in Microlicieae and the superior to partly inferior is the unique change in this character (Table 4). The basipetal dehiscence is the MRCA (Fig.

S4B) in Microliciea and the most common change is from basipetal to acropetal (Table

4). The deciduous columella is the MRCA in Microlicieae (Fig. S4C) with only two changes and the most common change is from caducous to persistent in Microlicieae

(Table 4).

45

DISCUSSION

Microlicia, as traditionally circumscribed, is paraphyletic. According to Larridon et al.

(2013), there are three main strategies that can be implemented to develop new classifications encompassing the concept of monophyly for large paraphyletic groups and their segregate genera: (1) accepting paraphyletic taxa, (2) splitting paraphyletic genera or (3) lumping segregate genera into the paraphyletic one. We understand that the first option is not acceptable (see Hennig, 1965; Welzen, 1997; Dias et al., 2005).

Since there was not found a set of morphological characteristics that allow us to define and segregate Microlicia into several genera, whereas some clades have medium or low support and there are ungrouped species in the phylogenetic tree, we are not able to propose strategy two. Therefore, we suggest lumping the genera Chaetostoma,

Lavoisiera, Trembleya, and Stenodon in Microlicia and to expand its circumscription because for a date this is a most likely way to achieve a stable classification classification within Microliceae. This has been a common approach in large and taxonomically complex groups in (e.g. Melastomataceae [Miconia Ruiz &

Pav. (Michelangeli et al., 2004)], Myrtaceae [Myrcia DC. ex Guill. (Lima, 2017)

Myrteae DC. (Vasconcelos et al., 2017) and Eugenia L. (Bünger et al., 2016)]

With this phylogenetic approach for Microlicia were also certified Trembleya as paraphyletic genera, as also related by Pacífcio et al. (2019). The results obtained by

Fritsch et al. (2004), in their studies with the tribe Microlicieae, were confirmed. Such as, Rhynchanthera as the sister group of all remaining genera in Microlicieae, and

Chaetostoma and Lavoisiera as monophyletic groups. Stenodon did not have its position indicated within Microlicieae by Fritsch et al. (2004), so its placement in a clade with other species of Microlicia is a novelty from this study (Fig. 4, clade D).

Backbones with short branch lengths and low support seem to be a common pattern across different levels of phylogenetic hypotheses in Melastomataceae

46

(Goldenberg et al., 2018; e.g., Goldenberg et al., 2008; Penneys et al., 2010; Reginato

& Michelangeli, 2016b). Here, to try improving the branch support and resolution for the relationships between the genera in Microlicieae, the number of sampled taxa was increased and more molecular markers (nrETS, waxy, atpF-atpH, trnS-trnG) were added when compared to the study of Fritsch et al. (2004).

Leaf anatomical studies have been indicating high variability between species traditionally recognized within Microlicia and Trembleya, making it impossible to delimit these genera through anatomical data (Silva et al., 2018b; Carmo et al., 2019;

Santos & Caetano, unpublished data). However, it seems that some anatomical characters may have potential diagnostic value for Lavoisiera (see Silva et al., 2018b;

Carmo et al., 2019; Santos & Caetano, unpublished data). Studies on flavonoids also confirm the difficulty to segregate these genera, and it has been shown that some species of Trembleya are related to Lavoisiera, while others are related to Microlicia

(Bonfim-Patrício et al., 2001).

MAJOR LINEAGES IN MICROLICIEAE

Rhynchanthera lineage (clade A)

The genus Rhynchanthera was placed in Microlicieae based on morphological characters (Naudin, 1849; Renner, 1990, 1993) and it was confirmed by molecular studies (Clausing & Renner, 2001; Fritsch et al., 2004; Michelangeli et al., 2013; Rocha et al., 2016a; Versiane et al., this paper). According to these studies the genus has always been recovered as the first-divergent lineage within Microlicieae.

Rhynchanthera has 16 species and the widest distribution in the tribe, occurring from southern Mexico to Paraguay and Santa Catarina state, in Brazil (Renner, 1990). Its species are subshrubs or shrubs, with flowers usually with five purple petals,

47 androecium with one whorl of fertile stamens, which can be dimorphic or subisomorphic, and also with purple filament and pedoconnective, and five whitish to purplish staminodia (Renner, 1990).

Poteranthera lineage (clade B) The genus Poteranthera was first included in the tribe Lasiandrales (=Melastomateae)

(Naudin, 1849), then transferred to Microlicieae (Triana, 1871) and kept there by

Cogniaux (1885, 1891). Renner (1993) placed the genus in Melastomateae according to morphological characters, later Rocha et al. (2016b) placed it within Microlicieae. Here were confirmed the result obtained by Rocha et al. (2016b), however it would be necessary to increase the sampling of Poteranthera to make inferences about its interspecific relationships. Poteranthera has five species with a disjunct distribution in

Venezuela, Brazil, Bolivia (Kriebel, 2012; Rocha et al., 2016a; Almeda & Pacifico,

2018). Its species can be recognized by being small and annual herbs with glandular hairs on the leaf margins, four or five petals, an apically constricted hypanthium and androecium with one or two whorls of fertile stamens, in the first with or without staminodia (Kriebel, 2012; Rocha et al., 2016a; Almeda & Pacifico, 2018).

Microlicia s.l. lineage (clade C-I)

Unresolved species

Here, the species were not placed in any clade (Fig. 4). Trembleya phlogiformis and T. pradosiana belong to one of the three sections delimited in Trembleya by Cogniaux

(1883): Trembleya sect. Heterogenae, which can be distinguished from the other sections by the urceolate hypanthium, dimorphic stamens and puberulent-glandular leaves with or without setose margins. This section also included T. neopyrenaica

Naudin and T. warmingii, both not sampled in our analyses. The first one is

48 morphologically similar to T. pradosiana and T. phlogiformis and the second was transferred to Poteranthera by Almeda & Pacifico (2018) based on morphological data.

Microlicia flava, has subshrub habit, sessile leaves, discolours (when dry), indument with short-stalked glands, conspicuously viscid and bright,dichasia reduced to one flower, flowers with yellow petals, dimorphic stamens with polysporangiate anthers, three locular ovary, and deciduous columella (Romero, 2000; Versiane et al., this paper). This species is endemic in the campos rupestres from Serra da Canastra, Minas

Gerais state (Romero, 2000). Trembleya hatschbachii (Martins, 1995) has shrub habit, petiolate leaves, discolours (when dry), indument with short-stalked glands, viscosous, flowers disposed in compound dichasia, yellow petals, dimorphic stamens with tetrasporangiate anthers, five locular ovary, and deciduous columella (Martins, 1995;

Versiane et al., this paper). Trembleya hastchbachii is considered endemic to Grão

Mogol region, Minas Gerais state (Martins, 1995).

Trembleya s.s. (clade C)

This clade includes species from the other two of Cogniaux’s sections (1883):

Trembleya sect. Trembleya Cogn. had campanulate hypanthia, dimorphic stamens, and glabrous or slightly pubescent leaves; T. sect. Erioleuca Cogn. had the leaves with a tomentose adaxial surface. In our analyses, the species from T. sect. Heterogenae Cogn. and some species with no sectional placement, described after Cogniaux, 1883) (see

Martins, 1995; Fidanza et al., 2013; Pacifico & Fidanza, 2015), were placed in the

Ericoides clade or as not belonging to any clade (unresolved species). Tembleya s.s. clade occur in the campos rupestres and cerrado in Brazil, with the most species endemic to Minas Gerais state, except by T. parviflora D.Don with wide distribution from Bahia to Paraná states (Flora do Brasil 2020). This clade is confirmed in the preliminary results obtained by Pacifico et al. (2019) in phylogenetic studies with the

49 genus Trembleya. Also, according to Pacifico et al. (2019) T. altoparaisensis R.

Pacifico, Almeda & Fidanza, endemic to Goiás state, is placed in this clade. The species in Trembleya s.s., can be recognized by the habit varying from subshrubs to shrubs and small trees, petiolate leaves with evident secondary and tertiary veins, the adaxial surface with short-petiolate glands, becoming green-blackish or blackish when dry, bracteolate flowers disposed in dichasia or reduced to one or two flowers, capsules with basipetal dehiscence and columella deciduous. Despite being diagnosable, the recognition of this clade as a distinct genus would render Microlicia as a paraphyletic genus. Therefore, we suggest that the species in Trembleya s.s. clade should be transferred to Microlicia.

Stenodon and allies (clade D)

Stenodon was proposed by Naudin (1844) with one species and characterized by its shrubby habit, twisted branches with thick rough bark, shortly petiolate, oblong- lanceolate leaves, axillary, solitary, and sessile flowers, with six to seven petals, subisomorphic stamens, and a 4-locular ovary. Later, Triana (1871) included a second species, S. gracilis (not sampled), and since then no more species have been described to the genus. The species of Microlicia (except M. stenodonoides Diniz & Silva) placed in this clade have a subshrubby to shrubby (but not twisted) habit, slender branches, flowers with five petals, subisomorphic to dimorphic stamens, and 3−5-locular ovary.

There is not a set of characters that would justify the transference of these species from

Microlicia to Stenodon or allow the recognition of a broadly circumscribed Stenodon.

However, the inclusion of Stenodon in Microlicia is possible if the Microlicia circunscription is expanded to include the thick branches with rough bark of Stenodon.

Most species in this clade have distribution in Goiás and Minas Gerais states, with some

50 species also occurring in Mato Grosso and São Paulo states (Flora do Brasil 2020), found in campo rupestre and/or cerrado.

Pinheiroi (clade E)

This clade is formed by two endemic species to the campos rupestre from Chapada

Diamantina in the Bahia state (Wurdack, 1983; Almeda & Martins, 2001; Flora do

Brasil 2020). Microlicia pinheiroi is characterized by its shrub habit, petiolated and discolours leaves, lanceolate blade, entire margins, indument with short-stalked glands, dimorphic and concolours stamens, polysporangiated anthers, three locular ovary, and deciduous columella (Wurdack, 1983; Versiane et al., this paper) and M. giuliettiana

A.B. Martins & Almeda has shrub habit, sessile leaves, obovate to broadly elliptic blade, crenate or serreate margins, indument with long-stalked glands, dimorphic and bicolours stamens, tetrasporangiated anthers, three locular ovary, and deciduous columella (Almeda & Martins, 2001; Versiane et al., this paper). According to our data, both species do not exhibit morphological features capable to segregate them in a new genus as the results suggest, nor because of their geographic distribution, since more species from Bahia are placed in clade I (Fig. 4). The state of Bahia is the second center of diversity for Microlicia with 55 species (Flora do Brasil 2020; Romero & Woodgyer,

2018). Taking it into account, we believe that an increased sampling of Microlicia from this state could give us better results regarding the relationships among these species, since here only seven species from Bahia were sampled.

Lavoisiera (clade F)

The genus Lavoisiera was first placed in the tribe Lavoisiereae while Microlicia was in the tribe Rhexieae (Candolle, 1828). When establishing Microlicieae, Naudin (1849) transferred Lavoisiera and Microlicia to this tribe and sustained through the years

51

(Triana, 1871; Cogniaux, 1891; Renner, 1993; Almeda & Martins, 2001; Martins &

Almeda 2017). Based on the circumscription of Lavoisiera with capsules dehiscing acropetally and ovary partly inferior with a persistent columella and laterally flattened lamelliform placental intrusion, Almeda & Martins (2001) transferred seven species of

Lavoisiera that did not show these characters to Microlicia. From these, five were sampled here and were kept apart from the Lavoisiera clade and placed in clades D (M. ordinata (Wurdack) Almeda & A.B. Martins), E (M. giullietiana), and I (M. longipedicellata (Cogn.) Almeda & A.B. Martins, M. noblickii (Wurdack) A.B. Martins

& Almeda, and M. mucugensis (Wurdack) Almeda & A.B. Martins). Lavoisiera clade is endemic to the campos rupestres and cerrado in Brazil (Martins & Almeda, 2017). It can be characterized subshrub to shrub, to small trees or trees habit, mostly with sessile leaves, flowers with five to eight or rarely nine to ten petals, partly inferior to inferior ovary, 2−4(−10)-locular, capsules usually acropetally dehiscent, and persistent columella (Martins & Almeda, 2017). Anatomical characters such as the spongy parenchyma with thickened primary walls and supporting tissue (collenchyma and/or sclerenchyma) in the leaf margin (Silva et al., 2018a; Carmo et al., 2019; Santos &

Caetano, unpublished data) are considered synapomorphies to Lavoisiera clade. In agreement with Fritsch et al. (2004), all species of Lavoisiera sampled here form a strong and consistent clade and could be kept as a distinct genus. However, it could render Microlicia as a paraphyletic genus, so we suggest transferring all species from

Lavoisiera to Microlicia.

Chaetostoma (clade G)

This clade can be recognized by its sessile, carenate, imbricate and pungent leaves without short-stalked glandular trichomes, flowers with a crown of bristles around the apex of the hypanthium, and five petals, or seldom six petals (Koschnitzke & Martins,

52

1999, 2006, 2007; Silva et al., 2018a). Its species are restricted to campos rupestres, campos de altitude, and cerrado in the Brazilian states of Bahia, Goiás, Mato Grosso,

Minas Gerais, São Paulo, Rio de Janeiro, and Paraná (Koschnitzke & Martins, 2006).

The position of Chaestostoma was weakly supported in Fritsch et al. (2004) because only one species was sampled. Here we added four more species and it was recovered as monophyletic, with bristles of the hypanthium apex as an exclusive feature to this clade but all other morphological characters mentioned to characterize the genus are shared with other clades in the tribe. In the same way as Lavoisiera, if we keep Chaetostoma as a distinct genus, it could render Microlicia as a paraphyletic genus. Therefore, we suggest transferring all its species to Microlicia.

Viminalis (clade H)

In the Viminalis clade, the species are subshrub to shrub, with sessile to petiolate leaves, white or pink petals, subisomorphic or dimorphic stamens, tetrasporangiated to polisporangiated anthers, ovary 3 or 4-locular (Gleason, 1931; Versiane et al., 2016;

Romero et al., 2017). However, all these features are also found in the other clades formed by Microlicia species. So, apart from the molecular evidence, no morphological characters or a set of it, to keep this clade as a new genus separated from the remaining

Microlicia. This clade has an interesting disjunct distribution, with some species from

Central Brazil grouped with extra-Brazilian (Peru) species occurring in similar habitats

(grassland areas). Most species have an endemic distribution in Goiás (Flora do Brasil

2020). However, M. viminalis also occurs in Minas Gerais, Bahia, Tocantins, and Mato

Grosso (Romero, 2003; pers. obs.), M. euphorbioides Mart. in Distrito Federal, Mato

Grosso, Minas Gerais, São Paulo (Flora do Brasil 2020), and M. sphagnicola Gleason occurs only in the Central Andes in Peru (Gleason, 1931).

53

Ericoides (clade I)

Ericoides is the clade largest recovered in this study, consequently, shows the wider range of morphological variation. Except by the autopomorphies (bristles on the hypanthium apex, partly inferior and/or inferior ovary, and fruit acropetal dehiscence), all character states found here overlap with the other clades formed by Microlicia species. In this clade, the species are endemic in the campos rupestes from Bahia and

Minas Gerais states. The few supported clades recovered here, despite no morphological sustain, tend to follow its species’ geographic distribution (e.g. Microlicia amplexicaulis

Cogn. and M. pusilla Cogn. from Minas Gerais state; M. hatschbachii and M. noblickii from Bahia state). Therefore, we reinforce the importance of sampling more species from Bahia to better inference about the biogeographic grouping in all clades.

MORPHOLOGY

In this paper, according to the available data some clades recovered in the molecular analyses could be supported as segregate genera by one morphological character (e.g.

Chaestostoma) or when combined (e.g. Trembleya s.s., Lavoisiera). However, the clades formed by species of Microlicia with Trembleya or Stenodon genera are impossible of delimitation due to a high degree of homoplasies in their morphological characters (e.g. Stenodon and allies, Pinheiroi, Viminalis, Ericoides).

Habit

The habit had few changes (Table 4, Fig. 5A), reinforcing it is as an uncontested character to segregate the lineages within Microlicieae. The herbaceous state is exclusive to Poteranthera being an innovation within Microlicieae. May it can be an adaptation to seasonal wetlands and swampy habitats where the Poteranthera species

54 grow (Almeda & Pacifico, 2018). The species in the Rhynchanthera and Microlicia s.l. lineages have woody habit, where in the first grow in moist places where water accumulates or along the margin’s streams or ponds (Renner, 1990), and the second one often occur in open, rocky, and nutrient-poor habitats (campos rupestres, cerrado).

Leaf venation

In Microlicieae this character has been used to segregate the genera Trembleya and

Microlicia, which in the first the secondary veins are present in the former, while only the primary vein is visible in the latter (Martins, 1997; Fidanza et al., 2013; Pacifico &

Fidanza, 2015). This character had many changes among and within the clades in

Microlicieae (Table 4, Fig. 5B), indicating its high degree of homoplasy. However, if combined with other characters it could be used to recognize Trembleya s.s. clade.

Flower arrangement

Until now, flower arrangement and inflorescence architecture have been used to distinguish genera in Microlicieae, mainly to segregate Trembleya from Microlicia

(Fidanza et al. 2013; Pacifico & Fidanza, 2015; Pacifico et al. 2019). The pyramidal inflorescences were present in Trembleya while in Microlicia the flowers were solitaries

(Martins, 1997; Almeda & Martins, 2001). However, this character seems to be very plastic given the many changes recovered along with the evolutionary history in the tribe (Table 4, Fig. 5C). Except by the Chaetostoma clade, which has only solitary flowers (Koschintzke, 2006), all remaining clades of Microlicieae have some kind of inflorescence (Martins & Almeda, 2017; Versiane et al., this paper). So, due to the plasticity of this character is not suitable to diagnose the majority groups in the

Microlicieae.

55

Petals number

In Microlicieae, occasionally, the number of petals can vary between species within the clades or on the same individual. However, pentamerous flowers are more common and it is present in all clades (Fig. 5D). Lavoisiera clade exhibits the highest variation in this character reaching five to ten petals (Fig. 5D, clade F). Within Ericoides clade, some species have more than five petals as Trembleya inversa Fidanza, A.B. Martins &

Almeda with six to seven petals (Flora do Brasil, 2020) and Microlicia longipedicellata with eight to nine; in Stenodon and allies, Microlicia stenodonoides D.O. Diniz-Neres

& M.J. Silva has four to six petal (Diniz-Neres & Silva, 2017) and Stenodon suberosus, five to seven (Fig. 5D). Therefore, this character cannot be useful as diagnostic for the clades recovered in this phylogenetic study.

Bristle on the hypanthium apex

In earlier studies (Koschnitzke & Martins, 1999; Koschnitzke & Martins, 2007) this character had already used to recognized Chaetostoma’s species from the other genera in the tribe, mainly from Microlicia. Here, with all species of Chaetostoma nested together in a monophyletic group (Fig. 4), the bristle on the hypanthium apex could be useful to characterize it, since it’s a diagnostic character to the Chaetostoma clade (Fig.

S3A).

Androecium

Within Microlicieae, dimorphic stamens are the most common condition when compared to subisomorphic ones, and isomorphic stamens are the less frequent character state (Fig. S3B). Subisomorphic stamens are present in some species of

Rhynchanthera, Stenodon and allies, and Viminalis clades (Fig. S3B). In our analyses only Poteranthera pusilla has isomorphic stamens, however, dimorphic stamens are

56 found in its remaining species, not sampled here (see Kriebel, 2012; Almeda & Pacifico,

2018). All species in our phylogenetic tree, nested in the clades Trembleya s.s.,

Lavoisiera (see Martins & Almeda, 2017), Chaetostoma (see Koschnitzke & Martins,

2007), Pinheiroi, and Ericoides have dimorphic stamens (Fig. S3B).

Most genera of Melastomataceae have androecia with one whorl of fertile stamens (diplostemonous flowers) (Rocha et al., 2016b). In the Rhynchanthera lineage all species have haplostemonous flowers and presence of staminodia (Renner 1990); in

Potheranthera lineage the species can be haplostemonous with or not staminodia and also diplostemonous (Kriebel, 2012; Rocha et al., 2016b; Almeda & Pacifico, 2018); and in Microlicia s.l lineage only one species is haplostemonous with presence of staminodia in the Lavoisiera clade (Fig. S3C) (see Almeda & Martins, 2017). In

Microlicieae, haplostemonous flowers have probably evolved via suppression of the antepetalous whorl from the diplostemonous flowers and the presence or not of staminodes is intimately involved with the number of androecium whorls.

Polysporangiate anthers were an exclusive character to the genera Microlicia within the tribe Microlicieae, however with this new approach it was reconstructed to the clades D, E, G, and I (Fig. S3D).

Ovary

In earlier studies, (Candolle, 1828) the ovary loci number was used to segregate the genera Chaetostoma, Lavoisiera, Microlicia, Rhynchanthera, and Trembleya, and it was followed by Naudin (1844, 1849), Triana (1871), Cogniaux (1883, 1891). However, some authors have been relating how this character is variable becoming not satisfactory to characterize the genera (see Renner, 1990; Martins, 1997; Koshnitzke & Martins,

2006; Rocha et al., 2016; Martins & Almeda, 2017). In our analysis, the ovary loci

57 number has many changes (Table 4) being quite variable among and within the clades

(Fig. 6B) so it reinforces the idea that it cannot be useful to support them.

Melastomataceae are among the families with ontogenetic characters variable enough to result in multiple events of transformation in ovary position (Basso-Alves et al., 2017). The partly inferior and/or inferior ovary position is a diagnostic character for

Lavoisiera clade (Table 4, Fig. S4A), while all remaing clades have superior ovary. In all Melastomataceae the stamens are inflexed while in bud, with the anthers accommodated between the style and walls of the hypanthium and ovary (Reginato &

Michelangeli, 2016c). Maybe the ovary position in Microlicieae is related in how long the anthers can involve the ovary, where in superior ovary all ovary is involve by the

5 anthers and in semi-iferior and inferior ovary the anther reach ½ to ⁄4 of the ovary.

Fruit

Species in Microlicieae have capsular fruits that may have basipetal or acropetal dehiscence (Almeda & Martins, 2001; Fritsch et al., 2004). This character has low degree of homoplasie and few changes (three times) in the tribe (Table 5). Only species in Lavoisiera clade have acropetal fruits, while the fruits in the remaining clades are basipetal (Fig. S4B). However, acccording to Martins & Almeda (2017) six species of

Lavoisiera have fruits dehiscing from the apex to the base. Studies with anatomical and ontogenetic fruit development in Microlicieae (Ribeiro et al. unpublished data) relate that the dehiscence region in the genera Lavoisiera and Microlicia is variable between them thus is not a consistent character to separate them. In most of capsular-fruit in

Melastomataceae the columella is deciduous, however in Lavoieira clade the columella is persistent and was used to segregate it from the other genera in Microlicieae (Fig.

S4C) (see Naudin, 1844; Almeda & Martins, 2001; Martins & Almeda, 2017).

However, Pataro et al. (2013) reported a persistent columella for M. pulchra Pataro &

58

R. Romero and, during this study, we found both persistent and deciduous columellae in

M. balsamifera Cogn. (UEC134323) and M. longipedicellata (HUFU66441; observation in field).

KEY TO GENERA IN MICROLICIEAE

1. Herbaceous plants ……………………………………………………… Poteranthera

1'. Woody plants

2. Haplostemonous flowers with five staminodes purple, white or cream-coloured

…………………………………………………………………………… Rhynchanthera

2'. Diplostemonous flowers, staminodes absent (except for L. confertiflora with five yellow staminodes) ..…….…………………………………………………… Microlicia

CONCLUSIONS

To date, this study represents the major understanding about the relationships within the tribe Microlicieae. Were included four different markers (ETS, waxy, atpF-atpH and trnS-trnG) and increased the sampling for Microlicieae when compared with the phylogenetic study proposed by Fritsch et al. (2004). From the data presented here, the phylogenetic analyses revealed three lineages in Microlicieae, Rhynchanthera,

Poteranthera, and Microlicia s.l. The latter is compound by the genera Chaetostoma,

Lavoisiera, Microlicia, Stenodon, and Trembleya, where Microlicia and Trembleya are paraphyletic and the remaining genera were recovered as monophyletic groups.

In Melastomataceae, homoplastic characters traditionally used for the circumscription of different taxon levels have been demonstrated across the family

(Goldenberg et al., 2018). So, apart from the difficulty to obtain well-resolved and supported trees, taxonomic re-circumscriptions in the family are also hindered by the

59 lack of synapomorphies or even of a set of morphological characters with diagnostic value for the clades identified in molecular analyses (Rocha et al., 2016b). In

Microlicieae, the most morphological characters traditionally used to diagnose the genera are homoplastic, therefore not appropriated as predictors for relationships within the group and genera delimitation.

Thus, according to the results and to try achieving the monophyly of Microlicia, here the four segregate genera (Chaetostoma, Lavoisiera, Stenodon, and Trembleya) are included in a broadly circumscribed Microlicia. From now, the tribe Microlicieae comprises three genera Rhynchanthera, Poteranthera, and Microlicia, and Microlicia includes 239 species exclusive from South America (Brazil, Bolivia, Colombia,

Guyana, Peru, and Venezuela) occurring in campos rupestes, cerrado, tepuis, and sub- paramos with most of them endemic to Brazil.

ACKNOWLEDGMENTS

We thank the curators and staff of BHCB, CEN, ESA, HUFU, IBGE, MBM, NY, RB,

SPF, UB, UEC, UEG, UPCB, and US herbaria for specimen loans or assistance during visits; Financial support was provided by Coordenacão de Aperfeiçoamento de Pessoal de Nível Superior – CAPES (Finance Code 001) and Conselho Nacional de

Desenvolvimento Científico e Tecnológico – CNPq (Process: 142384/2018-6). AFAV would like to thank Ana Carolina D. Castello, Suzana Maria Costa and Elisa Cândido for assistance in the laboratory; André Vito Scatigna and Danielle Diniz for sending specimens.

REFERENCES

60

Aberer AJ, Krompass D, Stamatakis A. 2013. Pruning rogue taxa improves

phylogenetic accuracy: an efficient algorithm and webservice. Systematic Biology

62(1): 162–166.

Almeda F, Martins AB. 2001. New combinations and new names in some Brazilian

Microlicieae (Melastomataceae), with notes on the delimitation of Lavoisiera,

Microlicia and Trembleya. Novon 11: 1–7.

Almeda F, Robinson OR. 2011. Systematics and phylogeny of Siphanthera

(Melastomataceae). Systematic Botany Monographs 93: 1–101.

Almeda F, Pacifico, R. 2018. Neotropical Poteranthera (Melastomataceae:

Microlicieae) revisited. Systematic Botany 43(2): 552−556. DOI

10.1600/036364418X697274

Baumgratz JFA, Souza MLDR, Woodgyer EM, Lughadha EMN. 1996.

Polysporangiate anthers: described for the first time in Melastomataceae. Kew

Bulletin 51: 133–144.

Bacci LF, Michelangeli FA, Goldenberg R. 2019. Revisiting the classification of

Melastomataceae: implications for habit and fruit evolution. Botanical Journal of the

Linnean Society 190: 1–24. doi.org/10.1093/botlinnean/boz006

Basso-Alves JP, Goldenberg R, Teixeira SP. 2017. The ontogenetic bases for

variation in ovary position in Melastomataceae. American Journal of Botany 104(8):

1142–1156.

Bochorny T, Michelangeli FA, Almeda F, Goldenberg R. 2019. Phylogenetics,

morphology and circumscription of Cambessedesieae: a new Neotropical tribe of

Melastomataceae. Botanical Journal of the Linnean Society 190(3): 281–302.

doi.org/10.1093/botlinnean/boz018

61

Bonfim-Patrício MC, Salatino A, Martins AB, Wurdack JJ, Salatino ML. 2001.

Flavonoids of Lavoisiera, Microlicia and Trembleya (Melatomataceae) and their

taxonomic meaning. Biochemical Systematics and Ecololgy 29: 711−726

Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu CH, Xie D, Suchard MA,

Rambaut A, Drummond AJ. 2014. BEAST 2: A Software Platform for Bayesian

Evolutionary Analysis. PLOS Computational Biology 10(4): e1003537. doi:

10.1371/journal.pcbi.1003537.

Bünger MO, Mazine FF, Forest F, Bueno ML, Stehmann JR, Lucas EJ. 2016. The

evolutionary history of Eugenia sect. Phyllocalyx (Myrtaceae) corroborates

historically stable areas in the southern Atlantic forests. Annals of Botany 118(7):

1209–1223. doi:10.1093/aob/mcw209

Candolle, ALPP. de 1828. Melastomaceae. In: Candolle, ALPP de, ed. Prodromus

systematis naturalis regni vegetabilis 3. Paris: Treuttel et Wurtz, 99–202.

Carmo A, Pacifico R, Gonçales-Silva R, Sbais PG, Fidanza K, Souza LA. 2019.

Epidermal micromorphology and venation patterns of Microlicieae

(Melastomataceae) leaves: Looking for new characters for the taxonomy of this

neotropical tribe. Flora. doi.org/10.1016/j.flora.2019.151494

Clausing G, Renner SS. 2001. Molecular phylogenetics of Melastomataceae and

Memecylaceae: implications for character evolution. American Journal of Botany

88(3): 486–498.

Cogniaux CA. 1883. Melastomataceae: Microlicieae. In: Martius, CFP, Eichler AW,

Urban I, eds. Flora brasiliensis. Fleischer. Leipzig, 14(3): 5–205.

Cogniaux CA. 1885. Poteranthera. In: Martius, CFP, Eichler AW, Urban I, eds. Flora

Brasiliensis. Fleischer. Leipzig, 14(3): 201–204.

Cogniaux CA. 1891. Melastomataceae. In: Candolle ALPP de, Candolle, ACP de, eds.

Monographiae Phanerogamarum. Paris: G. Masson, 7: 1–1256.

62

Dias P, Assis LCS, Udulutsch RG. 2005. Monophyly vs. paraphyly in plant

systematics. Taxon 54: 1039-1040

Diniz-Neres DO, Silva MJ. 2017. A remarkable new species of Microlicia

(Melastomataceae) including an identification key to species of the genus from the

Chapada dos Veadeiros National Park, Goiás, Brazil. Systematic Botany 42(3): 554–

561. DOI 10.1600/036364417X696014

Diniz DO, Silva MJ. 2018. A noteworthy new species of Microlicia (Melastomataceae,

Microlicieae) from the highlands of the state of Goiás, Brazil. Journal of the Torrey

Botanical Society 145(2): 175–181. doi.org/10.3159/TORREY-D-17-00027.1

Don D. 1823. An illustration of the natural family of plants called Melastomaceae.

Memoirs of the Wernerian Natural History Society 4(2): 276–329.

Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of

fresh leaf tissue. Phytochemical Bulletin 19: 11–15.

Dunn IS, Blattner FR. 1987. Charons 36 to 40: multi-enzyme, high capacity,

recombination deficient replacement vectors with polylinkers and polystuffers.

Nucleic Acids Research 15: 2677–2698.

Endress PK. 2010. Flower structure and trends of evolution in and their major

subclades. Annals of the Missouri Botanical Garden 97: 541–583.

Endress PK. 2011. Changing views of flower evolution and new questions. In:

Wanntorp L, Ronse De Craene LP, eds. Flowers on the tree of life. Cambridge:

Cambridge University Press, 120–141.

Fidanza K, Martins AB, Almeda F. 2013. Four new species of Trembleya

(Melastomataceae: Microlicieae) from Serra do Cabral, Minas Gerais, Brazil.

Brittonia. DOI 10.1007/s12228-012-9281-x.

Flora do Brasil 2020 under construction. Jardim Botânico do Rio de Janeiro.

Available at < http://floradobrasil.jbrj.gov.br/ > (accessed 06 January 2020).

63

Fritsch PW, Almeda F, Renner SS, Martins AB, Cruz BC. 2004. Phylogeny and

circumscription of the near-endemic Brazilian tribe Microlicieae (Melastomataceae).

American Journal of Botany 91(7): 1105–1114.

Galtier N, Daubin V. 2008. Dealing with incongruence in phylogenomic analyses.

Philosophical Transactions of the Royal Society B 363: 4023–4029.

Gleason HA. 1931. Studies on the Flora of Northern South America-XV. Recent

Collections of Melastomataceae from Peru and Amazonian Brazil. Bulletin of the

Torrey Botanical Club 58(4): 215–262.

Goldenberg R, Penneys DS, Almeda F, Judd WS, Michelangeli FA. 2008.

Phylogeny of Miconia (Melastomataceae): patterns of stamen diversification in a

megadiverse neotropical genus. International Journal of Plant Sciences 24: 315–327.

Goldenberg R, Reginato M, Michelangeli FA. 2018. Disentangling the infrageneric

classification of megadiverse taxa from Mata Atlantica: Phylogeny e and function as

the crucible of architectural diversity. Annals of Botany 112: 1477–1493.

Guimarães PJF, Michelangeli FA, Sosa K, Gómez JRS. 2019. Systematics of

Tibouchina and allies (Melastomataceae: Melastomateae): A new taxonomic

classification. Taxon. doi.org/10.1002/tax.12151

Harmon LJ, Weir JT, Brock CD, Glor RE, Challenger W. 2008. GEIGER:

Investigating evolutionary radiations. Bioinformatics 24: 129–131.

doi.org/10.1093/bioinformatics/btm538

Hennig W. 1965. Phylogenetic Systematics. Annual Review of Entomology 10: 97–116.

Hillis DM, Heath TA, St. John K. 2005. Analysis and visualization of Tree Space.

Systematic Biology 54(3): 471–482. DOI: 10.1080/10635150590946961

Jeffroy O, Brinkmann H, Delsuc F, Philippe H. 2006. Phylogenomics: the beginning

of incongruence? Trends in Genetics 22(4): 225–31.

64

Jombart T, Kendall M, Almagro-Garcia J, Colijn C. 2017. TREESPACE: Statistical

exploration of landscapes of phylogenetic trees. Molecular Ecology Resources 17:

1385–1392. DOI: 10.1111/1755-0998.12676

Katoh S. 2013. MAFFT multiple sequence alignment software version 7:

Improvements in performance and usability. Molecular Biology and Evolution 30:

772–780. doi.org/10.1093/molbev/mst010

Kress WJ, Prince LM, Williams KJ. 2002. The phylogeny and a new classification of

the gingers (Zingiberaceae): evidence from molecular data. American Journal of

Botany 89: 1682–1696.

Kriebel R. 2012. A synopsis of the genus Poteranthera (Melastomeae:

Melastomataceae) with the description of a new, apparently pollinator deceiving

species. Brittonia 62: 6–14.

Kriebel R, Michelangeli FA, Kelly LM. 2015. Discovery of unusual anatomical and

continuous characters in the evolutionary history of Conostegia (Miconieae:

Melastomataceae). Molecular Phylogenetics and Evolution 82: 289–313.

Koschnitzke C, Martins AB. 1999. New combinations and a new species in the

Brazilian genus Chaetostoma DC. (Microlicieae: Melastomataceae). Novon 9(2):

202–204.

Koschnitzke C, Martins AB. 2006. Revisão taxonômica de Chaetostoma DC.

(Melastomataceae, Microlicieae). Arquivos do Museu Nacional 64(2): 95–119.

Koschnitzke C, Martins AB. 2007. Nomenclatural Alterations in Microlicieae

(Melastomataceae) Novon 17(4): 472–475.

Kumar S, Filipski AJ, Battistuzzi FU, Pond SLK, Tamura K. 2012. Statistics and

truth in Phylogenomics. Molecular Biology and Evolution 29(2): 457–472.

doi:10.1093/molbev/msr202

65

Lanfear R, Calcott B, Ho SYW, Guindon S. 2012. PartitionFinder: Combined

selection of partitioning schemes and substitution models for phylogenetic analyses.

Molecular Biology and Evolution 29: 1695–1701. doi.org/10.1093/molbev/mss020

Larridon I, Bauters K, Reynders M, Huygh W, Muasya AM, Simpson DA,

Goetghebeur P. 2013. Towards a new classification of the giant paraphyletic genus

Cyperus (Cyperaceae): phylogenetic relationships and generic delimitation in C4

Cyperus. Botanical Journal of the Linnean Society 172: 106–126.

Lima, DFS. 2017. Estudos filogenéticos e taxonômicos em Myrcia DC. sensu lato

(Myrtaceae), com ênfase no clado Guianensis. Unpublished D. Phil. Thesis.

Universidade Estadual de Campinas.

Maddison WP, Maddison DR. 2001. MESQUITE: a modular system for evolutionary

analysis, Version 3.04. Available at: http://mesquiteproject.org.

Martins E. 1995. Nova espécie do gênero Trembleya DC. (Microlicieae -

Melastomatacea). Boletim de Botânica da Universidade de São Paulo 14: 39−42.

doi.org/10.11606/issn.2316-9052.v14i0p39-42

Martins E. 1997. Revisão taxonômica do gênero Trembleya DC. (Melastomataceae).

Unpublished D. Phil. Thesis. Universidade Estadual de Campinas.

Martins AB, Almeda F. 2017. A monograph of the Brazilian endemic genus

Lavoisiera (Melastomataceae: Microlicieae). Phytotaxa 135(1): 1–194.

Mendoza-Cifuentes H, Ariza W, Granados DE, Romero R. 2019. A new species of

Microlicia (Melastomataceae): first record of the genus for Colombia. PhytoKeys

122: 87–96. /doi.org/10.3897/phytokeys.122.34171

Michelangeli FA, Penneys DS, Giza J, Soltis D, Hils MH, Skean JD. 2004. A

preliminary phylogeny of the tribe Miconieae (Melastomataceae) based on nrITS

sequence data and its implications on inflorescence position. Taxon 53: 279–290.

66

Michelangeli FA, Guimarães PJF, Penneys DS, Almeda F, Kriebel R. 2013.

Phylogenetic relationships and distribution of New World Melastomeae

(Melastomataceae). Botanical Journal of the Linnean Society 171: 38–60.

Miller MA, Pfeiffer W, Schwartz T. 2010. Creating the CIPRES Science Gateway for

inference of large phylogenetic trees in proceedings of the gateway computing

environments workshop (GCE) New Orleans, LA. Pp 1–8. Available at:

http://www.phylo.org/.

Naudin C. 1844. Additions à la flore du Brésil méridional. Description de genres

nouveaux, et rectification de quelques anciens genres appartenant à la famille des

Mélastomacées. Annales des Sciences Naturelles, Botanique, série 3 2: 140–155.

Naudin C. 1849. Additions à la flore du Brésil méridional. Description de genres

nouveaux, et rectification de quelques anciens genres appartenant à la famille des

Mélastomacées. Annales des Sciences Naturelles, Botanique, série 3 12: 126–159.

Pacifico RB, Fidanza K. 2015. Three new endangered species of Trembleya DC.

(Melastomataceae: Microlicieae) from Minas Gerais, Brazil. Phytotaxa 238(2): 163–

173. dx.doi.org/10.11646/phytotaxa.238.2.4

Pacifico RB, Fidanza K. 2017. Microlicia sciophylla (Melastomataceae: Microlicieae),

a new micro-endemic species from the rocky fields of Minas Gerais, Brazil. Kew

Bulletin 72: 5. DOI 10.1007/S12225-017-9688-3

Pacifico RB, Fidanza K, Almeda F. 2017. Two new species of Microlicia

(Melastomataceae) from the Rupestrian Grasslands of Serra do Cipó, Minas Gerais,

Brazil. Phytotaxa 316(1): 39–50. doi.org/10.11646/phytotaxa.316.1.3

Pacifico R, Almeda F, Carmo AA, Fidanza K. 2019. A new species of Trembleya

(Melastomataceae: Microlicieae) with notes on leaf anatomy and generic

circumscription. Phytotaxa 391(5): 289–300.

67

Paradis E, Claude J, Strimmer K. 2004. APE: Analyses of phylogenetics and

evolution in R language. Bioinformatics 20: 289–290.

doi.org/10.1093/bioinformatics/btg412

Pataro L, Romero R, Roque N. 2013. Four new species of Microlicia

(Melastomataceae) from Chapada Diamantina, Bahia, Brazil. Kew Bulletin 68:

285Y–293.

Penneys DS, Michelangeli FA, Judd WS, Almeda F. 2010. Henrietteeae

(Melastomataceae): A New Neotropical Berry-Fruited Tribe. Systematic Botany

35(4): 783–800. DOI 10.1600/036364410X539862

Rambaut A, Drummond AJ. 2013 Tracer. Avaiable at:

.

Reginato M, Michelangeli FA, Goldenberg R. 2010. Phylogeny of Pleiochiton

(Melastomataceae, Miconieae): total evidence. Botanical Journal of the Linnean

Society 162(3): 423–434.

Reginato M, Michelangeli FA. 2016a. Primers for low-copy nuclear genes in the

Melastomataceae. Applications in Plant Sciences 4(1): 1500092.

Reginato M, Michelangeli FA. 2016b. Untangling the phylogeny of Leandra s.str.

(Melastomataceae, Miconieae). Molecular Phylogenetics and Evolution 96: 17–32.

Reginato M, Michelangeli FA. 2016c. Diversity and constraints in the floral morphological evolution of Leandra s.str. (Melastomataceae). Annals of Botany 118:

445–458. doi:10.1093/aob/mcw116,

Renner SS. 1990. A revision of Rhynchanthera (Melastomataceae). Nordic Journal of

Botany 9: 601–630.

Renner SS. 1993. Phylogeny and classification of the Melastomataceae and

Memecylaceae. Nordic Journal of Botany 13: 519–540.

68

Renner SS. 1994. A revision of Pterolepis (Melastomataceae: Melastomeae). Nordic

Journal of Botany 14: 73–104.

Revell LJ. 2012. phytools: An R package for phylogenetic comparative biology (and

other things). Methods in Ecology and Evolution 3: 217–223.

doi.org/10.1111/j.2041-210X.2011. 00169.x

Rocha MJR, Batista JAN, Guimarães PJF, Michelangeli FA. 2016a. Phylogenetic

relationships in the Marcetia alliance (Melastomeae, Melastomataceae) and

implications for generic circumscription. Botanical Journal of the Linnean Society

181: 585–609.

Rocha MJR, Batista JAN, Guimarães PJF, Michelangeli FA. 2016b. Phylogenetic

placement and a new circumscription of Poteranthera (Microlicieae;

Melastomataceae). Phytotaxa 263 (3): 219–232.

http://dx.doi.org/10.11646/phytotaxa.263.3.3

Romero R. 2000. A new species of Microlicia (Melastomataceae) from Brazil.

Brittonia 52(2): 142−144.

Romero R. 2003. Revisão taxonômica de Microlicia sect. Chaetostomoides

(Melastomataceae). Revista Brasileira de Botânica 26(4): 429–435.

Romero R. 2005. A new species of Microlicia (Melastomataceae) from Minas Gerais,

Brazil. Novon 15(2): 358–360.

Romero R. 2010. Uma nova espécie de Microlicia (Melastomataceae) do estado de

Minas Gerais, Brasil. Rodriguésia 61(Sup.): S15-S18.

Romero R. 2013a. A new species of Microlicia (Melastomataceae) from the Espinhaço

Range, Minas Gerais, Brazil. Phytotaxa 88: 1–5.

dx.doi.org/10.11646/phytotaxa.88.1.1.

Romero R. 2013b. Taxonomic notes in Microlicia (Melastomataceae, Microlicieae).

Phytotaxa 110(1): 48–54. dx.doi.org/10.11646/phytotaxa.110.1.4

69

Romero R, Caetano APS, Santos KF, De-Paula OC, Versiane AFA. 2019a. Two

new species of Microlicia D.Don (Melastomataceae) from campos rupestres in

eastern Brazil. Nordic Journal of Botany 37(10): 1–11.

https://doi.org/10.1111/njb.02350

Romero R, Fontelas JC, Versiane AFA, De-Paula OC. 2019b. Microlicia

longirostrata, a new species of Melastomataceae from Goiás in Central Brazil.

Systematic Botany 44(2): 394-354.

Romero R, Silva KR, Simão DG. 2014. Microlicia crassa and M. maculata spp. nov.

(Melastomataceae) from Minas Gerais, Brazil: morphology and leaf anatomy. Nordic

Jounal of Botany 33: 178–185. doi: 10.1111/njb.00635.

Romero R, Silva KR, Simão DG. 2015. Microlicia cogniauxiana and Microlicia

naudiniana (Melastomataceae), two new species from the Espinhaço Range, Brazil

Systematic Botany 40(4). DOI 10.1600/036364415X690058.

Romero R, Versiane AFA. 2014. A new species of Microlicia and a checklist of

Melastomataceae from the mountains of Capitólio municipality, Minas Gerais,

Brazil. Phytotaxa 170(2): 118–124.

Romero R, Versiane AFA. 2016. Microlicia candolleana (Melastomataceae): a new

endemic species to the Espinhaço range, Minas Gerais, Brazil. Phytotaxa 261(3):

275–281. http://dx.doi.org/10.11646/phytotaxa.261.3.7.

Romero R, Versiane AFA, Fontelas JC, Diniz-Neres DO, de Paula, CO. 2017. A

new species of Microlicia (Melastomataceae) with bright golden cuticle from

Chapada dos Veadeiros, Goiás (Central Brazil). Systematic Botany 42(3): 562–568.

DOI 10.1600/036364417X696069

Romero R, Woodgyer E. 2010. Validation of the name Microlicia petiolulata

(Melastomataceae), a new species from the Espinhaço Range, Minas Gerais, Brazil.

Kew Bulletin 65: 69–72.

70

Romero R, Woodgyer E. 2014. Rediscovery of two species of Microlicia

(Melastomataceae) in Minas Gerais, Brazil. Phytotaxa 173(1): 41–48.

Romero R, Woodgyer E. 2018. Six new species of Microlicia (Melastomataceae) from

Bahia, Brazil. Kew Bulletim 73: 22. DOI 10.1007/S12225-018-9747-4

Sanderson MJ, Shaffer HB. 2002. Troubleshooting Molecular Phylogenetic Analyses.

Annual Review of Ecology and Systematics 33: 49–72 doi: 10.1

146/annurev.ecolsys.33.010802.150509

Shaw J, Lickey EB, Beck JT, Farmer SB, Liu W, Miller J, Siripun KC, Winder

CT, Schilling EE, Small RL. 2005. The tortoise and the hare II: relative utility of 21

noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal

of Botany 92: 142–166.

Silva DN, Diniz DO, Koschnitzke C, Guimarães PJF, Silva MJ, Maia, VH. 2018a.

Chaetostoma hexapetalum (Microlicieae, Melastomataceae): A new species from the

Chapada dos Veadeiros, Goiás, Brazil. Systematic Botany 43(4): 986–992 DOI

10.1600/036364418X697805

Silva KR, Romero R, Simão DG. 2018b. Leaf characters of Lavoisiera, Microlicia and

Trembleya (Microlicieae, Melastomataceae) and their implications for taxonomy.

Feddes Repertorium 1–14. DOI: 10.1002/fedr.201700019

Simon MF, Grether R, Queiroz LP, Skema C, Pennington RT, Hughes CE. 2009.

Recent assembly of the Cerrado, a neotropical plant diversity Hotspot, by in situ

evolution of adaptations to fire. Proceedings of the National Academy of Sciences 06

(48): 20359–20364.

Som A. 2014. Causes, consequences and solutions of phylogenetic incongruence.

Briefings in Bioinformatics 16(3): 536–548. doi:10.1093/bib/bbu015

Stamatakis A. 2006. RAxML-VI-HPC: maximum likelihood based phylogenetic

analyses with thousands of taxa and mixed models. Bioinformatics 22: 2688–2690.

71

Thiers B. 2020. Index Herbariorum: A global directory of public herbaria and

associated staff. New York Botanical Garden’s Virtual Herbarium. Avaliable at:

. (accessed 06 January 2020).

Triana J. 1871. Les Melastomacées. Transactions of the Linnean Society of London 28:

1–188.

Vasconcelos TNC, Proença CEB, Ahmad B, Aguilar DS, Aguilar R, Amorim BS,

Campbell K, Costa IR, De-Carvalho PS, Faria JEQ, Giaretta A, Kooij PW,

Lima DF, Mazine FF, Peguero B, Prenner G, Santos MF, Soewarto J, Wingler

A, Lucas EJ. 2017. Myrteae phylogeny, calibration, biogeography and

diversification patterns: increased understanding in the most species rich tribe of

Myrtaceae. Molecular phylogenetics and evolution 109: 113-137.

Veranso-Libalah MC, Stone RD, Fongod, AGN, Couvreur TLP, Kadereit G. 2017.

Phylogeny and systematics of African Melastomateae (Melastomataceae) Taxon

66(3): 584–614. doi.org/10.12705/663.5

Versiane AFA, Santos ML, Romero R. 2016. Melastomataceae na Serra dos Pireneus,

Goiás, Brasil. Rodriguésia 67(3): 721–759. DOI: 10.1590/2175-7860201667314

Welzen PC van. 1997. Paraphyletic groups or what should a classification entail. Taxon

46(1): 99–103.

Woodgyer E. 2005. Multi-Access key and checklist to the species of Microlicia

(Melastomataceae) in Bahia, Brazil. Kew Bulletim 60: 441–448.

Wurdack JJ. 1983. Certamen Melastomataceis XXXVI. Phytologia 53(2): 121–13.

72

SUPPLEMENTARY MATERIAL

Table S1. Voucher information and GenBank accession numbers for taxa included in this study. The (x) indicates sequences generated in this study that will be deposited to GenBank before the manuscript submission. A dash (–) indicates that the DNA region was not sequenced. Taxon Locality Voucher nrITS nrETS waxy atpF-atpH trnS-trnG Microlicieae Chaetostoma armatum (Spreng.) Cogn. Minas Gerais R. Romero et al. 8671 (HUFU) x x x x x Chaetostoma cupressinum (D.Don) Kosch. & A.B. Martins Minas Gerais F. Almeda et al. 8367 (CAS) AY553747* – – – – Chaetostoma flavum Kosch. & A.B. Martins Goiás J.C. Fontelas et al. 137 (HUFU) x x x x x Chaetostoma scoparium Cogn. Goiás J.C. Fontelas et al. 130 (HUFU) x x x x x Chaetostoma stenocladon (Naudin) Kosch. & A.B. Martins Goiás A.F.A. Versiane et al. 490 (HUFU) x x x x – Lavoiseira adamantium Barreto ex Pedersoli Minas Gerais R. Romero et al. 8727 (HUFU) x x x x x Lavoiseira alba Mart. & Schrank ex DC. Minas Gerais A.F.A. Versiane et al. 111 (HUFU) x x x x x Lavoisiera bellineloi A.B. Martins & Almeda Minas Gerais R. Romero et al. 8843 (HUFU) x x x x x Lavoisiera caryophyllea Naudin Minas Gerais R. Romero et al. 8602 (HUFU) x x x x x Lavoiseira chamaepitys A. St.-Hil. ex Naudin Minas Gerais R. Romero et al. 8712 (HUFU) x x x x x Lavoiseira confertiflora Rich. ex Naudin Minas Gerais A.F.A. Versiane et al. 666 (UEC) x x x x x Lavoisiera cordata Cogn. Minas Gerais R. Romero et al. 8609 (HUFU) x x x x – Lavoisiera crassifolia Mart. & Schrank ex DC. Minas Gerais F. Almeda et al. 8442 (CAS) AY553727* – – – – Lavoiseira glandulifera Naudin Minas Gerais R. Romero et al. 8598 (HUFU) x x x x x Lavoisiera humilis Naudin Minas Gerais R. Romero et al. 8724 (HUFU) x – x – – Lavoisiera imbricata (Thunb.) DC. Minas Gerais R. Romero et al. 8604 (HUFU) x x x x x Lavoiseira macrocarpa Naudin Minas Gerais F. Almeda et al. 8545 (CAS) AY553733* – – – – Lavoisiera mucorifera Mart. & Schrank ex DC. Minas Gerais A.F.A. Versiane et al. 372 (HUFU) x x x x – Lavoisiera pulcherrima Mart. & Schrank ex DC. Minas Gerais M. Castro et al. 284 (HUFU) x x x x x Lavoisiera sampaioana Mell. Barr. Minas Gerais R. Romero et al. 8605 (HUFU) x x x x x Lavoiseira scaberula Naudin Minas Gerais R. Romero et al. 8667 (HUFU) x x x x – Lavoisiera subulata Triana Minas Gerais F. Almeda et al. 8550 (CAS) AY553732* – – – – Lavoisiera sp. nov. Minas Gerais M. Castro et al. 285 (HUFU) x x x x x

73

Taxon Locality Voucher nrITS nrETS waxy atpF-atpH trnS-trnG Microlicia acuminata Naudin Minas Gerais A.F.A. Versiane et al. 186 (HUFU) x x – x – Microlicia agrestis Cogn. Minas Gerais A.F.A. Versiane et al. 395 (HUFU) x x x x – Microlicia amblysepala Ule Bahia F. Almeda et al. 8308 (CAS) AY553738* – – – – Microlicia amplexicaulis Cogn. Minas Gerais D. Marques et al. 271 (HUFU) x x x x x Microlicia avicularis (Naudin) Mart. Minas Gerais R. Romero et al. 8635 (HUFU) x x x x x Microlicia canastrensis Naudin Minas Gerais A.F.A. Versiane et al. 62 (HUFU) x x x x – Microlicia candolleana R. Romero & Versiane Minas Gerais R. Romero et al. 8614 (HUFU) x x x x x Microlicia chrysoglandulosa R. Romero et al. Goiás J.C. Fontelas et al. 131 (HUFU) x x – – x Microlicia ciliatoglandulosa R. Romero Minas Gerais R. Romero et al. 8853 (HUFU) x x x x x Microlicia cogniauxiana R. Romero Minas Gerais R. Romero et al. 8650 (HUFU) x x x x x Microlicia confertiflora Naudin Minas Gerais R. Romero et al. 8593 (HUFU) x x – x x Microlicia consimilis Wurdack Goiás A.F.A. Versiane et al. 169 (HUFU) x x x – x Microlicia cordata (Spreng.) Cham. Minas Gerais A.F.A. Versiane et al. 27 (HUFU) x x x x x Microlicia crassa R. Romero Minas Gerais R. Romero et al. 8700 (HUFU) x x – x – Microlicia crebopunctata Pilg. Goiás J.C. Fontelas et al. 136 (HUFU) x x x x x Microlicia cryptandra Naudin Goiás R.A. Pacheco et al. 852 (HUFU) x x x – – Microlicia damazioi Brade Minas Gerais R. Romero et al. 8625 (HUFU) x x x x x Microlicia decipens Naudin Minas Gerais A.F.A. Versiane et al. 378 (HUFU) x x x x x Microlicia edmundoi Brade Minas Gerais A.F.A. Versiane et al. 365 (HUFU) x x x x x Microlicia elegans Naudin Minas Gerais R. Romero et al. 8616 (HUFU) x x – x x Microlicia ericoides D.Don Minas Gerais R. Romero et al. 8930 (HUFU) x x x x x Microlicia euphorbioides Mart. ex Naudin Minas Gerais A.F.A. Versiane et al. 213 (HUFU) x x x x x Microlicia fasciculata Mart. Minas Gerais A.F.A. Versiane et al. 247 (HUFU) x x – x x Microlicia flava R. Romero Minas Gerais J.N. Nakajima et al. 4445 (HUFU) x – x x – Microlicia giuliettiana A.B. Martins & Almeda Bahia R. Goldenberg 1830 (NYBG) – x x – – Microlicia graveolens DC. Minas Gerais R. Romero et al. 8594 (HUFU) x x x x x Microlicia hatschbachii Wurdack Bahia J. Coelho 73 (UPCB) x x x x x Microlicia helvola (Spreng.) Triana Minas Gerais A.F.A. Versiane et al. 30 (HUFU) x x – x x Microlicia hirticalyx R. Romero & Woodgyer Minas Gerais R. Romero et al. 8622 (HUFU) x x x x x Microlicia insignis Schltdl. Minas Gerais J.C. Fontelas et al. 124 (HUFU) x x x x x

74

Taxon Locality Voucher nrITS nrETS waxy atpF-atpH trnS-trnG Microlicia isophylla DC. Minas Gerais A.F.A. Versiane et al. 98 (HUFU) x x – x x Microlicia longicalycina R. Romero Minas Gerais R. Romero et al. 8729 (HUFU) x x – x x Microlicia longipedicelata (Cogn.) Almeda & A.B. Martins Minas Gerais J.C. Fontelas et al. 117 (HUFU) x x x x x Microlicia longirostrata R. Romero, J.C. Fontelas & Versiane Minas Gerais J.C. Fontelas et al. 142 (HUFU) x x x x x Microlicia macrophylla Naudin Minas Gerais A.F.A. Versiane et al. 342 (HUFU) x x x x x Microlicia maculata R. Romero Minas Gerais R. Romero et al. 8665 (HUFU) x x x x – Microlicia maximowicziana Cogniaux Minas Gerais R. Romero et al. 8839 (HUFU) x x x x x Microlicia minima Markgr. Bahia F. Hurbath 694 (UPCB) x x – x – Microlicia minutiflora Cogn. Minas Gerais R. Romero et al. 8617 (HUFU) x x – – x Microlicia mucugensis (Wurdack) Almeda & A.B. Martins Bahia R. Goldenberg 1810 (NYBG) x x – – – Microlicia multicaulis Mart. ex Naudin Minas Gerais R. Romero et al. 8669 (HUFU) x x x x x Microlicia naudiniana R. Romero Minas Gerais A.F.A. Versiane et al. 379 (HUFU) x x x x x Microlicia nervosa R. Romero Minas Gerais A.F.A. Versiane et al. 360 (HUFU) x x x x x Microlicia noblickii (Wurdack) Almeda & A.B. Martins Bahia R. Goldenberg 1832 (NYBG) x x – – – Microlicia obtusifolia Cogn. ex R. Romero Minas Gerais R. Romero et al. 8981 (HUFU) x x x x – Microlicia ordinata (Wurdack) Almeda & A.B. Martins Goiás J.C. Fontelas et al. 127 (HUFU) x x x x x Microlicia pabstii Brade Minas Gerais A.F.A. Versiane & K.S. Rezende 355 (HUFU) x x x x – Microlicia parvula (Markgr.) Kosch. & A.B. Martins Bahia F. Hurbarth 688 (UPCB) x – – – – Microlicia passerina Naudin Minas Gerais R. Romero et al. 8852 (HUFU) x x x x x Microlicia petiolulata Cogn. ex R. Romero & Woodgyer Minas Gerais A.F.A. Versiane et al. 665 (UEC) x x x x x Microlicia pilosissima Cogn. Minas Gerais R. Romero et al. 8643 (HUFU) x x x x x Microlicia pinheiroi Wurdack Bahia R. Goldenberg 1818 (NYBG) x x x x – Microlicia polystemma Naudin Minas Gerais A.F.A. Versiane et al. 56 (HUFU) x x x x x Microlicia psammophila Wurdack Goiás J.C. Fontelas et al. 133 (HUFU) x x x x x Microlicia pusilla Cogn. Minas Gerais A.F.A. Versiane et al. 351 (HUFU) x x x x x Microlicia ramosa Pilger Goiás J.N. Nakajima et al. 5049 (HUFU) x x x x x Microlicia regeliana Cogn. Minas Gerais R. Romero et al. 8637 (HUFU) x x x x – Microlicia riedeliana Cogn. Minas Gerais R. Romero et al. 8647 (HUFU) x x x x – Microlicia rugosa R. Romero & Versiane Minas Gerais R. Romero et al. 8596 (HUFU) – x x x x Microlicia serpyllifolia D.Don Minas Gerais R. Romero et al. 8611 (HUFU) x x x x –

75

Taxon Locality Voucher nrITS nrETS waxy atpF-atpH trnS-trnG Microlicia serrulata Cham. Minas Gerais R. Romero et al. 8634 (HUFU) x x x x – Microlicia setosa (Spreng.) DC. Minas Gerais A.F.A. Versiane et al. 661 (UEC) x x x x x Microlicia sphagnicola Gleason Peru F.A. Michelangeli 2954 (NYBG) x x x x – Microlicia stenodonoides D.O. Diniz-Neres & M.J. Silva Goiás Diniz-Neres & R.N. Ribeiro 795 (UFG) x x x x x Microlicia stricta Cogn. Minas Gerais A.F.A. Versiane et al. 371 (HUFU) x x x x – Microlicia tenuifolia R. Romero Minas Gerais J.C. Fontelas et al. 111 (HUFU) x x x x – Microlicia tetrasticha Cogn. Minas Gerais R. Romero et al. 8901 (HUFU) x x – x – Microlicia tomentella Naudin Minas Gerais R. Romero et al. 8898 (HUFU) x x x x – Microlicia trembleyeaformis Naudin Minas Gerais R. Romero et al. 7550 (HUFU) x – – – – Microlicia viminalis (DC.) Triana Minas Gerais J.C. Fontelas et al. 129 (HUFU) x x x x x Microlicia viscida R. Romero & Versiane Minas Gerais R. Romero et al. 8865 (HUFU) x x x x x Microlicia sp. nov. 01 Minas Gerais R. Romero et al. 8696 (HUFU) x x x x – Microlicia sp. nov. 02 Minas Gerais R. Romero et al. 8880 (HUFU) x x x x x Microlicia sp. nov. 03 Minas Gerais A.F.A. Versiane et al. 20 (HUFU) x x – x x Microlicia sp. nov. 04 Minas Gerais R. Romero & A.F.A Versiane 8919 (HUFU) x x – x x Microlicia sp. nov. 05 Minas Gerais R. Romero et al. 8975 (HUFU) x – x – – Microlicia sp. nov. 06 Minas Gerais R. Romero et al. 8891 (HUFU) x x – x x Microlicia sp. nov. 07 Minas Gerais R. Romero et al. 8589 (HUFU) x x – x x Microlicia sp. nov. 08 Minas Gerais A.F.A. Versiane et al. 344 (HUFU) x x – x x Microlicia sp. nov. 09 Minas Gerais R. Romero et al. 8842 (HUFU) x x – x x Poteranthera pusilla Bong. Minas Gerais M.J.R. Rocha et al. 940 (BHCB) KU845164* KU845167* – x – Rhynchanthera bracteata Triana Colômbia F. Zenteno 8801 (NY) JQ730137* KF462880* – KX826831* KX826831* Rhynchanthera brachyrryncha Chamisso São Paulo A.F.A. Versiane & A.C.D. Castello 671 (UEC) x x – – – Rhynchanthera grandiflora (Aubl.) DC. Minas Gerais A.F.A. Versiane et al. 03 (HUFU) x KF462881* – x x Rhynchanthera serrulata (Rich.) DC. Venezuela F.A. Michelangeli et al. 828 (NYBG) AY460435* – – – – Stenodon suberosus Naudin Goiás J.C. Fontelas et al. 141 (HUFU) x x x x x Trembleya acuminata R.B. Pacifico & Fidanza Minas Gerais R. Romero et al. 8926 (HUFU) x x x – x Trembleya chamissoana Naudin Minas Gerais R. Romero et al. 8627 (HUFU) x x x x x Trembleya hastchbachii Wurdack & E. Martins Minas Gerais A.V. Scatigna & J.C. Galvão 376 (UEC) x x x – – Trembleya inversa K. Fidanza, A.B. Martins & Almeda Minas Gerais R. Romero et al. 8845 (HUFU) x x x x –

76

Taxon Locality Voucher nrITS nrETS waxy atpF-atpH trnS-trnG Trembleya laniflora (D.Don.) Cogn. Minas Gerais R. Romero et al. 8615 (HUFU) x x x x x Trembleya parviflora (D.Don) Cogn. Minas Gerais A.F.A. Versiane et al. 240 (HUFU) x x x x x Trembleya pentagona Naudin Minas Gerais M. Castro et al. 282 (HUFU) x x x x x Trembleya phlogiformis DC. Minas Gerais A.F.A. Versiane et al. 14 (HUFU) – x x x x Trembleya pradosiana Netto Minas Gerais R. Romero et al. 8887 (HUFU) x x Trembleya purpurascens Fidanza, A.B. Martins & Almeda Minas Gerais A.F.A. Versiane et al. 657 (UEC) x x x x x Trembleya pythioides Cham. Minas Gerais M. Castro et al. 283 (HUFU) x x x x x Trembleya rosmarinoides DC. Minas Greais A.F.A. Versiane & A.C.D. Castello 677 (UEC) x x x x x Trembleya thomazii R.B. Pacifico & K. Fidanza Minas Gerais R. Romero et al. 8894 (HUFU) x x x x x Trembleya tridentata Naudin Minas Gerais M. Castro et al. 281 (HUFU) x x x x x Trembleya sp. nov. Minas Gerais R. Romero et al. 8630 (HUFU) x x x x x Marcetieae Marcetia formosa Wurdack Bahia M.J.R. Rocha et al. 345 (BHCB) KU501090* KU501035* - KU501144* KU500965* Marcetia grandiflora Wurdack Bahia M.J.R. Rocha et al. 319 (BHCB) KU501091* KU501036* - KU501145* KU500966* Siphanthera cordata Pohl ex DC. Minas Gerais M.J.R. Rocha et al. 964 (BHCB) KU501102* KU501048* - - KU500977* Siphanthera gracillima (Naudin) Wurdack Minas Gerais M.J.R. Rocha et al. 941 (BHCB) KU501101* KU501047* - KU501154* KU500976* Rhexieae Rhexia aristosa Britton Estados Unidos R.F.C. Naczi 12065 (NY) JQ730134* KF462878* - - KU500986* Rhexia virginica L. Estados Unidos F.A. Michelangeli 1448 (NY) JQ730136* KF462879* - KX826830* KX826830*

77

Table S2. Morphological characters and character states scored in this study as primary homology hypotheses. A. Habit: herbaceous (0), woody (1); B. Leaf secondary veins: absent (0), present (1); (C) Flower arrangement: solitary (0); dichasium (1); paired (2); glomerule (3); (D) Petal number: less than five (0); five (1); more than five (2); (E) Bristles on the hypanthium apex: absent (0); present (1); (F) Stamens dimorphism: isomorphic (0); subisomorphic (1); dimorphic (2); (G) Stamens number: androecium of one whorl (0); androecium of two whorls (1); (H) Anthers, number of sporangia: tetrasporangiate (0); polysporangiate (1); (I) Ovary locule number: two (0); three (1); four (2); five (3); more than five (4); (J) Ovary position: (0) superior; (1) partly inferior; (2) inferior; (K) Fruit dehiscence: basipetal (0); acropetal (1); (L) Columella: persistent (0); caducous (1).

Character Species A B C D E F G H I J K L Chaetostoma armatum (Spreng.) Cogn. 1 0 0 1 1 2 1 0 1 0 0 1 Chaetostoma cupressinum (D.Don) Kosch. & A.B. Martins 1 0 0 1 1 2 1 0 2,3 0 0 1 Chaetostoma flavum Kosch. & A.B. Martins 1 0 0 1 1 2 1 0 1 0 0 1 Chaetostoma scoparium Cogn. 1 0 0 1 1 2 1 0 1 0 0 1 Chaetostoma stenocladon (Naudin) Kosch. & A.B. Martins 1 0 0 1 1 2 1 0 3 0 0 1 Lavoiseira adamantium Barreto ex Pedersoli 1 0 0 2 0 2 1 0 4 1 1 0 Lavoiseira alba Mart. & Schrank ex DC. 1 0,1 0,1,2 1 0 2 1 0 3 1 1 0 Lavoisiera bellineloi A.B. Martins & Almeda 1 0 0 2 0 2 1 0 4 1 1 0 Lavoisiera caryophyllea Naudin 1 0 0,1 1 0 2 1 0 2,3 1 0 0 Lavoiseira chamaepitys A. St.-Hil. ex Naudin 1 0 0 1,2 0 2 1 0 0,2,3 1 1 0 Lavoiseira confertiflora Rich. ex Naudin 1 0 0,1,2 1 0 2 0 0 3 1 1 0 Lavoisiera cordata Cogn. 1 0 1 2 0 2 1 0 3,4 1 0 0 Lavoisiera crassifolia Mart. & Schrank ex DC. 1 0 0,1 1,2 0 2 1 0 3,4 1 1 0 Lavoiseira glandulifera Naudin 1 0 0,2 1 0 2 1 0 3 1,2 1 0 Lavoisiera humilis Naudin 1 0 0 1,2 0 2 1 0 2 1 1 0 Lavoisiera imbricata (Thunb.) DC. 1 0 0 2 0 2 1 0 4 1,2 1 0

78

Lavoiseira macrocarpa Naudin 1 0 0,1,2 2 0 2 1 0 4 1 0 0 Lavoisiera mucorifera Mart. & Schrank ex DC. 1 0 0 1,2 0 2 1 0 4 1 1 0 Lavoisiera pulcherrima Mart. & Schrank ex DC. 1 0 0,2 2 0 2 1 0 4 1 1 0 Lavoisiera sampaioana Mell. Barr. 1 0 0 2 0 2 1 0 4 1 1 0 Lavoiseira scaberula Naudin 1 0 0 2 0 2 1 0 0,2 1 1 0 Lavoisiera subulata Triana 1 0 0 1,2 0 2 1 0 4 1 1 0 Lavoisiera sp. nov. 1 0 0 1 0 2 1 0 1 1 1 0 Microlicia acuminata Naudin 1 0,1 0 1 0 1,2 1 0 1 0 0 1 Microlicia agrestis Cogn. 1 0,1 0 1 0 2 1 0 1 0 0 1 Microlicia amblysepala Ule 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia amplexicaulis Cogn. 1 0 0,1 1 0 2 1 0 1 0 0 1 Microlicia avicularis (Naudin) Mart. 1 0 0,1 1 0 2 1 0 1 0 0 1 Microlicia canastrensis Naudin 1 0,1 0 1 0 2 1 0 1 0 0 1 Microlicia candolleana R. Romero & Versiane 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia chrysoglandulosa R. Romero et al. 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia ciliatoglandulosa R. Romero 1 0 0,1 1 0 2 1 0 1,2 0 0 1 Microlicia cogniauxiana R. Romero 1 0 1 1 0 2 1 0 1 0 0 1 Microlicia confertiflora Naudin 1 0 0 1 0 1 1 0 1 0 0 1 Microlicia consimilis Wurdack 1 0 0,1 1 0 2 1 0 1 0 0 1 Microlicia cordata (Spreng.) Cham. 1 0 1,2 1 0 2 1 0 1 0 0 1 Microlicia crassa R. Romero 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia crebopunctata Pilg. 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia cryptandra Naudin 1 0 0 1 0 1 1 0 1 0 0 1 Microlicia damazioi Brade 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia decipens Naudin 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia edmundoi Brade 1 1 0 1 0 2 1 0 1 0 0 1

79

Microlicia elegans Naudin 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia ericoides D.Don 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia euphorbioides Mart. ex Naudin 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia fasciculata Mart. 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia flava R. Romero 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia giuliettiana A.B. Martins & Almeda 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia graveolens DC. 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia hatschbachii Wurdack 1 0 0,1,2 1 0 1 1 1 1 0 0 1 Microlicia helvola (Spreng.) Triana 1 0,1 1 1 0 2 1 1 1 0 0 1 Microlicia hirticalyx R. Romero & Woodgyer 1 0 0,1 1 0 2 1 0 1 0 0 1 Microlicia insignis Schltdl. 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia isophylla DC. 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia longicalycina R. Romero 1 0 0,2 1 0 2 1 0 1 0 0 1 Microlicia longipedicelata (Cogn.) Almeda & A.B. Martins 1 0 0 2 0 2 1 0 4 0 0 0,1 Microlicia longirostrata R. Romero, J.C. Fontelas & Versiane 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia macrophylla Naudin 1 0&1 0 1 0 2 1 0 1 0 0 1 Microlicia maculata R. Romero 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia maximowicziana Cogniaux 1 0 0 1 0 2 1 1 1,2 0 0 1 Microlicia minima Markgr. 1 0 0,1,2 1 0 1 1 1 1 0 0 1 Microlicia minutiflora Cogn. 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia mucugensis (Wurdack) Almeda & A.B. Martins 1 0 1 1 0 2 1 1 1 0 0 1 Microlicia multicaulis Mart. ex Naudin 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia naudiniana R. Romero 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia nervosa R. Romero 1 0 0,1 1 0 2 1 0 1 0 0 1 Microlicia noblickii (Wurdack) Almeda & A.B. Martins 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia obtusifolia Cogn. ex R. Romero 1 0 0 1 0 2 1 0 1 0 0 1

80

Microlicia ordinata (Wurdack) Almeda & A.B. Martins 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia pabstii Brade 1 1 0,1,2 1 0 2 1 0 1 0 0 1 Microlicia parvula (Markgr.) Kosch. & A.B. Martins 1 0 0,1 1 0 1 1 1 1 0 0 1 Microlicia passerina Naudin 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia petiolulata Cogn. ex R. Romero & Woodgyer 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia pilosissima Cogn. 1 0 0,1 1 0 2 1 0 1 0 0 1 Microlicia pinheiroi Wurdack 1 0 0,1 1 0 2 1 1 1 0 0 1 Microlicia polystemma Naudin 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia psammophila Wurdack 1 0 0 1 0 1 1 0 1 0 0 1 Microlicia pusilla Cogn. 1 0 0 1 0 2 1 0 1,2 0 0 1 Microlicia ramosa Pilger 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia regeliana Cogn. 1 0 1 1 0 2 1 1 1 0 0 1 Microlicia riedeliana Cogn. 1 0 1 1 0 2 1 0 1 0 0 1 Microlicia rugosa R. Romero & Versiane 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia serpyllifolia D.Don 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia serrulata Cham. 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia setosa (Spreng.) DC. 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia sphagnicola Gleason 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia stenodonoides D.O. Diniz-Neres & M.J. Silva 1 0 0 0,1,2 0 2 1 1 2 0 0 1 Microlicia stricta Cogn. 1 0 0 1 0 2 1 1 1 0 0 1 Microlicia tenuifolia R. Romero 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia tetrasticha Cogn. 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia tomentella Naudin 1 0 0,1 1 0 2 1 0 1 0 0 1 Microlicia trembleyeaformis Naudin 1 1 0 1 0 2 1 0 1 0 0 1 Microlicia viminalis (DC.) Triana 1 0 1 1 0 2 1 0 1 0 0 1 Microlicia viscida R. Romero & Versiane 1 0 0 1 0 2 1 0 1 0 0 1

81

Microlicia sp. nov. 01 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia sp. nov. 02 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia sp. nov. 03 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia sp. nov. 04 1 0 0 1 0 2 1 0 2,3 0 0 1 Microlicia sp. nov. 05 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia sp. nov. 06 1 0 0 1 0 2 1 0 1 0 0 1 Microlicia/Trembleya sp. 01 1 0 1 1,2 0 2 1 0 1 0 0 1 Microlicia/Trembleya sp. 02 1 0 3 1 0 2 1 0 1 0 0 1 Microlicia/Trembleya sp. 03 1 0 3 1 0 2 1 0 1 0 0 1 Poteranthera pusilla Bong. 0 0 0 1 0 0 0,1 0 1 0 0 1 Rhynchanthera braceata Triana 1 1 1 1 0 2 0 0 1,2 0 0 1 Rhynchanthera brachyrryncha Cham. 1 1 1 1 0 2 0 0 1,2 0 0 1 Rhynchanthera grandiflora (Aubl.) DC. 1 1 1,2 1 0 2 0 0 1,2,3 0 0 1 Rhynchanthera serrulata (Rich.) DC. 1 0 2 1 0 1 0 0 1,2 0 0 1 Stenodon suberosus Naudin 1 1 0,1 1,2 0 1 1 0 2,3,4 0 0 1 Trembleya acuminata R.B. Pacifico & Fidanza 1 1 1 2 0 2 1 0 2 0 0 1 Trembleya chamissoana Naudin 1 1 2 1 0 2 1 0 3 0 0 1 Trembleya hastchbachii Wurdack & E. Martins 1 1 1 1 0 2 1 0 2 0 0 1 Trembleya inversa K. Fidanza, A.B. Martins & Almeda 1 0 1 1,2 0 2 1 0 2 0 0 1 Trembleya laniflora (D.Don.) Cogn. 1 1 2 1 0 2 1 0 3 0 0 1 Trembleya parviflora (D.Don) Cogn. 1 1 1 1 0 2 1 0 3 0 0 1 Trembleya pentagona Naudin 1 1 2 1 0 2 1 0 3 0 0 1 Trembleya phlogiformis DC. 1 1 1,2 1 0 2 1 0 1,2,3 0 0 1 Trembleya purpurascens Fidanza, A.B. Martins & Almeda 1 0 1 1 0 2 1 0 1 0 0 1 Trembleya pythioides Cham. 1 1 2 1 0 2 1 0 2,3 0 0 1 Trembleya pradosiana Netto 1 1 1,2 1 0 2 1 0 3 0 0 1

82

Trembleya rosmarinoides DC. 1 1 2 1 0 2 1 0 2,3 0 0 1 Trembleya thomazii R.B. Pacifico & K. Fidanza 1 0 1 1 0 2 1 0 1 0 0 1 Trembleya tridentata Naudin 1 1 2 1 0 2 1 0 3 0 0 1 Trembleya sp. nov. 1 1 2 1 0 2 1 0 3 0 0 1 Rhexia aristosa Britton 1 0 1,2 0 0 1 1 0 2 2 0 1 Rhexia virginica L 1 1 1.2 0 0 1 1 0 2 2 0 1 Marcetia formosa Wurdack 1 1 2 0 0 0 1 0 2 0 0 1 Marcetia grandiflora Wurdack 1 0 2 0 0 0 1 0 2 0 0 1 Siphanthera cordata Pohl ex DC. 0 1 1 0 0 0 1 0 0 0 0 1 Siphanthera gracillima (Naudin) Wurdack 0 1 1 0 0 0 0 0 0 0 0 1

83

Figure S1. Majority consensus tree from Bayesian inference analysis of nuclear sequences. Numbers are posterior probabilities (PP) from Bayesian inference analysis (only PP ≥ 0.90). The type species of each genus is marked with an asterisk (*).

84

Figure S2. Majority consensus tree from Bayesian inference analysis of plastidial sequences. Numbers are posterior probabilities (PP) from Bayesian inference analysis (only PP ≥ 0.90). The type species of each genus is marked with an asterisk (*).

85

Table S5. All changes between all morphological character states.

Changes between states Character 0→1 0→2 0→3 0→4 1→0 1→2 1→3 1→4 2→0 2→1 2→3 2→4 3→0 3→1 3→2 3→4 4→0 4→1 4→2 4→3 Habit 0.238 ─ ─ ─ 2.001 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Leaf secondary veins 11.908 ─ ─ ─ 4.893 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Flower arrangement 14.83 1.647 0.995 ─ 2.42 2.285 0.104 ─ 0.479 1.917 0.013 ─ 0.034 0.034 0.052 ─ ─ ─ ─ ─ Petals number 0.955 0.142 ─ ─ 0.116 4.559 ─ ─ 0.059 3.355 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Bristles on the hypanthium apex 1.034 ─ ─ ─ 0.032 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Stamens dimorphism 0.594 0.78 ─ ─ 0.355 0.972 ─ 0.905 9.08 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Stamen number 1.205 ─ ─ ─ 3.735 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Anthers, number of sporangia 12.428 ─ ─ ─ 0.51 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Ovary locule number 0.876 1.077 0.605 0.605 0.653 5.049 4.874 2.912 0.432 0.927 0.703 0.717 0.483 0.937 1.274 0.776 1.143 1.553 2.804 3.798 Ovary position 1.062 0.983 ─ ─ 0.035 0.046 ─ ─ 0.073 0.007 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Fruit dehiscence 1.122 ─ ─ ─ 3.268 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ Columella 0.03 ─ ─ ─ 1.027 ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─ ─

86

Figure S3. Stochastic mapping of selected characters in the Microlicieae tribe. A, Bristles on the hypanthium apex; B, Stamens shape; C, Stamens number; D, Anthers, number of sporangia.

87

Figure S4. Stochastic mapping of selected characters in the Microlicieae tribe. A, Ovary position; B,

Fruit dehiscence; C, Columella.

88

CAPÍTULO II

Taxonomy and Nomenclatural notes in Microlicieae

(Melastomataceae): combinations and new names in

Microlicia D.Don2

Ana Flávia Alves Versiane, Renato Goldenberg and Rosana Romero

2 Trabalho será submetido à revista Phytotaxa

89

Abstract – Currently, Microlicia D.Don has 161 species with the majority occurring in Brazil and few are found in Bolivia, Colombia, Guiana, Peru, and Venezuela. A recent molecular phylogenetic study showed the genus as a paraphyletic with Chaetostoma DC.,

Lavoisiera DC., Stenodon Naudin, and Trembleya DC. nested inside it. Therefore, the authors suggested that to recognize a monophyletic Microlicia, these five genera should be merged in a broadly genus. Thus, Microlicia is here circumscribed to include Chaetostoma, Lavoisiera,

Stenodon, and Trembleya; and the nomenclatural alterations are also presented. A total of 67 taxa are transferred to Microlicia, from which 48 are new combinations and 19 are new names, and five names are re-established. Microlicia consists now by 236 taxa.

Keywords: campos rupestres, Chaetostoma, Lavoisiera, phylogeny, Stenodon, Trembleya

Resumo – Microlicia D.Don atualmente inclui 161 espécies com a maioria ocorrendo no Brasil e poucas sendo encontradas na Bolívia, Colômbia, Guiana, Peru e Venezuela. Uma recente análise molecular mostrou o gênero como parafilético com Chaetostoma DC.,

Lavoisiera DC., Stenodon Naudin e Trembleya DC. agrupados dentro dele. Portanto, os autores sugeriram que para reconhecer Microlicia como monofilético, estes cinco gêneros deveriam ser incluídos em um amplo gênero. Desta forma, Microlicia é aqui circunscrito para incluir Chaetostoma, Lavoisiera, Stenodon e Trembleya; e as alterações nomenclaturais são também apresentadas. Um total de 67 táxons são transferidos para Microlicia, dos quais 48 são novas combinações, 19 são novos nomes e cinco nomes foram reestabelicidos. Com isso,

Microlicia consiste agora de 236 táxons.

Palavras-chave: campos rupestres, Chaetostoma, filogenia, Lavoisiera, Stenodon, Trembleya

90

Introduction Microlicieae Naudin (1849: 203) comprises a cohesive tribe in Melastomataceae based both on morphological and molecular evidence (Clausing & Renner 2001; Fritsch et al. 2004;

Versiane et al., chapter 1). It has 256 species with a near-endemic distribution in South

America, from which 90% occur in cerrado and campos rupestres (Fritsch et al. 2004;

Versiane et al. [first chapter]).

Until now, the tribe was composed of seven genera: Chaetostoma De Candolle (1828:

112), Lavoisiera De Candolle (1828: 102), Microlicia D.Don (1823: 301), Poteranthera

Bongard (1838: 137), Rhynchanthera De Candolle (1828: 106), Stenodon Naudin (1844: 146) and Trembleya De Candolle (1828: 125) (Rocha et al. 2016). All these share stamens with prolonged pedoconnective, rostrate anthers, capsular fruits and reniform to ellipsoid seeds with foveolate or lacunate-reticulated testa (Clausing & Renner 2001; Almeda & Martins

2001; Fritsch et al. 2004; Rocha et al. 2016; Martins & Almeda 2017; Versiane et al. [first chapter]).

Microlicia is the largest genus in the tribe with 161 species (Flora do Brasil 2020;

Mendoza-Cifuentes et al. 2019; Romero et al. 2019), but this number is probably higher since several known new species are still to be published. The recognition of novelties in Microlicia has been made with caution, because the morphological characters used to define species are generally very slight, hindering delimitation from closely related species (Romero et al.

2014). The genus has been circumscribed based on the capsule with longitudinal dehiscence from the apex to the base, a superior ovary with three locules (rarely four locules), and flowers typically with five petals (rarely six or eight petals) (Romero et al. 2015).

The uncertainty concerning the generic delimitation in Microlicieae has been an object of discussion through the years (e.g. Koschniztke & Martins 1999, 2007; Almeda & Martins

2001; Versiane et al. [first chapter]) to minimize and clarify the distinction among these genera. Thereby, some authors have been proposing some nomenclatural changes based on

91

morphological characters, where species of Microlicia have been transferred to Chaetostoma

(Koschniztke & Martins 1999) and Lavoisiera (Almeda & Martins 2001); Chaetostoma to

Microlicia (Koschniztke & Martins 2007), Lavoisiera to Microlicia and Trembleya (Almeda

& Martins 2001). Taxonomic revisions have been done for Microlicia sect. Chaetostomoides

(Romero 2003), Chaetostoma (Koschniztke & Martins 2007), Lavoisiera (Martins & Almeda

2017), and Trembleya (Martins 1991) or are in preparation (Trembleya: Pacífico et al. in prep.; Stenodon: Alves & Romero in prep.).

The last recent molecular study focusing on the delimitation of Microlicia was undertaken by Versiane et al. (first chapter), where they increased the number of samples of

Microlicieae and included others different molecular markers, regarding previous studies (i.e.

Fritsch et al. 2004) and utilized ancestral state reconstruction analyses in order to assess an appropriate strategy for the classification of Microlicia. The results revealed that the genus

Microlicia as traditionally defined based on morphology (see Cogniaux 1883; 189; Renner

1993; Almeda & Martins 2001) is not monophyletic, with Trembleya, Stenodon, Lavoisiera and Chaetostoma nested inside it (Fig. 1). Therefore, the authors suggested that, to recognize a monophyletic Microlicia, these five genera should be merged in a broadly circumscribed genus.

Microlicia has priority of publication over the other four genera, according to Principle III of the International Code of Nomenclature (Turland et al. 2018). Therefore, we synonymized

Chaetostoma, Lavoisiera, Stenodon, and Trembleya under Microlicia and all taxonomic and nomenclatural changes necessary to recognize the genus as monophyletic are provided.

92

Figure 1. Phylogenetic relationships within the Microlicieae tribe, based on majority consensus tree from Bayesian Inference of nuclear and plastidial concatenate sequences, showing Chaetostoma, Lavoisiera, Stenodon and Trembleya nested within Microlicia (Adaptated from Versiane et al. [first chapter]).

Material and Methods

The procedures we used here to choose whether a species should receive a new combination, a new name or a synonymization under another epithet followed Michelangeli et al. (2018). Data on the names treated here were based on the taxonomic revisions for each genus (Cogniaux 1891; Martins 1991, Koschniztke & Martins 2007; Martins & Almeda

2017), and on the protologues of the species (Martins 1995; Fidanza et al. 2013; Pacífico &

Fidanza 2015; Pacífico et al. 2019). Whenever new names had to be created, we tried to keep

93

the same rationale used by the original author in the previous epithet (e.g. referred to the same locality, honored person, descriptive characters).

Databases International Plant Names Index (http://www.ipni.org), TROPICOS

(http://www.tropicos.org), JSTOR Global Plants (https://plants.jstor.org/), and Flora do Brasil

2020 were consulted to include all species of Lavoisiera, Trembleya, Stenodon and

Chaetostoma in Microlicia. We provide information only in the new names to justify the new epithets. Author names and the abbreviated publication names follow IPNI.

Results

A total of 67 taxa in Lavoisiera, Trembleya, Stenodon, and Chaetostoma were transferred to Microlicia, of these 48 are new combinations and 19 are new names.

Additionaly five names were re-established in Microlicia.

NEW CIRCUMSCRIPTION OF MICROLICIA

Microlicia D.Don (1823: 302). Type:—Microlicia ericoides D.Don.

=Lavoisiera De Candolle (1828: 102), syn. nov. Type:— Lavoisiera imbricata (Thunberg) De

Candolle (1828: 103). [≡ Rhexia imbricada Thunberg (1817: 10)]. syn. nov.

=Chaetostoma De Candolle (1828: 112), syn. nov. Type:— Chaetostoma armatum (Sprengel)

Cogniaux in Martius, Eichler & Urban (1883: 31), [≡ Rhexia armata Sprengel (1825: 308)]. syn. nov.

=Trembleya De Candolle (1828: 125), syn. nov. Type:—Trembleya rosmarinoides De

Candolle (1828: 125). syn. nov.

=Stenodon Naudin (1844: 146), syn. nov. Type:— Stenodon suberosus Naudin (1844: 146). syn. nov.

94

Subshrubs, shrubs or small trees, rarely dwarf tree. Much branched, cespitose. Branches variously pubescent or glabrous, quadrangular to terete; older branches usually becoming subterete, defoliating and decorticating basally with age. Leaves sessile, subsessile or petiolate, ascending or horizontal, spreading or imbricate, flat or keeled, membranaceous, chartaceous or coriaceous, entire, serrulate or crenulate, narrowly or callose-thickened or not, variously pubescent or glabrous. Inflorescence with solitary flower, paired, resulting from the reduction of simple or compound dichasia, lax to congested (glomerulate head); bracts present or not, sessile, subsessile or petiolate, variously pubescent or glabrous. Flower sessile or pedicellate; bristles present or not towards hypanthium apex, sepals persistent to early, tardily caducous or not. Petals 5(–9–10), pink, magenta, cream, white or yellow, sometimes with a yellow, white, or greenish patch at the base. Stamens (5−)10−18(−20), diplostemonous, rare haplostemonous, staminodia predominantly absent, rare present (L. confertiflora), commoly dimorphic, subisomorphic or rare isomorphic, anthers tetra or polysporangiate, bicolor or concolor; pedoconnective prolonged below the thecae, ventrally dilated into poorly or well- developed appendages. Ovary superior or ½−¾-inferior or completely inferior, (2−)3−5(−10)- locular, glabrous. Style glabrous, straight or curved at the apex; stigma punctiform. Capsule dry, loculicidal, basipetal or acropetal dehiscence, columella caducous or persistent. Seeds half slightly curved to one side, testa foveolate.

COMBINATIONS OR NEW NAMES IN MICROLICIA

Microlicia acuminifolia Versiane & R. Romero, nom. nov. Trembleya acuminata R.B.

Pacífico & K. Fidanza (2015: 164). Non Microlicia acuminata Naudin (1845: 179), nec

Microlicia acuminata Cogniaux (1891: 55).

Note: The new epithet refers to the acuminate-cuspidate apex of the leaves, typical for this

species.

95

Microlicia adamantium (Mello Barreto ex Pedersoli) Versiane & R. Romero, comb. nov.

Lavoiseira adamantium Mello Barreto ex Pedersoli (1979/1980: 21–22).

Microlicia alba (Martius & Schrank ex DC.) Versiane & R. Romero, comb. nov. Lavoiseira

alba Martius & Schrank ex De Candolle. (1828: 103–104).

Microlicia albiflora (Naudin) Versiane & R. Romero, comb. nov. Chaetostoma pungens var.

albiflorum Naudin (1845: 191). Chaetostoma albiflorum (Naudin) Koschnitzke & A.B.

Martins (1999: 202).

Microlicia altoparaisensis (R. Pacífico, Almeda & Fidanza) Versiane & R. Romero, comb.

nov. Trembleya altoparaisensis R. Pacífico, Almeda & Fidanza (2019: 289).

Microlicia angustifolia (Cogniaux) Versiane & R. Romero, comb. nov. Lavoisiera

angustifolia Cogniaux in Martius, Eichler & Urban (1888: 595).

Microlicia arachnoidea (Almeda & A.B. Martins) Versiane & R. Romero, comb. nov.

Lavoisiera arachnoidea Almeda & A.B. Martins (2017: 51).

Microlicia armata (Sprengel) Versiane & R. Romero, comb. nov. Rhexia armata Sprengel

(1825: 308). Chaetostoma armatum (Spreng.) Cogniaux in Martius, Eichler & Urban

(1883: 31).

Microlicia australis (Naudin) Versiane & R. Romero, comb. nov. Lavoisiera australis A.

Saint-Hilaire ex Naudin (1844: 151).

Lavoisiera pulchella Chamisso, Linnaea (1834: 370). Non Microlicia pulchella Chamisso

(1834:392), nec Microlicia pulchella Martius.

Note: Since the epithet “pulchella” was already in use in Microlicia, we chose to make a

new combination for a name in a heterotypic synonym, L. australis (see Martins & Almeda

2017).

Microlicia botaensis (R.B. Pacífico & K. Fidanza) Versiane & R. Romero, comb. nov.

Trembleya botaensis R.B. Pacífico & K. Fidanza (2015: 167).

96

Microlicia belinelloi (A.B. Martins & Almeda) Versiane & R. Romero, comb. nov.

Lavoisiera belinelloi A.B. Martins & Almeda (2017: 54).

Microlicia calycina (Chamisso) Versiane & R. Romero, comb. nov. Trembleya calycina

Chamisso (1834: 430).

Microlicia caryophyllea (Naudin) Versiane & R. Romero, comb. nov. Lavoisiera

caryophyllea Naudin (1844: 150).

Microlicia cataphracta (De Candolle) Versiane & R. Romero, comb. nov. Lavoisiera

cataphracta Martius & Schrank ex Candolle (1828: 102).

Lavoisiera imbricata (Thunberg) De Candolle (1828:103). Rhexia imbricata Thunberg

(1817:10). Non Microlicia imbricata Chamisso (1834:395).

Note: Since the epithet “imbricata” was already used in Microlicia, we chose to make a

new combination for a name in a heterotypic synonym, L. cataphracta (see Martins &

Almeda 2017).

Microlicia chamissoana (Naudin) Versiane & R. Romero, comb. nov. Trembleya

chamissoana Naudin (1849: 270).

Microlicia congestiflora Versiane & R. Romero, nom. nov. Lavoisiera confertiflora Naudin

(1844: 149). Non Microlicia confertiflora Naudin (1845: 176).

Note: The new epithet refers to the flowers that are crowded in this species.

Microlicia cordifolia Versiane & R. Romero, nom. nov. Lavoisiera cordata Cogniaux in

Martius, Eichler & Urban (1883: 140). Non Microlicia cordata (Sprengel) Chamisso

(1834: 390).

Note: The new epithet refers to the leaf shape in this species.

Microlicia crassifolia (Martius & Schrank ex De Candolle) Versiane & R. Romero, comb.

nov. Lavoiseira crassifolia Martius & Schrank ex De Candolle (1828: 104).

97

Microlicia curtiana Versiane & R. Romero nom. nov. Lavoiseira bradeana Barreto (1936:

70). Non Microlicia bradeana Hoehne (1922: 28).

Note: The new epithet honors Alexander Curt Brade, a German botanist, who worked with

Melastomataceae.

Microlicia daviesiana (Almeda & A.B. Martins) Versiane & R. Romero, comb. nov.

Lavoisiera daviesiana Almeda & A.B. Martins (2017: 87).

Microlicia fastigiata (Naudin) Versiane & R. Romero, comb. nov. Chaetostoma fastigiatum

Naudin (1845: 191).

Microlicia firmula (Martius & Schrank ex De Candolle) Versiane & R. Romero, comb. nov.

Lavoiseira firmula Martius & Schrank ex De Candolle (1828: 104).

Microlicia flavopetala Versiane & R. Romero, nom. nov. Chaetostoma flavum Koschnitzke

& A.B. Martins (1999: 204). Non Microlicia flava R. Romero (2000: 142).

Note: The new epithet refers to the yellow petals of this species.

Microlicia gentianoides (De Candolle) Versiane & R. Romero, comb. nov. Lavoisiera

gentianoides De Candolle (1828: 104).

Microlicia glandulifolia Versiane & R. Romero, nom. nov. Lavoiseira glandulifera Naudin

(1844:149). Non Microlicia glandulifera Naudin (1891: 69).

Note: The new epithet refers to the glandular trichomes on both leaf surfaces in this

species.

Microlicia glaziovii (Cogniaux) Versiane & R. Romero, comb. nov. Chaetostoma glaziovii

Cogniaux in Martius, Eichler & Urban (1883: 30).

Microlicia graomogolensis Versiane & R. Romero, nom. nov. Trembleya hatschbachii

Wurdack & E. Martins (1995: 40). Non Microlicia hastchbachii Wurdack (1981: 149).

Note: This epithet refers to Grão Mogol, northern Minas Gerais, to which this species is

endemic.

98

Microlicia hexapetala (D. Nunes, D.O. Diniz, Koschnitzke. & M.J. Silva) Versiane & R.

Romero, comb. nov. Chaetostoma hexapetalum D. Nunes, D.O. Diniz, Koschnitzke. &

M.J. Silva (2018: 986).

Microlicia hilarii Versiane & R. Romero, nom. nov. Lavoisiera chamaepitys A. Saint-Hilaire

ex Naudin (153: 1844). Non Microlicia chamaepitys Naudin (1845: 185).

Note: The new epithet honors Auguste de Saint-Hilaire, the French botanist that collected

the type of this species.

Microlicia inerme (Naudin) Versiane & R. Romero, comb. nov. Chaetostoma inerme Naudin

(1845: 191).

Microlicia inversa (K. Fidanza, A.B. Martins & Almeda) Versiane & R. Romero, comb. nov.

Trembleya inversa K. Fidanza, A.B. Martins & Almeda (2013: 281).

Microlicia itambana (Martius & Schrank ex De Candolle) Versiane & R. Romero, comb.

nov. Lavoisiera itambana Martius & Schrank ex De Candolle (1828: 104).

Microlicia macrantha Versiane & R. Romero, nom. nov. Lavoisiera grandiflora A. Saint-

Hilaire ex Naudin (1844: 148). Non Microlicia grandiflora Baillon (1877: 34).

Note: The new epithet refers to the large flowers of this species.

Microlicia macrocarpa (Naudin) Versiane & R. Romero, comb. nov. Lavoisiera macrocarpa

Naudin (1844: 148).

Microlicia matogrossensis Versiane & R. Romero, nom. nov. Chaetostoma riedelianum

Cogniaux in Martius, Eichler & Urban (1883: 33). Non Microlicia riedeliana Cogniaux

(1883: 80).

Note: The new epithet is derived from the state of Mato Grosso, Brazil, to which this

species is endemic.

Microlicia mellobarretoi (Markgraf) Versiane & R. Romero, comb. nov. Lavoisiera

mellobarretoi Markgraf (1940: 220).

99

Microlicia minor Versiane & R. Romero, nom. nov. Lavoisiera humilis Naudin (1844: 153).

Non Microlicia humilis Naudin (1845: 179).

Note: The new epithet refers to the small size of the plants in this species.

Microlicia mucorifera (Martius & Schrank ex De Candolle) Versiane & R. Romero, comb.

nov. Lavoisiera mucorifera Martius & Schrank ex De Candolle (1828: 103).

Microlicia neopyrenaica (Naudin) Versiane & R. Romero, comb. nov. Trembleya

neopyrenaica Naudin (1844:154).

Microlicia nervulosa (Naudin) Versiane & R. Romero, comb. nov. Lavoisiera nervulosa

Naudin (1844: 149).

Microlicia parviflora (D.Don) Versiane & R. Romero, comb. nov. Meriania parviflora

D.Don (1823: 323). Trembleya parviflora (D.Don) Cogniaux (1883: 127).

Microlicia pentagona (Naudin) Versiane & R. Romero, comb. nov. Trembleya pentagona

Naudin (1844: 154).

Microlicia phlogiformis (De Candolle) Versiane & R. Romero, comb. nov. Trembleya

phlogiformis De Candolle (1828: 126).

Microlicia pilosa Versiane & R. Romero, nom. nov. Lavoisiera vestita Almeda & A.B.

Martins (2017: 171). Non Microlicia vestita Naudin (1828: 119).

Note: The new epithet refers to the dense indumentum on branches, leaves, hypanthia, and

sepals.

Microlicia pohliana (O. Berg ex Triana) Versiane & R. Romero, comb. nov. Lavoisiera

pohliana O. Berg ex Triana (1871: 30).

Microlicia plinervia Versiane & R. Romero, nom. nov. Lavoisiera cogniauxana Barreto

(1936: 63). Non Microlicia cogniauxiana R. Romero (2015: 1012).

Note: The new epithet refers to the 7–9-plinerved leaf blades, exclusive for this species.

100

Microlicia pradosiana (Netto) Versiane & R. Romero, comb. nov. Trembleya pradosiana

Netto (1865: 378).

Microlicia pulcherrima (Martius & Schrank ex De Candolle) Versiane & R. Romero, comb.

nov. Lavoisiera pulcherrima Martius & Schrank ex De Candolle (1828: 104).

Microlicia punctata (Martius & Schrank ex De Candolle) Versiane & R. Romero, comb. nov.

Lavoiseira punctata Martius & Schrank ex De Candolle (1828: 104).

Microlicia pusillifolia Versiane & R. Romero, nom. nov. Lavoisiera minima A.B. Martins &

Almeda (2013: 87). Non Microlicia minima Markgraf (1927: 46).

Note: The new epithet refers to the tiny size of the leaf blades in this species.

Microlicia pythioides (Chamisso) Versiane & R. Romero, comb. nov. Trembleya pythioides

Chamisso (1834: 428).

Microlicia quinquenervis (Wurdack) Versiane & R. Romero, comb. nov. Lavoisiera

quinquenervis Wurdack (1974: 136).

Microlicia rigida (Cogniaux) Versiane & R. Romero, comb. nov. Lavoisiera rigida Cogniaux

in Martius, Eichler & Urban (1883: 144).

Microlicia rosmarinoides (De Candolle) Versiane & R. Romero, comb. nov. Trembleya

rosmarinoides De Candolle (1828: 125).

Microlicia rundeliana (Almeda & A.B. Martins) Versiane & R. Romero, comb. nov.

Lavoisiera rundeliana Almeda & A.B. Martins (2017: 154).

Microlicia sampaioana (Mello Barreto) Versiane & R. Romero, comb. nov. Lavoisiera

sampaioana Mello Barreto (1935: 10).

Microlicia sanroqueana Versiane & R. Romero, nom. nov. Lavoisiera canastrensis Almeda

& A.B. Martins (2017: 60). Non Microlicia canastrensis Naudin (1845: 179).

101

Note: The new specific epithet is derived from the municipality of São Roque de Minas,

Minas Gerais, one of the main entrances to Serra da Canastra, to which this species appears

to be endemic.

Microlicia scaberula (Naudin) Versiane & R. Romero, comb. nov. Lavoiseira scaberula

Naudin (1844: 151).

Microlicia senaei (Schwacke) Versiane & R. Romero, comb. nov. Lavoisiera senaei

Schwacke (1900: 3).

Microlicia serratifolia Versiane & R. Romero, nom. nov. Trembleya serrulata K. Fidanza,

A.B. Martins & Almeda (2013: 288). Non Microlicia serrulata Chamisso (1834: 390).

Note: The new epithet refers to the conspicuously serrulate margins of the leaf blades.

Microlicia setifolia Versiane & R. Romero, nom. nov. Lavoisiera setosa A. B. Martins &

Almeda (2017: 164). Non Microlicia setosa (Sprengel) De Candolle (1828: 120), nec

Microlicia setosa Martius (1831: 253).

Note: The new epithet refers to the prominent bristle on the apex of each leaf blade.

Microlicia speciosa Versiane & R. Romero, nom. nov. Lavoisiera elegans Cogniaux in

Martius, Eichler & Urban (1883:160). Trembleya elegans (Cogniaux) Almeda & A.B.

Martins (2001: 6). Non Microlicia elegans Naudin (1845: 187)

Note: The new epithet means splendid, which has a similar meaning as the old name for

this species (elegans = elegant).

Microlicia suberosa (Naudin) Versiane & R. Romero, comb. nov. Stenodon suberosus

Naudin (1844: 146).

Microlicia subulata (Triana) Versiane & R. Romero, comb. nov. Lavoisiera subulata Triana

(1871: 30).

Microlicia tetragona (Martius & Schrank ex De Candolle) Versiane & R. Romero, comb

nov. Lavoisiera tetragona Martius & Schrank ex De Candolle (1828: 103).

102

Microlicia thomazii (R.B. Pacífico & Fidanza) Versiane & R. Romero, comb. nov.

Trembleya thomazii R.B. Pacífico & Fidanza (2015: 171).

Microlicia tridentata (Naudin) Versiane & R. Romero, comb. nov. Trembleya tridentata

Naudin (1844: 154).

RE-ESTABLISHED NAMES

Microlicia clavillosa Wurdack (1982: 297).

Chaetostoma scoparium Cogniaux in Engler (1896: 447). Non Microlicia scoparia De

Candolle (1828: 120), nec Microlicia scoparia Chamisso (1834: 394).

Note: Microlicia clavillosa was described by Wurdack (1982) based on the gathering Gates

& Estabrook 215. During the taxonomic revision of Chaetostoma, Koschnitzke & Martins

(2006) realized that this species was already published by Cogniaux (1896) as

Chaestostoma scoparium, based in a specimen collected by Ule. Then, M. clavillosa was

synomized under C. scoparium as heterotypic synomyn. Since the epithet “scoparia” was

already used in Microlicia, here we decide to re-establish Microlicia clavillosa as the

correct name for this species.

Microlicia cupressina D.Don (1823 :302). Chaetostoma cupressinum (D.Don) Koschnitzke

& A.B. Martins (1999: 202).

Microlicia laniflora (D.Don) Baillon (1879: 95). Melastoma laniflora D.Don (1823: 292).

Trembleya laniflora (D.Don.) Cogniaux in Martius, Eichler & Urban (1883: 130).

Microlicia selaginea Naudin (1845: 188). Chaetostoma selagineum (Naudin) Koschnitzke &

A.B. Martins (1999: 202).

Microlicia stenocladon Naudin (1845: 188). Chaetostoma stenocladon (Naudin)

Koschnitzke & A.B. Martins (1999: 202).

103

Acknowledgments

The first author thanks Coordenacão de Aperfeiçoamento de Pessoal de Nível Superior –

CAPES and Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPq

(Process: 142384/2018-6) for the financial support provided during her doctorate degree.

References

Almeda, F. & Martins, A.B. (2001) New combinations and new names in some Brazilian

Microlicieae (Melastomataceae), with notes on the delimitation of Lavoisiera, Microlicia

and Trembleya. Novon 11: 1–7.

Bongard, H.G. (1838) Genera duo Melastomacearum ordine nova. Mémoires de l’académie

Impériale des Sciences de Saint–Pétersbourg Ser. 6., Sciences Naturalles 4(2): 137–142.

Candolle, A.L.P.P. de (1828) Melastomaceae. In: Candolle, A.L.P.P. de (Ed.) Prodromus

systematis naturalis regni vegetabilis 3. Paris, Treuttel et Wurtz, pp. 99–202.

Clausing, G. & Renner S.S. (2001) Molecular phylogenetics of Melastomataceae and

Memecylaceae: implications for character evolution. American Journal of Botany 88(3):

486–498.

Cogniaux, C.A. (1891) Melastomataceae. In De Candolle ALPP, De Candolle, ACP (eds.).

Monographiae Phanerogamarum. Paris: G. Masson 7: 1–1256.

Don, D. (1823) An illustration of the natural family of plants called Melastomaceae. Memoirs

of the Wernerian Natural History Society 4(2): 276–329.

Fidanza, K., Martins, A.B. & Almeda, F. (2013) Four new species of Trembleya

(Melastomataceae: Microlicieae) from Serra do Cabral, Minas Gerais, Brazil. Brittonia.

DOI 10.1007/s12228-012-9281-x.

Flora do Brasil 2020, unde construction. Jardim Botânico do Rio de Janeiro. Available at <

http://floradobrasil.jbrj.gov.br/ > (accessed 01 May 2019).

104

Fritsch, P.W., Almeda, F., Renner, S.S., Martins, A.B. & Cruz, B.C. (2004) Phylogeny and

circumscription of the near-endemic Brazilian tribe Microlicieae (Melastomataceae).

American Journal of Botany 91(7): 1105–1114.

Koschnitzke, C & Martins A.B. (1999) New combinations and a new species in the Brazilian

genus Chaetostoma DC. (Microlicieae: Melastomataceae). Novon 9(2): 202–204.

Koschnitzke, C & Martins A.B. (2006) Revisão taxonômica de Chaetostoma DC.

(Melastomataceae, Microlicieae). Arquivos do Museu Nacional 64(2): 95–119.

Martins, E. (1997) Revisão taxonômica do gênero Trembleya DC. (Melastomataceae).

Unpublished D. Phil. Thesis. Universidade Estadual de Campinas.

Martins, E. (1995) Nova espécie do eênero Trembleya DC. (Microlicieae - Melastomataceae).

Boletim de Botânica 14: 39−42. doi.org/10.11606/issn.2316-9052.v14i0p39-42

Martins, A.B. & Almeda, F. (2017) A monograph of the Brazilian endemic genus Lavoisiera

(Melastomataceae: Microlicieae). Phytotaxa 135(1): 1–194.

Mendoza-Cifuentes, H., Ariza, W., Granados, D.E. & Romero, R. (2019). A new species of

Microlicia (Melastomataceae): first record of the genus for Colombia. PhytoKeys 122: 87–

96. /doi.org/10.3897/phytokeys.122.34171

Michelangeli, F.A., Goldenberg, R., Almeda, F., Judd, W.S., Bécquer, E.R., Ocampo, G.,

Ionta, G.M., Skean, Jr., J.D., Majure, L.C. & Penneys, D.S. (2018) Nomenclatural

novelties in Miconia (Melastomataceae: Miconieae). Brittonia DOI 10.1007/s12228-018-

9546-0

Naudin, C. (1849) Additions à la flore du Brésil méridional. Description de genres nouveaux,

et rectification de quelques anciens genres appartenant à la famille des Mélastomacées.

Annales des Sciences Naturelles, Botanique, série 3 12: 126–159.

105

Pacífico, R.B., Fidanza, K. (2015) Three new endangered species of Trembleya DC.

(Melastomataceae: Microlicieae) from Minas Gerais, Brazil. Phytotaxa 238 (2): 163–173.

dx.doi.org/10.11646/phytotaxa.238.2.4

Pacífico, R.B. & Fidanza, K. (2018) Flora da Serra do Cipó, Minas Gerais: Microlicieae

(Melastomataceae). Boletim de Botânica da Universidade de São Paulo 36:29−95.

DOI10.11606/issn.2316-9052.v36ip29-95

Pacífico, R., Almeda, F., Carmo, A.A. & Fidanza, K. 2019. A new species of Trembleya

(Melastomataceae: Microlicieae) with notes on leaf anatomy and generic circumscription.

Phytotaxa 391(5): 289–300.

Rocha, M.J.R., Batista, J.A.N., Guimarães, P.J.F. & Michelangeli, F.A. (2016) Phylogenetic

placement and a new circumscription of Poteranthera (Microlicieae; Melastomataceae).

Phytotaxa 263 (3): 219–232. http://dx.doi.org/10.11646/phytotaxa.263.3.3

Romero, R. (2003) Revisão taxonômica de Microlicia sect. Chaetostomoides

(Melastomataceae). Revista Brasileira de Botânica 26(4): 429–435.

Romero, R., Silva, K.R., Simão, D.G. (2014) Microlicia crassa and M. maculata spp. nov.

(Melastomataceae) from Minas Gerais, Brazil: morphology and leaf anatomy. Nordic

Jounal of Botany 33: 178–185. doi: 10.1111/njb.00635.

Romero, R., Silva, K.R., Simão, D.G. (2015) Microlicia cogniauxiana and Microlicia

naudiniana (Melastomataceae), two new species from the Espinhaço Range, Brazil

Systematic Botany 40(4). DOI 10.1600/036364415X690058.

Romero, R., Caetano, A.P.S., Santos, K.F., De-Paula, O.C. & Versiane, A.F.A. 2019. Two

new species of Microlicia D.Don (Melastomataceae) from campos rupestres in eastern

Brazil. Nordic Journal of Botany. 37(10): 1–17 doi: 10.1111/njb.02350

Turland, N.J., Wiersema, J.H., Barrie, F.R., Greuter, W., Hawksworth, D.L., Herendeen, P.S.,

Knapp, S., Kusber, W.-H., Li, D.-Z., Marhold, K., May, T.W., McNeill, J., Monro, A.M.,

106

Prado, J., Price, M.J. & Smith, G.F. (eds.) 2018. International Code of Nomenclature for algae, fungi, and plants (Shenzhen Code) adopted by the Nineteenth International

Botanical Congress Shenzhen, China, July 2017. Regnum Vegetabile 159. Glashütten:

Koeltz Botanical Books. DOI https://doi.org/10.12705/Code.2018

107

CONSIDERAÇÕES FINAIS

Encontrar grupos monofiléticos com caracteres morfológicos pouco homoplásticos e que os delimitem com clareza é um dos principais objetivos dos estudos filogenéticos. Já há algum tempo, o uso de ferramentas moleculares, juntamente com técnicas de estudos morfológicos, vem trazendo uma nova forma de estudar e entender a biodiversidade. Além disso, compreender a relação dos seres vivos também reflete no entendimento da distribuição e ocupação deles nos mais variados ambientes do globo. Grupos de plantas com alta diversidade costumam ser um grande desafio para quem usa dados moleculares e morfológicos. Muitas vezes os caracteres morfológicos tradicionalmente utilizados podem ser poucos informativos, não sustentando os clados recuperados em análises moleculares, e revelando grupos para- ou polifiléticos. Espécies de Melastomataceae ocupam vários tipos de ambientes, desde florestais até campos abertos, úmidos e/ou alagados ou completamente secos. Toda essa capacidade de adaptação contribui para sua expressiva diversidade, tornando-a uma família de grande importância nos Neotrópicos. Um dos problemas encontrados na classificação de Melastomataceae é que caracteres tradicionalmente usados na segregação de tribos, seções, gêneros e/ou espécies mostram-se incompatíveis com os resultados filogenéticos, devido ao alto grau de homoplasias e os tornando, assim, ineficientes na delimitação de tais grupos. Desta forma, muitas vezes a resposta pode estar em caracteres nunca ou pouco explorados (e.g. anatômicos, fisiológicos, ecológicos e/ou biogeográficos). Microlicieae é uma das maiores tribos de Melastomataceae. Apesar de ser considerada monofilética, vários conflitos em relação à delimitação dos gêneros que a compõem (Rhynchanthera, Poteranthera, Chaetostoma, Lavoisiera, Microlicia, Stenodon e Trembleya) persistiam até o presente momento. Esta dificuldade em delimitar os gêneros ocorreu pela ampla variação morfológica encontrada em Microlicia fazendo sobreposição com os demais gêneros na tribo, principalmente, com Chaetostoma, Lavoisiera e Trembleya. Alguns autores tentaram esclarecer essas relações, tanto com estudos morfológicos quanto moleculares, mas algumas lacunas foram deixadas devido a uma amostragem insuficiente de Microlicia. Diante do exposto, verificou-se a necessidade de aprofundar os estudos filogenéticos em Microlicieae, a fim de esclarecer as relações de Microlicia com os gêneros restantes. Para tanto, foi feito um aumento na amostragem de espécies de todos os gêneros em Microlicieae e sequências de quatro diferentes marcadores moleculares foram analisadas em relação ao

108

estudo anterior. Além disso, uma análise minunciosa dos caracteres morfológicos tradicionalmente usados para a delimitação dos gêneros foi feita. De acordo com os resultados filogenéticos moleculares confirmamos que Microlicia, da forma como tradicionalmente circunscrito, é um gênero parafilético. Também encontramos um alto grau de homoplasia foi encontrado na maioria dos caracteres morfológicos tradicionalmente utilizados na circunscrição dos gêneros. Diante dos resultados obtidos no presente estudo, três alternativas se apresentam: manter o gênero parafilético; dividí-lo em vários gêneros menores; ou considerar Microlicia, Trembleya, Lavoisiera, Stenodon e Chaetostoma como um único gênero. Levando em consideração o princípio da sistemática filogenética que norteou nosso estudo e os caracteres morfológicos aqui estudados não foram eficazes para o reconhecimento das linhagens em Microlicia em gêneros distintos, decidimos expandir a circunscrição do gênero para inclusão dos demais (Trembleya, Lavoisiera, Stenodon e Chaetostoma), e assim tornando o gênero Microlicia monofilético. Rhynchanthera e Poteranthera, provavelmente, são as linhagens que primeiro divergiram em Microlicieae. Um total de 67 nomes previamente reconhecidos em Chaetostoma, Lavoisiera, Stenodon e Trembleya foram transferidos para Microlicia, sendo 48 novas combinações e 19 novos nomes. Cinco nomes que foram primeiramente descritos em Microlicia devem retornar ao gênero. Além de contribuir para a compreensão e ampliação do conhecimento em Melastomataceae, apresentando uma nova classificação para Microlicieae e uma nova circunscrição para Microlicia, este estudo poderá contribuir com estudos biogeográficos futuros sobre o Cerrado, principalmente sobre a história evolutiva de Microlicieae nos campos rupestres do Brasil. Estudos futuros envolvendo datação e mudanças em padrões de diversificação em Microlicieae também possibilitarão o entendimento da evolução do grupo. Cabe destacar que durante a elaboração do presente estudo uma nova espécie de Trembleya, uma espécie de Lavoisiera e pelo menos 12 Microlicia foram descobertas, das quais algumas já foram publicadas (Romero et al., 2017; Romero et al., 2019a; Romero et al., 2019b) e outras estão em fase de publicação. Fato que demonstra quão diverso é o grupo, ainda não totalmente conhecido indicando a necessidade de estudos com o mesmo.

109

REFERÊNCIAS GERAIS

Aberer AJ, Krompass D, Stamatakis A. 2013. Pruning rogue taxa improves phylogenetic

accuracy: an efficient algorithm and webservice. Systematic Biology 62(1): 162–166.

Almeda F, Martins AB. 2001. New combinations and new names in some Brazilian

Microlicieae (Melastomataceae), with notes on the delimitation of Lavoisiera, Microlicia

and Trembleya. Novon 11: 1–7.

Almeda F, Robinson OR. 2011. Systematics and phylogeny of Siphanthera

(Melastomataceae). Systematic Botany Monographs 93: 1–101.

Almeda F, Pacifico, R. 2018. Neotropical Poteranthera (Melastomataceae: Microlicieae)

revisited. Systematic Botany 43(2): 552−556. DOI 10.1600/036364418X697274

APG - The Angiosperm Phylogeny Group 2016. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG IV. Botanical Journal of the Linnean Society 181: 1–20. Baumgratz JFA, Souza MLDR, Woodgyer EM, Lughadha EMN. 1996. Polysporangiate

anthers: described for the first time in Melastomataceae. Kew Bulletin 51: 133–144.

Bacci LF, Michelangeli FA, Goldenberg R. 2019. Revisiting the classification of

Melastomataceae: implications for habit and fruit evolution. Botanical Journal of the

Linnean Society 190: 1–24. doi.org/10.1093/botlinnean/boz006

Basso-Alves JP, Goldenberg R, Teixeira SP. 2017. The ontogenetic bases for variation in

ovary position in Melastomataceae. American Journal of Botany 104(8): 1142–1156.

Bochorny T, Michelangeli FA, Almeda F, Goldenberg R. 2019. Phylogenetics,

morphology and circumscription of Cambessedesieae: a new Neotropical tribe of

Melastomataceae. Botanical Journal of the Linnean Society 190(3): 281–302.

doi.org/10.1093/botlinnean/boz018

Bongard HG 1838. Genera duo Melastomacearum ordine nova. Mémoires de l’académie

Impériale des Sciences de Saint–Pétersbourg Ser. 6., Sciences Naturalles 4(2): 137–142.

110

Bonfim-Patrício MC, Salatino A, Martins AB, Wurdack JJ, Salatino ML. 2001.

Flavonoids of Lavoisiera, Microlicia and Trembleya (Melatomataceae) and their

taxonomic meaning. Biochemical Systematics and Ecololgy 29: 711−726

Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu CH, Xie D, Suchard MA, Rambaut

A, Drummond AJ. 2014. BEAST 2: A Software Platform for Bayesian Evolutionary

Analysis. PLOS Computational Biology 10(4): e1003537. doi:

10.1371/journal.pcbi.1003537.

Bünger MO, Mazine FF, Forest F, Bueno ML, Stehmann JR, Lucas EJ. 2016. The

evolutionary history of Eugenia sect. Phyllocalyx (Myrtaceae) corroborates historically

stable areas in the southern Atlantic forests. Annals of Botany 118(7): 1209–1223.

doi:10.1093/aob/mcw209

Candolle, ALPP. de 1828. Melastomaceae. In: Candolle, ALPP de, ed. Prodromus

systematis naturalis regni vegetabilis 3. Paris: Treuttel et Wurtz, 99–202.

Carmo A, Pacifico R, Gonçales-Silva R, Sbais PG, Fidanza K, Souza LA. 2019.

Epidermal micromorphology and venation patterns of Microlicieae (Melastomataceae)

leaves: Looking for new characters for the taxonomy of this neotropical tribe. Flora.

doi.org/10.1016/j.flora.2019.151494

Clausing G, Renner SS. 2001. Molecular phylogenetics of Melastomataceae and

Memecylaceae: implications for character evolution. American Journal of Botany 88(3):

486–498.

Cogniaux CA. 1883. Melastomataceae: Microlicieae. In: Martius, CFP, Eichler AW, Urban I,

eds. Flora brasiliensis. Fleischer. Leipzig, 14(3): 5–205.

Cogniaux CA. 1885. Poteranthera. In: Martius, CFP, Eichler AW, Urban I, eds. Flora

Brasiliensis. Fleischer. Leipzig, 14(3): 201–204.

111

Cogniaux CA. 1891. Melastomataceae. In: Candolle ALPP de, Candolle, ACP de, eds.

Monographiae Phanerogamarum. Paris: G. Masson, 7: 1–1256.

Dias P, Assis LCS, Udulutsch RG. 2005. Monophyly vs. paraphyly in plant systematics.

Taxon 54: 1039-1040

Diniz-Neres DO, Silva MJ. 2017. A remarkable new species of Microlicia

(Melastomataceae) including an identification key to species of the genus from the

Chapada dos Veadeiros National Park, Goiás, Brazil. Systematic Botany 42(3): 554–561.

DOI 10.1600/036364417X696014

Diniz DO, Silva MJ. 2018. A noteworthy new species of Microlicia (Melastomataceae,

Microlicieae) from the highlands of the state of Goiás, Brazil. Journal of the Torrey

Botanical Society 145(2): 175–181. doi.org/10.3159/TORREY-D-17-00027.1

Don D. 1823. An illustration of the natural family of plants called Melastomaceae. Memoirs

of the Wernerian Natural History Society 4(2): 276–329.

Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf

tissue. Phytochemical Bulletin 19: 11–15.

Dunn IS, Blattner FR. 1987. Charons 36 to 40: multi-enzyme, high capacity, recombination

deficient replacement vectors with polylinkers and polystuffers. Nucleic Acids Research

15: 2677–2698.

Endress PK. 2010. Flower structure and trends of evolution in eudicots and their major

subclades. Annals of the Missouri Botanical Garden 97: 541–583.

Endress PK. 2011. Changing views of flower evolution and new questions. In: Wanntorp L,

Ronse De Craene LP, eds. Flowers on the tree of life. Cambridge: Cambridge University

Press, 120–141.

112

Fidanza K, Martins AB, Almeda F. 2013. Four new species of Trembleya

(Melastomataceae: Microlicieae) from Serra do Cabral, Minas Gerais, Brazil. Brittonia.

DOI 10.1007/s12228-012-9281-x.

Flora do Brasil 2020 under construction. Jardim Botânico do Rio de Janeiro. Available at <

http://floradobrasil.jbrj.gov.br/ > (accessed 06 January 2020).

Fritsch PW, Almeda F, Renner SS, Martins AB, Cruz BC. 2004. Phylogeny and

circumscription of the near-endemic Brazilian tribe Microlicieae (Melastomataceae).

American Journal of Botany 91(7): 1105–1114.

Galtier N, Daubin V. 2008. Dealing with incongruence in phylogenomic analyses.

Philosophical Transactions of the Royal Society B 363: 4023–4029.

Giulietti AM & Pirani JR. 1988. Patterns of geographic distribution of some plant species from the Espinhaço Range, Minas Gerais and Bahia, Brazil. In: Vanzolini, P.E., & Heyer, W.R. (eds) Proceedings of a workshop on Neotropical distribution patterns. Academia Brasileira de Ciências, Rio de Janeiro, 39–69 Gleason HA. 1931. Studies on the Flora of Northern South America-XV. Recent Collections

of Melastomataceae from Peru and Amazonian Brazil. Bulletin of the Torrey Botanical

Club 58(4): 215–262.

Goldenberg R, Fraga CN, Fontana AP, Nicolas AN, Michelangeli FA. 2012. Taxonomy and phylogeny of Merianthera (Melastomataceae). Taxon 61(5): 1040−1056. Goldenberg R, Penneys DS, Almeda F, Judd WS, Michelangeli FA. 2008. Phylogeny of

Miconia (Melastomataceae): patterns of stamen diversification in a megadiverse

neotropical genus. International Journal of Plant Sciences 24: 315–327.

Goldenberg R, Reginato M, Michelangeli FA. 2018. Disentangling the infrageneric

classification of megadiverse taxa from Mata Atlantica: Phylogeny e and function as the

crucible of architectural diversity. Annals of Botany 112: 1477–1493.

113

Guimarães PJF, Michelangeli FA, Sosa K, Gómez JRS. 2019. Systematics of Tibouchina

and allies (Melastomataceae: Melastomateae): A new taxonomic classification. Taxon.

doi.org/10.1002/tax.12151

Harmon LJ, Weir JT, Brock CD, Glor RE, Challenger W. 2008. GEIGER: Investigating

evolutionary radiations. Bioinformatics 24: 129–131.

doi.org/10.1093/bioinformatics/btm538

Hennig W. 1965. Phylogenetic Systematics. Annual Review of Entomology 10: 97–116.

Hillis DM, Heath TA, St. John K. 2005. Analysis and visualization of Tree Space.

Systematic Biology 54(3): 471–482. DOI: 10.1080/10635150590946961

Jeffroy O, Brinkmann H, Delsuc F, Philippe H. 2006. Phylogenomics: the beginning of

incongruence? Trends in Genetics 22(4): 225–31.

Jombart T, Kendall M, Almagro-Garcia J, Colijn C. 2017. TREESPACE: Statistical

exploration of landscapes of phylogenetic trees. Molecular Ecology Resources 17: 1385–

1392. DOI: 10.1111/1755-0998.12676

Katoh S. 2013. MAFFT multiple sequence alignment software version 7: Improvements in

performance and usability. Molecular Biology and Evolution 30: 772–780.

doi.org/10.1093/molbev/mst010

Koschnitzke C, Martins AB. 1999. New combinations and a new species in the Brazilian

genus Chaetostoma DC. (Microlicieae: Melastomataceae). Novon 9(2): 202–204.

Koschnitzke C, Martins AB. 2006. Revisão taxonômica de Chaetostoma DC.

(Melastomataceae, Microlicieae). Arquivos do Museu Nacional 64(2): 95–119.

Koschnitzke C, Martins AB. 2007. Nomenclatural Alterations in Microlicieae

(Melastomataceae) Novon 17(4): 472–475.

114

Kress WJ, Prince LM, Williams KJ. 2002. The phylogeny and a new classification of the

gingers (Zingiberaceae): evidence from molecular data. American Journal of Botany 89:

1682–1696.

Kriebel R. 2012. A synopsis of the genus Poteranthera (Melastomeae: Melastomataceae)

with the description of a new, apparently pollinator deceiving species. Brittonia 62: 6–14.

Kriebel R, Michelangeli FA, Kelly LM. 2015. Discovery of unusual anatomical and

continuous characters in the evolutionary history of Conostegia (Miconieae:

Melastomataceae). Molecular Phylogenetics and Evolution 82: 289–313.

Kumar S, Filipski AJ, Battistuzzi FU, Pond SLK, Tamura K. 2012. Statistics and truth in

Phylogenomics. Molecular Biology and Evolution 29(2): 457–472.

doi:10.1093/molbev/msr202

Lanfear R, Calcott B, Ho SYW, Guindon S. 2012. PartitionFinder: Combined selection of

partitioning schemes and substitution models for phylogenetic analyses. Molecular Biology

and Evolution 29: 1695–1701. doi.org/10.1093/molbev/mss020

Larridon I, Bauters K, Reynders M, Huygh W, Muasya AM, Simpson DA, Goetghebeur

P. 2013. Towards a new classification of the giant paraphyletic genus Cyperus

(Cyperaceae): phylogenetic relationships and generic delimitation in C4 Cyperus.

Botanical Journal of the Linnean Society 172: 106–126.

Lima, DFS. 2017. Estudos filogenéticos e taxonômicos em Myrcia DC. sensu lato

(Myrtaceae), com ênfase no clado Guianensis. Unpublished D. Phil. Thesis. Universidade

Estadual de Campinas.

Maddison WP, Maddison DR. 2001. MESQUITE: a modular system for evolutionary

analysis, Version 3.04. Available at: http://mesquiteproject.org.

115

Martins E. 1995. Nova espécie do gênero Trembleya DC. (Microlicieae - Melastomatacea).

Boletim de Botânica da Universidade de São Paulo 14: 39−42.

doi.org/10.11606/issn.2316-9052.v14i0p39-42

Martins E. 1997. Revisão taxonômica do gênero Trembleya DC. (Melastomataceae).

Unpublished D. Phil. Thesis. Universidade Estadual de Campinas.

Martins AB, Almeda F. 2017. A monograph of the Brazilian endemic genus Lavoisiera

(Melastomataceae: Microlicieae). Phytotaxa 135(1): 1–194.

Mendoza-Cifuentes H, Ariza W, Granados DE, Romero R. 2019. A new species of

Microlicia (Melastomataceae): first record of the genus for Colombia. PhytoKeys 122: 87–

96. /doi.org/10.3897/phytokeys.122.34171

Michelangeli FA, Goldenberg R, Almeda F, Judd WS, Bécquer ER, Ocampo G, Ionta

GM, Skean Jr, JD, Majure LC, Penneys DS. 2018. Nomenclatural novelties in Miconia

(Melastomataceae: Miconieae). Brittonia DOI 10.1007/s12228-018-9546-0

Michelangeli FA, Guimarães PJF, Penneys DS, Almeda F, Kriebel R. 2013. Phylogenetic

relationships and distribution of New World Melastomeae (Melastomataceae). Botanical

Journal of the Linnean Society 171: 38–60.

Michelangeli FA, Penneys DS, Giza J, Soltis D, Hils MH, Skean JD. 2004. A preliminary

phylogeny of the tribe Miconieae (Melastomataceae) based on nrITS sequence data and its

implications on inflorescence position. Taxon 53: 279–290.

Miller MA, Pfeiffer W, Schwartz T. 2010. Creating the CIPRES Science Gateway for

inference of large phylogenetic trees in proceedings of the gateway computing

environments workshop (GCE) New Orleans, LA. Pp 1–8. Available at:

http://www.phylo.org/.

116

Naudin C. 1844. Additions à la flore du Brésil méridional. Description de genres nouveaux,

et rectification de quelques anciens genres appartenant à la famille des Mélastomacées.

Annales des Sciences Naturelles, Botanique, série 3 2: 140–155.

Naudin C. 1849. Additions à la flore du Brésil méridional. Description de genres nouveaux,

et rectification de quelques anciens genres appartenant à la famille des Mélastomacées.

Annales des Sciences Naturelles, Botanique, série 3 12: 126–159.

Pacifico RB, Fidanza K. 2015. Three new endangered species of Trembleya DC.

(Melastomataceae: Microlicieae) from Minas Gerais, Brazil. Phytotaxa 238(2): 163–173.

dx.doi.org/10.11646/phytotaxa.238.2.4

Pacifico RB, Fidanza K. 2017. Microlicia sciophylla (Melastomataceae: Microlicieae), a

new micro-endemic species from the rocky fields of Minas Gerais, Brazil. Kew Bulletin

72: 5. DOI 10.1007/S12225-017-9688-3

Pacífico RB, Fidanza K. 2018. Flora da Serra do Cipó, Minas Gerais: Microlicieae

(Melastomataceae). Boletim de Botânica da Universidade de São Paulo 36: 29−95.

DOI10.11606/issn.2316-9052.v36ip29-95

Pacifico RB, Fidanza K, Almeda F. 2017. Two new species of Microlicia

(Melastomataceae) from the Rupestrian Grasslands of Serra do Cipó, Minas Gerais, Brazil.

Phytotaxa 316(1): 39–50. doi.org/10.11646/phytotaxa.316.1.3

Pacifico R, Almeda F, Carmo AA, Fidanza K. 2019. A new species of Trembleya

(Melastomataceae: Microlicieae) with notes on leaf anatomy and generic circumscription.

Phytotaxa 391(5): 289–300.

Paradis E, Claude J, Strimmer K. 2004. APE: Analyses of phylogenetics and evolution in

R language. Bioinformatics 20: 289–290. doi.org/10.1093/bioinformatics/btg412

Pataro L, Romero R, Roque N. 2013. Four new species of Microlicia (Melastomataceae)

from Chapada Diamantina, Bahia, Brazil. Kew Bulletin 68: 285Y–293.

117

Penneys DS, Michelangeli FA, Judd WS, Almeda F. 2010. Henrietteeae

(Melastomataceae): A New Neotropical Berry-Fruited Tribe. Systematic Botany 35(4):

783–800. DOI 10.1600/036364410X539862

Rambaut A, Drummond AJ. 2013 Tracer. Avaiable at:

.

Rapini A, Ribeiro PL, Lambert S. & Pirani JR. 2008. A flora dos campos rupestres da Cadeia do Espinhaço. Megadiversidade 4: 15–23. Reginato M, Michelangeli FA, Goldenberg R. 2010. Phylogeny of Pleiochiton

(Melastomataceae, Miconieae): total evidence. Botanical Journal of the Linnean Society

162(3): 423–434.

Reginato M, Michelangeli FA. 2016a. Primers for low-copy nuclear genes in the

Melastomataceae. Applications in Plant Sciences 4(1): 1500092.

Reginato M, Michelangeli FA. 2016b. Untangling the phylogeny of Leandra s.str.

(Melastomataceae, Miconieae). Molecular Phylogenetics and Evolution 96: 17–32.

Reginato M, Michelangeli FA. 2016c. Diversity and constraints in the floral morphological evolution of Leandra s.str. (Melastomataceae). Annals of Botany 118: 445–458. doi:10.1093/aob/mcw116

Reginato M, Neubig KM, Majure LC, Michelangeli FA. 2016. The first complete plastid genomes of Melastomataceae are highly structurally conserved. PeerJ 4: e2715. DOI 10.7717/peerj.2715 Renner SS. 1990. A revision of Rhynchanthera (Melastomataceae). Nordic Journal of Botany

9: 601–630.

Renner SS. 1993. Phylogeny and classification of the Melastomataceae and Memecylaceae.

Nordic Journal of Botany 13: 519–540.

Renner SS. 1994. A revision of Pterolepis (Melastomataceae: Melastomeae). Nordic Journal

of Botany 14: 73–104.

118

Revell LJ. 2012. phytools: An R package for phylogenetic comparative biology (and other

things). Methods in Ecology and Evolution 3: 217–223. doi.org/10.1111/j.2041-

210X.2011. 00169.x

Rocha MJR, Batista JAN, Guimarães PJF, Michelangeli FA. 2016a. Phylogenetic

relationships in the Marcetia alliance (Melastomeae, Melastomataceae) and implications

for generic circumscription. Botanical Journal of the Linnean Society 181: 585–609.

Rocha MJR, Batista JAN, Guimarães PJF, Michelangeli FA. 2016b. Phylogenetic

placement and a new circumscription of Poteranthera (Microlicieae; Melastomataceae).

Phytotaxa 263 (3): 219–232. http://dx.doi.org/10.11646/phytotaxa.263.3.3

Rocha MJR, Guimarães PJF, Michelangeli FA, Batista JAN. 2018. Taxonomy of Marcetieae: A new Neotropical tribe of Melastomataceae. International Journal of Plant Sciences 179(1): 50−74. DOI: 10.1086/694932 Romero R. 2000. A new species of Microlicia (Melastomataceae) from Brazil. Brittonia

52(2): 142−144.

Romero R. 2003. Revisão taxonômica de Microlicia sect. Chaetostomoides

(Melastomataceae). Revista Brasileira de Botânica 26(4): 429–435.

Romero R. 2005. A new species of Microlicia (Melastomataceae) from Minas Gerais, Brazil.

Novon 15(2): 358–360.

Romero R. 2010. Uma nova espécie de Microlicia (Melastomataceae) do estado de Minas

Gerais, Brasil. Rodriguésia 61(Sup.): S15-S18.

Romero R. 2013a. A new species of Microlicia (Melastomataceae) from the Espinhaço

Range, Minas Gerais, Brazil. Phytotaxa 88: 1–5. dx.doi.org/10.11646/phytotaxa.88.1.1.

Romero R. 2013b. Taxonomic notes in Microlicia (Melastomataceae, Microlicieae).

Phytotaxa 110(1): 48–54. dx.doi.org/10.11646/phytotaxa.110.1.4

119

Romero R, Caetano APS, Santos KF, De-Paula OC, Versiane AFA. 2019a. Two new

species of Microlicia D.Don (Melastomataceae) from campos rupestres in eastern Brazil.

Nordic Journal of Botany 37(10): 1–11. https://doi.org/10.1111/njb.02350

Romero R, Fontelas JC, Versiane AFA, De-Paula OC. 2019b. Microlicia longirostrata, a

new species of Melastomataceae from Goiás in Central Brazil. Systematic Botany 44(2):

394-354.

Romero R, Silva KR, Simão DG. 2014. Microlicia crassa and M. maculata spp. nov.

(Melastomataceae) from Minas Gerais, Brazil: morphology and leaf anatomy. Nordic

Jounal of Botany 33: 178–185. doi: 10.1111/njb.00635.

Romero R, Silva KR, Simão DG. 2015. Microlicia cogniauxiana and Microlicia naudiniana

(Melastomataceae), two new species from the Espinhaço Range, Brazil Systematic Botany

40(4). DOI 10.1600/036364415X690058.

Romero R, Versiane AFA. 2014. A new species of Microlicia and a checklist of

Melastomataceae from the mountains of Capitólio municipality, Minas Gerais, Brazil.

Phytotaxa 170(2): 118–124.

Romero R, Versiane AFA. 2016. Microlicia candolleana (Melastomataceae): a new endemic

species to the Espinhaço range, Minas Gerais, Brazil. Phytotaxa 261(3): 275–281.

http://dx.doi.org/10.11646/phytotaxa.261.3.7.

Romero R, Versiane AFA, Fontelas JC, Diniz-Neres DO, de Paula, CO. 2017. A new

species of Microlicia (Melastomataceae) with bright golden cuticle from Chapada dos

Veadeiros, Goiás (Central Brazil). Systematic Botany 42(3): 562–568. DOI

10.1600/036364417X696069

Romero R, Woodgyer E. 2010. Validation of the name Microlicia petiolulata

(Melastomataceae), a new species from the Espinhaço Range, Minas Gerais, Brazil. Kew

Bulletin 65: 69–72.

120

Romero R, Woodgyer E. 2014. Rediscovery of two species of Microlicia (Melastomataceae)

in Minas Gerais, Brazil. Phytotaxa 173(1): 41–48.

Romero R, Woodgyer E. 2018. Six new species of Microlicia (Melastomataceae) from

Bahia, Brazil. Kew Bulletim 73: 22. DOI 10.1007/S12225-018-9747-4

Sanderson MJ, Shaffer HB. 2002. Troubleshooting Molecular Phylogenetic Analyses.

Annual Review of Ecology and Systematics 33: 49–72 doi: 10.1

146/annurev.ecolsys.33.010802.150509

Shaw J, Lickey EB, Beck JT, Farmer SB, Liu W, Miller J, Siripun KC, Winder CT,

Schilling EE, Small RL. 2005. The tortoise and the hare II: relative utility of 21

noncoding chloroplast DNA sequences for phylogenetic analysis. American Journal of

Botany 92: 142–166.

Silva DN, Diniz DO, Koschnitzke C, Guimarães PJF, Silva MJ, Maia, VH. 2018a.

Chaetostoma hexapetalum (Microlicieae, Melastomataceae): A new species from the

Chapada dos Veadeiros, Goiás, Brazil. Systematic Botany 43(4): 986–992 DOI

10.1600/036364418X697805

Silva KR, Romero R, Simão DG. 2018b. Leaf characters of Lavoisiera, Microlicia and

Trembleya (Microlicieae, Melastomataceae) and their implications for taxonomy. Feddes

Repertorium 1–14. DOI: 10.1002/fedr.201700019

Simon MF, Grether R, Queiroz LP, Skema C, Pennington RT, Hughes CE. 2009. Recent

assembly of the Cerrado, a neotropical plant diversity Hotspot, by in situ evolution of

adaptations to fire. Proceedings of the National Academy of Sciences 06 (48): 20359–

20364.

Som A. 2014. Causes, consequences and solutions of phylogenetic incongruence. Briefings in

Bioinformatics 16(3): 536–548. doi:10.1093/bib/bbu015

121

Stamatakis A. 2006. RAxML-VI-HPC: maximum likelihood based phylogenetic analyses

with thousands of taxa and mixed models. Bioinformatics 22: 2688–2690.

Thiers B. 2020. Index Herbariorum: A global directory of public herbaria and associated

staff. New York Botanical Garden’s Virtual Herbarium. Avaliable at:

. (accessed 06 January 2020).

Turland NJ, Wiersema JH, Barrie FR, Greuter W, Hawksworth DL, Herendeen PS,

Knapp S, Kusber WH, Li DZ, Marhold K, May TW, McNeill J, Monro AM, Prado J,

Price MJ, Smith GF. 2018. International Code of Nomenclature for algae, fungi, and

plants (Shenzhen Code) adopted by the Nineteenth International Botanical Congress

Shenzhen, China, July 2017. Regnum Vegetabile 159. Glashütten: Koeltz Botanical Books.

DOI https://doi.org/10.12705/Code.2018

Triana J. 1871. Les Melastomacées. Transactions of the Linnean Society of London 28: 1–

188.

Vasconcelos TNC, Proença CEB, Ahmad B, Aguilar DS, Aguilar R, Amorim BS,

Campbell K, Costa IR, De-Carvalho PS, Faria JEQ, Giaretta A, Kooij PW, Lima DF,

Mazine FF, Peguero B, Prenner G, Santos MF, Soewarto J, Wingler A, Lucas EJ.

2017. Myrteae phylogeny, calibration, biogeography and diversification patterns: increased

understanding in the most species rich tribe of Myrtaceae. Molecular phylogenetics and

evolution 109: 113-137.

Veranso-Libalah MC, Stone RD, Fongod, AGN, Couvreur TLP, Kadereit G. 2017.

Phylogeny and systematics of African Melastomateae (Melastomataceae) Taxon 66(3):

584–614. doi.org/10.12705/663.5

Versiane AFA, Santos ML, Romero R. 2016. Melastomataceae na Serra dos Pireneus,

Goiás, Brasil. Rodriguésia 67(3): 721–759. DOI: 10.1590/2175-7860201667314

122

Welzen PC van. 1997. Paraphyletic groups or what should a classification entail. Taxon

46(1): 99–103.

Woodgyer E. 2005. Multi-Access key and checklist to the species of Microlicia

(Melastomataceae) in Bahia, Brazil. Kew Bulletim 60: 441–448.

Wurdack JJ. 1983. Certamen Melastomataceis XXXVI. Phytologia 53(2): 121–13.

123

Anexo 1 - Documento referente a Bioética e/ou Biossegurança.

124

125

126

127

128

Anexo 2 – Declaração de que tese não infringe os dispositivos da lei nº 9610/98, nem o direito autoral de qualquer editora.