Evolutionary history and systematics of Campylocentrum (Orchidaceae: Vandeae: Angraecinae): a phylogenetic and biogeographical approach

Evolutionary history and systematics of Campylocentrum (Orchidaceae: Vandeae: Angraecinae): a... Abstract Subtribe Angraecinae (Orchidaceae: Vandeae) are mainly distributed in Africa, but with two genera, Campylocentrum and Dendrophylax, restricted to the Neotropics. As a widespread Neotropical genus, Campylocentrum constitutes an appropriate model for revealing biogeographical patterns in this area and investigating routes of colonization and dispersal. In this study, we reconstructed phylogenetic relationships of the genus with Bayesian inference and maximum parsimony analyses of combined nuclear (ITS rDNA and Xdh) and plastid (matK exon, rpl32-trnL spacer, trnL intron, trnL-trnF spacer and ycf1 exon) DNA datasets, aimed at establishing a new infrageneric classification for this taxonomically complex genus. Based on the most comprehensive phylogenetic tree, we investigated the biogeographical history of Campylocentrum by estimating divergence times, inferred using fossil and secondary calibrations applying a relaxed-clock model approach, and reconstructing ancestral areas of distribution under a time-stratified likelihood model. The phylogenetic analyses provided strong support for the majority of the clades. Campylocentrum is monophyletic, and we recognize five sections based upon strongly supported clades. We conclude that the African angraecoid ancestor of Campylocentrum and Dendrophylax dispersed to the Antilles. Campylocentrum is estimated to be a relatively young genus (late Miocene, c. 8.2 Mya) and its most recent common ancestor had a disjunct distribution in the Antilles and Parana dominion. During the Pliocene, the five sections diverged and expanded their distributions in the Neotropics, and in the Pleistocene diversification was experienced by some of the terminal clades. We hypothesize that the evolutionary history of Campylocentrum was strongly influenced by orogenic events during the Pliocene and climatic fluctuations during the Pleistocene. INTRODUCTION Orchidaceae are one of the two largest plant families and occur on all continents except Antarctica (Dressler, 2005). They are particularly diverse in the Neotropics and Southeast Asia (Dressler, 1993). The Neotropical region houses the highest terrestrial biodiversity in the planet (Myers et al., 2000; Maiti & Maiti, 2011), including some hyper-diverse orchid groups such as subtribes Laeliinae, Maxillariinae, Oncidiinae and Pleurothallidinae, all endemic to this region (Pridgeon et al., 2005, 2009). Other taxa, such as tribe Vandeae, are almost entirely Palaeotropical, including the mostly Afro-Malagasy subtribe Angraecinae (Vandeae; Carlsward et al., 2006a; Pridgeon et al., 2014), but which includes two Neotropical genera, Campylocentrum Benth. and Dendrophylax Rchb.f. (Carlsward et al., 2006a; Pridgeon et al., 2014). Angraecinae consist of 49 genera (Pridgeon et al., 2014; but see also Szlachetko et al., 2013, who recognized many more). Variation in the number of genera recognized depends on the concept of the large genus Angraecum Bory, which in recent molecular phylogenetic studies was found to be non-monophyletic (Carlsward et al., 2006a; Szlachetko et al., 2013; Andriananjamanantsoa et al., 2016). Campylocentrum and Dendrophylax are embedded in Angraecum s.l. as sister to Angraecum sections Conchoglossum Schltr. and Arachnangraecum Schltr. or to the genera Angraecoides (Corden.) Szlach. Mytrik & Grochocka and Eichlerangraecum Szlach. Mytrik & Grochocka, as proposed by Szlachetko et al. (2013). Because the study by Szlachetko et al. (2013) did not include many of the other genera in Angraecinae, it cannot be considered definitive in terms of resolving the taxonomic complications of this subtribe. We prefer instead to follow the taxonomy of Pridgeon et al. (2014), although we admit that this is not a satisfactory taxonomic arrangement and must be considered a temporary solution. In Angraecinae, Neotropical Campylocentrum and Dendrophylax include the only leafless taxa, the phylogenetic relationships of which were studied by Carlsward, Whitten & Williams (2003). They found that leaves were lost twice in this clade. Campylocentrum has both leafy and leafless species, whereas Dendrophylax comprises only leafless species (Carlsward et al., 2003). Campylocentrum currently includes 70 species (Kolanowska & Szlachetko, 2013; Carlsward, 2014a; Pessoa & Alves, 2015a, b, 2016a, b), occurring in all Neotropical countries except Chile (Primavera, 2013; Carlsward, 2014a). Campylocentrum and Dendrophylax are most easily distinguished by their inflorescences. Campylocentrum has racemes with numerous, relatively small flowers that open simultaneously, whereas in Dendrophylax the flowers are fewer and often much larger. The latter has inflorescences that are often fractiflex, occasionally branched, with one to few flowers opening successively. Dendrophylax flowers vary in size among species from a few millimetres to several centimetres, including some with long nectar spurs (Carlsward & Cribb, 2014). Cogniaux (1906) proposed an infrageneric classification of Campylocentrum, organizing it in three sections based on vegetative characters (Table 1; Bogarín & Pupulin, 2010). Table 1. Sections of Campylocentrum sensuCogniaux (1906) and their main vegetative features   Roots  Stem  Leaves  C. section Campylocentrum  Smooth/granulose  Elongated  Conduplicate/terete and developed  C. section Dendrophylopsis  Smooth  Reduced  Absent  C. section Pseudocampylocentrum  Granulose  Elongated  Terete and reduced    Roots  Stem  Leaves  C. section Campylocentrum  Smooth/granulose  Elongated  Conduplicate/terete and developed  C. section Dendrophylopsis  Smooth  Reduced  Absent  C. section Pseudocampylocentrum  Granulose  Elongated  Terete and reduced  View Large Although previous molecular analyses have clarified higher taxonomic levels in Orchidaceae (reviewed by Chase et al., 2015), the majority of Neotropical genera remain unstudied at the species level. Campylocentrum constitutes an appropriate model for revealing biogeographical patterns in this area, as it is a widespread genus. Previously published phylogenetic analyses (Carlsward et al., 2003, 2006a; Neubig et al., 2009) included only a few Campylocentrum spp. Therefore, a more comprehensive and well-supported phylogenetic hypothesis is needed to provide a robust background for studying the evolution of traits and systematics of the group. The main goals of this study are: (1) to provide a phylogenetic tree based on nuclear (ribosomal ITS and low-copy Xdh exon) and plastid (matK exon, rpl32-trnL spacer, trnL intron, trnL-F spacer and ycf1 exon) DNA data; (2) to establish a new infrageneric classification for Campylocentrum; (3) to estimate divergence times of its clades and relate these dates to events in the history of the Earth; and (4) to produce a likelihood-based biogeographical analysis for reconstruction of ancestral areas of distribution and for estimation of dispersal, vicariance and peripheral isolation events that may have occurred. MATERIAL AND METHODS Taxon sampling This study includes samples for 30 of the 70 recognized Campylocentrum spp. (Carlsward, 2014a; Kolanowska & Szlachetko, 2013; Pessoa & Alves, 2015a, 2015b, 2016a, b), including all habits in the genus and all taxonomic sections proposed by Cogniaux (1906). Multiple accessions were used for 14 species with wide distributions to encompass their morphological variation (total of 55 samples of the genus). Unsampled species are all narrowly distributed and clearly closely related to species we have included. Their absence is highly unlikely to affect the conclusions drawn here. Outgroups were chosen based on Carlsward et al. (2003, 2006a) and Szlachetko et al. (2013) and included three Dendrophylax spp., shown in previous studies to be the sister group of Campylocentrum, and three representatives of African Angraecum s.l. [Angraecoides (Cordem.) Garay, Dolabrifolia Pfitzer ex Rchb.f. and Humblotiangraecum (Schltr.) Szlach. Mytrik & Grochocka sensuSzlachetko et al. (2013)], for a total of nine samples in the outgroup ( Appendix 1). Samples were obtained during fieldwork except for two herbarium specimens, some living collections (all dried and stored in silica gel) and some DNA samples available from the DNA Bank at the University of Florida, which were used to produce sequences of the additional markers used in this study (rpl32-trnL, Xdh, ycf1). The sequences of Campylocentrum previously published by Carlsward et al. (2003, 2006a) available in GenBank (ITS, matK and trnL intron, and trnL-trnF spacer) were also included, although we have updated the taxonomic names used in that study by examining the vouchers housed at FLAS. DNA extraction, amplification and sequencing DNA was extracted following a CTAB procedure (Doyle & Doyle, 1987, 1990). Total DNA was then purified using QIAquick silica columns (Qiagen, Crawley, UK). The primers of Sun et al. (1994) were used to amplify the internal transcribed spacer region of nuclear ribosomal DNA (ITS1 + 5.8S rDNA + ITS2). Amplifications were performed in a volume of 25 µL, with 12.5 µL DreamTaq Green Master Mix (Thermo Fisher Scientific, Loughborough, UK), 4 µL TBT-PAR, 0.5 µL dimethylsulphoxide (DMSO), 6.0 µL nuclease-free water, 0.5 µL each primer (10 µM) and 1 µL template DNA (30–90 ng/μL). The reaction conditions were: initial denaturation of 94 °C for 1 min, followed by 30 cycles of 94 °C for 1 min, 48 °C for 1 min and 72 °C for 90 s and a final extension of 72 °C for 4 min. The low copy nuclear gene xanthine dehydrogenase (Xdh) was amplified using primers X551 and X1591 of Górniak, Paun & Chase (2010). Amplifications were performed in the same volume and used the same reagents as for ITS. The PCR programme followed the touchdown procedure proposed by those authors: initial denaturation of 94 °C for 2 min, followed by six cycles of 94 °C for 45 s, 55–49 °C (reducing 1 °C per cycle) for 45 s and 72 °C for 90 s, then 28 cycles of 94 °C for 45 s, 49 °C for 45 s and 72 °C for 90 s and a final extension of 72 °C for 5 min. The plastid spacer rpl32-trnL was amplified with the primers UAG and F of Shaw et al. (2007); the trnL intron and trnL-F spacer were amplified as one fragment with the primers c and f or separately in some cases with the pairs e/f and c/d of Taberlet et al. (1991). Portions of the plastid matK gene were amplified using the primers 5R and XF of Ford et al. (2009). Portions of the plastid ycf1 gene were amplified with the primers 3720F and 5500R and the internal IntF and IntR of Neubig et al. (2009). Amplification of the five plastid regions was performed in a volume of 25 µL containing 12.5 µL DreamTaq Green Master Mix, 4 µL trehalose-based (TBT-PAR), 6.5 µL nuclease-free water, 0.5 µL each primer (10 µM) and 1 µL template DNA (30–90 ng/μL). The PCR programme used to amplify rpl32-trnL and trnL-F (both intron and spacer) was: initial denaturation step of 80 °C for 5 min, followed by 30 cycles of 95 °C for 1 min, 50 °C for 1 min and 65 °C for 4 min, and a final extension of 65 °C for 5 min. Amplification of matK had an initial denaturation step of 94 °C for 1 min, followed by 35 cycles of 94 °C for 30 s, 46–48 °C for 40 s and 72 °C for 40 s, and a final extension of 72 °C for 5 min. Amplification of ycf1 was carried out using a touchdown protocol following Neubig et al. (2009): 94 °C for 3 min, followed by eight cycles of 94 °C for 30 s, 60–52 °C (reducing 1 °C per cycle) for 1 min and 72 °C for 3 min, then 30 cycles of 94 °C for 30 s, 50°C for 1 min and 72 °C for 3 min, and a final extension of 72 °C for 3 min. All PCR products were purified using the QIAquick PCR purification kit (Qiagen) following the manufacturer’s protocol. Amplifications from DNA of herbarium specimens collected less than 5 years ago were all successful. Cycle sequencing was carried out using a Big Dye Terminator v. 3.1 Cycle Sequencing Kit (Applied Biosystems, ABI, Warrington, UK) using the same primers as for the amplifications. The reaction mix for the nuclear markers contained 1.5 μL 5× sequencing buffer, 0.25 μL Big Dye terminator, 0.75 μL 10 μM primer (1.5 pmol), 1–2 μL amplification product (30–90 ng/μL), 0.2 μL DMSO and 1.3 μL H2O in a total reaction volume of 5 μL. The reaction mix for the plastid markers contained 1.5 μL 5× sequencing buffer, 0.25 μL Big Dye terminator, 0.75 μL 10 μM primer, 1–2 μL amplified product (30–90 ng/μL) and 1.5 μL H2O in a total reaction volume of 5 μL. The cycle sequencing programme comprised 25 cycles of denaturation of 94 °C for 15 s, annealing of 50 °C for 5 s and elongation of 60 °C for 4 min. The cycle sequencing products were sequenced on an ABI 3720 automated DNA sequencer according to the manufacturer’s protocol. Chromatograms were edited and contigs were assembled using Geneious 8.0.4 (Biomatters, Auckland, New Zealand). Phylogenetic analysis Sequences were aligned initially using Geneious 8.0.4 (Biomatters), and a second alignment was carried out using MUSCLE (Edgar, 2004), which was subsequently manually optimized. Independent sequence matrices were compiled for each DNA region, including the sequences from GenBank. Bayesian inference (BI) and maximum parsimony (MP) analyses were performed for the combined (ITS, Xdh and plastid markers) data set using, respectively, MrBayes 3.1.2 (Ronquist & Huelsenbeck, 2003) on the CIPRES Science Gateway portal (Miller et al., 2010) and PAUP 4.0b10 (Swofford, 2002). Angraecum leonis A.H.Kent (A. section Humblotiangraecum Schltr.) was selected as the single outgroup based on the previous phylogenetic studies of Carlsward et al. (2006a) and Szlachetko et al. (2013). The MP analyses were performed via heuristic searches using 40000 random taxa additions and tree bisection reconnection (TBR) branch swapping. Bootstrap percentages (BP) were estimated with 1000 non-parametric replicates and TBR swapping. Clades with BP ≥ 85 and 0.95 posterior probability (PP; Cummings et al., 2003; Simmons et al., 2004) were considered strongly supported (Erixon et al., 2003). The best-fitting nucleotide substitution model for BI was selected using JModelTest 0.1.1 (Posada, 2008) under the Bayesian information criterion (BIC; Brown & Lemmon, 2007). The most appropriate models for each region were TrNef+G for nrITS, TrN+G for Xdh, TPM1uf+G for matK, TPM1uf for rpl32-trnL, TPM1uf+G for trnL-F and TVM+G for ycf1 (Table 2). The best model was adapted from the options available in MrBayes 3.1. for each partition (see Table 2). Table 2. Features of DNA datasets used in this study, and best fit models used to Bayesian inference (BI) for each marker   ITS  Xdh  matK  rpl32-trnL  trnL-F  ycf1  Plastid combined  All data combined  Number of taxa  63  59  62  64  63  63  64  64  Aligned length (bp)  660  908  647  688  1496  1729  4560  6128  Number of variable positions  218 (33%)  179 (19.7%)  112 (17.1%)  68 (9.8%)  453 (22.3%)  419 (24.2%)  935 (20.5%)  1326 (21.6%)  Number of potentially parsimony-informative sites  132 (20.1%)  89 (9.8%)  53 (8.2%)  30 (4.3%)  227 (15.1%)  301 (17.4%)  561 (12.3%)  782 (12.7%)  Number of changes/ variable sites  1.66  1.25  1.35  1.13  1.34  1.58  1.47  1.51  Fitch tree length  361  225  151  77  609  664  1374  1997  Consistency index  0.72  0.86  0.82  0.91  0.80  0.72  0.75  0.75  Retention index  0.90  0.93  0.92  0.95  0.90  0.91  0.90  0.90  Best fit model (BI)  HKY+G  HKY+G  GTR+G  GTR  GTR+G  GTR+I+G  –  –    ITS  Xdh  matK  rpl32-trnL  trnL-F  ycf1  Plastid combined  All data combined  Number of taxa  63  59  62  64  63  63  64  64  Aligned length (bp)  660  908  647  688  1496  1729  4560  6128  Number of variable positions  218 (33%)  179 (19.7%)  112 (17.1%)  68 (9.8%)  453 (22.3%)  419 (24.2%)  935 (20.5%)  1326 (21.6%)  Number of potentially parsimony-informative sites  132 (20.1%)  89 (9.8%)  53 (8.2%)  30 (4.3%)  227 (15.1%)  301 (17.4%)  561 (12.3%)  782 (12.7%)  Number of changes/ variable sites  1.66  1.25  1.35  1.13  1.34  1.58  1.47  1.51  Fitch tree length  361  225  151  77  609  664  1374  1997  Consistency index  0.72  0.86  0.82  0.91  0.80  0.72  0.75  0.75  Retention index  0.90  0.93  0.92  0.95  0.90  0.91  0.90  0.90  Best fit model (BI)  HKY+G  HKY+G  GTR+G  GTR  GTR+G  GTR+I+G  –  –  View Large Bayesian inference included two independent runs with four chains each with the Markov chain Monte Carlo (MCMC) parameters set to 20 million generations sampling every 10000 trees. We discarded as burn-in the first 10000 trees. Convergence between the two independent runs was checked using Tracer 1.6 (Rambaut et al., 2013). To assess possible conflicts between the independent DNA data matrices, congruence was evaluated by looking for well-supported incongruent clades among the phylogenetic trees obtained for each matrix separately. All plastid regions were treated as a single matrix since the plastid genome is inherited as a unit, and thus it is not subject to recombination (Palmer et al., 1988), whereas the two nuclear regions, ITS and Xdh, were analysed separately. Time divergence estimates and biogeographical analysis Due the lack of orchid fossils assigned to closely related species of Angraecinae (Iles et al., 2015), we built a phylogenetic tree including five additional species of orchids ( Appendix 3) that represent the major clades between Angraecinae and the closest available fossil genus (Earina Lindl.; Conran, Bannister & Lee, 2009). This matrix included only the nuclear region ITS and two plastid regions, matK and ycf1, due to the unavailability of the remaining regions included in this study in GenBank for the additional five species. One specimen per species of Campylocentrum and Dendrophylax was included, except for two widespread species, C. fasciola (Lindl.) Cogn. and C. pachyrrhizum (Rchb.f.) Rolfe, for which were included two specimens to represent the whole distribution based on the phylogenetic results; we included one sample from each of the two subclades in the combined analysis (Fig. 1). Figure 1. View largeDownload slide Phylogenetic relationships of Campylocentrum produced by Bayesian inference of nuclear (ITS and Xdh) and plastid (matK, rpl32-trnL, trnL-F, ycf1) regions combined. Posterior probabilities (≥ 0.9) are indicated above branches and maximum parsimony bootstrap percentages (≥ 80) are indicated below branches. Lateral bars: green – leaves conduplicate; red – leafless; blue – leaves cylindrical; yellow – capsules six-ribbed; pink – capsules unribbed; purple – roots smooth; orange – granulose. Figure 1. View largeDownload slide Phylogenetic relationships of Campylocentrum produced by Bayesian inference of nuclear (ITS and Xdh) and plastid (matK, rpl32-trnL, trnL-F, ycf1) regions combined. Posterior probabilities (≥ 0.9) are indicated above branches and maximum parsimony bootstrap percentages (≥ 80) are indicated below branches. Lateral bars: green – leaves conduplicate; red – leafless; blue – leaves cylindrical; yellow – capsules six-ribbed; pink – capsules unribbed; purple – roots smooth; orange – granulose. Absolute divergence times were estimated with a Bayesian approach in BEAST 1.8.0 (Drummond et al., 2012) on CIPRES. A relaxed molecular clock analysis with uncorrelated log-normal model was performed, which takes into consideration rate heterogeneity between lineages with substitution rates uncorrelated across the tree (Drummond et al., 2006), allowing the mutation rate to vary among partitions. This was supported by a coefficient of variation (CV) of 0.525; if the CV ranges between 0.1 and 1 then the relaxed clock model we applied in this analysis is appropriate (Drummond & Bouckaert, 2015). We used BEAUti 1.8 to create input files, assigning the same best-fitting models used for BI (Table 2), and constraints in topology were applied as necessary (Drummond et al., 2012) to match the topology of Epidendroideae in Freudenstein & Chase (2015) and the topology found for Campylocentrum spp. in the previous results (Fig. 1). The tree speciation prior followed the Yule process, and two MCMC chains were run for 100 million generations sampling every 10000 trees. Convergence and mixing were assessed using the effective sampling size criterion (ESS > 200) in Tracer 1.6, and processing of post-burn-in trees was performed with TreeAnnotator 1.8 (Drummond & Rambaut, 2007) to obtain a maximum clade credibility tree with mean values and 95% confidence intervals for nodal ages. Three nodes were calibrated (see Fig. 2, α, β and γ). Both primary and secondary calibrations were applied in BEAST. Only one fossil, Earina fouldenensis Conran, Bannister & D.E.Lee, from the early Miocene (23–20 Mya) of New Zealand was used (Conran et al., 2009), and it was assigned a log-normal prior distribution (Fig. 2, α) with a mean value of 23.2 Mya and standard deviation of 0.1. In addition, we used two secondary calibrations obtained from Givnish et al. (2015): the divergences of the Angraecinae + Aeridinae clades from their sister clade in Vandeae (21.21 ± 4.20 Mya) and Angraecinae and Aeridinae (13.25 ± 4.00 Mya), which were both assigned normal prior distributions (Fig. 2, β, γ). Figure 2. View largeDownload slide Divergence time estimates for Campylocentrum and related genera based on nuclear ITS, and plastid matK and ycf1 performed with BEAST. Bars represent 95% highest posterior density (HPD) estimates. The asterisk indicates a fossil calibration (α), diamonds indicate secondary calibrations, β = divergence of the clade Angraecinae + Aeridinae in the tribe Vandeae; γ = divergence of subtribes Angraecinae and Aeridinae. A geological timescale is placed at the top; vertical discontinuous grey lines separate two time slices (TSI−TSII) used in the biogeographical analysis. Figure 2. View largeDownload slide Divergence time estimates for Campylocentrum and related genera based on nuclear ITS, and plastid matK and ycf1 performed with BEAST. Bars represent 95% highest posterior density (HPD) estimates. The asterisk indicates a fossil calibration (α), diamonds indicate secondary calibrations, β = divergence of the clade Angraecinae + Aeridinae in the tribe Vandeae; γ = divergence of subtribes Angraecinae and Aeridinae. A geological timescale is placed at the top; vertical discontinuous grey lines separate two time slices (TSI−TSII) used in the biogeographical analysis. The biogeographical history of Campylocentrum was analysed using the likelihood-based dispersal–extinction–cladogenesis (DEC) model implemented in LAGRANGE v.20151128 (Ree et al., 2005; Ree & Smith, 2008), calculating global extinction and dispersal rates and ancestral range reconstructions for each node in the maximum clade credibility tree obtained from BEAST, excluding the five additional orchid species added for calibration purposes. The distribution of Campylocentrum and related genera was split into seven operational areas (Fig. 3) adapted from Morrone (2014): Sub-Saharan Africa (G); Antillean sub-region (A); Brazilian dominion (D); Chacoan dominion (E); Mesoamerica dominion (B); Pacific dominion (C); and Parana dominion (F). Species distribution ranges (presence/absence) included a maximum of three areas per species except for two widespread species, C. fasciola and C. pachyrrhizum, which included two specimens from the extremes of their distributions. Figure 3. View largeDownload slide Ancestral area estimation inferred by LAGRANGE. Pie charts at nodes indicate the probabilities of ancestral areas, when < 0.1 where combined in the ‘remainder’ category (black sections). A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. A geological timescale is placed at the top, and node numbers are indicated below branches. Vertical discontinuous grey lines separate two time slices (TSI = 7−0 Mya and TSII = 7−11 Mya), and inset maps represent the palaeogeographical configuration of the Neotropical region and Amazon Basin in these time slices. Inferred dispersal (X→Y), vicariance (X/X) and peripheral isolated speciation (X\Y) events are represented also in the tree. Figure 3. View largeDownload slide Ancestral area estimation inferred by LAGRANGE. Pie charts at nodes indicate the probabilities of ancestral areas, when < 0.1 where combined in the ‘remainder’ category (black sections). A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. A geological timescale is placed at the top, and node numbers are indicated below branches. Vertical discontinuous grey lines separate two time slices (TSI = 7−0 Mya and TSII = 7−11 Mya), and inset maps represent the palaeogeographical configuration of the Neotropical region and Amazon Basin in these time slices. Inferred dispersal (X→Y), vicariance (X/X) and peripheral isolated speciation (X\Y) events are represented also in the tree. We performed a stratified model that contained two time slices (TS): TSI (11–7 Mya, Tortonian to late Miocene) reflecting the presence of a large lake in the Amazon basin (the Acre System); and TSII (7–0 Mya, late Miocene to present) capturing the final drainage of the Acre System forming the modern Amazon and Orinoco Basins (Fig. 3; Hoorn et al., 2010). For each TS, a different matrix of bidirectional dispersal rates reproducing the geographical connectivity among the areas was applied (Viruel et al., 2015;  Appendix 4). RESULTS Molecular datasets This study produced 318 new sequences for Campylocentrum and related genera ( Appendix 1), which were combined with 57 previously published sequences (Carlsward et al., 2003, 2006a) with their species names updated in line with current taxonomic concepts. All six DNA regions were generated for all accessions except nine: five lacking only Xdh and none lacking more than one region. The complete matrix has a length of 6128 bp, of which 782 (12.7%) were potentially parsimony informative. ITS (20.1% of positions), trnL-F (15.1%) and ycf1 (17.4%) were the most informative. Due to its greater length (1729 bp aligned), ycf1 provided the greatest absolute number of informative sites (301; Table 2). Phylogenetic relationships In the parsimony analysis, the complete dataset (ITS, Xdh and plastid regions) produced 16 most-parsimonious trees of length 1997 steps with a consistency index (CI) = 0.75 and retention index (RI) = 0.90. Analyses performed separately for each region show similar indices (Table 2). The BI and MP trees obtained from all data combined had similar topologies (Figs 1, S1). Most relationships among clades were similar also to the combined plastid dataset (Fig. S4); however, the independent nuclear analyses (ITS and Xdh) were less well resolved (Figs S2, S3). The most congruent topology among all analyses resolved Campylocentrum (BP 100, PP 1.00) and Dendrophylax (BP 97, PP 1.00) as sister clades (Fig. 1). Successively diverging from the base are Angraecum moandense De Wild. (BP 100, PP 1.00), A. distichum Lindl. (BP 100 %, PP 1) and A. leonis. When performed separately, the Xdh and ITS analyses show A. moandense in a polytomy with Campylocentrum and Dendrophylax, so consequently these nuclear regions did not identify the Neotropical clade (Campylocentrum + Dendrophylax) observed with the plastid regions (Figs S2–S4). Two major clades occur in Campylocentrum (A and B), both strongly supported in the combined BI and MP trees (BP 100, PP 1.00). A similarly strongly supported topology is observed in the separate ITS, Xdh and plastid trees (Figs 1, 2, S2–S4). However, the ITS analysis did not produce a monophyletic Campylocentrum; clades A and B fall in a polytomy with Dendrophylax and A. moandense (Fig. S2). Clade A comprises C. section Campylocentrum as sister to C. sections Dendrophylopsis Cogn. in Mart. + Pseudocampylocentrum Cogn. in Mart. in the combined BI tree, a result also observed in the plastid tree (Figs 1, S4), but the Xdh and ITS (individual) trees make these into polytomies (Figs 2, 3). Although section Campylocentrum has a high PP (0.98) with all data combined, in the combined MP analysis it is split into two clades along a polytomy: one includes C. calostachyum (Barb.Rodr.) E.M.Pessoa & M.Alves, C. serranum E.Pessoa & Alves and C. ulaei Cogn. (BP 100, PP 1.00) and other with the remaining species (BP 92, PP 1.00). Among these two clades in the same polytomy is also C. sections Dendrophylopsis + Pseudocampylocentrum with moderate to strong support (BP 83, PP 0.99; Figs 1, S1). Conversely, the plastid tree provides good resolution for C. section Campylocentrum with strong support in both BI and MP analyses (BP 97, PP 1.00; Fig. S4). Clade B includes two new sections described in the taxonomic treatment below, C. sections Laevigatum E.Pessoa & M.W.Chase and Teretifolium E.Pessoa & M.W.Chase, both strongly supported in the combined BI and MP analysis (BP 99, PP 1.00 and BP 100, PP 1.00, respectively; Figs 1, S1). Both sections are also strongly supported in the plastid tree (BP 99, PP 1.00 and BP 100, PP 1.00, respectively; Fig. S4). The ITS and Xdh trees include only C. section Teretifolium as monophyletic with good support (BP 88, PP 1.00 and BP 86, PP 1.00, respectively) (Figs S2, S3). The datasets used in this study were not able to clarify relationships among taxa in C. section Teretifolium, except for C. sellowii (Rchb.f.) Rolfe as sister to the rest of this clade (BP 100, PP 1.00), which is internally unresolved due to low genetic variation. A similar topology is provided by the combined plastid tree (Fig. S4). Dating and biogeography The divergence time between African A. moandense and the Neotropical clade formed by Campylocentrum and Dendrophylax was estimated as late Miocene at 8.53 Mya [95% highest posterior density (HPD) = 10.7–6.4 Mya]. The ancestral area estimation suggests a first long-distance dispersal from Africa (G) to the Antilles (A) (node 3, probability = 0.7, Figs 2, 3). The most recent common ancestor (MRCA) of Campylocentrum and Dendrophylax was most probably restricted to the Antilles (A), and the estimated split for these two genera was almost contemporary with the previous node, 8.21 Mya (95% HPD = 10.5–5.9 Mya), supporting a hypothesis of rapid divergence after colonization of the Neotropics (node 4, probability = 0.61, Figs 2, 3). The MRCA of Campylocentrum had a possible disjunct distribution between the Antilles (A) and Parana (F) dominions, suggesting a second long-distance dispersal (A→F; node 7, probability = 0.2, Figs 2, 3). Clades A (C. section Campylocentrum + C. section Dendrophylopsis + C. section Pseudocampylocentrum) and B (C. section Laevigatum + C. section Teretifolium) also diverged in the late Miocene, 7.22 Mya (95% HPD = 9.2–5.2 Mya), with a subsequent dispersal and expansion to the Chacoan dominion (E) and Central America (B), respectively (node 8, probability = 0.4; node 20, probability = 0.26, Figs 2, 3). These clades started their diversification in the Miocene/Pliocene and late Miocene, 2.9 Mya (95% HPD = 4.2–1.7 Mya) and 5.38 Mya (95% HPD = 6.79–3.97 Mya), respectively (nodes 8 and 20, Figs 2, 3). Divergence of the MRCA of C. sections Dendrophylopsis + Pseudocampylocentrum was probably 4.79 Mya (95% HPD = 6.10–3.48 Mya), and it was most probably distributed exclusively in Central America (B), which is also the case for the MCRAs for each of these sections. The wide distribution of some species in these sections was probably reached in the Pleistocene (node 21, probability = 0.43; node 22, probability = 0.47, Figs 2, 3). In C. section Campylocentrum, two main clades also diverged in the late Miocene/Pliocene by vicariance, 5.16 Mya (95% HPD = 6.6–3.7 Mya; node 27, probability = 0.42, Figs 2, 3), one clade formed by C. calostachyum, C. serranum and C. ulaei with an origin in the Parana dominion (F) and an MRCA from 2.69 Mya (95% HPD = 3.8–1.6 Mya; node 28, probability = 0.96, Figs 2, 3), and the second clade with an MCRA from 4.28 Mya (95% HPD = 4.60–2.32 Mya) originating in Central America (B), after which it diversified rapidly in the Pleistocene (node 30, probability = 0.35, Figs 2, 3). In clade B, the MRCA of both C. sections Laevigatum and Teretifolium also had an estimated origin and diversification in the Pleistocene, 1.76 Mya (95% HPD = 2.7–0.8 Mya) and 2.04 Mya (95% HPD = 2.9–1.18 Mya), respectively, mainly in Parana (F) (node 9, probability = 0.83; node 13, probability = 0.46, Figs 2, 3). Low resolution in reconstructing distributions in the terminal clades of C. section Dendrophylopsis and the ‘C. micranthum (Lindl.) Maury complex’ was observed due to the wide distribution of the species and poor sampling of the leafless species. Only the original area is known at this point, and considering the putative subsequent expansions more detailed studies focused on these groups might provide a better understanding of their evolutionary history. Divergence times, 95% HPDs and ancestral areas with their probabilities for all tree nodes are presented in  Appendix 2. DISCUSSION Phylogenetic relationships Phylogenetic analysis of the combined data provides good resolution for the majority of clades, which allowed us to discuss and define with a high level of confidence some taxonomic groups. The results support the African A. moandense [cited as A. chevalieri Summerh. by Carlsward et al., 2006a; see Pessoa & Alves, 2016c; A. section Angraecoides (Cordem.) Garay sensuGaray (1973) (= genus Angraecoides sensuSzlachetko et al., 2013)], as sister to the Atlantic clade (Campylocentrum +Dendrophylax; Carlsward et al., 2006a; Szlachetko et al., 2013; Andriananjamanantsoa et al., 2016). Angraecum section Angraecoides includes species mainly distributed in West Tropical Africa (Govaerts et al., 2016) and is characterized by an often one-flowered inflorescence, green to greenish yellow flowers and cylindrical or clavate spurs (Garay, 1973). Their floral morphology resembles that of Dendrophylax (e.g. D. barrettiae Fawc. & Rendle), although their leafy habit is similar to some Campylocentrum spp. (e.g. C. crassirhizum Hoehne). Campylocentrum and Dendrophylax are demonstrated to be monophyletic, as first shown by Carlsward et al. (2003, 2006a) and confirmed by Szlachetko et al. (2013) and Andriananjamanantsoa et al. (2016) who included a better sampling of Angraecum s.l. Despite the shared leafless condition, the genera are easily distinguished by their inflorescence arrangement, mainly many-flowered with flowers densely and distichously arranged in Campylocentrum in contrast to few-flowered, loosely and not distichously arranged in Dendrophylax (Carlsward, 2014a, b). Inflorescence structure and flower size can be considered as synapomorphies for Campylocentrum in Angraecinae. This study presents the first phylogenetic analysis of Campylocentrum including a representative sample of its morphological diversity and distribution. The missing taxa are probably either synonyms of widely distributed species or narrow endemics closely related to the species included. In the genus, two major clades are strongly supported (A and B). Both were also recovered by Carlsward et al. (2003). However, their study included only six species, of which samples of C. stenanthum Schltr. were mis- identified as C. micranthum, C. robustum Cogn. and C. schiedei (Rchb.f.) Benth. ex Hemsl. (vouchers checked at FLAS). For clade B, those authors used three samples including C. jamaicense (Rchb.f. ex Griseb.) Benth. ex Fawc. (misidentified as C. micranthum) and C. crassirhizum [as C. lansbergii (Rchb.f.) Schltr.] (vouchers checked at FLAS). The lack of a generic revision is the main reason for this misinterpretation of species identities. Pessoa, Maciel & Alves (2015) and Pessoa & Alves (2015a, b) studied C. micranthum, which was recognized as a ‘nomenclatural complex’ requiring new combinations, re-establishments of some old names and re-circumscription of most species. Most vegetative and floral features previously used in taxonomic treatments in the genus (Cogniaux, 1906; Todzia, 1980; Bogarín & Pupulin, 2010; Pessoa & Alves, 2016a, b) are here homoplastic, consequently having limited use for delimiting natural taxa. However, capsule morphology clearly distinguishes clade A (six-ribbed capsule) from clade B (non-ribbed). A non-ribbed capsule is thus interpreted here as the synapomorphy for clade B, whereas a six-ribbed capsule is a symplesiomorphy since it is similar to those of Dendrophylax and Angraecum. Clade A is widespread in the Neotropics and composed of the three sections proposed by Cogniaux (1906). The sister of the rest of this clade, C. section Campylocentrum, is the most diverse section in the genus and includes species with conduplicate leaves and elongate stems, but its circumscription sensuCogniaux (1906) is found not to be monophyletic. It is re-circumscribed here, excluding species with non-ribbed capsules (clade B). With this new circumscription C. section Campylocentrum is strongly supported in the combined BI analysis and in MP and BI for the plastid data. However, it is not supported in the combined MP analysis, a result of the polytomy including the C. ulaei group (C. calostachyum, C. serranum and C. ulaei). Three strongly supported major subclades are recognized in C. section Campylocentrum. Sister to the rest is the C. ulaei group, endemic to Parana (F) and characterized by relatively longer inflorescences. The sister group of this clade is mainly distributed outside the Parana dominion (except for C. micranthum) and is split into two subclades. The first (C. tenellum Todzia, C. brenesii Schtlr.) includes species from the Antilles (A), Central America (B) and the Pacific dominions (C). The other is the C. micranthum complex (Pessoa et al., 2015), which includes seven species sampled here with relatively shorter and congested inflorescences. In the C. micranthum complex, C. huebneri Mansf. + C. mattogrossense Hoene are sister to the rest, which includes the other five species sampled. One of the latter is C. schiedei, the type of the genus. Among C. panamense Ames, C. micranthum, C. kuntzei Cogn. ex Kuntze and C. stenanthum, there is a clear difference in spur morphology; C. huebneri has the longest spur in the genus (1.0–2.0 cm long vs. 0.2–0.6 cm long), and C. mattogrossense has a straight spur (vs. commonly slightly curved to inflexed in the other species; Pessoa & Alves, 2015a, b). The concept of C. micranthum has been widely and incorrectly applied (Bogarín & Pupulin, 2010; Pessoa & Alves, 2015a, b;,Pessoa et al., 2015), and our results support some of the proposed names in the complex, although many others were not included in this study. A broad study of the group is required to better understand the species boundaries. Campylocentrum section Campylocentrum is sister to a clade formed by C. sections Dendrophylopsis + Pseudocampylocentrum. The leafless species of Campylocentrum are included in C. section Dendrophylopsis, which is also distinguished by having a viscidium comprising a single part (Pessoa & Alves, 2016b) and roots with thicker cell walls in the endovel- amen, exoderm and endoderm (Pessoa et al., 2017). All these features are considered synapomorphies of C. section Dendrophylopsis. Two subclades occur in this section, one formed by C. grisebachii Cogn. + C. pachyrrhizum that is characterized by the apex of the anther cap rounded or truncate and a second one with C. fasciola + C. tyrridion Garay & Dunst. ex Foldats that has a bilobed anther cap (Pessoa & Alves, 2016b). Both clades are supported in the BI analyses but not by MP. The monospecific C. section Pseudocampylocentrum morphologically resembles species placed in this study in C. section Teretifolium (clade B) based on its cylindrical leaves and granulose root exterior. Although externally similar, the root surface is composed of unicellular absorbent hairs, whereas in C. section Teretifolium roots have tufts of epivelamen (Pessoa et al., 2017). Terete leaves seem to be an adaptive convergence and are also found in other genera in Angraecinae [e.g. Nephrangis (Schltr.) Summerh. and Solenangis Schltr.]. Our results confirm that leaflessness and terete leaves are homoplastic, arising more than once in the subtribe (Carlsward et al., 2006a; Pessoa & Alves, 2016a). Clade B is almost restricted to the Parana dominion (F), except for C. jamaicense endemic to the Antilles. It includes C. sections Laevigatum and Teretifolium, both strongly supported and characterized by non-ribbed capsules. Campylocentrum section Teretifolium is distinguished from its sister by having cylindrical leaves and granulose to tuberculate root surfaces. Although a taxonomic treatment exists for the group (Pessoa & Alves, 2016a), relationships among its species are still poorly resolved. Our study was unable to disentangle it, except for recognizing C. sellowii as sister to the other four species included. Low interspecific genetic divergence probably reflects rapid and recent diversification producing a lack of resolution (Vatanparast et al., 2011; Matos-Maraví et al., 2013; Snak et al., 2016; Liu et al., 2016). Two species of C. section Laevigatum have been previously considered synonyms of C. micranthum (C. section Campylocentrum), C. brevifolium and C. jamaicense. Both are similar in habit, especially their conduplicate leaves with asymmetrically bilobed apices, as observed in the majority of genera in Angraecinae. However, contrary to our expectation, species of these sections are not closely related to each other in the tree (Fig. 1). Campylocentrum section Laevigatum is distinguished mainly by non-ribbed capsules in contrast to six-ribbed capsules in C. section Campylocentrum. It also differs by roots with ○-thickened cell walls in the exoderm instead of ∩-thickened cell walls as in other species of the genus (Pessoa et al., 2017). Although only two species have been anatomically analysed [C. jamaicense by Carlsward, Stern & Bytebier (2006b) and C. crassirhizum by Pessoa et al. (2017)], this character is suggested here as a possible synapomorphy for the section. Campylocentrum section Laevigatum has two subclades. One of them is C. pauloense Hoehne & Schltr. + C. spannagelii Hoehne + C. densiflorum Cogn. + C. brachycarpum Cogn., all four endemic to the southern Parana dominion (northern Argentina, southern Brazil and Paraguay). Campylocentrum brachycarpum and C. densiflorum represent in this study the C. organense (Rchb.f.) Rolfe group (C. organense absent from this study) characterized by relatively large, ovate floral bracts that cover all or most of the ovary (vs. relatively shorter and deltoid in other leafy species). The second subclade includes C. neglectum + C. robustum + C. crassirhizum + C. jamaicense, mainly distributed in the Atlantic Forest and the South American ‘dry diagonal’ (formed by caatinga, cerrado and chaco vegetation sensuPrado & Gibbs, 1993; the Chacoan domain), except for C. jamaicense, which occurs in the Antilles. Campylocentrum crassirhizum and C. jamaicense, despite their large disjunction, are similar morphologically and have been distinguished only by spur length. Our analyses did not provide a clear understanding of the relationship between these species. The two accessions of C. crassirhizum did not fall together: the one from southeastern Brazil is sister to the other accession from northeastern Brazil plus the specimens of C. jamaicense. However, this topology is not supported by MP, and more studies including population genetics will be important to understand relationships of these two species. Divergence time and biogeography The MRCA of A. moandense and the Neotropical genera of Angraecinae arrived in the Antilles (A) during the late Miocene (c. 8.53 Mya; Fig. 3). This date implies a relatively recent trans-Atlantic long- distance dispersal event like that of Pitcairnia feliciana (A.Chev.) Harms & Mildbraed (Bromeliaceae) c. 9.3 Mya (Givnish et al., 2011) and Maschalocephalus Gilg. & K.Schum. (Rapateaceae) c. 7.3 Mya (Givnish et al., 2004). However, for these two species the migration events occurred in the opposite direction from that proposed here for Angraecinae and were not followed by diversification (Givnish et al., 2004, 2011). Moreover, although it occurred much earlier in the Oligocene/Miocene (c. 35 Mya), Caricaceae (the papaya family) are a good example of an African taxon that experienced a radiation in the Neotropics (Carvalho & Renner, 2012; Christenhusz & Chase, 2013). Carvalho & Renner (2012) suggested that the ancestor of the group arrived first in the Antilles, in agreement with the results here for Angraecinae. In the Antilles (A), divergence of Campylocentrum and Dendrophylax (c. 8.22 Mya) occurred soon after dispersal of their MRCA from Africa (c. 8.53 Mya). The MRCA of Campylocentrum reached the Parana dominion (F) through a second long-distance dispersal also in the late Miocene (c. 7.22 Mya; Fig. 3). According to Pound et al. (2011) in this epoch a portion of the present northern coast of Brazil (Maranhão and Piauí States) was inundated. The authors indicated that, based on a data/model biome reconstruction, the ancient coast was covered by tropical evergreen forests. It is possible that the MRCA of Campylocentrum arrived in that region instead of the coast of the Guyanas, which even though closest to the Antilles was covered at that time by tropical savannas. The once-flooded area and the ancient coast were probably where the ancestor arrived first in South America and are currently occupied by cerrado and caatinga vegetation (the Chacoan dominion sensuMorrone, 2014). Extinction events caused by climate and consequently vegetation changes (Hoorn et al., 2010) could explain the current absence of the genus in this area. Migrations from Mesoamerica (B) to Paraná (F) took place during the late Miocene, Pliocene and Pleistocene in clade A, when the Andes experienced accelerated rates of uplift (Hoorn et al., 2010). This could suggest that the Andes might have not been a completely impassable geographical barrier for genetic exchange, as reported by Sánchez-Baracaldo (2004) for ferns and Pérez-Escobar et al. (2017) for Cymbidieae and Pleurothallidinae (Orchidaceae), who identified multiple migrations and re-colonizations over the Andes across long timescales. In the Pliocene, clade A expanded its distribution from the Antilles (A) to Mesoamerica (B) and clade B from Parana (F) to Chacoan dominion (E). In both cases, they were followed by a rapid diversification, originating by peripheral isolation events in the five main clades recognized in the genus (here considered as C. sections Campylocentrum, Dendrophylopsis, Laevigatum, Pseudocampylocentrum and Teretifolium; Fig. 3). According to Hoorn et al. (2010) and Turchetto-Zolet et al. (2013), climate changes and orogenic events with consequent changes in the drainage systems and vegetation (Salzmann et al., 2008) occurred in the Pliocene and contributed to shaping the current diversity of plant lineages in the Neotropics. In this region, several taxonomic groups exhibited diversification in this epoch, including bellflowers (Campanulaceae; Lagomarsino et al., 2016), tank-epiphytic bromeliads (Givnish et al., 2011, 2014), some Fabaceae such as Chamaecrista (L.) Moench. series Coriaceae (Benth.) H.S.Irwin & Barneby (Rando et al., 2016), Platymiscium Vogel (Saslis-Lagoudakis et al., 2008) and Canavalia Adans. (Snak et al., 2016), and Solanaceae subtribe Physalinae (Zamora-Tavares et al., 2016). Central America (B) was an important centre of origin for some clades of Campylocentrum, especially those in clade A. During the Pliocene, sections Campylocentrum (in part, c. 4.28 Mya, node 30, except for the C. ulaei group), Dendrophylopsis (c. 4.06 Mya) and Pseudocampylocentrum (c. 4.79 Mya) expanded their distributions from this area followed by diversification (Fig. 3). Orogenic events occurring in the Pliocene, such as the definitive closure of the Isthmus of Panama (Knowlton & Weigt, 1998; although other authors have stated recently that it was earlier, Montes et al., 2015) and volcanism (Weinberg, 1992; Vogel et al., 2006; Mann, 2007) seem to have been relevant to the diversification of these groups in the area. Nevertheless, Parana (F) was the centre of origin for C. section Teretifolium (c. 1.76 Mya) and part of C. section Laevigatum (node 17, c. 1.23 Mya), with evidence of in situ diversification in the Pleistocene. The C. ulaei group (clade A, C. section Campylocentrum) is also endemic to Parana (F). It diverged from its sister by a vicariant event (B/F) in the early Pliocene, and the species diverged during the Plio-Pleistocene, which matches groups in clade B (Fig. 3). Pleistocene climatic fluctuations (Behling & Negrelle, 2001) played an important role in speciation in the Parana dominion (F), as argued by Hooghiemstra & van der Hammen (1998). At that time, the glacial and interglacial periods may have resulted in successive fragmentation and expansion of forests, promoting allopatric and parapatric speciation, also termed as the refuge theory (Haffer, 1969; Prance, 1973). The climatic fluctuations that occurred during the Pleistocene were also important in the evolution of the C. micranthum complex (clade A, C. section Campylocentrum). It experienced an extensive diversification with expansion of its distribution from Mesoamerica (B), Pacific (C) and Brazil (D), ranging to the Antilles (A), Chacoan (E) and Paraná dominions (F) in the last 1 My (Fig. 3). In a study of ecological niche modelling, Kolanowska (2015) suggested several refugia for two species of the C. micranthum complex at the last glacial maximum. Although with a restricted approach, they cited the presence of several refugia for the group in Mesoamerica (B) and Pacific (C), a result supported by our study. Similar rapid diversification in the last 1 My was experienced by some Andean clades of Campanulaceae (Lagomarsino et al., 2016) and Brazilian saxicolous Chamaecrista series Coriaceae (Rando et al., 2016). However, the diversification of the latter was not followed by range expansion, and their diversity more probably arose in situ. CONCLUSIONS Phylogenetic reconstruction of Campylocentrum confirmed it as monophyletic in its current circumscription. It is composed of two major clades, one that includes the three sections previously proposed by Cogniaux (1906) and the second with two new sections, C. sections Laevigatum and Teretifolium, first proposed in this study (see taxonomic treatment below; these species had been included previously in C. section Campylocentrum). Although the majority of the morphological characters are here considered homoplastic, the two main clades included in the genus are clearly distinguished by their capsules (six-ribbed vs. non-ribbed). The sections are recognized only by sets of features, except for C. section Dendrophylopsis that exhibits clear synapomorphies such as leaflessness, one-parted viscidia and roots with endovelamen, exoderm and endoderm cell walls thicker than those in the leafy species. The divergence time analysis and ancestral distribution reconstruction indicated that Campylocentrum is relatively young and arose in the late Miocene. Its MRCA probably had a disjunct distribution between the Antilles and Paraná dominions. During the Pliocene, the five sections of the genus had already diverged and expanded their distributions to the Mesoamerican, Chacoan, Pacific and Brazilian dominions. Climatic fluctuations in the Pleistocene probably played an important role in spurring diversification, especially in the sections of clade B in Parana (F) and the C. micranthum complex (clade A) in the Mesoamerican, Pacific and Brazilian dominions. Therefore, the evolutionary history of Campylocentrum would seem to confirm the importance of orogenic events occurring in the Pliocene and climatic fluctuations in the Pleistocene for diversification in the Neotropics. TAXONOMIC TREATMENT 1 Campylocentrum Benth., J. Linn. Soc. Bot. 18: 337. 1881. Type species: Campylocentrum schiedei (Rchb.f.) Benth. ex Hemsl. (basionym: Angraecum schiedei Rchb.f.; originally published as Todaroa micrantha A.Richard & Galeotti). Todaroa A.Rich. & Galeotti, Ann. Sci. Nat. Bot. 3: 28. 1845. nom. illeg. [non Todaroa Parl. Hist. Nat. ìles Canaries 2: 155. 1843. Apiaceae (= Umbelliferae)]. Type species: Todaroa micrantha A.Rich. & Galeotti, nom. illeg. 1.1. Campylocentrum section Campylocentrum This widespread section is characterized by leafy species with cylindrical and smooth roots, conduplicate leaves and six-ribbed capsules. It includes about the half of the species of the genus (38 species). List of species: C. amistadense Bogarín, C. antioquiense Kolan. & Szlach., C. apiculatum Schltr., C. asplundii Dodson, C. brenesii Schltr., C. calostachyum (Barb.Rodr.) ined., C. carlos-parrae Kolan. & Szlach., C. cornejoi Dodson, C. ecuadorense Schltr., C. embreei Dodson, C. escobariae Kolan. & Szlach., C. hirtellum Cogn., C. hirtzii Dodson, C. hondurense Ames, C. huebneri Mansf., C. huebnerioides D.E.Benn. & Christenson, C. lansbergii (Rchb.f.) Schltr., C. kuntzei Cogn. ex Kuntze, C. madisonii Dodson, C. mattogrossense Hoehne, C. micranthum (Lindl.) Maury, C. microphyllum Ames & Correll, C. minus Fawc. & Rendle, C. natalieae Carnevali & I.Ramírez, C. palominoi Kolan., O.Pérez & E.Parra, C. panamense Ames, C. peniculus Schltr., C. polystachyum (Lindl.) Rolfe, C. pugioniforme (Klotzsch) Rolfe, C. pygmaeum Cogn., C. queremalense Kolan. & Szlach., C. serranum E.M.Pessoa & M.Alves, C. schiedei (Rchb.f.) Benth. ex Hemsl., C. schneeanum Foldats, C. stenanthum Schltr., C. steyermarkii Foldats, C. tenellum Todzia, C. ulaei Cogn. 1.2. Campylocentrum section Dendrophylopsis Cogn. in Mart., Fl. Bras. 3(6): 504. 1906. Type species: Campylocentrum fasciola (Lindl.) Cogn., designated by Pessoa & Alves (2016b). This widespread section includes the leafless species with dorsiventrally flattened or cylindrical, smooth roots, leaves always reduced to achlorophyllous scales and six-ribbed capsules. It comprises13 species. List of species: C. amazonicum Cogn., C. benelliae E.M.Pessoa & M.Alves, C. fasciola (Lindl.) Cogn., C. fernandezii Kolan. & Szlach., C. generalense D.Bogarín & F.Pupulin, C. insulare Siqueira & E.M.Pessoa, C. minutum C.Schweinf., C. pachyrrhizum (Rchb.f.) Rolfe, C. paludosum E.M.Pessoa & M.Miranda, C. pubirhachis Schltr., C. tenue (Lindl.) Rolfe, C. tyrridion Garay & Dunst. ex Foldats 1.3. Campylocentrum section Laevigatum E.M.Pessoa & M.W.Chase, sect. nov. Type species: Campylocentrum brevifolium (Lindl.) E.M.Pessoa & M.Alves, Kew Bull. 70: 43. 2015. Similar to C. section Campylocentrum but differs by its unribbed capsules (vs. six-ribbed). This section is almost restricted to eastern South America, except for C. jamaicense from the Antilles; it includes leafy species with cylindrical, smooth roots, conduplicate leaves and unribbed capsules. It is composed of 15 species. List of species: C. brevifolium (Lindl.) E.M.Pessoa & M.Alves, C. brachycarpum Cogn., C. carvalhoi E.M.Pessoa & M.Alves, C. crassirhizum Hoehne, C. densiflorum Cogn., C. hasslerianum Hoehne, C. itatiaiae E.M.Pessoa & M.Alves, C. intermedium (Rchb.f. & Warm.) Cogn., C. jamaicense (Rchb.f. ex Griseb.) Benth. ex Fawc., C. neglectum (Rchb.f. & Warm.) Cogn., C. organense (Rchb.f.) Rolfe, C. pauloense Hoehne & Schltr., C. robustum Cogn., C. schlechterianum E.M.Pessoa & M.Alves, C. spannagelii Hoehne 1.4. Campylocentrum section Pseudocampylocentrum Cogn. in Mart., Fl. Bras. 3(6): 504. 1906. Type species: Campylocentrum poeppigii (Rchb.f.) Rolfe. This monospecific section is distributed in the Antilles, Central America and northern South America. The only taxon recognized so far is a leafy species, with cylindrical, minutely granulose roots, terete, reduced leaves and six-ribbed capsules. List of species: C. poeppigii (Rchb.f.) Rolfe 1.5. Campylocentrum section Teretifolium E.M.Pessoa & M.W.Chase, sect. nov. Type species: Campylocentrum ornithorrhynchum (Lindl.) Rolfe, Orchid Review 11: 246. 1903. Similar to C. section Pseudocampylocentrum, but differs by its granulose root surface, comprising tufts of epivelamen (vs. comprising unicellular absorbent hairs) and unribbed capsules (vs. six-ribbed). This section is endemic to eastern South America. It includes leafy species with cylindrical, minutely granulose to tuberculate roots, terete leaves and unribbed capsules. It includes six species. List of species: C. labiakii E.M.Pessoa & M.Alves, C. ornithorrhynchum (Lindl.) Rolfe, C. parahybunense (Barb.Rodr.) Rolfe, C. pernambucense Hoehne, C. sellowii (Rchb.f.) Rolfe, C. wawrae (Rchb.f. ex Beck) Rolfe SUPPLEMENTARY INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Figure S1. Strict consensus tree from MP analysis of nuclear (ITS and Xdh) and plastid (matK, rpl32-trnL, trnL-F, ycf1) regions combined (length = 1997 steps, CI = 0.75 and RI = 0.90). Numbers below the branches are bootstrap percentages (≥50). Figure S2. Phylogenetic relationships of Campylocentrum obtained by Bayesian inference of ITS data. Posterior probabilities (≥0.5) are indicated above branches and maximum parsimony bootstrap percentages (≥50) are indicated below branches. Figure S3. Phylogenetic relationships of Campylocentrum obtained from a Bayesian inference of Xdh data. Posterior probabilities (≥0.5) are indicated above branches and maximum parsimony bootstrap percentages (≥50) are indicated below branches. Figure S4. Phylogenetic relationships of Campylocentrum obtained from by Bayesian inference of combined plastid regions (matK, rpl32-trnL, trnL-F, ycf1). Posterior probabilities (≥0.5) are indicated above branches and maximum parsimony bootstrap percentages (≥50) are indicated below branches. ACKNOWLEDGMENTS We thank the staff of the Jodrell Laboratory (Royal Botanic Gardens, Kew) for help and useful feedback. The directors/curators of living collections at the Jardim Botânico do Rio de Janeiro, Instituto de Botânica de São Paulo, University of Costa Rica (project number 814-B5-A87) and Heidelberg University allowed us to collect tissue for DNA extractions. Thanks to those who funded our fieldwork including the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), the US National Science Foundation (DEB-0946618), Velux Stiftung and the Beneficia Foundation. The first author thanks CNPq and CAPES for a PhD fellowship. REFERENCES Andriananjamanantsoa HN, Engberg S, Louis EEJr, Brouillet L. 2016. Diversification of Angraecum (Orchidaceae, Vandeae) in Madagascar: revised phylogeny reveals species accumulation through time rather than rapid radiation. PLoS One  11: e0163194. Google Scholar CrossRef Search ADS PubMed  Behling H, Negrelle RRB. 2001. Tropical rain forest and climate dynamics of the Atlantic lowland, southern Brazil, during the late Quaternary. Quaternary Research  56: 383– 389. Google Scholar CrossRef Search ADS   Bentham G. 1881. Notes on Orchidaceae. Journal of the Linnean Society, Botany  18: 281– 361. Google Scholar CrossRef Search ADS   Bogarín D, Pupulin F. 2010. The genus Campylocentrum (Orchidaceae: Angraecinae) in Costa Rica: a revision. Harvard Papers in Botany  15: 353– 414. Google Scholar CrossRef Search ADS   Brown JM, Lemmon AR. 2007. The importance of data partitioning and the utility of Bayes factors in Bayesian phylogenetics. Systematic Biology  56: 643– 655. Google Scholar CrossRef Search ADS PubMed  Carlsward BS. 2014a. Campylocentrum. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN, eds. Genera orchidacearum, Epidendroideae (part 3) . New York: Oxford University Press, 370– 372. Carlsward BS. 2014b. Dendrophylax. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN, eds. Genera orchidacearum, Epidendroideae (part 3) . New York: Oxford University Press, 383– 385. Carlsward BS, Cribb P. 2014. Angraeciinae. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN, eds. Genera orchidacearum, Epidendroideae (Part 3) . New York: Oxford University Press, 338– 344. Carlsward BS, Stern WL, Bytebier B. 2006b. Comparative vegetative anatomy and systematics of the angraecoids (Vandae, Orchidaceae) with an emphasis on the leafless habit. Botanical Journal of the Linnean Society  151: 165– 218. Google Scholar CrossRef Search ADS   Carlsward BS, Whitten WM, Williams NH. 2003. Molecular phylogenetics of Neotropical leafless Angraecinae (Orchidaceae): reevaluation of generic concepts. International Journal of Plant Sciences  164: 43– 51. Google Scholar CrossRef Search ADS   Carlsward BS, Whitten WM, Williams NH, Bytebier B. 2006a. Molecular phylogenetics of Vandeae (Orchidaceae) and the evolution of leaflessness. American Journal of Botany  93: 770– 786. Google Scholar CrossRef Search ADS   Carvalho FA, Renner SS. 2012. A dated phylogeny of the papaya family (Caricaceae) reveals the crop’s closest relatives and the family’s biogeographic history. Molecular Phylogenetics and Evolution  65: 46– 53. Google Scholar CrossRef Search ADS PubMed  Chase MW, Cameron KM, Freudenstein JV, Pridgeon AM, Salazar G, Van den Berg C, Schuiteman A. 2015. An updated classification of Orchidaceae. Botatical Journal of the Linnean Society  177: 151– 174. Google Scholar CrossRef Search ADS   Christenhusz MJM, Chase MW. 2013. Biogeographical patterns of plants in the Neotropics – dispersal rather than plate tectonics is most explanatory. Botanical Journal of the Linnean Society  171: 277– 286. Google Scholar CrossRef Search ADS   Cogniaux CA. 1906. Campylocentrum. In: Martius CFP, Eichlerand AW, Urban I, eds. Flora Brasiliensis 3(6) . Munich and Leipzig: Typographia Regia Wolf et fil., 503– 524. Conran GC, Bannister JM, Lee DE. 2009. Earliest orchid macrofossils: early Miocene Dendrobium and Earina (Orchidaceae: Epidendroideae) from New Zealand. American Journal of Botany  96: 466– 474. Google Scholar CrossRef Search ADS PubMed  Cummings MP, Handley SA, Myers DS, Reed DL, Rokas A, Winka K. 2003. Comparing bootstrap and posterior probability values in the four-taxon case. Systematic Biology  52: 477– 487. Google Scholar CrossRef Search ADS PubMed  Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin, Botanical Society of America  19: 11– 15. Doyle JJ, Doyle JL. 1990. Isolation of plant DNA from fresh tissue. Focus  12: 13− 15. Dressler RL. 1993. The orchids: natural history and classification . Cambridge: Harvard University Press. Dressler RL. 2005. How many orchid species? Selbyana  26: 155– 158. Drummond AJ, Bouckaert RR. 2015. Computational evolution with BEAST . Cambridge: Cambridge University Press. Drummond AJ, Ho SY, Phillips MJ, Rambaut A. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biology  4: e88. Google Scholar CrossRef Search ADS PubMed  Drummond AJ, Rambaut A. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology  7: 214. Google Scholar CrossRef Search ADS PubMed  Drummond AJ, Suchard MA, Xie D, Rambaut A. 2012. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution  29: 1969– 1973. Google Scholar CrossRef Search ADS PubMed  Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic acids research  32: 1792– 1797. Google Scholar CrossRef Search ADS PubMed  Erixon P, Svennblad B, Britton T, Oxelman B. 2003. Reliability of Bayesian posterior probabilities and bootstrap frequencies in phylogenetics. Systematic Biology  52: 665– 673. Google Scholar CrossRef Search ADS PubMed  Ford CS, Ayres KL, Toomey N, Haider N, Stahl JA, Kelly LJ, Wikström N, Hollingsworth PM, Duff RJ, Hoot SB, Cowan RS, Chase MW, Wilkinson MJ. 2009. Selection of candidate coding DNA barcoding regions for use on land plants. Botanical Journal of the Linnean Society  159: 1– 11. Google Scholar CrossRef Search ADS   Freudenstein JV, Chase MW. 2015. Phylogenetic relationships in Epidendroideae (Orchidaceae), one of the great flowering plant radiations: progressive specialization and diversification. Annals of Botany  115: 665– 681. Google Scholar CrossRef Search ADS PubMed  Garay LA. 1973. Systematics of the genus Angraecum (Orchidaceae). Kew Bulletin  28: 495– 516. Google Scholar CrossRef Search ADS   Givnish TJ, Barfuss MH, Van Ee B, Riina R, Schulte K, Horres R, Gonsiska PA, Jabaily RS, Crayn DM, Smith JA, Winter K, Brown GK, Evans TM, Holst BK, Luther H, Till W, Zizka G, Berry PE, Sytsma KJ. 2011. Phylogeny, adaptive radiation, and historical biogeography in Bromeliaceae: insights from an eight-locus plastid phylogeny. American Journal of Botany  98: 872– 895. Google Scholar CrossRef Search ADS PubMed  Givnish TJ, Barfuss MH, Van Ee B, Riina R, Schulte K, Horres R, Gonsiska PA, Jabaily RS, Crayn DM, Smith JA, Winter K, Brown GK, Evans TM, Holst BK, Luther H, Till W, Zizka G, Berry PE, Sytsma KJ. 2014. Adaptive radiation and diversification in Bromeliaceae: insights from a 7-locus plastid phylogeny. Molecular Phylogenetics and Evolution  71: 55– 78. Google Scholar CrossRef Search ADS PubMed  Givnish TJ, Millam KC, Evans TM, Hall JC, Pires JC, Berry PE, Sytsma KJ. 2004. Ancient vicariance or recent long-distance dispersal? Inferences about phylogeny and South American-African disjunction in Rapateaceae and Bromeliaceae based on ndhF sequence data. International Journal of Plant Sciences  165: S35– S54. Google Scholar CrossRef Search ADS   Givnish TJ, Spalink D, Ames M, Lyon SP, Hunter SJ, Zuluaga A, Iles WJD, Clements MA, Arroyo MTK, Leebens-Mack J, Endara L, Kriebel R, Neubig KM, Whitten WM, Williams NH, Cameron KM. 2015. Orchid phylogenomics and multiple drivers of their extraordinary diversification. Proceedings of the Royal Society of London B  282: 20151553. Google Scholar CrossRef Search ADS   Górniak M, Paun O, Chase MW. 2010. Phylogenetic relationships within Orchidaceae based on a low-copy nuclear coding gene, Xdh: congruence with plastid DNA results. Molecular Phylogenetics and Evolution  56: 784– 795. Google Scholar CrossRef Search ADS PubMed  Govaerts R, Pfahl J, Campacci MA, Baptista DH, Tigges H, Shaw J, Cribb P, George A, Kreuz K, Wood J. 2016. World checklist of Orchidaceae. Richmond: The Board of Trustees of the Royal Botanic Gardens, Kew. Available at: http://apps.kew.org/wcsp/ (accessed 7 August 2016). Haffer J. 1969. Speciation in Amazonian forest birds. Science (New York, N.Y.)  165: 131– 137. Google Scholar CrossRef Search ADS PubMed  Hooghiemstra H, van der Hammen T. 1998. Neogene and Quaternary development of the Neotropical rain forest: the forest refugia hypothesis, and a literature overview. Earth Science Reviews  44: 147– 183. Google Scholar CrossRef Search ADS   Hoorn C, Wesselingh FP, ter Steege H, Bermudez MA, Mora A, Sevink J, Sanmartín I, Sanchez-Meseguer A, Anderson CL, Figueiredo JP, Jaramillo C, Riff D, Negri FR, Hooghiemstra H, Lundberg J, Stadler T, Särkinen T, Antonelli A. 2010. Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity. Science (New York, N.Y.)  330: 927– 931. Google Scholar CrossRef Search ADS PubMed  Huson DH, Bryant D. 2006. Application of phylogenetic networks in evolutionary studies. Molecular Biology and Evolution  23: 254– 267. Google Scholar CrossRef Search ADS PubMed  Iles WJD, Smith SY, Gandolfo MA, Graham SW. 2015. Monocot fossils suitable for molecular dating analyses. Botanical Journal of the Linnean Society  178: 346– 374. Google Scholar CrossRef Search ADS   Knowlton N, Weigt LA. 1998. New dates and new rates for divergence across the Isthmus of Panama. Proceedings of the Royal Society of London B  265: 2257– 2263. Google Scholar CrossRef Search ADS   Kolanowska M. 2015. Determination of potential glacial refugia and possible migration routes of Campylocentrum (Vandeae, Orchidaceae) species through the Darién Gap. Acta Societatis Botanicorum Poloniae  84: 97– 102. Google Scholar CrossRef Search ADS   Kolanowska M, Szlachetko D. 2013. New species of the genus Campylocentrum (Orchidaceae – Vandae) from Colombia. Phyton  53: 221– 239. Kuntze CEO. 1898. Campylocentrum kuntzei. Revisio generum plantarum  3: 298. Lagomarsino LP, Condamine FL, Antonelli A, Mulch A, Davis CC. 2016. The abiotic and biotic drivers of rapid diversification in Andean bellflowers (Campanulaceae). The New Phytologist  210: 1430– 1442. Google Scholar CrossRef Search ADS PubMed  Lindley J. 1840. Angraecum gladiifolium. Edwards’s Botanical Register  26: 68. Liu XQ, Ickert-Bond SM, Nie ZL, Zhou Z, Chen LQ, Wen J. 2016. Phylogeny of the Ampelocissus-Vitis clade in Vitaceae supports the New World origin of the grape genus. Molecular Phylogenetics and Evolution  95: 217– 228. Google Scholar CrossRef Search ADS PubMed  Maiti PK, Maiti P. 2011. Biodiversity: perception, peril and preservation . New Delhi: PHI Learning Pvt. Mann P. 2007. Overview of the tectonic history of northern Central America. Geological Society of America Special Papers  428: 1– 19. Matos-Maraví PF, Peña C, Willmott KR, Freitas AVL, Wahlberg N. 2013. Systematics and evolutionary history of butterflies in the “Taygetis clade” (Nymphalidae: Satyrinae: Euptychiina): towards a better understanding of Neotropical biogeography. Molecular Phylogenetics and Evolution  66: 54– 68. Google Scholar CrossRef Search ADS PubMed  Miller M, Holder M, Vos R, Midford P, Liebowitz T, Chan L, Hoover P, Warnow T. 2010. The CIPRES Portals . CIPRES, Available at: http://www.phylo.org/sub_sections/portal (accessed 6 April 2016). Montes C, Cardona A, Jaramillo C, Pardo A, Silva JC, Valencia V, Ayala C, Pérez-Angel LC, Rodriguez-Parra LA, Ramirez V, Niño H. 2015. Middle Miocene closure of the Central American seaway. Science  348: 226− 229. Google Scholar CrossRef Search ADS PubMed  Morrone JJ. 2014. Biogeographical regionalization of the Neotropical region. Zootaxa  3782: 1− 110. Google Scholar CrossRef Search ADS PubMed  Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GA, Kent J. 2000. Biodiversity hotspots for conservation priorities. Nature  403: 853– 858. Google Scholar CrossRef Search ADS PubMed  Nei M, Li WH. 1979. Mathematical models for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy Sciences USA  76: 5269– 5273. Google Scholar CrossRef Search ADS   Neubig KM, Whitten WM, Carlsward BS, Blanco MA, Endara L, Williams NH, Moore M. 2009. Phylogenetic utility of ycf1 in orchids: a plastid gene more variable than matK. Plant Systematics and Evolution  277: 75– 84. Google Scholar CrossRef Search ADS   Palmer JD, Jansen RK, Michaels HJ, Chase MW, Manhart JR. 1988. Chloroplast DNA variation and plant phylogeny. Annals of the Missouri Botanical Garden  75: 1180– 1206. Google Scholar CrossRef Search ADS   Pérez-Escobar OA, Chomicki G, Condamine FL, Karremans AP, Bogarín D, Matzke NJ, Silvestro D, Antonelli A. 2017. Recent origin and rapid speciation of Neotropical orchids in the world’s richest plant biodiversity hotspot. The New Phytologist  215: 891– 905. Google Scholar CrossRef Search ADS PubMed  Pessoa EM, Alves M. 2015a. Three new species of Campylocentrum (Vandeae, Orchidaceae) from Brazil. Phytotaxa  217: 265– 272. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2015b. Notes on Campylocentrum (Vandeae; Epidendroideae; Orchidaceae): re-establishment of two South American species. Phytotaxa  204: 85– 90. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2016a. Taxonomic revision of Campylocentrum (Orchidaceae–Vandeae–Angraecinae): species with terete leaves. Systematic Botany  41: 700– 713. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2016b. Taxonomical revision of Campylocentrum sect. Dendrophylopsis Cogn. (Orchidaceae–Vandae–Angraecinae). Phytotaxa  286: 131– 152. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2016c. A new combination in Angraecoides (Agraecinae, Orchidaceae): a forgotten name. Phytotaxa  269: 131– 132. Google Scholar CrossRef Search ADS   Pessoa EM, Arruda E, Pereira FFSD, Alves M. 2017. The odd roots of Campylocentrum (Angraecinae-Orchidaceae): a descriptive anatomical study of its morphologically variable roots. Rodriguésia  68: 1207– 1215. Google Scholar CrossRef Search ADS   Pessoa EM, Maciel JR, Alves M. 2015. Campylocentrum brevifolium (Lindl.) E.M. Pessoa & M. Alves, a neglected and critically endangered orchid from the Atlantic Forest of Brazil. Kew Bulletin  70: 43. Google Scholar CrossRef Search ADS   Posada D. 2008. jModelTest: phylogenetic model averaging. Molecular Biology and Evolution  25: 1253– 1256. Google Scholar CrossRef Search ADS PubMed  Pound MJ, Haywood AM, Salzmann U, Riding JB, Lunt DJ, Hunter SJ. 2011. A Tortonian (late Miocene, 11.61–7.25 Ma) global vegetation reconstruction. Paleogeography, Paleoclimatology, Paleoecology  2000: 29– 45. Google Scholar CrossRef Search ADS   Prado DE, Gibbs PE. 1993. Patterns of species distribution in the dry seasonal forests of South America. Annals of the Missouri Botanical Gardens  80: 902– 927. Google Scholar CrossRef Search ADS   Prance GT. 1973. Phytogeographic support for the theory of Pleistocene forest refuges in the Amazon Basin, based upon evidence from distribution patterns in Caryocaraceae, Chrysobalanaceae, Dichapetalaceae and Lecythidaceae. Acta Amazonica  3: 5– 28. Pridgeon AM, Cribb PJ, Chase MC, Rasmussen FN. 2005. Genera orchidacearum, Epidendroideae (part 1) . New York: Oxford University Press. Pridgeon AM, Cribb PJ, Chase MC, Rasmussen FN. 2009. Genera orchidacearum, Epidendroideae (part 2) . New York: Oxford University Press. Pridgeon AM, Cribb PJ, Chase MC, Rasmussen FN. 2014. Genera orchidacearum, Epidendroideae (part 3) . New York: Oxford University Press. Primavera I. 2013. Novedades en orquídeas para Uruguay: segunda contribución. Plantas epífitas. Agrociencia Uruguay  17: 22– 35. Rando JG, Zuntini AR, Conceição AS, Berg C, Pirani JR, Queiroz LP. 2016. Phylogeny of Chamaecrista ser. Coriaceae (Leguminosae) unveils a lineage recently diversified in Brazilian campo rupestre vegetation. International Journal of Plant Sciences  177: 3– 17. Google Scholar CrossRef Search ADS   Rambaut A, Suchard MA, Xie D, Drummond AJ. 2013. Tracer 1.6. Available at: http://tree.bio.ed.ac.uk/software/tracer/ (accessed 6 April 2016). Ree RH, Moore BR, Webb CO, Donoghue MJ. 2005. A likelihood framework for inferring the evolution of geographic range on phylogenetic trees. Evolution  59: 2299– 2311. Google Scholar CrossRef Search ADS PubMed  Ree RH, Smith SA. 2008. Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Systematic Biology  57: 4– 14. Google Scholar CrossRef Search ADS PubMed  Rolfe RA. 1903. The genus Campylocentrum. Orchid Review  11: 245– 247. Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics (Oxford, England)  19: 1572– 1574. Google Scholar CrossRef Search ADS PubMed  Salzmann U, Haywood AM, Lunt DJ, Valdes PJ, Hill DJ. 2008. A new global biome reconstruction and data-model comparison for the middle Pliocene. Global Ecology and Biogeography  17: 432– 447. Google Scholar CrossRef Search ADS   Sánchez-Baracaldo P. 2004. Phylogenetics and biogeography of the Neotropical fern genera Jamesonia and Eriosorus (Pteridaceae). American Journal of Botany  91: 274– 284. Google Scholar CrossRef Search ADS PubMed  Saslis-Lagoudakis C, Chase MW, Robinson DN, Russel SJ, Klirgaard BB. 2008. Phylogenetics of Neotropical Platymiscium (Leguminosae: Dalbergieae): systematics, divergence times, and biogeography inferred from nuclear ribosomal and plastid DNA sequence data. American Journal of Botany  95: 1270– 1286. Google Scholar CrossRef Search ADS PubMed  Simmons MP, Pickett KM, Miya M. 2004. How meaningful are Bayesian support values? Molecular Biology and Evolution  21: 188– 199. Google Scholar CrossRef Search ADS PubMed  Shaw J, Lickey EB, Schilling EE, Small RL. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American Journal of Botany  94: 275– 288. Google Scholar CrossRef Search ADS PubMed  Snak C, Vatanparast M, Silva C, Lewis GP, Lavin M, Kajita T, Queiroz LP. 2016. A dated phylogeny of the papilionoid legume genus Canavalia reveals recent diversification by a pantropical liana lineage. Molecular Phylogenetics and Evolution  98: 133– 146. Google Scholar CrossRef Search ADS PubMed  Sun Y, Skinner DZ, Liang GH, Hulbert SH. 1994. Phylogenetic analysis of Sorghum and related taxa using internal transcribed spacers of nuclear ribosomal DNA. Theoretical and Applied Genetics  89: 26– 32. Google Scholar CrossRef Search ADS PubMed  Swofford D. 2002. PAUP*. Phylogenetic analysis using parsimony and other methods. Version 4 . Sunderland: Sinauer. Szlachetko DL, Tukałło P, Mytnik-Ejsmont J, Grochoka E. 2013. Reclassification of the Angraecum-alliance (Orchidaceae, Vandoideae) based on molecular and morphological data. Biodiversity Research and Conservation  29: 1− 23. Google Scholar CrossRef Search ADS   Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology  17: 1105– 1109. Google Scholar CrossRef Search ADS PubMed  Todzia CA. 1980. A revision of the Costa Rican species of Campylocentrum (Orchidaceae). Brenesia  18: 117– 136. Turchetto-Zolet AC, Pinheiro F, Salgueiro F, Palma-Silva C. 2013. Phylogeographical patterns shed light on evolutionary process in South America. Molecular Ecology  22: 1193– 1213. Google Scholar CrossRef Search ADS PubMed  Vatanparast M, Takayana K, Sousa MS, Tateishi Y, Kajita T. 2011. Origin of Hawaiian endemic species of Canavalia (Fabaceae) from sea-dispersed species revealed by chloroplast and nuclear DNA sequences. Journal of Japanese Botany  86: 15– 25. Viruel J, Segarra-Moragues JG, Raz L, Forest F, Wilkin P, Sanmartín I, Catalán P. 2015. Late Cretaceous-early Eocene origin of yams (Dioscorea, Dioscoreaceae) in the Laurasion Palaearctic and their subsequent Oligocene-Miocene diversification. Journal of Biogeography  43: 750– 762. Google Scholar CrossRef Search ADS   Vogel TA, Patino LC, Eaton JK, Valley JW, Rose WI, Alvarado GE, Viray EL. 2006. Origin of silicic magmas along the Central American volcanic front: genetic relationship to mafic melts. Journal of Volcanology and Geothermal Resources  156: 217− 228. Google Scholar CrossRef Search ADS   Weinberg RF. 1992. Neotectonic development of western Nicaragua. Tectonics  11: 1010– 1017. Google Scholar CrossRef Search ADS   Zamora-Tavares MP, Martínez M, Magallón S, Guzmán-Dávalos L, Vargas-Ponce O. 2016. Physalis and physaloids: a recent and complex evolutionary history. Molecular Phylogenetics and Evolution  100: 41– 50. Google Scholar CrossRef Search ADS PubMed  APPENDIX 1 Taxa, locality, voucher information (collector, number and herbarium acronym) and GenBank accession numbers provided in the following order: ITS, Xdh, matK, rpl32-trnL, trnL-trnF, ycf1. GenBank accessions of previously published sequences are indicated with an asterisk. Angraecum distichum Lindl., Garden origin (Cal-Orchid), B. Carlsward 237 (FLAS), LT706519, KY514267, KY510533, LT732577, LT724124, KY558780. Angraecum leonis Rchb.f., Garden Origin, B. Carlsward 390 (FLAS), LT706518, KY514266, KY510532, LT732576, LT724123, KY558779. Angraecum moandense de Wild., Garden origin (Selby Botanical Gardens), B.Carlsward 208 (FLAS), AF506320*, KY514268, AF506363*, LT732609, AF506339*, KY558781. Campylocentrum brachycarpum Cogn., Brazil, Espírito Santo, E. Pessoa 1191 (UFP), LT706520, KY514284, KY510543, LT732586, LT724135, KY558799. C. brachycarpum Cogn., Brazil, São Paulo: Orquidário do Estado, LZ53 (SP), LT706521, KY514286, KY510544, LT732587, LT724136, KY558798. C. brenesii Schltr., Costa Rica, M. Blanco 2139 (USJ), LT706522, KY514294, KY510549, LT732639, LT724145, KY558829. C. brenesii Schltr., Costa Rica, D. Bogarín 11119 (CR), LT706523, KY514296, KY510551, LT732638, LT724144, KY558828. C. brenesii Schltr., Costa Rica, D. Bogarín 6363 (CR), LT706524, KY514295, KY510550, LT732637, LT724143, KY558827. C. calostachyum (Barb. Rodr.) ined., Brazil, São Paulo, M.R. Miranda 02 (UFP), LT706525, KY514316, KY510556, LT732620, LT724160, KY558809. C. crassirhizum Hoehne, Brazil, Sergipe, E. Pessoa 1243 (UFP), LT706526, KY514291, KY510545, LT732595, LT724140, KY558805. C. crassirhizum Hoehne, Brazil, Espírito Santo, E. Pessoa 1192 (UFP), LT706527, KY514292, KY510548, LT732584, LT724139, KY558806. C. densiflorum Cogn., Brazil, Santa Catarina, E. Pessoa 1196 (UFP), LT718666, KY514285, KY510542, LT732588, LT724134, KY558797. C. fasciola (Lindl.) Cogn., Jamaica, Claude Hamilton, B. Carlsward 301 (FLAS), DQ091564*, KY514322, DQ091321*, LT732608, DQ091445*, KY558817. C. fasciola (Lindl.) Cogn., Brazil, Mato Grosso, A.P. Beneli 966 (UFP), LT718688, KY514320, KY510561, LT732607, LT724165, KY558815. C. fasciola (Lindl.) Cogn., Jamaica, M. Witten 711, LT718690, KY514324, n.s., LT732604, LT724166, KY558816. C. fasciola (Lindl.) Cogn., Jamaica, Claude Hamilton, B. Carlsward 185 (FLAS), AF506294*, KY514323, AF506342*, LT732596, AF147226*, KY558813. C. fasciola (Lindl.) Cogn., Costa Rica, D. Bogarín 10415 (CR), LT718689, KY514321, KY510560, LT732605, LT724167, KY558818. C. fasciola (Lindl.) Cogn., Ecuador, M. Witten 1933 (QCNE), AF506295*, KY514319, AF506343*, LT732606, LT724164, KY558814. C. grisebachii Cogn., Brazil, Minas Gerais, E. Pessoa 1188 (UFP), LT718680, KY514298, KY510558, LT732597, LT724162, KY558821. C. huebneri Mansf., Brazil, Roraima, E. Pessoa 701 (UFP), LT718665, n.s., KY510565, LT732629, LT724147, KY558830. C. jamaicense (Rchbf. ex Griseb.) Benth. ex Fawc, Jamaica, M. Witten 1934 (FLAS), AF506299*, KY514289, AF506348*, LT732590, AF506326*, KY558800. C. jamaicense (Rchbf. ex Griseb.) Benth. ex Fawc., Puerto Rico, D. Ackerman 3341 (UPRRP), AY147219*, KY514288, AF506325*, LT732591, AF506346*, KY558801. C. kuntzei Cogn. ex Kuntze, Paraguay, Guairá, G. Esser 14519 (HEID), LT718676, n.s., KY510571, LT732634, LT724149, KY558834. C. labiakii Pessoa & Alves, Brazil, Espírito Santo, J. Meireles 545 (ESA), LT718660, n.s., KY510534, LT732619, LT724129, KY558794. C. mattogrossense Hoehne, Brazil, Mato Grosso, A. P. Beneli 968 (UFP), LT718687, KY514300, KY510567, LT732631, LT724148, KY558831. C. mattogrossense Hoehne, Brazil, Pará, A. K. Koch 566 (SP), LT718686, KY514299, KY510566, LT732630, LT724246, n.s. C. micranthum (Lindl.) Maury, Brazil, Roraima, E. Pessoa 1001 (UFP), LT718674, KY514311, KY510570, LT732628, LT724156, KY558841. C. micranthum (Lindl.) Maury, Brazil, Permambuco, E. Pessoa 1062 (UFP), LT718673, KY514309, KY510568, LT732625, LT724154, KY558839. C. micranthum (Lindl.) Maury, Brazil, Ceará, E. Pessoa 1115 (UFP), LT718672, KY514310, KY510569, LT732632, LT724155, KY558840. C. multiflorum Schltr., Costa Rica, D. Bogarín 1486 (CR), LT718681, KY514297, KY510552, LT732602, LT724163, KY558819. C. neglectum (Rchb.f. & Warm.) Cogn., Brazil, Distrito Federal, E. Pessoa 1278 (UFP), LT718663, KY514283, KY510546, LT732593, LT724137, KY558803. C. neglectum (Rchb.f. & Warm.) Cogn., Brazil, B. Carlsward 272 (FLAS), AF506297*, KY514287, AF506345*, LT732592, AF506324*, KY558804. C. ornithorrhinchum (Lindl.) Rolfe, Brazil, Permambuco, E. Pessoa 1239 (UFP), LT718661, KY514279, KY510536, LT732615, LT724130, KY558791. C. ornithorrhinchum (Lindl.) Rolfe, Brazil, São Paulo, R.Romanini18135 (SP), LT718659, KY514280, KY510539, LT732616, LT724131, KY558790. C. pachyrrhizum (Rchb.f.) Rolfe, Brazil, Sergipe, E. Pessoa 1245 (UFP), LT718682, KY514304, KY510554, LT732600, LT724142, KY558822. C. pachyrrhizum (Rchb.f.) Rolfe, Brazil, Pará, A.K. Koch 565 (SP), LT718684, KY514305, KY510553, LT732599, LT724141, KY558824. C. pachyrrhizum (Rchb.f.) Rolfe, USA, Florida, No voucher, LT718683, KY514307, AF506349*, LT732598, AF506327*, KY558823. C. pachyrrhizum (Rchb.f.) Rolfe, Puerto Rico, D. Ackerman s.n. (UPRRP), AF506301*, KY514306, AF506350*, LT732601, AF506328*, KY558825. C. panamense Ames, Costa Rica, D. Bogarín 871 (CR), LT718675, n.s., KY510573, LT732624, LT724153, KY558833. C. pauloense Hoehne & Schltr., Brazil, Santa Catarina, E. Pessoa 1197 (UFP), LT718664, KY514281, KY510541, LT732585, LT724133, KY558796. C. pernambucense Hoehne, Brazil, Sergipe, E. Pessoa 1244 (UFP), n.s., KY514277, KY510537, LT732618, LT724127, KY558792. C. pernambucense Hoehne, Brazil, Pernambuco, E. Pessoa 1287 (UFP), LT718662, KY514276, KY510538, LT732613, LT724126, KY558793. C. poeppigii (Rchb.f.) Rolfe, Brazil, Roraima, E. Pessoa 968 (UFP), LT718678, KY514303, KY510563, LT732612, LT724157, KY558812. C. poeppigii (Rchb.f.) Rolfe, Mexico, B. Carlsward 307(FLAS), AF506302*, KY514301, AF506351*, LT732610, AF506329*, KY558811. C. poeppigii (Rchb.f.) Rolfe, Costa Rica, D. Bogarín 2218 (CR), LT718679, KY514302, KY510564, LT732611, LT724158, KY558810. C. robustum Cogn., Brazil, Minas Gerais, Museu de História Natural 0967 (MHN), LT718667, KY514290, KY510547, LT732594, LT724138, KY558802. C. schiedei (Rchb.f.) Benth. ex Hemsl., Costa Rica, D. Bogarín 494 (CR), LT718671, KY514308, KY510572, LT732633, LT724152, KY558832. C. sellowii (Rchb.f.) Rolfe, Brazil, Minas Gerais, E. Pessoa 1189 (UFP), LT718658, KY514275, KY510535, LT732614, LT724125, KY558788. C. serranum Pessoa & Alves, Brazil, Permambuco, E. Pessoa 945 (UFP), LT718670, KY514317, KY510557, LT732621, LT724161, KY558808. C. spannagelli Hoehne, Brazil, Rio de Janeiro, E. Pessoa 1195 (UFP), LT718665, KY514282, KY510540, LT732589, LT724132, KY558795. C. stenanthum Schltr., Costa Rica, M. Witten 1822 (FLAS), AF506304, KY514312, AF506352*, LT732635, LT724150, KY558835. C. stenanthum Schltr., Mexico, B. Carlsward 180 (FLAS), AF506298*, KY514315, AF506347*, LT732636, AY147227*, KY558838. C. stenanthum Schltr., Panama, B. Carlsward 315, (FLAS), AY147220*, KY514314, AY147228*, LT732623, AY147235*, KY558837. C. stenanthum Schltr., Costa Rica, D. Bogarín 1263 (CR), LT718677, KY514313, n.s., LT732627, LT724163, KY558836. C. tenellum Todzia, Costa Rica, D. Bogarín 5844 (CR), LT718668, KY514293, KY510562, LT732626, n.s., KY558826. C. tyrridion Garay & Dunst. ex Foldats, Mexico, G. Carnevali 5145 (CICY), AF506305*, n.s., DQ091322*, LT732603, DQ091446*, KY558820. C. ulaei Cogn., Brazil, São Paulo, M.R. Miranda 01 (UFP), LT718669, KY514318, KY510555, LT732622, LT724159, KY558807. C. wawrae (Rchb.f. ex Beck) Rolfe, Brazil, Minas Gerais, D. Barbosa 13 (CESJ), LT718657, KY514278, n.s., LT732617, LT724128, KY558789. Dendrophylax funalis (Sw.) Benth. ex Rolfe, Jamaica, B. Carlsward 302 (FLAS), AY147221*, KY514269, AF506355*, LT732583, AY147229*, KY558782. D. porrectus (Rchb.f.) Carlsward & Whitten, Mexico, Yucatán, G. Carnevali 5907 (CICY), AF506314*, KY514271, AF506357*, LT732581, AF506335*, KY558785. D. porrectus (Rchb.f.) Carlsward & Whitten, USA, Florida, B. Carlsward 329 (FLAS), AY147223*, KY514274, AY147237*, LT732579, AY147232*, KY558784. D. porrectus (Rchb.f.) Carlsward & Whitten, Dominican Republic, M. Witten 1950 (JBSD), AY147224*, KY514272, AY147238*, LT732578, AY147233*, KY558786.D. porrectus (Rchb.f.) Carlsward & Whitten, Jamaica, B. Carlsward 184 (FLAS), Af506315*, KY514273, Af506358*, LT732580, AY147231*, KY558787. D. varius (J.F. Gmel.) Urb., Dominican Republic, M. Whitten 1960 (JBSD), AY147222*, KY514270, AY147236*, LT732582, AY147230*, KY558783. APPENDIX 2 Divergence time estimates and ancestral areas for nodes indicated in Figures 2 and 3. HPD = highest posterior density; A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. Node  Ages (Mya)  Ancestral areas    Mean  95% HPD  Area  Probability  1  10.93  13.7−8.2  G  0.21  2  8.47  11.4−5.6  G  0.97  3  8.53  10.7−6.4  AG  0.70  4  8.21  10.5−5.9  A  0.61  5  4.7  6.7−2.7  A  0.74  6  4.01  6.0−2.1  A  0.91  7  7.22  9.2−5.2  AF  0.20  8  2.9  4.2−1.7  AEF  0.40  9  1.76  2.7−0.8  F  0.83  10  0.93  1.53−0.4  F  1  11  0.46  0.76−0.26  F  1  12  0.2  0.27−0.13  F  1  13  2.04  2.9−1.18  AEF  0.46  14  0.92  1.4−0.44  AEF  0.59  15  0.71  1.07−0.35  AEF  0.63  16  0.64  1.08−0.2  EF  0.95  17  1.23  1.83−0.63  F  1  18  1.04  1.63−0.45  F  1  19  0.56  0.95−0.17  F  1  20  5.38  6.79−3.97  BF  0.26  21  4.79  6.1−3.48  B  0.43  22  4.06  5.2−2.92  B  0.47  23  1.25  2.0−0.5  ABF  0.1  24  3.4  4.64−2.16  B  0.47  25  0.61  1.2−0.2  ABF  0.1  26  2.4  3.4−1.4  BF  0.84  27  5.16  6.6−3.7  BF  0.42  28  2.69  3.8−1.6  F  0.96  29  1.82  2.7−0.94  F  0.99  30  4.28  5.5−3.06  B  0.35  31  3.29  4.4−2.18  B  0.99  32  3.46  4.6−2.32  BD  0.27  33  0.52  1.0−0.1  D  0.42  34  1.56  2.2−0.92  BCD  0.2  35  1.11  1.7−0.52  BCD  0.21  36  0.81  1.3−0.32  BCD  0.21  37  0.73  1.2−0.26  BCD  0.15  Node  Ages (Mya)  Ancestral areas    Mean  95% HPD  Area  Probability  1  10.93  13.7−8.2  G  0.21  2  8.47  11.4−5.6  G  0.97  3  8.53  10.7−6.4  AG  0.70  4  8.21  10.5−5.9  A  0.61  5  4.7  6.7−2.7  A  0.74  6  4.01  6.0−2.1  A  0.91  7  7.22  9.2−5.2  AF  0.20  8  2.9  4.2−1.7  AEF  0.40  9  1.76  2.7−0.8  F  0.83  10  0.93  1.53−0.4  F  1  11  0.46  0.76−0.26  F  1  12  0.2  0.27−0.13  F  1  13  2.04  2.9−1.18  AEF  0.46  14  0.92  1.4−0.44  AEF  0.59  15  0.71  1.07−0.35  AEF  0.63  16  0.64  1.08−0.2  EF  0.95  17  1.23  1.83−0.63  F  1  18  1.04  1.63−0.45  F  1  19  0.56  0.95−0.17  F  1  20  5.38  6.79−3.97  BF  0.26  21  4.79  6.1−3.48  B  0.43  22  4.06  5.2−2.92  B  0.47  23  1.25  2.0−0.5  ABF  0.1  24  3.4  4.64−2.16  B  0.47  25  0.61  1.2−0.2  ABF  0.1  26  2.4  3.4−1.4  BF  0.84  27  5.16  6.6−3.7  BF  0.42  28  2.69  3.8−1.6  F  0.96  29  1.82  2.7−0.94  F  0.99  30  4.28  5.5−3.06  B  0.35  31  3.29  4.4−2.18  B  0.99  32  3.46  4.6−2.32  BD  0.27  33  0.52  1.0−0.1  D  0.42  34  1.56  2.2−0.92  BCD  0.2  35  1.11  1.7−0.52  BCD  0.21  36  0.81  1.3−0.32  BCD  0.21  37  0.73  1.2−0.26  BCD  0.15  View Large APPENDIX 3 Taxa, voucher information and GenBank accession numbers used for the dating analysis in this study in the following order: ITS, matK, ycf1. Sequences in GenBank but not published yet in papers are indicated with an asterisk, and the voucher information is not available. Earina autumnalis (G.Forst.) Hook.f., M. W. Chase O-298 (K), AF260149, AF263656, none. Cattleya bicolor Lindl., F.G. Brieger 4333 (ESA), JN600949, none, JN600718. C. bicolor Lindl., F.G. Brieger 895 (ESA), none, EU139961, none. Maxillaria splendens Poepp. & Endl., W. M. Whitten 2940 (FLAS), none, none, KF660506. M. splendens Poepp. & Endl., S. Koehler 144 (UEC), DQ210252, DQ210684, none. Cymbidium eburneum Lindl., J. Luo*, none, KF361650*, KF361650*, C. eburneum Lindl., M. W. Chase 1505 (K), AF470503, none, none. Phalaenopsis wilsonii Rolfe, B. Carlsward 331 (FLAS), DQ091672, none, EU490763. P. wilsonii Rolfe, TBG 144214*, none, AB217751*, none. APPENDIX 4 Palaeogeographical models for dispersal rates across areas implemented in the biogeographical analysis. A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. Time Slice 1: 11−7 Mya; Time Slice 2: 7−0 Mya.   A  B  C  D  E  F  G  Time Slice 1  A  1  0.1  0.1  0.05  0.0001  0.001  0.001  B    1  0.1  0.01  0.0001  0.0001  0.0001  C      1  0.1  0.0001  0.0001  0.0001  D        1  0.1  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1  Time Slice 2  A  1  0.5  0.5  0.1  0.005  0.001  0.001  B    1  0.5  0.05  0.001  0.001  0.0001  C      1  0.5  0.01  0.001  0.0001  D        1  0.5  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1    A  B  C  D  E  F  G  Time Slice 1  A  1  0.1  0.1  0.05  0.0001  0.001  0.001  B    1  0.1  0.01  0.0001  0.0001  0.0001  C      1  0.1  0.0001  0.0001  0.0001  D        1  0.1  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1  Time Slice 2  A  1  0.5  0.5  0.1  0.005  0.001  0.001  B    1  0.5  0.05  0.001  0.001  0.0001  C      1  0.5  0.01  0.001  0.0001  D        1  0.5  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1  View Large © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Botanical Journal of the Linnean Society Oxford University Press

Evolutionary history and systematics of Campylocentrum (Orchidaceae: Vandeae: Angraecinae): a phylogenetic and biogeographical approach

Loading next page...
 
/lp/ou_press/evolutionary-history-and-systematics-of-campylocentrum-orchidaceae-z4vrZFq0Co
Publisher
The Linnean Society of London
Copyright
© 2018 The Linnean Society of London, Botanical Journal of the Linnean Society
ISSN
0024-4074
eISSN
1095-8339
D.O.I.
10.1093/botlinnean/box089
Publisher site
See Article on Publisher Site

Abstract

Abstract Subtribe Angraecinae (Orchidaceae: Vandeae) are mainly distributed in Africa, but with two genera, Campylocentrum and Dendrophylax, restricted to the Neotropics. As a widespread Neotropical genus, Campylocentrum constitutes an appropriate model for revealing biogeographical patterns in this area and investigating routes of colonization and dispersal. In this study, we reconstructed phylogenetic relationships of the genus with Bayesian inference and maximum parsimony analyses of combined nuclear (ITS rDNA and Xdh) and plastid (matK exon, rpl32-trnL spacer, trnL intron, trnL-trnF spacer and ycf1 exon) DNA datasets, aimed at establishing a new infrageneric classification for this taxonomically complex genus. Based on the most comprehensive phylogenetic tree, we investigated the biogeographical history of Campylocentrum by estimating divergence times, inferred using fossil and secondary calibrations applying a relaxed-clock model approach, and reconstructing ancestral areas of distribution under a time-stratified likelihood model. The phylogenetic analyses provided strong support for the majority of the clades. Campylocentrum is monophyletic, and we recognize five sections based upon strongly supported clades. We conclude that the African angraecoid ancestor of Campylocentrum and Dendrophylax dispersed to the Antilles. Campylocentrum is estimated to be a relatively young genus (late Miocene, c. 8.2 Mya) and its most recent common ancestor had a disjunct distribution in the Antilles and Parana dominion. During the Pliocene, the five sections diverged and expanded their distributions in the Neotropics, and in the Pleistocene diversification was experienced by some of the terminal clades. We hypothesize that the evolutionary history of Campylocentrum was strongly influenced by orogenic events during the Pliocene and climatic fluctuations during the Pleistocene. INTRODUCTION Orchidaceae are one of the two largest plant families and occur on all continents except Antarctica (Dressler, 2005). They are particularly diverse in the Neotropics and Southeast Asia (Dressler, 1993). The Neotropical region houses the highest terrestrial biodiversity in the planet (Myers et al., 2000; Maiti & Maiti, 2011), including some hyper-diverse orchid groups such as subtribes Laeliinae, Maxillariinae, Oncidiinae and Pleurothallidinae, all endemic to this region (Pridgeon et al., 2005, 2009). Other taxa, such as tribe Vandeae, are almost entirely Palaeotropical, including the mostly Afro-Malagasy subtribe Angraecinae (Vandeae; Carlsward et al., 2006a; Pridgeon et al., 2014), but which includes two Neotropical genera, Campylocentrum Benth. and Dendrophylax Rchb.f. (Carlsward et al., 2006a; Pridgeon et al., 2014). Angraecinae consist of 49 genera (Pridgeon et al., 2014; but see also Szlachetko et al., 2013, who recognized many more). Variation in the number of genera recognized depends on the concept of the large genus Angraecum Bory, which in recent molecular phylogenetic studies was found to be non-monophyletic (Carlsward et al., 2006a; Szlachetko et al., 2013; Andriananjamanantsoa et al., 2016). Campylocentrum and Dendrophylax are embedded in Angraecum s.l. as sister to Angraecum sections Conchoglossum Schltr. and Arachnangraecum Schltr. or to the genera Angraecoides (Corden.) Szlach. Mytrik & Grochocka and Eichlerangraecum Szlach. Mytrik & Grochocka, as proposed by Szlachetko et al. (2013). Because the study by Szlachetko et al. (2013) did not include many of the other genera in Angraecinae, it cannot be considered definitive in terms of resolving the taxonomic complications of this subtribe. We prefer instead to follow the taxonomy of Pridgeon et al. (2014), although we admit that this is not a satisfactory taxonomic arrangement and must be considered a temporary solution. In Angraecinae, Neotropical Campylocentrum and Dendrophylax include the only leafless taxa, the phylogenetic relationships of which were studied by Carlsward, Whitten & Williams (2003). They found that leaves were lost twice in this clade. Campylocentrum has both leafy and leafless species, whereas Dendrophylax comprises only leafless species (Carlsward et al., 2003). Campylocentrum currently includes 70 species (Kolanowska & Szlachetko, 2013; Carlsward, 2014a; Pessoa & Alves, 2015a, b, 2016a, b), occurring in all Neotropical countries except Chile (Primavera, 2013; Carlsward, 2014a). Campylocentrum and Dendrophylax are most easily distinguished by their inflorescences. Campylocentrum has racemes with numerous, relatively small flowers that open simultaneously, whereas in Dendrophylax the flowers are fewer and often much larger. The latter has inflorescences that are often fractiflex, occasionally branched, with one to few flowers opening successively. Dendrophylax flowers vary in size among species from a few millimetres to several centimetres, including some with long nectar spurs (Carlsward & Cribb, 2014). Cogniaux (1906) proposed an infrageneric classification of Campylocentrum, organizing it in three sections based on vegetative characters (Table 1; Bogarín & Pupulin, 2010). Table 1. Sections of Campylocentrum sensuCogniaux (1906) and their main vegetative features   Roots  Stem  Leaves  C. section Campylocentrum  Smooth/granulose  Elongated  Conduplicate/terete and developed  C. section Dendrophylopsis  Smooth  Reduced  Absent  C. section Pseudocampylocentrum  Granulose  Elongated  Terete and reduced    Roots  Stem  Leaves  C. section Campylocentrum  Smooth/granulose  Elongated  Conduplicate/terete and developed  C. section Dendrophylopsis  Smooth  Reduced  Absent  C. section Pseudocampylocentrum  Granulose  Elongated  Terete and reduced  View Large Although previous molecular analyses have clarified higher taxonomic levels in Orchidaceae (reviewed by Chase et al., 2015), the majority of Neotropical genera remain unstudied at the species level. Campylocentrum constitutes an appropriate model for revealing biogeographical patterns in this area, as it is a widespread genus. Previously published phylogenetic analyses (Carlsward et al., 2003, 2006a; Neubig et al., 2009) included only a few Campylocentrum spp. Therefore, a more comprehensive and well-supported phylogenetic hypothesis is needed to provide a robust background for studying the evolution of traits and systematics of the group. The main goals of this study are: (1) to provide a phylogenetic tree based on nuclear (ribosomal ITS and low-copy Xdh exon) and plastid (matK exon, rpl32-trnL spacer, trnL intron, trnL-F spacer and ycf1 exon) DNA data; (2) to establish a new infrageneric classification for Campylocentrum; (3) to estimate divergence times of its clades and relate these dates to events in the history of the Earth; and (4) to produce a likelihood-based biogeographical analysis for reconstruction of ancestral areas of distribution and for estimation of dispersal, vicariance and peripheral isolation events that may have occurred. MATERIAL AND METHODS Taxon sampling This study includes samples for 30 of the 70 recognized Campylocentrum spp. (Carlsward, 2014a; Kolanowska & Szlachetko, 2013; Pessoa & Alves, 2015a, 2015b, 2016a, b), including all habits in the genus and all taxonomic sections proposed by Cogniaux (1906). Multiple accessions were used for 14 species with wide distributions to encompass their morphological variation (total of 55 samples of the genus). Unsampled species are all narrowly distributed and clearly closely related to species we have included. Their absence is highly unlikely to affect the conclusions drawn here. Outgroups were chosen based on Carlsward et al. (2003, 2006a) and Szlachetko et al. (2013) and included three Dendrophylax spp., shown in previous studies to be the sister group of Campylocentrum, and three representatives of African Angraecum s.l. [Angraecoides (Cordem.) Garay, Dolabrifolia Pfitzer ex Rchb.f. and Humblotiangraecum (Schltr.) Szlach. Mytrik & Grochocka sensuSzlachetko et al. (2013)], for a total of nine samples in the outgroup ( Appendix 1). Samples were obtained during fieldwork except for two herbarium specimens, some living collections (all dried and stored in silica gel) and some DNA samples available from the DNA Bank at the University of Florida, which were used to produce sequences of the additional markers used in this study (rpl32-trnL, Xdh, ycf1). The sequences of Campylocentrum previously published by Carlsward et al. (2003, 2006a) available in GenBank (ITS, matK and trnL intron, and trnL-trnF spacer) were also included, although we have updated the taxonomic names used in that study by examining the vouchers housed at FLAS. DNA extraction, amplification and sequencing DNA was extracted following a CTAB procedure (Doyle & Doyle, 1987, 1990). Total DNA was then purified using QIAquick silica columns (Qiagen, Crawley, UK). The primers of Sun et al. (1994) were used to amplify the internal transcribed spacer region of nuclear ribosomal DNA (ITS1 + 5.8S rDNA + ITS2). Amplifications were performed in a volume of 25 µL, with 12.5 µL DreamTaq Green Master Mix (Thermo Fisher Scientific, Loughborough, UK), 4 µL TBT-PAR, 0.5 µL dimethylsulphoxide (DMSO), 6.0 µL nuclease-free water, 0.5 µL each primer (10 µM) and 1 µL template DNA (30–90 ng/μL). The reaction conditions were: initial denaturation of 94 °C for 1 min, followed by 30 cycles of 94 °C for 1 min, 48 °C for 1 min and 72 °C for 90 s and a final extension of 72 °C for 4 min. The low copy nuclear gene xanthine dehydrogenase (Xdh) was amplified using primers X551 and X1591 of Górniak, Paun & Chase (2010). Amplifications were performed in the same volume and used the same reagents as for ITS. The PCR programme followed the touchdown procedure proposed by those authors: initial denaturation of 94 °C for 2 min, followed by six cycles of 94 °C for 45 s, 55–49 °C (reducing 1 °C per cycle) for 45 s and 72 °C for 90 s, then 28 cycles of 94 °C for 45 s, 49 °C for 45 s and 72 °C for 90 s and a final extension of 72 °C for 5 min. The plastid spacer rpl32-trnL was amplified with the primers UAG and F of Shaw et al. (2007); the trnL intron and trnL-F spacer were amplified as one fragment with the primers c and f or separately in some cases with the pairs e/f and c/d of Taberlet et al. (1991). Portions of the plastid matK gene were amplified using the primers 5R and XF of Ford et al. (2009). Portions of the plastid ycf1 gene were amplified with the primers 3720F and 5500R and the internal IntF and IntR of Neubig et al. (2009). Amplification of the five plastid regions was performed in a volume of 25 µL containing 12.5 µL DreamTaq Green Master Mix, 4 µL trehalose-based (TBT-PAR), 6.5 µL nuclease-free water, 0.5 µL each primer (10 µM) and 1 µL template DNA (30–90 ng/μL). The PCR programme used to amplify rpl32-trnL and trnL-F (both intron and spacer) was: initial denaturation step of 80 °C for 5 min, followed by 30 cycles of 95 °C for 1 min, 50 °C for 1 min and 65 °C for 4 min, and a final extension of 65 °C for 5 min. Amplification of matK had an initial denaturation step of 94 °C for 1 min, followed by 35 cycles of 94 °C for 30 s, 46–48 °C for 40 s and 72 °C for 40 s, and a final extension of 72 °C for 5 min. Amplification of ycf1 was carried out using a touchdown protocol following Neubig et al. (2009): 94 °C for 3 min, followed by eight cycles of 94 °C for 30 s, 60–52 °C (reducing 1 °C per cycle) for 1 min and 72 °C for 3 min, then 30 cycles of 94 °C for 30 s, 50°C for 1 min and 72 °C for 3 min, and a final extension of 72 °C for 3 min. All PCR products were purified using the QIAquick PCR purification kit (Qiagen) following the manufacturer’s protocol. Amplifications from DNA of herbarium specimens collected less than 5 years ago were all successful. Cycle sequencing was carried out using a Big Dye Terminator v. 3.1 Cycle Sequencing Kit (Applied Biosystems, ABI, Warrington, UK) using the same primers as for the amplifications. The reaction mix for the nuclear markers contained 1.5 μL 5× sequencing buffer, 0.25 μL Big Dye terminator, 0.75 μL 10 μM primer (1.5 pmol), 1–2 μL amplification product (30–90 ng/μL), 0.2 μL DMSO and 1.3 μL H2O in a total reaction volume of 5 μL. The reaction mix for the plastid markers contained 1.5 μL 5× sequencing buffer, 0.25 μL Big Dye terminator, 0.75 μL 10 μM primer, 1–2 μL amplified product (30–90 ng/μL) and 1.5 μL H2O in a total reaction volume of 5 μL. The cycle sequencing programme comprised 25 cycles of denaturation of 94 °C for 15 s, annealing of 50 °C for 5 s and elongation of 60 °C for 4 min. The cycle sequencing products were sequenced on an ABI 3720 automated DNA sequencer according to the manufacturer’s protocol. Chromatograms were edited and contigs were assembled using Geneious 8.0.4 (Biomatters, Auckland, New Zealand). Phylogenetic analysis Sequences were aligned initially using Geneious 8.0.4 (Biomatters), and a second alignment was carried out using MUSCLE (Edgar, 2004), which was subsequently manually optimized. Independent sequence matrices were compiled for each DNA region, including the sequences from GenBank. Bayesian inference (BI) and maximum parsimony (MP) analyses were performed for the combined (ITS, Xdh and plastid markers) data set using, respectively, MrBayes 3.1.2 (Ronquist & Huelsenbeck, 2003) on the CIPRES Science Gateway portal (Miller et al., 2010) and PAUP 4.0b10 (Swofford, 2002). Angraecum leonis A.H.Kent (A. section Humblotiangraecum Schltr.) was selected as the single outgroup based on the previous phylogenetic studies of Carlsward et al. (2006a) and Szlachetko et al. (2013). The MP analyses were performed via heuristic searches using 40000 random taxa additions and tree bisection reconnection (TBR) branch swapping. Bootstrap percentages (BP) were estimated with 1000 non-parametric replicates and TBR swapping. Clades with BP ≥ 85 and 0.95 posterior probability (PP; Cummings et al., 2003; Simmons et al., 2004) were considered strongly supported (Erixon et al., 2003). The best-fitting nucleotide substitution model for BI was selected using JModelTest 0.1.1 (Posada, 2008) under the Bayesian information criterion (BIC; Brown & Lemmon, 2007). The most appropriate models for each region were TrNef+G for nrITS, TrN+G for Xdh, TPM1uf+G for matK, TPM1uf for rpl32-trnL, TPM1uf+G for trnL-F and TVM+G for ycf1 (Table 2). The best model was adapted from the options available in MrBayes 3.1. for each partition (see Table 2). Table 2. Features of DNA datasets used in this study, and best fit models used to Bayesian inference (BI) for each marker   ITS  Xdh  matK  rpl32-trnL  trnL-F  ycf1  Plastid combined  All data combined  Number of taxa  63  59  62  64  63  63  64  64  Aligned length (bp)  660  908  647  688  1496  1729  4560  6128  Number of variable positions  218 (33%)  179 (19.7%)  112 (17.1%)  68 (9.8%)  453 (22.3%)  419 (24.2%)  935 (20.5%)  1326 (21.6%)  Number of potentially parsimony-informative sites  132 (20.1%)  89 (9.8%)  53 (8.2%)  30 (4.3%)  227 (15.1%)  301 (17.4%)  561 (12.3%)  782 (12.7%)  Number of changes/ variable sites  1.66  1.25  1.35  1.13  1.34  1.58  1.47  1.51  Fitch tree length  361  225  151  77  609  664  1374  1997  Consistency index  0.72  0.86  0.82  0.91  0.80  0.72  0.75  0.75  Retention index  0.90  0.93  0.92  0.95  0.90  0.91  0.90  0.90  Best fit model (BI)  HKY+G  HKY+G  GTR+G  GTR  GTR+G  GTR+I+G  –  –    ITS  Xdh  matK  rpl32-trnL  trnL-F  ycf1  Plastid combined  All data combined  Number of taxa  63  59  62  64  63  63  64  64  Aligned length (bp)  660  908  647  688  1496  1729  4560  6128  Number of variable positions  218 (33%)  179 (19.7%)  112 (17.1%)  68 (9.8%)  453 (22.3%)  419 (24.2%)  935 (20.5%)  1326 (21.6%)  Number of potentially parsimony-informative sites  132 (20.1%)  89 (9.8%)  53 (8.2%)  30 (4.3%)  227 (15.1%)  301 (17.4%)  561 (12.3%)  782 (12.7%)  Number of changes/ variable sites  1.66  1.25  1.35  1.13  1.34  1.58  1.47  1.51  Fitch tree length  361  225  151  77  609  664  1374  1997  Consistency index  0.72  0.86  0.82  0.91  0.80  0.72  0.75  0.75  Retention index  0.90  0.93  0.92  0.95  0.90  0.91  0.90  0.90  Best fit model (BI)  HKY+G  HKY+G  GTR+G  GTR  GTR+G  GTR+I+G  –  –  View Large Bayesian inference included two independent runs with four chains each with the Markov chain Monte Carlo (MCMC) parameters set to 20 million generations sampling every 10000 trees. We discarded as burn-in the first 10000 trees. Convergence between the two independent runs was checked using Tracer 1.6 (Rambaut et al., 2013). To assess possible conflicts between the independent DNA data matrices, congruence was evaluated by looking for well-supported incongruent clades among the phylogenetic trees obtained for each matrix separately. All plastid regions were treated as a single matrix since the plastid genome is inherited as a unit, and thus it is not subject to recombination (Palmer et al., 1988), whereas the two nuclear regions, ITS and Xdh, were analysed separately. Time divergence estimates and biogeographical analysis Due the lack of orchid fossils assigned to closely related species of Angraecinae (Iles et al., 2015), we built a phylogenetic tree including five additional species of orchids ( Appendix 3) that represent the major clades between Angraecinae and the closest available fossil genus (Earina Lindl.; Conran, Bannister & Lee, 2009). This matrix included only the nuclear region ITS and two plastid regions, matK and ycf1, due to the unavailability of the remaining regions included in this study in GenBank for the additional five species. One specimen per species of Campylocentrum and Dendrophylax was included, except for two widespread species, C. fasciola (Lindl.) Cogn. and C. pachyrrhizum (Rchb.f.) Rolfe, for which were included two specimens to represent the whole distribution based on the phylogenetic results; we included one sample from each of the two subclades in the combined analysis (Fig. 1). Figure 1. View largeDownload slide Phylogenetic relationships of Campylocentrum produced by Bayesian inference of nuclear (ITS and Xdh) and plastid (matK, rpl32-trnL, trnL-F, ycf1) regions combined. Posterior probabilities (≥ 0.9) are indicated above branches and maximum parsimony bootstrap percentages (≥ 80) are indicated below branches. Lateral bars: green – leaves conduplicate; red – leafless; blue – leaves cylindrical; yellow – capsules six-ribbed; pink – capsules unribbed; purple – roots smooth; orange – granulose. Figure 1. View largeDownload slide Phylogenetic relationships of Campylocentrum produced by Bayesian inference of nuclear (ITS and Xdh) and plastid (matK, rpl32-trnL, trnL-F, ycf1) regions combined. Posterior probabilities (≥ 0.9) are indicated above branches and maximum parsimony bootstrap percentages (≥ 80) are indicated below branches. Lateral bars: green – leaves conduplicate; red – leafless; blue – leaves cylindrical; yellow – capsules six-ribbed; pink – capsules unribbed; purple – roots smooth; orange – granulose. Absolute divergence times were estimated with a Bayesian approach in BEAST 1.8.0 (Drummond et al., 2012) on CIPRES. A relaxed molecular clock analysis with uncorrelated log-normal model was performed, which takes into consideration rate heterogeneity between lineages with substitution rates uncorrelated across the tree (Drummond et al., 2006), allowing the mutation rate to vary among partitions. This was supported by a coefficient of variation (CV) of 0.525; if the CV ranges between 0.1 and 1 then the relaxed clock model we applied in this analysis is appropriate (Drummond & Bouckaert, 2015). We used BEAUti 1.8 to create input files, assigning the same best-fitting models used for BI (Table 2), and constraints in topology were applied as necessary (Drummond et al., 2012) to match the topology of Epidendroideae in Freudenstein & Chase (2015) and the topology found for Campylocentrum spp. in the previous results (Fig. 1). The tree speciation prior followed the Yule process, and two MCMC chains were run for 100 million generations sampling every 10000 trees. Convergence and mixing were assessed using the effective sampling size criterion (ESS > 200) in Tracer 1.6, and processing of post-burn-in trees was performed with TreeAnnotator 1.8 (Drummond & Rambaut, 2007) to obtain a maximum clade credibility tree with mean values and 95% confidence intervals for nodal ages. Three nodes were calibrated (see Fig. 2, α, β and γ). Both primary and secondary calibrations were applied in BEAST. Only one fossil, Earina fouldenensis Conran, Bannister & D.E.Lee, from the early Miocene (23–20 Mya) of New Zealand was used (Conran et al., 2009), and it was assigned a log-normal prior distribution (Fig. 2, α) with a mean value of 23.2 Mya and standard deviation of 0.1. In addition, we used two secondary calibrations obtained from Givnish et al. (2015): the divergences of the Angraecinae + Aeridinae clades from their sister clade in Vandeae (21.21 ± 4.20 Mya) and Angraecinae and Aeridinae (13.25 ± 4.00 Mya), which were both assigned normal prior distributions (Fig. 2, β, γ). Figure 2. View largeDownload slide Divergence time estimates for Campylocentrum and related genera based on nuclear ITS, and plastid matK and ycf1 performed with BEAST. Bars represent 95% highest posterior density (HPD) estimates. The asterisk indicates a fossil calibration (α), diamonds indicate secondary calibrations, β = divergence of the clade Angraecinae + Aeridinae in the tribe Vandeae; γ = divergence of subtribes Angraecinae and Aeridinae. A geological timescale is placed at the top; vertical discontinuous grey lines separate two time slices (TSI−TSII) used in the biogeographical analysis. Figure 2. View largeDownload slide Divergence time estimates for Campylocentrum and related genera based on nuclear ITS, and plastid matK and ycf1 performed with BEAST. Bars represent 95% highest posterior density (HPD) estimates. The asterisk indicates a fossil calibration (α), diamonds indicate secondary calibrations, β = divergence of the clade Angraecinae + Aeridinae in the tribe Vandeae; γ = divergence of subtribes Angraecinae and Aeridinae. A geological timescale is placed at the top; vertical discontinuous grey lines separate two time slices (TSI−TSII) used in the biogeographical analysis. The biogeographical history of Campylocentrum was analysed using the likelihood-based dispersal–extinction–cladogenesis (DEC) model implemented in LAGRANGE v.20151128 (Ree et al., 2005; Ree & Smith, 2008), calculating global extinction and dispersal rates and ancestral range reconstructions for each node in the maximum clade credibility tree obtained from BEAST, excluding the five additional orchid species added for calibration purposes. The distribution of Campylocentrum and related genera was split into seven operational areas (Fig. 3) adapted from Morrone (2014): Sub-Saharan Africa (G); Antillean sub-region (A); Brazilian dominion (D); Chacoan dominion (E); Mesoamerica dominion (B); Pacific dominion (C); and Parana dominion (F). Species distribution ranges (presence/absence) included a maximum of three areas per species except for two widespread species, C. fasciola and C. pachyrrhizum, which included two specimens from the extremes of their distributions. Figure 3. View largeDownload slide Ancestral area estimation inferred by LAGRANGE. Pie charts at nodes indicate the probabilities of ancestral areas, when < 0.1 where combined in the ‘remainder’ category (black sections). A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. A geological timescale is placed at the top, and node numbers are indicated below branches. Vertical discontinuous grey lines separate two time slices (TSI = 7−0 Mya and TSII = 7−11 Mya), and inset maps represent the palaeogeographical configuration of the Neotropical region and Amazon Basin in these time slices. Inferred dispersal (X→Y), vicariance (X/X) and peripheral isolated speciation (X\Y) events are represented also in the tree. Figure 3. View largeDownload slide Ancestral area estimation inferred by LAGRANGE. Pie charts at nodes indicate the probabilities of ancestral areas, when < 0.1 where combined in the ‘remainder’ category (black sections). A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. A geological timescale is placed at the top, and node numbers are indicated below branches. Vertical discontinuous grey lines separate two time slices (TSI = 7−0 Mya and TSII = 7−11 Mya), and inset maps represent the palaeogeographical configuration of the Neotropical region and Amazon Basin in these time slices. Inferred dispersal (X→Y), vicariance (X/X) and peripheral isolated speciation (X\Y) events are represented also in the tree. We performed a stratified model that contained two time slices (TS): TSI (11–7 Mya, Tortonian to late Miocene) reflecting the presence of a large lake in the Amazon basin (the Acre System); and TSII (7–0 Mya, late Miocene to present) capturing the final drainage of the Acre System forming the modern Amazon and Orinoco Basins (Fig. 3; Hoorn et al., 2010). For each TS, a different matrix of bidirectional dispersal rates reproducing the geographical connectivity among the areas was applied (Viruel et al., 2015;  Appendix 4). RESULTS Molecular datasets This study produced 318 new sequences for Campylocentrum and related genera ( Appendix 1), which were combined with 57 previously published sequences (Carlsward et al., 2003, 2006a) with their species names updated in line with current taxonomic concepts. All six DNA regions were generated for all accessions except nine: five lacking only Xdh and none lacking more than one region. The complete matrix has a length of 6128 bp, of which 782 (12.7%) were potentially parsimony informative. ITS (20.1% of positions), trnL-F (15.1%) and ycf1 (17.4%) were the most informative. Due to its greater length (1729 bp aligned), ycf1 provided the greatest absolute number of informative sites (301; Table 2). Phylogenetic relationships In the parsimony analysis, the complete dataset (ITS, Xdh and plastid regions) produced 16 most-parsimonious trees of length 1997 steps with a consistency index (CI) = 0.75 and retention index (RI) = 0.90. Analyses performed separately for each region show similar indices (Table 2). The BI and MP trees obtained from all data combined had similar topologies (Figs 1, S1). Most relationships among clades were similar also to the combined plastid dataset (Fig. S4); however, the independent nuclear analyses (ITS and Xdh) were less well resolved (Figs S2, S3). The most congruent topology among all analyses resolved Campylocentrum (BP 100, PP 1.00) and Dendrophylax (BP 97, PP 1.00) as sister clades (Fig. 1). Successively diverging from the base are Angraecum moandense De Wild. (BP 100, PP 1.00), A. distichum Lindl. (BP 100 %, PP 1) and A. leonis. When performed separately, the Xdh and ITS analyses show A. moandense in a polytomy with Campylocentrum and Dendrophylax, so consequently these nuclear regions did not identify the Neotropical clade (Campylocentrum + Dendrophylax) observed with the plastid regions (Figs S2–S4). Two major clades occur in Campylocentrum (A and B), both strongly supported in the combined BI and MP trees (BP 100, PP 1.00). A similarly strongly supported topology is observed in the separate ITS, Xdh and plastid trees (Figs 1, 2, S2–S4). However, the ITS analysis did not produce a monophyletic Campylocentrum; clades A and B fall in a polytomy with Dendrophylax and A. moandense (Fig. S2). Clade A comprises C. section Campylocentrum as sister to C. sections Dendrophylopsis Cogn. in Mart. + Pseudocampylocentrum Cogn. in Mart. in the combined BI tree, a result also observed in the plastid tree (Figs 1, S4), but the Xdh and ITS (individual) trees make these into polytomies (Figs 2, 3). Although section Campylocentrum has a high PP (0.98) with all data combined, in the combined MP analysis it is split into two clades along a polytomy: one includes C. calostachyum (Barb.Rodr.) E.M.Pessoa & M.Alves, C. serranum E.Pessoa & Alves and C. ulaei Cogn. (BP 100, PP 1.00) and other with the remaining species (BP 92, PP 1.00). Among these two clades in the same polytomy is also C. sections Dendrophylopsis + Pseudocampylocentrum with moderate to strong support (BP 83, PP 0.99; Figs 1, S1). Conversely, the plastid tree provides good resolution for C. section Campylocentrum with strong support in both BI and MP analyses (BP 97, PP 1.00; Fig. S4). Clade B includes two new sections described in the taxonomic treatment below, C. sections Laevigatum E.Pessoa & M.W.Chase and Teretifolium E.Pessoa & M.W.Chase, both strongly supported in the combined BI and MP analysis (BP 99, PP 1.00 and BP 100, PP 1.00, respectively; Figs 1, S1). Both sections are also strongly supported in the plastid tree (BP 99, PP 1.00 and BP 100, PP 1.00, respectively; Fig. S4). The ITS and Xdh trees include only C. section Teretifolium as monophyletic with good support (BP 88, PP 1.00 and BP 86, PP 1.00, respectively) (Figs S2, S3). The datasets used in this study were not able to clarify relationships among taxa in C. section Teretifolium, except for C. sellowii (Rchb.f.) Rolfe as sister to the rest of this clade (BP 100, PP 1.00), which is internally unresolved due to low genetic variation. A similar topology is provided by the combined plastid tree (Fig. S4). Dating and biogeography The divergence time between African A. moandense and the Neotropical clade formed by Campylocentrum and Dendrophylax was estimated as late Miocene at 8.53 Mya [95% highest posterior density (HPD) = 10.7–6.4 Mya]. The ancestral area estimation suggests a first long-distance dispersal from Africa (G) to the Antilles (A) (node 3, probability = 0.7, Figs 2, 3). The most recent common ancestor (MRCA) of Campylocentrum and Dendrophylax was most probably restricted to the Antilles (A), and the estimated split for these two genera was almost contemporary with the previous node, 8.21 Mya (95% HPD = 10.5–5.9 Mya), supporting a hypothesis of rapid divergence after colonization of the Neotropics (node 4, probability = 0.61, Figs 2, 3). The MRCA of Campylocentrum had a possible disjunct distribution between the Antilles (A) and Parana (F) dominions, suggesting a second long-distance dispersal (A→F; node 7, probability = 0.2, Figs 2, 3). Clades A (C. section Campylocentrum + C. section Dendrophylopsis + C. section Pseudocampylocentrum) and B (C. section Laevigatum + C. section Teretifolium) also diverged in the late Miocene, 7.22 Mya (95% HPD = 9.2–5.2 Mya), with a subsequent dispersal and expansion to the Chacoan dominion (E) and Central America (B), respectively (node 8, probability = 0.4; node 20, probability = 0.26, Figs 2, 3). These clades started their diversification in the Miocene/Pliocene and late Miocene, 2.9 Mya (95% HPD = 4.2–1.7 Mya) and 5.38 Mya (95% HPD = 6.79–3.97 Mya), respectively (nodes 8 and 20, Figs 2, 3). Divergence of the MRCA of C. sections Dendrophylopsis + Pseudocampylocentrum was probably 4.79 Mya (95% HPD = 6.10–3.48 Mya), and it was most probably distributed exclusively in Central America (B), which is also the case for the MCRAs for each of these sections. The wide distribution of some species in these sections was probably reached in the Pleistocene (node 21, probability = 0.43; node 22, probability = 0.47, Figs 2, 3). In C. section Campylocentrum, two main clades also diverged in the late Miocene/Pliocene by vicariance, 5.16 Mya (95% HPD = 6.6–3.7 Mya; node 27, probability = 0.42, Figs 2, 3), one clade formed by C. calostachyum, C. serranum and C. ulaei with an origin in the Parana dominion (F) and an MRCA from 2.69 Mya (95% HPD = 3.8–1.6 Mya; node 28, probability = 0.96, Figs 2, 3), and the second clade with an MCRA from 4.28 Mya (95% HPD = 4.60–2.32 Mya) originating in Central America (B), after which it diversified rapidly in the Pleistocene (node 30, probability = 0.35, Figs 2, 3). In clade B, the MRCA of both C. sections Laevigatum and Teretifolium also had an estimated origin and diversification in the Pleistocene, 1.76 Mya (95% HPD = 2.7–0.8 Mya) and 2.04 Mya (95% HPD = 2.9–1.18 Mya), respectively, mainly in Parana (F) (node 9, probability = 0.83; node 13, probability = 0.46, Figs 2, 3). Low resolution in reconstructing distributions in the terminal clades of C. section Dendrophylopsis and the ‘C. micranthum (Lindl.) Maury complex’ was observed due to the wide distribution of the species and poor sampling of the leafless species. Only the original area is known at this point, and considering the putative subsequent expansions more detailed studies focused on these groups might provide a better understanding of their evolutionary history. Divergence times, 95% HPDs and ancestral areas with their probabilities for all tree nodes are presented in  Appendix 2. DISCUSSION Phylogenetic relationships Phylogenetic analysis of the combined data provides good resolution for the majority of clades, which allowed us to discuss and define with a high level of confidence some taxonomic groups. The results support the African A. moandense [cited as A. chevalieri Summerh. by Carlsward et al., 2006a; see Pessoa & Alves, 2016c; A. section Angraecoides (Cordem.) Garay sensuGaray (1973) (= genus Angraecoides sensuSzlachetko et al., 2013)], as sister to the Atlantic clade (Campylocentrum +Dendrophylax; Carlsward et al., 2006a; Szlachetko et al., 2013; Andriananjamanantsoa et al., 2016). Angraecum section Angraecoides includes species mainly distributed in West Tropical Africa (Govaerts et al., 2016) and is characterized by an often one-flowered inflorescence, green to greenish yellow flowers and cylindrical or clavate spurs (Garay, 1973). Their floral morphology resembles that of Dendrophylax (e.g. D. barrettiae Fawc. & Rendle), although their leafy habit is similar to some Campylocentrum spp. (e.g. C. crassirhizum Hoehne). Campylocentrum and Dendrophylax are demonstrated to be monophyletic, as first shown by Carlsward et al. (2003, 2006a) and confirmed by Szlachetko et al. (2013) and Andriananjamanantsoa et al. (2016) who included a better sampling of Angraecum s.l. Despite the shared leafless condition, the genera are easily distinguished by their inflorescence arrangement, mainly many-flowered with flowers densely and distichously arranged in Campylocentrum in contrast to few-flowered, loosely and not distichously arranged in Dendrophylax (Carlsward, 2014a, b). Inflorescence structure and flower size can be considered as synapomorphies for Campylocentrum in Angraecinae. This study presents the first phylogenetic analysis of Campylocentrum including a representative sample of its morphological diversity and distribution. The missing taxa are probably either synonyms of widely distributed species or narrow endemics closely related to the species included. In the genus, two major clades are strongly supported (A and B). Both were also recovered by Carlsward et al. (2003). However, their study included only six species, of which samples of C. stenanthum Schltr. were mis- identified as C. micranthum, C. robustum Cogn. and C. schiedei (Rchb.f.) Benth. ex Hemsl. (vouchers checked at FLAS). For clade B, those authors used three samples including C. jamaicense (Rchb.f. ex Griseb.) Benth. ex Fawc. (misidentified as C. micranthum) and C. crassirhizum [as C. lansbergii (Rchb.f.) Schltr.] (vouchers checked at FLAS). The lack of a generic revision is the main reason for this misinterpretation of species identities. Pessoa, Maciel & Alves (2015) and Pessoa & Alves (2015a, b) studied C. micranthum, which was recognized as a ‘nomenclatural complex’ requiring new combinations, re-establishments of some old names and re-circumscription of most species. Most vegetative and floral features previously used in taxonomic treatments in the genus (Cogniaux, 1906; Todzia, 1980; Bogarín & Pupulin, 2010; Pessoa & Alves, 2016a, b) are here homoplastic, consequently having limited use for delimiting natural taxa. However, capsule morphology clearly distinguishes clade A (six-ribbed capsule) from clade B (non-ribbed). A non-ribbed capsule is thus interpreted here as the synapomorphy for clade B, whereas a six-ribbed capsule is a symplesiomorphy since it is similar to those of Dendrophylax and Angraecum. Clade A is widespread in the Neotropics and composed of the three sections proposed by Cogniaux (1906). The sister of the rest of this clade, C. section Campylocentrum, is the most diverse section in the genus and includes species with conduplicate leaves and elongate stems, but its circumscription sensuCogniaux (1906) is found not to be monophyletic. It is re-circumscribed here, excluding species with non-ribbed capsules (clade B). With this new circumscription C. section Campylocentrum is strongly supported in the combined BI analysis and in MP and BI for the plastid data. However, it is not supported in the combined MP analysis, a result of the polytomy including the C. ulaei group (C. calostachyum, C. serranum and C. ulaei). Three strongly supported major subclades are recognized in C. section Campylocentrum. Sister to the rest is the C. ulaei group, endemic to Parana (F) and characterized by relatively longer inflorescences. The sister group of this clade is mainly distributed outside the Parana dominion (except for C. micranthum) and is split into two subclades. The first (C. tenellum Todzia, C. brenesii Schtlr.) includes species from the Antilles (A), Central America (B) and the Pacific dominions (C). The other is the C. micranthum complex (Pessoa et al., 2015), which includes seven species sampled here with relatively shorter and congested inflorescences. In the C. micranthum complex, C. huebneri Mansf. + C. mattogrossense Hoene are sister to the rest, which includes the other five species sampled. One of the latter is C. schiedei, the type of the genus. Among C. panamense Ames, C. micranthum, C. kuntzei Cogn. ex Kuntze and C. stenanthum, there is a clear difference in spur morphology; C. huebneri has the longest spur in the genus (1.0–2.0 cm long vs. 0.2–0.6 cm long), and C. mattogrossense has a straight spur (vs. commonly slightly curved to inflexed in the other species; Pessoa & Alves, 2015a, b). The concept of C. micranthum has been widely and incorrectly applied (Bogarín & Pupulin, 2010; Pessoa & Alves, 2015a, b;,Pessoa et al., 2015), and our results support some of the proposed names in the complex, although many others were not included in this study. A broad study of the group is required to better understand the species boundaries. Campylocentrum section Campylocentrum is sister to a clade formed by C. sections Dendrophylopsis + Pseudocampylocentrum. The leafless species of Campylocentrum are included in C. section Dendrophylopsis, which is also distinguished by having a viscidium comprising a single part (Pessoa & Alves, 2016b) and roots with thicker cell walls in the endovel- amen, exoderm and endoderm (Pessoa et al., 2017). All these features are considered synapomorphies of C. section Dendrophylopsis. Two subclades occur in this section, one formed by C. grisebachii Cogn. + C. pachyrrhizum that is characterized by the apex of the anther cap rounded or truncate and a second one with C. fasciola + C. tyrridion Garay & Dunst. ex Foldats that has a bilobed anther cap (Pessoa & Alves, 2016b). Both clades are supported in the BI analyses but not by MP. The monospecific C. section Pseudocampylocentrum morphologically resembles species placed in this study in C. section Teretifolium (clade B) based on its cylindrical leaves and granulose root exterior. Although externally similar, the root surface is composed of unicellular absorbent hairs, whereas in C. section Teretifolium roots have tufts of epivelamen (Pessoa et al., 2017). Terete leaves seem to be an adaptive convergence and are also found in other genera in Angraecinae [e.g. Nephrangis (Schltr.) Summerh. and Solenangis Schltr.]. Our results confirm that leaflessness and terete leaves are homoplastic, arising more than once in the subtribe (Carlsward et al., 2006a; Pessoa & Alves, 2016a). Clade B is almost restricted to the Parana dominion (F), except for C. jamaicense endemic to the Antilles. It includes C. sections Laevigatum and Teretifolium, both strongly supported and characterized by non-ribbed capsules. Campylocentrum section Teretifolium is distinguished from its sister by having cylindrical leaves and granulose to tuberculate root surfaces. Although a taxonomic treatment exists for the group (Pessoa & Alves, 2016a), relationships among its species are still poorly resolved. Our study was unable to disentangle it, except for recognizing C. sellowii as sister to the other four species included. Low interspecific genetic divergence probably reflects rapid and recent diversification producing a lack of resolution (Vatanparast et al., 2011; Matos-Maraví et al., 2013; Snak et al., 2016; Liu et al., 2016). Two species of C. section Laevigatum have been previously considered synonyms of C. micranthum (C. section Campylocentrum), C. brevifolium and C. jamaicense. Both are similar in habit, especially their conduplicate leaves with asymmetrically bilobed apices, as observed in the majority of genera in Angraecinae. However, contrary to our expectation, species of these sections are not closely related to each other in the tree (Fig. 1). Campylocentrum section Laevigatum is distinguished mainly by non-ribbed capsules in contrast to six-ribbed capsules in C. section Campylocentrum. It also differs by roots with ○-thickened cell walls in the exoderm instead of ∩-thickened cell walls as in other species of the genus (Pessoa et al., 2017). Although only two species have been anatomically analysed [C. jamaicense by Carlsward, Stern & Bytebier (2006b) and C. crassirhizum by Pessoa et al. (2017)], this character is suggested here as a possible synapomorphy for the section. Campylocentrum section Laevigatum has two subclades. One of them is C. pauloense Hoehne & Schltr. + C. spannagelii Hoehne + C. densiflorum Cogn. + C. brachycarpum Cogn., all four endemic to the southern Parana dominion (northern Argentina, southern Brazil and Paraguay). Campylocentrum brachycarpum and C. densiflorum represent in this study the C. organense (Rchb.f.) Rolfe group (C. organense absent from this study) characterized by relatively large, ovate floral bracts that cover all or most of the ovary (vs. relatively shorter and deltoid in other leafy species). The second subclade includes C. neglectum + C. robustum + C. crassirhizum + C. jamaicense, mainly distributed in the Atlantic Forest and the South American ‘dry diagonal’ (formed by caatinga, cerrado and chaco vegetation sensuPrado & Gibbs, 1993; the Chacoan domain), except for C. jamaicense, which occurs in the Antilles. Campylocentrum crassirhizum and C. jamaicense, despite their large disjunction, are similar morphologically and have been distinguished only by spur length. Our analyses did not provide a clear understanding of the relationship between these species. The two accessions of C. crassirhizum did not fall together: the one from southeastern Brazil is sister to the other accession from northeastern Brazil plus the specimens of C. jamaicense. However, this topology is not supported by MP, and more studies including population genetics will be important to understand relationships of these two species. Divergence time and biogeography The MRCA of A. moandense and the Neotropical genera of Angraecinae arrived in the Antilles (A) during the late Miocene (c. 8.53 Mya; Fig. 3). This date implies a relatively recent trans-Atlantic long- distance dispersal event like that of Pitcairnia feliciana (A.Chev.) Harms & Mildbraed (Bromeliaceae) c. 9.3 Mya (Givnish et al., 2011) and Maschalocephalus Gilg. & K.Schum. (Rapateaceae) c. 7.3 Mya (Givnish et al., 2004). However, for these two species the migration events occurred in the opposite direction from that proposed here for Angraecinae and were not followed by diversification (Givnish et al., 2004, 2011). Moreover, although it occurred much earlier in the Oligocene/Miocene (c. 35 Mya), Caricaceae (the papaya family) are a good example of an African taxon that experienced a radiation in the Neotropics (Carvalho & Renner, 2012; Christenhusz & Chase, 2013). Carvalho & Renner (2012) suggested that the ancestor of the group arrived first in the Antilles, in agreement with the results here for Angraecinae. In the Antilles (A), divergence of Campylocentrum and Dendrophylax (c. 8.22 Mya) occurred soon after dispersal of their MRCA from Africa (c. 8.53 Mya). The MRCA of Campylocentrum reached the Parana dominion (F) through a second long-distance dispersal also in the late Miocene (c. 7.22 Mya; Fig. 3). According to Pound et al. (2011) in this epoch a portion of the present northern coast of Brazil (Maranhão and Piauí States) was inundated. The authors indicated that, based on a data/model biome reconstruction, the ancient coast was covered by tropical evergreen forests. It is possible that the MRCA of Campylocentrum arrived in that region instead of the coast of the Guyanas, which even though closest to the Antilles was covered at that time by tropical savannas. The once-flooded area and the ancient coast were probably where the ancestor arrived first in South America and are currently occupied by cerrado and caatinga vegetation (the Chacoan dominion sensuMorrone, 2014). Extinction events caused by climate and consequently vegetation changes (Hoorn et al., 2010) could explain the current absence of the genus in this area. Migrations from Mesoamerica (B) to Paraná (F) took place during the late Miocene, Pliocene and Pleistocene in clade A, when the Andes experienced accelerated rates of uplift (Hoorn et al., 2010). This could suggest that the Andes might have not been a completely impassable geographical barrier for genetic exchange, as reported by Sánchez-Baracaldo (2004) for ferns and Pérez-Escobar et al. (2017) for Cymbidieae and Pleurothallidinae (Orchidaceae), who identified multiple migrations and re-colonizations over the Andes across long timescales. In the Pliocene, clade A expanded its distribution from the Antilles (A) to Mesoamerica (B) and clade B from Parana (F) to Chacoan dominion (E). In both cases, they were followed by a rapid diversification, originating by peripheral isolation events in the five main clades recognized in the genus (here considered as C. sections Campylocentrum, Dendrophylopsis, Laevigatum, Pseudocampylocentrum and Teretifolium; Fig. 3). According to Hoorn et al. (2010) and Turchetto-Zolet et al. (2013), climate changes and orogenic events with consequent changes in the drainage systems and vegetation (Salzmann et al., 2008) occurred in the Pliocene and contributed to shaping the current diversity of plant lineages in the Neotropics. In this region, several taxonomic groups exhibited diversification in this epoch, including bellflowers (Campanulaceae; Lagomarsino et al., 2016), tank-epiphytic bromeliads (Givnish et al., 2011, 2014), some Fabaceae such as Chamaecrista (L.) Moench. series Coriaceae (Benth.) H.S.Irwin & Barneby (Rando et al., 2016), Platymiscium Vogel (Saslis-Lagoudakis et al., 2008) and Canavalia Adans. (Snak et al., 2016), and Solanaceae subtribe Physalinae (Zamora-Tavares et al., 2016). Central America (B) was an important centre of origin for some clades of Campylocentrum, especially those in clade A. During the Pliocene, sections Campylocentrum (in part, c. 4.28 Mya, node 30, except for the C. ulaei group), Dendrophylopsis (c. 4.06 Mya) and Pseudocampylocentrum (c. 4.79 Mya) expanded their distributions from this area followed by diversification (Fig. 3). Orogenic events occurring in the Pliocene, such as the definitive closure of the Isthmus of Panama (Knowlton & Weigt, 1998; although other authors have stated recently that it was earlier, Montes et al., 2015) and volcanism (Weinberg, 1992; Vogel et al., 2006; Mann, 2007) seem to have been relevant to the diversification of these groups in the area. Nevertheless, Parana (F) was the centre of origin for C. section Teretifolium (c. 1.76 Mya) and part of C. section Laevigatum (node 17, c. 1.23 Mya), with evidence of in situ diversification in the Pleistocene. The C. ulaei group (clade A, C. section Campylocentrum) is also endemic to Parana (F). It diverged from its sister by a vicariant event (B/F) in the early Pliocene, and the species diverged during the Plio-Pleistocene, which matches groups in clade B (Fig. 3). Pleistocene climatic fluctuations (Behling & Negrelle, 2001) played an important role in speciation in the Parana dominion (F), as argued by Hooghiemstra & van der Hammen (1998). At that time, the glacial and interglacial periods may have resulted in successive fragmentation and expansion of forests, promoting allopatric and parapatric speciation, also termed as the refuge theory (Haffer, 1969; Prance, 1973). The climatic fluctuations that occurred during the Pleistocene were also important in the evolution of the C. micranthum complex (clade A, C. section Campylocentrum). It experienced an extensive diversification with expansion of its distribution from Mesoamerica (B), Pacific (C) and Brazil (D), ranging to the Antilles (A), Chacoan (E) and Paraná dominions (F) in the last 1 My (Fig. 3). In a study of ecological niche modelling, Kolanowska (2015) suggested several refugia for two species of the C. micranthum complex at the last glacial maximum. Although with a restricted approach, they cited the presence of several refugia for the group in Mesoamerica (B) and Pacific (C), a result supported by our study. Similar rapid diversification in the last 1 My was experienced by some Andean clades of Campanulaceae (Lagomarsino et al., 2016) and Brazilian saxicolous Chamaecrista series Coriaceae (Rando et al., 2016). However, the diversification of the latter was not followed by range expansion, and their diversity more probably arose in situ. CONCLUSIONS Phylogenetic reconstruction of Campylocentrum confirmed it as monophyletic in its current circumscription. It is composed of two major clades, one that includes the three sections previously proposed by Cogniaux (1906) and the second with two new sections, C. sections Laevigatum and Teretifolium, first proposed in this study (see taxonomic treatment below; these species had been included previously in C. section Campylocentrum). Although the majority of the morphological characters are here considered homoplastic, the two main clades included in the genus are clearly distinguished by their capsules (six-ribbed vs. non-ribbed). The sections are recognized only by sets of features, except for C. section Dendrophylopsis that exhibits clear synapomorphies such as leaflessness, one-parted viscidia and roots with endovelamen, exoderm and endoderm cell walls thicker than those in the leafy species. The divergence time analysis and ancestral distribution reconstruction indicated that Campylocentrum is relatively young and arose in the late Miocene. Its MRCA probably had a disjunct distribution between the Antilles and Paraná dominions. During the Pliocene, the five sections of the genus had already diverged and expanded their distributions to the Mesoamerican, Chacoan, Pacific and Brazilian dominions. Climatic fluctuations in the Pleistocene probably played an important role in spurring diversification, especially in the sections of clade B in Parana (F) and the C. micranthum complex (clade A) in the Mesoamerican, Pacific and Brazilian dominions. Therefore, the evolutionary history of Campylocentrum would seem to confirm the importance of orogenic events occurring in the Pliocene and climatic fluctuations in the Pleistocene for diversification in the Neotropics. TAXONOMIC TREATMENT 1 Campylocentrum Benth., J. Linn. Soc. Bot. 18: 337. 1881. Type species: Campylocentrum schiedei (Rchb.f.) Benth. ex Hemsl. (basionym: Angraecum schiedei Rchb.f.; originally published as Todaroa micrantha A.Richard & Galeotti). Todaroa A.Rich. & Galeotti, Ann. Sci. Nat. Bot. 3: 28. 1845. nom. illeg. [non Todaroa Parl. Hist. Nat. ìles Canaries 2: 155. 1843. Apiaceae (= Umbelliferae)]. Type species: Todaroa micrantha A.Rich. & Galeotti, nom. illeg. 1.1. Campylocentrum section Campylocentrum This widespread section is characterized by leafy species with cylindrical and smooth roots, conduplicate leaves and six-ribbed capsules. It includes about the half of the species of the genus (38 species). List of species: C. amistadense Bogarín, C. antioquiense Kolan. & Szlach., C. apiculatum Schltr., C. asplundii Dodson, C. brenesii Schltr., C. calostachyum (Barb.Rodr.) ined., C. carlos-parrae Kolan. & Szlach., C. cornejoi Dodson, C. ecuadorense Schltr., C. embreei Dodson, C. escobariae Kolan. & Szlach., C. hirtellum Cogn., C. hirtzii Dodson, C. hondurense Ames, C. huebneri Mansf., C. huebnerioides D.E.Benn. & Christenson, C. lansbergii (Rchb.f.) Schltr., C. kuntzei Cogn. ex Kuntze, C. madisonii Dodson, C. mattogrossense Hoehne, C. micranthum (Lindl.) Maury, C. microphyllum Ames & Correll, C. minus Fawc. & Rendle, C. natalieae Carnevali & I.Ramírez, C. palominoi Kolan., O.Pérez & E.Parra, C. panamense Ames, C. peniculus Schltr., C. polystachyum (Lindl.) Rolfe, C. pugioniforme (Klotzsch) Rolfe, C. pygmaeum Cogn., C. queremalense Kolan. & Szlach., C. serranum E.M.Pessoa & M.Alves, C. schiedei (Rchb.f.) Benth. ex Hemsl., C. schneeanum Foldats, C. stenanthum Schltr., C. steyermarkii Foldats, C. tenellum Todzia, C. ulaei Cogn. 1.2. Campylocentrum section Dendrophylopsis Cogn. in Mart., Fl. Bras. 3(6): 504. 1906. Type species: Campylocentrum fasciola (Lindl.) Cogn., designated by Pessoa & Alves (2016b). This widespread section includes the leafless species with dorsiventrally flattened or cylindrical, smooth roots, leaves always reduced to achlorophyllous scales and six-ribbed capsules. It comprises13 species. List of species: C. amazonicum Cogn., C. benelliae E.M.Pessoa & M.Alves, C. fasciola (Lindl.) Cogn., C. fernandezii Kolan. & Szlach., C. generalense D.Bogarín & F.Pupulin, C. insulare Siqueira & E.M.Pessoa, C. minutum C.Schweinf., C. pachyrrhizum (Rchb.f.) Rolfe, C. paludosum E.M.Pessoa & M.Miranda, C. pubirhachis Schltr., C. tenue (Lindl.) Rolfe, C. tyrridion Garay & Dunst. ex Foldats 1.3. Campylocentrum section Laevigatum E.M.Pessoa & M.W.Chase, sect. nov. Type species: Campylocentrum brevifolium (Lindl.) E.M.Pessoa & M.Alves, Kew Bull. 70: 43. 2015. Similar to C. section Campylocentrum but differs by its unribbed capsules (vs. six-ribbed). This section is almost restricted to eastern South America, except for C. jamaicense from the Antilles; it includes leafy species with cylindrical, smooth roots, conduplicate leaves and unribbed capsules. It is composed of 15 species. List of species: C. brevifolium (Lindl.) E.M.Pessoa & M.Alves, C. brachycarpum Cogn., C. carvalhoi E.M.Pessoa & M.Alves, C. crassirhizum Hoehne, C. densiflorum Cogn., C. hasslerianum Hoehne, C. itatiaiae E.M.Pessoa & M.Alves, C. intermedium (Rchb.f. & Warm.) Cogn., C. jamaicense (Rchb.f. ex Griseb.) Benth. ex Fawc., C. neglectum (Rchb.f. & Warm.) Cogn., C. organense (Rchb.f.) Rolfe, C. pauloense Hoehne & Schltr., C. robustum Cogn., C. schlechterianum E.M.Pessoa & M.Alves, C. spannagelii Hoehne 1.4. Campylocentrum section Pseudocampylocentrum Cogn. in Mart., Fl. Bras. 3(6): 504. 1906. Type species: Campylocentrum poeppigii (Rchb.f.) Rolfe. This monospecific section is distributed in the Antilles, Central America and northern South America. The only taxon recognized so far is a leafy species, with cylindrical, minutely granulose roots, terete, reduced leaves and six-ribbed capsules. List of species: C. poeppigii (Rchb.f.) Rolfe 1.5. Campylocentrum section Teretifolium E.M.Pessoa & M.W.Chase, sect. nov. Type species: Campylocentrum ornithorrhynchum (Lindl.) Rolfe, Orchid Review 11: 246. 1903. Similar to C. section Pseudocampylocentrum, but differs by its granulose root surface, comprising tufts of epivelamen (vs. comprising unicellular absorbent hairs) and unribbed capsules (vs. six-ribbed). This section is endemic to eastern South America. It includes leafy species with cylindrical, minutely granulose to tuberculate roots, terete leaves and unribbed capsules. It includes six species. List of species: C. labiakii E.M.Pessoa & M.Alves, C. ornithorrhynchum (Lindl.) Rolfe, C. parahybunense (Barb.Rodr.) Rolfe, C. pernambucense Hoehne, C. sellowii (Rchb.f.) Rolfe, C. wawrae (Rchb.f. ex Beck) Rolfe SUPPLEMENTARY INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Figure S1. Strict consensus tree from MP analysis of nuclear (ITS and Xdh) and plastid (matK, rpl32-trnL, trnL-F, ycf1) regions combined (length = 1997 steps, CI = 0.75 and RI = 0.90). Numbers below the branches are bootstrap percentages (≥50). Figure S2. Phylogenetic relationships of Campylocentrum obtained by Bayesian inference of ITS data. Posterior probabilities (≥0.5) are indicated above branches and maximum parsimony bootstrap percentages (≥50) are indicated below branches. Figure S3. Phylogenetic relationships of Campylocentrum obtained from a Bayesian inference of Xdh data. Posterior probabilities (≥0.5) are indicated above branches and maximum parsimony bootstrap percentages (≥50) are indicated below branches. Figure S4. Phylogenetic relationships of Campylocentrum obtained from by Bayesian inference of combined plastid regions (matK, rpl32-trnL, trnL-F, ycf1). Posterior probabilities (≥0.5) are indicated above branches and maximum parsimony bootstrap percentages (≥50) are indicated below branches. ACKNOWLEDGMENTS We thank the staff of the Jodrell Laboratory (Royal Botanic Gardens, Kew) for help and useful feedback. The directors/curators of living collections at the Jardim Botânico do Rio de Janeiro, Instituto de Botânica de São Paulo, University of Costa Rica (project number 814-B5-A87) and Heidelberg University allowed us to collect tissue for DNA extractions. Thanks to those who funded our fieldwork including the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), the US National Science Foundation (DEB-0946618), Velux Stiftung and the Beneficia Foundation. The first author thanks CNPq and CAPES for a PhD fellowship. REFERENCES Andriananjamanantsoa HN, Engberg S, Louis EEJr, Brouillet L. 2016. Diversification of Angraecum (Orchidaceae, Vandeae) in Madagascar: revised phylogeny reveals species accumulation through time rather than rapid radiation. PLoS One  11: e0163194. Google Scholar CrossRef Search ADS PubMed  Behling H, Negrelle RRB. 2001. Tropical rain forest and climate dynamics of the Atlantic lowland, southern Brazil, during the late Quaternary. Quaternary Research  56: 383– 389. Google Scholar CrossRef Search ADS   Bentham G. 1881. Notes on Orchidaceae. Journal of the Linnean Society, Botany  18: 281– 361. Google Scholar CrossRef Search ADS   Bogarín D, Pupulin F. 2010. The genus Campylocentrum (Orchidaceae: Angraecinae) in Costa Rica: a revision. Harvard Papers in Botany  15: 353– 414. Google Scholar CrossRef Search ADS   Brown JM, Lemmon AR. 2007. The importance of data partitioning and the utility of Bayes factors in Bayesian phylogenetics. Systematic Biology  56: 643– 655. Google Scholar CrossRef Search ADS PubMed  Carlsward BS. 2014a. Campylocentrum. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN, eds. Genera orchidacearum, Epidendroideae (part 3) . New York: Oxford University Press, 370– 372. Carlsward BS. 2014b. Dendrophylax. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN, eds. Genera orchidacearum, Epidendroideae (part 3) . New York: Oxford University Press, 383– 385. Carlsward BS, Cribb P. 2014. Angraeciinae. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN, eds. Genera orchidacearum, Epidendroideae (Part 3) . New York: Oxford University Press, 338– 344. Carlsward BS, Stern WL, Bytebier B. 2006b. Comparative vegetative anatomy and systematics of the angraecoids (Vandae, Orchidaceae) with an emphasis on the leafless habit. Botanical Journal of the Linnean Society  151: 165– 218. Google Scholar CrossRef Search ADS   Carlsward BS, Whitten WM, Williams NH. 2003. Molecular phylogenetics of Neotropical leafless Angraecinae (Orchidaceae): reevaluation of generic concepts. International Journal of Plant Sciences  164: 43– 51. Google Scholar CrossRef Search ADS   Carlsward BS, Whitten WM, Williams NH, Bytebier B. 2006a. Molecular phylogenetics of Vandeae (Orchidaceae) and the evolution of leaflessness. American Journal of Botany  93: 770– 786. Google Scholar CrossRef Search ADS   Carvalho FA, Renner SS. 2012. A dated phylogeny of the papaya family (Caricaceae) reveals the crop’s closest relatives and the family’s biogeographic history. Molecular Phylogenetics and Evolution  65: 46– 53. Google Scholar CrossRef Search ADS PubMed  Chase MW, Cameron KM, Freudenstein JV, Pridgeon AM, Salazar G, Van den Berg C, Schuiteman A. 2015. An updated classification of Orchidaceae. Botatical Journal of the Linnean Society  177: 151– 174. Google Scholar CrossRef Search ADS   Christenhusz MJM, Chase MW. 2013. Biogeographical patterns of plants in the Neotropics – dispersal rather than plate tectonics is most explanatory. Botanical Journal of the Linnean Society  171: 277– 286. Google Scholar CrossRef Search ADS   Cogniaux CA. 1906. Campylocentrum. In: Martius CFP, Eichlerand AW, Urban I, eds. Flora Brasiliensis 3(6) . Munich and Leipzig: Typographia Regia Wolf et fil., 503– 524. Conran GC, Bannister JM, Lee DE. 2009. Earliest orchid macrofossils: early Miocene Dendrobium and Earina (Orchidaceae: Epidendroideae) from New Zealand. American Journal of Botany  96: 466– 474. Google Scholar CrossRef Search ADS PubMed  Cummings MP, Handley SA, Myers DS, Reed DL, Rokas A, Winka K. 2003. Comparing bootstrap and posterior probability values in the four-taxon case. Systematic Biology  52: 477– 487. Google Scholar CrossRef Search ADS PubMed  Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin, Botanical Society of America  19: 11– 15. Doyle JJ, Doyle JL. 1990. Isolation of plant DNA from fresh tissue. Focus  12: 13− 15. Dressler RL. 1993. The orchids: natural history and classification . Cambridge: Harvard University Press. Dressler RL. 2005. How many orchid species? Selbyana  26: 155– 158. Drummond AJ, Bouckaert RR. 2015. Computational evolution with BEAST . Cambridge: Cambridge University Press. Drummond AJ, Ho SY, Phillips MJ, Rambaut A. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biology  4: e88. Google Scholar CrossRef Search ADS PubMed  Drummond AJ, Rambaut A. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology  7: 214. Google Scholar CrossRef Search ADS PubMed  Drummond AJ, Suchard MA, Xie D, Rambaut A. 2012. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution  29: 1969– 1973. Google Scholar CrossRef Search ADS PubMed  Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic acids research  32: 1792– 1797. Google Scholar CrossRef Search ADS PubMed  Erixon P, Svennblad B, Britton T, Oxelman B. 2003. Reliability of Bayesian posterior probabilities and bootstrap frequencies in phylogenetics. Systematic Biology  52: 665– 673. Google Scholar CrossRef Search ADS PubMed  Ford CS, Ayres KL, Toomey N, Haider N, Stahl JA, Kelly LJ, Wikström N, Hollingsworth PM, Duff RJ, Hoot SB, Cowan RS, Chase MW, Wilkinson MJ. 2009. Selection of candidate coding DNA barcoding regions for use on land plants. Botanical Journal of the Linnean Society  159: 1– 11. Google Scholar CrossRef Search ADS   Freudenstein JV, Chase MW. 2015. Phylogenetic relationships in Epidendroideae (Orchidaceae), one of the great flowering plant radiations: progressive specialization and diversification. Annals of Botany  115: 665– 681. Google Scholar CrossRef Search ADS PubMed  Garay LA. 1973. Systematics of the genus Angraecum (Orchidaceae). Kew Bulletin  28: 495– 516. Google Scholar CrossRef Search ADS   Givnish TJ, Barfuss MH, Van Ee B, Riina R, Schulte K, Horres R, Gonsiska PA, Jabaily RS, Crayn DM, Smith JA, Winter K, Brown GK, Evans TM, Holst BK, Luther H, Till W, Zizka G, Berry PE, Sytsma KJ. 2011. Phylogeny, adaptive radiation, and historical biogeography in Bromeliaceae: insights from an eight-locus plastid phylogeny. American Journal of Botany  98: 872– 895. Google Scholar CrossRef Search ADS PubMed  Givnish TJ, Barfuss MH, Van Ee B, Riina R, Schulte K, Horres R, Gonsiska PA, Jabaily RS, Crayn DM, Smith JA, Winter K, Brown GK, Evans TM, Holst BK, Luther H, Till W, Zizka G, Berry PE, Sytsma KJ. 2014. Adaptive radiation and diversification in Bromeliaceae: insights from a 7-locus plastid phylogeny. Molecular Phylogenetics and Evolution  71: 55– 78. Google Scholar CrossRef Search ADS PubMed  Givnish TJ, Millam KC, Evans TM, Hall JC, Pires JC, Berry PE, Sytsma KJ. 2004. Ancient vicariance or recent long-distance dispersal? Inferences about phylogeny and South American-African disjunction in Rapateaceae and Bromeliaceae based on ndhF sequence data. International Journal of Plant Sciences  165: S35– S54. Google Scholar CrossRef Search ADS   Givnish TJ, Spalink D, Ames M, Lyon SP, Hunter SJ, Zuluaga A, Iles WJD, Clements MA, Arroyo MTK, Leebens-Mack J, Endara L, Kriebel R, Neubig KM, Whitten WM, Williams NH, Cameron KM. 2015. Orchid phylogenomics and multiple drivers of their extraordinary diversification. Proceedings of the Royal Society of London B  282: 20151553. Google Scholar CrossRef Search ADS   Górniak M, Paun O, Chase MW. 2010. Phylogenetic relationships within Orchidaceae based on a low-copy nuclear coding gene, Xdh: congruence with plastid DNA results. Molecular Phylogenetics and Evolution  56: 784– 795. Google Scholar CrossRef Search ADS PubMed  Govaerts R, Pfahl J, Campacci MA, Baptista DH, Tigges H, Shaw J, Cribb P, George A, Kreuz K, Wood J. 2016. World checklist of Orchidaceae. Richmond: The Board of Trustees of the Royal Botanic Gardens, Kew. Available at: http://apps.kew.org/wcsp/ (accessed 7 August 2016). Haffer J. 1969. Speciation in Amazonian forest birds. Science (New York, N.Y.)  165: 131– 137. Google Scholar CrossRef Search ADS PubMed  Hooghiemstra H, van der Hammen T. 1998. Neogene and Quaternary development of the Neotropical rain forest: the forest refugia hypothesis, and a literature overview. Earth Science Reviews  44: 147– 183. Google Scholar CrossRef Search ADS   Hoorn C, Wesselingh FP, ter Steege H, Bermudez MA, Mora A, Sevink J, Sanmartín I, Sanchez-Meseguer A, Anderson CL, Figueiredo JP, Jaramillo C, Riff D, Negri FR, Hooghiemstra H, Lundberg J, Stadler T, Särkinen T, Antonelli A. 2010. Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity. Science (New York, N.Y.)  330: 927– 931. Google Scholar CrossRef Search ADS PubMed  Huson DH, Bryant D. 2006. Application of phylogenetic networks in evolutionary studies. Molecular Biology and Evolution  23: 254– 267. Google Scholar CrossRef Search ADS PubMed  Iles WJD, Smith SY, Gandolfo MA, Graham SW. 2015. Monocot fossils suitable for molecular dating analyses. Botanical Journal of the Linnean Society  178: 346– 374. Google Scholar CrossRef Search ADS   Knowlton N, Weigt LA. 1998. New dates and new rates for divergence across the Isthmus of Panama. Proceedings of the Royal Society of London B  265: 2257– 2263. Google Scholar CrossRef Search ADS   Kolanowska M. 2015. Determination of potential glacial refugia and possible migration routes of Campylocentrum (Vandeae, Orchidaceae) species through the Darién Gap. Acta Societatis Botanicorum Poloniae  84: 97– 102. Google Scholar CrossRef Search ADS   Kolanowska M, Szlachetko D. 2013. New species of the genus Campylocentrum (Orchidaceae – Vandae) from Colombia. Phyton  53: 221– 239. Kuntze CEO. 1898. Campylocentrum kuntzei. Revisio generum plantarum  3: 298. Lagomarsino LP, Condamine FL, Antonelli A, Mulch A, Davis CC. 2016. The abiotic and biotic drivers of rapid diversification in Andean bellflowers (Campanulaceae). The New Phytologist  210: 1430– 1442. Google Scholar CrossRef Search ADS PubMed  Lindley J. 1840. Angraecum gladiifolium. Edwards’s Botanical Register  26: 68. Liu XQ, Ickert-Bond SM, Nie ZL, Zhou Z, Chen LQ, Wen J. 2016. Phylogeny of the Ampelocissus-Vitis clade in Vitaceae supports the New World origin of the grape genus. Molecular Phylogenetics and Evolution  95: 217– 228. Google Scholar CrossRef Search ADS PubMed  Maiti PK, Maiti P. 2011. Biodiversity: perception, peril and preservation . New Delhi: PHI Learning Pvt. Mann P. 2007. Overview of the tectonic history of northern Central America. Geological Society of America Special Papers  428: 1– 19. Matos-Maraví PF, Peña C, Willmott KR, Freitas AVL, Wahlberg N. 2013. Systematics and evolutionary history of butterflies in the “Taygetis clade” (Nymphalidae: Satyrinae: Euptychiina): towards a better understanding of Neotropical biogeography. Molecular Phylogenetics and Evolution  66: 54– 68. Google Scholar CrossRef Search ADS PubMed  Miller M, Holder M, Vos R, Midford P, Liebowitz T, Chan L, Hoover P, Warnow T. 2010. The CIPRES Portals . CIPRES, Available at: http://www.phylo.org/sub_sections/portal (accessed 6 April 2016). Montes C, Cardona A, Jaramillo C, Pardo A, Silva JC, Valencia V, Ayala C, Pérez-Angel LC, Rodriguez-Parra LA, Ramirez V, Niño H. 2015. Middle Miocene closure of the Central American seaway. Science  348: 226− 229. Google Scholar CrossRef Search ADS PubMed  Morrone JJ. 2014. Biogeographical regionalization of the Neotropical region. Zootaxa  3782: 1− 110. Google Scholar CrossRef Search ADS PubMed  Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GA, Kent J. 2000. Biodiversity hotspots for conservation priorities. Nature  403: 853– 858. Google Scholar CrossRef Search ADS PubMed  Nei M, Li WH. 1979. Mathematical models for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy Sciences USA  76: 5269– 5273. Google Scholar CrossRef Search ADS   Neubig KM, Whitten WM, Carlsward BS, Blanco MA, Endara L, Williams NH, Moore M. 2009. Phylogenetic utility of ycf1 in orchids: a plastid gene more variable than matK. Plant Systematics and Evolution  277: 75– 84. Google Scholar CrossRef Search ADS   Palmer JD, Jansen RK, Michaels HJ, Chase MW, Manhart JR. 1988. Chloroplast DNA variation and plant phylogeny. Annals of the Missouri Botanical Garden  75: 1180– 1206. Google Scholar CrossRef Search ADS   Pérez-Escobar OA, Chomicki G, Condamine FL, Karremans AP, Bogarín D, Matzke NJ, Silvestro D, Antonelli A. 2017. Recent origin and rapid speciation of Neotropical orchids in the world’s richest plant biodiversity hotspot. The New Phytologist  215: 891– 905. Google Scholar CrossRef Search ADS PubMed  Pessoa EM, Alves M. 2015a. Three new species of Campylocentrum (Vandeae, Orchidaceae) from Brazil. Phytotaxa  217: 265– 272. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2015b. Notes on Campylocentrum (Vandeae; Epidendroideae; Orchidaceae): re-establishment of two South American species. Phytotaxa  204: 85– 90. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2016a. Taxonomic revision of Campylocentrum (Orchidaceae–Vandeae–Angraecinae): species with terete leaves. Systematic Botany  41: 700– 713. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2016b. Taxonomical revision of Campylocentrum sect. Dendrophylopsis Cogn. (Orchidaceae–Vandae–Angraecinae). Phytotaxa  286: 131– 152. Google Scholar CrossRef Search ADS   Pessoa EM, Alves M. 2016c. A new combination in Angraecoides (Agraecinae, Orchidaceae): a forgotten name. Phytotaxa  269: 131– 132. Google Scholar CrossRef Search ADS   Pessoa EM, Arruda E, Pereira FFSD, Alves M. 2017. The odd roots of Campylocentrum (Angraecinae-Orchidaceae): a descriptive anatomical study of its morphologically variable roots. Rodriguésia  68: 1207– 1215. Google Scholar CrossRef Search ADS   Pessoa EM, Maciel JR, Alves M. 2015. Campylocentrum brevifolium (Lindl.) E.M. Pessoa & M. Alves, a neglected and critically endangered orchid from the Atlantic Forest of Brazil. Kew Bulletin  70: 43. Google Scholar CrossRef Search ADS   Posada D. 2008. jModelTest: phylogenetic model averaging. Molecular Biology and Evolution  25: 1253– 1256. Google Scholar CrossRef Search ADS PubMed  Pound MJ, Haywood AM, Salzmann U, Riding JB, Lunt DJ, Hunter SJ. 2011. A Tortonian (late Miocene, 11.61–7.25 Ma) global vegetation reconstruction. Paleogeography, Paleoclimatology, Paleoecology  2000: 29– 45. Google Scholar CrossRef Search ADS   Prado DE, Gibbs PE. 1993. Patterns of species distribution in the dry seasonal forests of South America. Annals of the Missouri Botanical Gardens  80: 902– 927. Google Scholar CrossRef Search ADS   Prance GT. 1973. Phytogeographic support for the theory of Pleistocene forest refuges in the Amazon Basin, based upon evidence from distribution patterns in Caryocaraceae, Chrysobalanaceae, Dichapetalaceae and Lecythidaceae. Acta Amazonica  3: 5– 28. Pridgeon AM, Cribb PJ, Chase MC, Rasmussen FN. 2005. Genera orchidacearum, Epidendroideae (part 1) . New York: Oxford University Press. Pridgeon AM, Cribb PJ, Chase MC, Rasmussen FN. 2009. Genera orchidacearum, Epidendroideae (part 2) . New York: Oxford University Press. Pridgeon AM, Cribb PJ, Chase MC, Rasmussen FN. 2014. Genera orchidacearum, Epidendroideae (part 3) . New York: Oxford University Press. Primavera I. 2013. Novedades en orquídeas para Uruguay: segunda contribución. Plantas epífitas. Agrociencia Uruguay  17: 22– 35. Rando JG, Zuntini AR, Conceição AS, Berg C, Pirani JR, Queiroz LP. 2016. Phylogeny of Chamaecrista ser. Coriaceae (Leguminosae) unveils a lineage recently diversified in Brazilian campo rupestre vegetation. International Journal of Plant Sciences  177: 3– 17. Google Scholar CrossRef Search ADS   Rambaut A, Suchard MA, Xie D, Drummond AJ. 2013. Tracer 1.6. Available at: http://tree.bio.ed.ac.uk/software/tracer/ (accessed 6 April 2016). Ree RH, Moore BR, Webb CO, Donoghue MJ. 2005. A likelihood framework for inferring the evolution of geographic range on phylogenetic trees. Evolution  59: 2299– 2311. Google Scholar CrossRef Search ADS PubMed  Ree RH, Smith SA. 2008. Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Systematic Biology  57: 4– 14. Google Scholar CrossRef Search ADS PubMed  Rolfe RA. 1903. The genus Campylocentrum. Orchid Review  11: 245– 247. Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics (Oxford, England)  19: 1572– 1574. Google Scholar CrossRef Search ADS PubMed  Salzmann U, Haywood AM, Lunt DJ, Valdes PJ, Hill DJ. 2008. A new global biome reconstruction and data-model comparison for the middle Pliocene. Global Ecology and Biogeography  17: 432– 447. Google Scholar CrossRef Search ADS   Sánchez-Baracaldo P. 2004. Phylogenetics and biogeography of the Neotropical fern genera Jamesonia and Eriosorus (Pteridaceae). American Journal of Botany  91: 274– 284. Google Scholar CrossRef Search ADS PubMed  Saslis-Lagoudakis C, Chase MW, Robinson DN, Russel SJ, Klirgaard BB. 2008. Phylogenetics of Neotropical Platymiscium (Leguminosae: Dalbergieae): systematics, divergence times, and biogeography inferred from nuclear ribosomal and plastid DNA sequence data. American Journal of Botany  95: 1270– 1286. Google Scholar CrossRef Search ADS PubMed  Simmons MP, Pickett KM, Miya M. 2004. How meaningful are Bayesian support values? Molecular Biology and Evolution  21: 188– 199. Google Scholar CrossRef Search ADS PubMed  Shaw J, Lickey EB, Schilling EE, Small RL. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American Journal of Botany  94: 275– 288. Google Scholar CrossRef Search ADS PubMed  Snak C, Vatanparast M, Silva C, Lewis GP, Lavin M, Kajita T, Queiroz LP. 2016. A dated phylogeny of the papilionoid legume genus Canavalia reveals recent diversification by a pantropical liana lineage. Molecular Phylogenetics and Evolution  98: 133– 146. Google Scholar CrossRef Search ADS PubMed  Sun Y, Skinner DZ, Liang GH, Hulbert SH. 1994. Phylogenetic analysis of Sorghum and related taxa using internal transcribed spacers of nuclear ribosomal DNA. Theoretical and Applied Genetics  89: 26– 32. Google Scholar CrossRef Search ADS PubMed  Swofford D. 2002. PAUP*. Phylogenetic analysis using parsimony and other methods. Version 4 . Sunderland: Sinauer. Szlachetko DL, Tukałło P, Mytnik-Ejsmont J, Grochoka E. 2013. Reclassification of the Angraecum-alliance (Orchidaceae, Vandoideae) based on molecular and morphological data. Biodiversity Research and Conservation  29: 1− 23. Google Scholar CrossRef Search ADS   Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology  17: 1105– 1109. Google Scholar CrossRef Search ADS PubMed  Todzia CA. 1980. A revision of the Costa Rican species of Campylocentrum (Orchidaceae). Brenesia  18: 117– 136. Turchetto-Zolet AC, Pinheiro F, Salgueiro F, Palma-Silva C. 2013. Phylogeographical patterns shed light on evolutionary process in South America. Molecular Ecology  22: 1193– 1213. Google Scholar CrossRef Search ADS PubMed  Vatanparast M, Takayana K, Sousa MS, Tateishi Y, Kajita T. 2011. Origin of Hawaiian endemic species of Canavalia (Fabaceae) from sea-dispersed species revealed by chloroplast and nuclear DNA sequences. Journal of Japanese Botany  86: 15– 25. Viruel J, Segarra-Moragues JG, Raz L, Forest F, Wilkin P, Sanmartín I, Catalán P. 2015. Late Cretaceous-early Eocene origin of yams (Dioscorea, Dioscoreaceae) in the Laurasion Palaearctic and their subsequent Oligocene-Miocene diversification. Journal of Biogeography  43: 750– 762. Google Scholar CrossRef Search ADS   Vogel TA, Patino LC, Eaton JK, Valley JW, Rose WI, Alvarado GE, Viray EL. 2006. Origin of silicic magmas along the Central American volcanic front: genetic relationship to mafic melts. Journal of Volcanology and Geothermal Resources  156: 217− 228. Google Scholar CrossRef Search ADS   Weinberg RF. 1992. Neotectonic development of western Nicaragua. Tectonics  11: 1010– 1017. Google Scholar CrossRef Search ADS   Zamora-Tavares MP, Martínez M, Magallón S, Guzmán-Dávalos L, Vargas-Ponce O. 2016. Physalis and physaloids: a recent and complex evolutionary history. Molecular Phylogenetics and Evolution  100: 41– 50. Google Scholar CrossRef Search ADS PubMed  APPENDIX 1 Taxa, locality, voucher information (collector, number and herbarium acronym) and GenBank accession numbers provided in the following order: ITS, Xdh, matK, rpl32-trnL, trnL-trnF, ycf1. GenBank accessions of previously published sequences are indicated with an asterisk. Angraecum distichum Lindl., Garden origin (Cal-Orchid), B. Carlsward 237 (FLAS), LT706519, KY514267, KY510533, LT732577, LT724124, KY558780. Angraecum leonis Rchb.f., Garden Origin, B. Carlsward 390 (FLAS), LT706518, KY514266, KY510532, LT732576, LT724123, KY558779. Angraecum moandense de Wild., Garden origin (Selby Botanical Gardens), B.Carlsward 208 (FLAS), AF506320*, KY514268, AF506363*, LT732609, AF506339*, KY558781. Campylocentrum brachycarpum Cogn., Brazil, Espírito Santo, E. Pessoa 1191 (UFP), LT706520, KY514284, KY510543, LT732586, LT724135, KY558799. C. brachycarpum Cogn., Brazil, São Paulo: Orquidário do Estado, LZ53 (SP), LT706521, KY514286, KY510544, LT732587, LT724136, KY558798. C. brenesii Schltr., Costa Rica, M. Blanco 2139 (USJ), LT706522, KY514294, KY510549, LT732639, LT724145, KY558829. C. brenesii Schltr., Costa Rica, D. Bogarín 11119 (CR), LT706523, KY514296, KY510551, LT732638, LT724144, KY558828. C. brenesii Schltr., Costa Rica, D. Bogarín 6363 (CR), LT706524, KY514295, KY510550, LT732637, LT724143, KY558827. C. calostachyum (Barb. Rodr.) ined., Brazil, São Paulo, M.R. Miranda 02 (UFP), LT706525, KY514316, KY510556, LT732620, LT724160, KY558809. C. crassirhizum Hoehne, Brazil, Sergipe, E. Pessoa 1243 (UFP), LT706526, KY514291, KY510545, LT732595, LT724140, KY558805. C. crassirhizum Hoehne, Brazil, Espírito Santo, E. Pessoa 1192 (UFP), LT706527, KY514292, KY510548, LT732584, LT724139, KY558806. C. densiflorum Cogn., Brazil, Santa Catarina, E. Pessoa 1196 (UFP), LT718666, KY514285, KY510542, LT732588, LT724134, KY558797. C. fasciola (Lindl.) Cogn., Jamaica, Claude Hamilton, B. Carlsward 301 (FLAS), DQ091564*, KY514322, DQ091321*, LT732608, DQ091445*, KY558817. C. fasciola (Lindl.) Cogn., Brazil, Mato Grosso, A.P. Beneli 966 (UFP), LT718688, KY514320, KY510561, LT732607, LT724165, KY558815. C. fasciola (Lindl.) Cogn., Jamaica, M. Witten 711, LT718690, KY514324, n.s., LT732604, LT724166, KY558816. C. fasciola (Lindl.) Cogn., Jamaica, Claude Hamilton, B. Carlsward 185 (FLAS), AF506294*, KY514323, AF506342*, LT732596, AF147226*, KY558813. C. fasciola (Lindl.) Cogn., Costa Rica, D. Bogarín 10415 (CR), LT718689, KY514321, KY510560, LT732605, LT724167, KY558818. C. fasciola (Lindl.) Cogn., Ecuador, M. Witten 1933 (QCNE), AF506295*, KY514319, AF506343*, LT732606, LT724164, KY558814. C. grisebachii Cogn., Brazil, Minas Gerais, E. Pessoa 1188 (UFP), LT718680, KY514298, KY510558, LT732597, LT724162, KY558821. C. huebneri Mansf., Brazil, Roraima, E. Pessoa 701 (UFP), LT718665, n.s., KY510565, LT732629, LT724147, KY558830. C. jamaicense (Rchbf. ex Griseb.) Benth. ex Fawc, Jamaica, M. Witten 1934 (FLAS), AF506299*, KY514289, AF506348*, LT732590, AF506326*, KY558800. C. jamaicense (Rchbf. ex Griseb.) Benth. ex Fawc., Puerto Rico, D. Ackerman 3341 (UPRRP), AY147219*, KY514288, AF506325*, LT732591, AF506346*, KY558801. C. kuntzei Cogn. ex Kuntze, Paraguay, Guairá, G. Esser 14519 (HEID), LT718676, n.s., KY510571, LT732634, LT724149, KY558834. C. labiakii Pessoa & Alves, Brazil, Espírito Santo, J. Meireles 545 (ESA), LT718660, n.s., KY510534, LT732619, LT724129, KY558794. C. mattogrossense Hoehne, Brazil, Mato Grosso, A. P. Beneli 968 (UFP), LT718687, KY514300, KY510567, LT732631, LT724148, KY558831. C. mattogrossense Hoehne, Brazil, Pará, A. K. Koch 566 (SP), LT718686, KY514299, KY510566, LT732630, LT724246, n.s. C. micranthum (Lindl.) Maury, Brazil, Roraima, E. Pessoa 1001 (UFP), LT718674, KY514311, KY510570, LT732628, LT724156, KY558841. C. micranthum (Lindl.) Maury, Brazil, Permambuco, E. Pessoa 1062 (UFP), LT718673, KY514309, KY510568, LT732625, LT724154, KY558839. C. micranthum (Lindl.) Maury, Brazil, Ceará, E. Pessoa 1115 (UFP), LT718672, KY514310, KY510569, LT732632, LT724155, KY558840. C. multiflorum Schltr., Costa Rica, D. Bogarín 1486 (CR), LT718681, KY514297, KY510552, LT732602, LT724163, KY558819. C. neglectum (Rchb.f. & Warm.) Cogn., Brazil, Distrito Federal, E. Pessoa 1278 (UFP), LT718663, KY514283, KY510546, LT732593, LT724137, KY558803. C. neglectum (Rchb.f. & Warm.) Cogn., Brazil, B. Carlsward 272 (FLAS), AF506297*, KY514287, AF506345*, LT732592, AF506324*, KY558804. C. ornithorrhinchum (Lindl.) Rolfe, Brazil, Permambuco, E. Pessoa 1239 (UFP), LT718661, KY514279, KY510536, LT732615, LT724130, KY558791. C. ornithorrhinchum (Lindl.) Rolfe, Brazil, São Paulo, R.Romanini18135 (SP), LT718659, KY514280, KY510539, LT732616, LT724131, KY558790. C. pachyrrhizum (Rchb.f.) Rolfe, Brazil, Sergipe, E. Pessoa 1245 (UFP), LT718682, KY514304, KY510554, LT732600, LT724142, KY558822. C. pachyrrhizum (Rchb.f.) Rolfe, Brazil, Pará, A.K. Koch 565 (SP), LT718684, KY514305, KY510553, LT732599, LT724141, KY558824. C. pachyrrhizum (Rchb.f.) Rolfe, USA, Florida, No voucher, LT718683, KY514307, AF506349*, LT732598, AF506327*, KY558823. C. pachyrrhizum (Rchb.f.) Rolfe, Puerto Rico, D. Ackerman s.n. (UPRRP), AF506301*, KY514306, AF506350*, LT732601, AF506328*, KY558825. C. panamense Ames, Costa Rica, D. Bogarín 871 (CR), LT718675, n.s., KY510573, LT732624, LT724153, KY558833. C. pauloense Hoehne & Schltr., Brazil, Santa Catarina, E. Pessoa 1197 (UFP), LT718664, KY514281, KY510541, LT732585, LT724133, KY558796. C. pernambucense Hoehne, Brazil, Sergipe, E. Pessoa 1244 (UFP), n.s., KY514277, KY510537, LT732618, LT724127, KY558792. C. pernambucense Hoehne, Brazil, Pernambuco, E. Pessoa 1287 (UFP), LT718662, KY514276, KY510538, LT732613, LT724126, KY558793. C. poeppigii (Rchb.f.) Rolfe, Brazil, Roraima, E. Pessoa 968 (UFP), LT718678, KY514303, KY510563, LT732612, LT724157, KY558812. C. poeppigii (Rchb.f.) Rolfe, Mexico, B. Carlsward 307(FLAS), AF506302*, KY514301, AF506351*, LT732610, AF506329*, KY558811. C. poeppigii (Rchb.f.) Rolfe, Costa Rica, D. Bogarín 2218 (CR), LT718679, KY514302, KY510564, LT732611, LT724158, KY558810. C. robustum Cogn., Brazil, Minas Gerais, Museu de História Natural 0967 (MHN), LT718667, KY514290, KY510547, LT732594, LT724138, KY558802. C. schiedei (Rchb.f.) Benth. ex Hemsl., Costa Rica, D. Bogarín 494 (CR), LT718671, KY514308, KY510572, LT732633, LT724152, KY558832. C. sellowii (Rchb.f.) Rolfe, Brazil, Minas Gerais, E. Pessoa 1189 (UFP), LT718658, KY514275, KY510535, LT732614, LT724125, KY558788. C. serranum Pessoa & Alves, Brazil, Permambuco, E. Pessoa 945 (UFP), LT718670, KY514317, KY510557, LT732621, LT724161, KY558808. C. spannagelli Hoehne, Brazil, Rio de Janeiro, E. Pessoa 1195 (UFP), LT718665, KY514282, KY510540, LT732589, LT724132, KY558795. C. stenanthum Schltr., Costa Rica, M. Witten 1822 (FLAS), AF506304, KY514312, AF506352*, LT732635, LT724150, KY558835. C. stenanthum Schltr., Mexico, B. Carlsward 180 (FLAS), AF506298*, KY514315, AF506347*, LT732636, AY147227*, KY558838. C. stenanthum Schltr., Panama, B. Carlsward 315, (FLAS), AY147220*, KY514314, AY147228*, LT732623, AY147235*, KY558837. C. stenanthum Schltr., Costa Rica, D. Bogarín 1263 (CR), LT718677, KY514313, n.s., LT732627, LT724163, KY558836. C. tenellum Todzia, Costa Rica, D. Bogarín 5844 (CR), LT718668, KY514293, KY510562, LT732626, n.s., KY558826. C. tyrridion Garay & Dunst. ex Foldats, Mexico, G. Carnevali 5145 (CICY), AF506305*, n.s., DQ091322*, LT732603, DQ091446*, KY558820. C. ulaei Cogn., Brazil, São Paulo, M.R. Miranda 01 (UFP), LT718669, KY514318, KY510555, LT732622, LT724159, KY558807. C. wawrae (Rchb.f. ex Beck) Rolfe, Brazil, Minas Gerais, D. Barbosa 13 (CESJ), LT718657, KY514278, n.s., LT732617, LT724128, KY558789. Dendrophylax funalis (Sw.) Benth. ex Rolfe, Jamaica, B. Carlsward 302 (FLAS), AY147221*, KY514269, AF506355*, LT732583, AY147229*, KY558782. D. porrectus (Rchb.f.) Carlsward & Whitten, Mexico, Yucatán, G. Carnevali 5907 (CICY), AF506314*, KY514271, AF506357*, LT732581, AF506335*, KY558785. D. porrectus (Rchb.f.) Carlsward & Whitten, USA, Florida, B. Carlsward 329 (FLAS), AY147223*, KY514274, AY147237*, LT732579, AY147232*, KY558784. D. porrectus (Rchb.f.) Carlsward & Whitten, Dominican Republic, M. Witten 1950 (JBSD), AY147224*, KY514272, AY147238*, LT732578, AY147233*, KY558786.D. porrectus (Rchb.f.) Carlsward & Whitten, Jamaica, B. Carlsward 184 (FLAS), Af506315*, KY514273, Af506358*, LT732580, AY147231*, KY558787. D. varius (J.F. Gmel.) Urb., Dominican Republic, M. Whitten 1960 (JBSD), AY147222*, KY514270, AY147236*, LT732582, AY147230*, KY558783. APPENDIX 2 Divergence time estimates and ancestral areas for nodes indicated in Figures 2 and 3. HPD = highest posterior density; A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. Node  Ages (Mya)  Ancestral areas    Mean  95% HPD  Area  Probability  1  10.93  13.7−8.2  G  0.21  2  8.47  11.4−5.6  G  0.97  3  8.53  10.7−6.4  AG  0.70  4  8.21  10.5−5.9  A  0.61  5  4.7  6.7−2.7  A  0.74  6  4.01  6.0−2.1  A  0.91  7  7.22  9.2−5.2  AF  0.20  8  2.9  4.2−1.7  AEF  0.40  9  1.76  2.7−0.8  F  0.83  10  0.93  1.53−0.4  F  1  11  0.46  0.76−0.26  F  1  12  0.2  0.27−0.13  F  1  13  2.04  2.9−1.18  AEF  0.46  14  0.92  1.4−0.44  AEF  0.59  15  0.71  1.07−0.35  AEF  0.63  16  0.64  1.08−0.2  EF  0.95  17  1.23  1.83−0.63  F  1  18  1.04  1.63−0.45  F  1  19  0.56  0.95−0.17  F  1  20  5.38  6.79−3.97  BF  0.26  21  4.79  6.1−3.48  B  0.43  22  4.06  5.2−2.92  B  0.47  23  1.25  2.0−0.5  ABF  0.1  24  3.4  4.64−2.16  B  0.47  25  0.61  1.2−0.2  ABF  0.1  26  2.4  3.4−1.4  BF  0.84  27  5.16  6.6−3.7  BF  0.42  28  2.69  3.8−1.6  F  0.96  29  1.82  2.7−0.94  F  0.99  30  4.28  5.5−3.06  B  0.35  31  3.29  4.4−2.18  B  0.99  32  3.46  4.6−2.32  BD  0.27  33  0.52  1.0−0.1  D  0.42  34  1.56  2.2−0.92  BCD  0.2  35  1.11  1.7−0.52  BCD  0.21  36  0.81  1.3−0.32  BCD  0.21  37  0.73  1.2−0.26  BCD  0.15  Node  Ages (Mya)  Ancestral areas    Mean  95% HPD  Area  Probability  1  10.93  13.7−8.2  G  0.21  2  8.47  11.4−5.6  G  0.97  3  8.53  10.7−6.4  AG  0.70  4  8.21  10.5−5.9  A  0.61  5  4.7  6.7−2.7  A  0.74  6  4.01  6.0−2.1  A  0.91  7  7.22  9.2−5.2  AF  0.20  8  2.9  4.2−1.7  AEF  0.40  9  1.76  2.7−0.8  F  0.83  10  0.93  1.53−0.4  F  1  11  0.46  0.76−0.26  F  1  12  0.2  0.27−0.13  F  1  13  2.04  2.9−1.18  AEF  0.46  14  0.92  1.4−0.44  AEF  0.59  15  0.71  1.07−0.35  AEF  0.63  16  0.64  1.08−0.2  EF  0.95  17  1.23  1.83−0.63  F  1  18  1.04  1.63−0.45  F  1  19  0.56  0.95−0.17  F  1  20  5.38  6.79−3.97  BF  0.26  21  4.79  6.1−3.48  B  0.43  22  4.06  5.2−2.92  B  0.47  23  1.25  2.0−0.5  ABF  0.1  24  3.4  4.64−2.16  B  0.47  25  0.61  1.2−0.2  ABF  0.1  26  2.4  3.4−1.4  BF  0.84  27  5.16  6.6−3.7  BF  0.42  28  2.69  3.8−1.6  F  0.96  29  1.82  2.7−0.94  F  0.99  30  4.28  5.5−3.06  B  0.35  31  3.29  4.4−2.18  B  0.99  32  3.46  4.6−2.32  BD  0.27  33  0.52  1.0−0.1  D  0.42  34  1.56  2.2−0.92  BCD  0.2  35  1.11  1.7−0.52  BCD  0.21  36  0.81  1.3−0.32  BCD  0.21  37  0.73  1.2−0.26  BCD  0.15  View Large APPENDIX 3 Taxa, voucher information and GenBank accession numbers used for the dating analysis in this study in the following order: ITS, matK, ycf1. Sequences in GenBank but not published yet in papers are indicated with an asterisk, and the voucher information is not available. Earina autumnalis (G.Forst.) Hook.f., M. W. Chase O-298 (K), AF260149, AF263656, none. Cattleya bicolor Lindl., F.G. Brieger 4333 (ESA), JN600949, none, JN600718. C. bicolor Lindl., F.G. Brieger 895 (ESA), none, EU139961, none. Maxillaria splendens Poepp. & Endl., W. M. Whitten 2940 (FLAS), none, none, KF660506. M. splendens Poepp. & Endl., S. Koehler 144 (UEC), DQ210252, DQ210684, none. Cymbidium eburneum Lindl., J. Luo*, none, KF361650*, KF361650*, C. eburneum Lindl., M. W. Chase 1505 (K), AF470503, none, none. Phalaenopsis wilsonii Rolfe, B. Carlsward 331 (FLAS), DQ091672, none, EU490763. P. wilsonii Rolfe, TBG 144214*, none, AB217751*, none. APPENDIX 4 Palaeogeographical models for dispersal rates across areas implemented in the biogeographical analysis. A = Antillean sub-region; B = Mesoamerica dominion; C = Pacific dominion; D = Brazilian dominion, E = Chacoan dominion; F = Parana dominion; G = Sub-Saharan Africa. Time Slice 1: 11−7 Mya; Time Slice 2: 7−0 Mya.   A  B  C  D  E  F  G  Time Slice 1  A  1  0.1  0.1  0.05  0.0001  0.001  0.001  B    1  0.1  0.01  0.0001  0.0001  0.0001  C      1  0.1  0.0001  0.0001  0.0001  D        1  0.1  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1  Time Slice 2  A  1  0.5  0.5  0.1  0.005  0.001  0.001  B    1  0.5  0.05  0.001  0.001  0.0001  C      1  0.5  0.01  0.001  0.0001  D        1  0.5  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1    A  B  C  D  E  F  G  Time Slice 1  A  1  0.1  0.1  0.05  0.0001  0.001  0.001  B    1  0.1  0.01  0.0001  0.0001  0.0001  C      1  0.1  0.0001  0.0001  0.0001  D        1  0.1  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1  Time Slice 2  A  1  0.5  0.5  0.1  0.005  0.001  0.001  B    1  0.5  0.05  0.001  0.001  0.0001  C      1  0.5  0.01  0.001  0.0001  D        1  0.5  0.1  0.0001  E          1  0.5  0.0001  F            1  0.001  G              1  View Large © 2018 The Linnean Society of London, Botanical Journal of the Linnean Society

Journal

Botanical Journal of the Linnean SocietyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off