Phylogenetics and cytology of a pantropical orchid genus Polystachya (Polystachyinae, Vandeae, Orchidaceae): Evidence from plastid DNA sequence data

Michael H J Barfuss; Barbara  Turner

† Background and Aims Here evidence for reticulation in the pantropical orchid genus Polystachya is presented, using gene trees from five nuclear and plastid DNA data sets, first among only diploid samples (homoploid hybridization) and then with the inclusion of cloned tetraploid sequences (allopolyploids). Two groups of tetra-ploids are compared with respect to their origins and phylogenetic relationships. † Methods Sequences from plastid regions, three low-copy nuclear genes and ITS nuclear ribosomal DNA were analysed for 56 diploid and 17 tetraploid accessions using maximum parsimony and Bayesian inference. Reticulation was inferred from incongruence between gene trees using supernetwork and consensus network analyses and from cloning and sequencing duplicated loci in tetraploids. † Key Results Diploid trees from individual loci showed considerable incongruity but little reticulation signal when support from more than one gene tree was required to infer reticulation. This was coupled with generally low support in the individual gene trees. Sequencing the duplicated gene copies in tetraploids showed clearer evidence of hybrid evolution, including multiple origins of one group of tetraploids included in the study. † Conclusions A combination of cloning duplicate gene copies in allotetraploids and consensus network comparison of gene trees allowed a phylogenetic framework for reticulation in Polystachya to be built. There was little evidence for homoploid hybridization, but our knowledge of the origins and relationships of three groups of allo-tetraploids are greatly improved by this study. One group showed evidence of multiple long-distance dispersals to achieve a pantropical distribution; another showed no evidence of multiple origins or long-distance dispersal but had greater morphological variation, consistent with hybridization between more distantly related parents.

Amplified fragment length polymorphism (AFLP) markers were used to investigate the relationships among Polystachya accessions from a group of closely related pantropical tetraploids. Before starting with the fingerprinting analyses, the polyploid accessions were first included in a ...

A new genus of Orchidaceae (Polystachyinae), Isochilostachya, is described. It is morphologically similar to Polystachya Hook. but is distinct by the narrower, grass-like leaves arranged in the upper part of the stem, long and acuminate floral bracts and sepals, and long clavate or capitate hairs densely arranged on the lip. A comprehensive description of the new genus is provided. A list of examined specimens and information about distribution, habitat, and altitude for each species are presented. A position of the members of the new genus in the cladogram recently obtained from molecular analyses of Polystachyinae is briefly discussed. A taxonomic key to Polystachyinae is included. Eleven new combinations on species level are validated. Eight lectotypes are designated. Most of the species of Isochilostachya are endemic to particular mountains of the Eastern Arc (Eastern Afromontane hotspot) in north-east Tanzania therefore a brief discussion about their restricted distribution is provided. Keywords: Polystachya sect. Isochiloides; taxonomy; endemism; Eastern Afromontane hotspot; Africa

TAXON 59 (2) • April 2010: 389–404 Russell & al. • Phylogenetics and cytology of Polystachya Phylogenetics and cytology of a pantropical orchid genus Polystachya (Polystachyinae, Vandeae, Orchidaceae): Evidence from plastid DNA sequence data Anton Russell,1 Rosabelle Samuel,1 Barbara Rupp,1 Michael H.J. Barfuss,1 Marko Šafran, 2 Visnja Besendorfer2 & Mark W. Chase3 1 Department of Systematic & Evolutionary Botany, University of Vienna, Rennweg 14, Vienna 1030, Austria 2 Department of Molecular Biology, University of Zagreb, Horvatovac 102a, 10000 Zagreb, Croatia 3 Jodrell Laboratory, Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3DS, U.K. Author for correspondence: Anton Russell, anton.russell@univie.ac.at Abstract The pantropical orchid genus Polystachya is the subject of ongoing taxonomic work. We inferred phylogenetic relationships in the genus using 5.3 kb of plastid DNA, for 83 out of ca. 240 species and 12 out of 15 sections, as well as five outgroup species. We also collected ploidy data using chromosome counts and genome size estimates. Bayesian and parsimony trees were congruent with each other and well resolved. Polystachya appears monophyletic based on current sampling, provided that the name P. neobenthamia is used instead of Neobenthamia gracilis for that species. The current sectional classification does not define monophyletic groups, but the present study can be used as the basis for a future sectional classification. Areas postulated as Pleistocene refugia for wet tropical forests in Africa also form centres of diversity for the genus. Biogeographical analyses using DIVA and Lagrange show an early radiation in eastern Africa, followed by separate radiations in eastern and western Africa. Subsequent dispersal from western to eastern Africa has occurred at a much higher rate than from east to west. Dispersal to the Neotropics occurred more than once, and one lineage has spread recently and rapidly throughout the tropics. Polyploidy has occurred several times during the diversification of the genus, most notably in association with the recent pantropical dispersal. Keywords long-distance dispersal; phylogenetics; plastid DNA sequences; polyploidy; Polystachya; tropical Africa INTRODUCTION Polystachya Hook. (approximately 240 species), is an unusually broadly distributed orchid genus that ranges throughout the tropics (Govaerts & al., 2009), with some species extending into subtropical southern Africa. Dressler (1993) placed it in the subtribe Polystachyinae along with Hederorkis Thou., Imerinaea Schltr. and Neobenthamia Rolfe. Imerinaea, a monotypic genus endemic to Madagascar, has been shown with DNA data to be a member of Eulophiinae (Cymbideae) (Pridgeon & al., in press), and material of Hederorkis (two species, one each on the Seychelles and Mauritius) has not been available for molecular analyses. The monotypic Neobenthamia has been confirmed by DNA sequence data to have a close relationship with Polystachya (Reich, 2006) and in this study we use the synonym Polystachya neobenthamia Schltr. instead of Neobenthamia gracilis Rolfe. In an analysis of morphological data, Freudenstein & Rasmussen (1999) found a close relationship between Polystachyinae and the large tribe Vandeae; this has been corroborated by Cameron (2001, as cited in Carlsward & al., 2006b) and van den Berg & al. (2005) as a sister-group relationship. Chase & al. (2003) and Pridgeon & al. (2005) included Polystachyinae within a larger Vandeae. Carlsward & al. (2006a,b) maintained a stricter definition of Vandeae in their studies, citing several morphological characters of Vandeae, including monopodial habit, loss of mucilage and tilosomes, and the presence of spherical silica bodies in leaf sclerenchyma, to differentiate it from Polystachyinae. Characteristics of the genus include a terminal inflorescence bearing one to many flowers arranged in a raceme or panicle. Flowers are usually nonresupinate with a distinctive conical mentum or spur formed from the fusion of the column foot and bases of the lateral sepals. The labellum is usually not much larger than the sepals and is partially enclosed by them when the flower is at its maximum expansion. It is usually three-lobed and often bears a basal callus and food hairs (farinaceous, clavate or setose pubescence with nutritional value, for attracting pollinating insects: Davies & al., 2002). The free part of the column is short and bears two pollinia and a single stipe. Most Polystachya species are epiphytes or epiliths, but a few species including P. neobenthamia are terrestrial. Variation is found in the plant size, shoot and pseudobulb structure, inflorescence size and shape, flower number, density, size and colour, and the size and shape of floral organs, especially the labellum and mentum. Flower morphology in orchids often appears labile, with the variation traditionally explained by adaptations to particular pollinator species, as reviewed for example by Waterman & Bidartondo (2008). In Polystachya, pollination has not been observed often, but halictid bees have been seen visiting P. rosea despite the lack of nectar or food hairs in this species (Pettersson & Nilsson, 1993), and syrphid flies have been observed visiting P. concreta on Réunion (T. Pailler, pers. comm.) The last taxonomic work to attempt an account of the entire genus was Kraenzlin’s (1926) monograph. Since 1926, 389 Russell & al. • Phylogenetics and cytology of Polystachya taxonomic accounts have dealt only with limited geographic areas for floras, or with smaller sets of taxa (e.g., Cribb, 1978; Geerinck, 1979; Podzorski & Cribb, 1979; Stévart & Nguema, 2004). In this time new species have been described and taxonomic concepts have changed considerably as more herbarium material has become available and more taxonomic work has been carried out. According to the system developed by Kraenzlin (1926), Summerhayes (Summerhayes, 1942, 1947; Brenan, 1954) and Cribb (1978), the genus is divided into 15 sections, circumscribed by vegetative and floral characters such as the presence of only a single apical leaf on the pseudobulb for sect. Cultriformes Kraenzl. or a dense, smallflowered inflorescence with setose bracts for sect. Polychaete P.J. Cribb. Some sections are defined by several characters and include only one or a few species; others use only a few characters to group a large portion of the genus. The Flore du Cameroun (Szlachetko & Olszewski, 2001) and Flore du Gabon (Szlachetko & al., 2004) used a different sectional classification from that of Flora Zambesiaca (La Croix & Cribb, 1998) and Flora of Tropical East Africa (Cribb, 1984). Some morphologically divergent species have been segregated into separate genera (Mytnik-Ejsmont, 2007; Mytnik-Ejsmont & Szlachetko, 2007a,b, 2008a,b,c), but there is no phylogenetic evidence to support these changes. A well-developed phylogenetic hypothesis should help clarify the infrageneric classification of the genus and might be used to redefine sections as a step towards a much-needed generic revision. For an orchid genus, Polystachya has an unusually wide distribution (Pridgeon & al., 2005). The majority of species occur on the African mainland, with a further 20 in tropical America, 25 on the Indian Ocean Islands, and 5 species endemic to the Gulf of Guinea islands (Govaerts & al., 2009). In tropical Africa, most species are more or less restricted in their geographical range, but several are common and widespread. The range of one species, Polystachya concreta, extends throughout the Neotropical and tropical African forests to the Indian Ocean islands, southern India, Sri Lanka and Southeast Asia. It should be noted that this broad definition of P. concreta is not universally accepted. Populations of the species from different areas tend to differ morphologically, and one of the taxonomic problems in Polystachya is the delimitation of species boundaries among these widely dispersed populations. The geographical and ecological range of the genus, combined with variation in habit and morphology make its study highly worthwhile. In particular, although the greatest species diversity occurs in Africa, we can look at the relationships of African plants to those in Madagascar, the Neotropics, and Asia. The availability of software for analysing geographical distributions with phylogenetic trees (Ronquist, 1996; Ree & Smith, 2008) allows us to start looking at large-scale patterns in the distribution of the genus and potential hypotheses to explain its current distribution. Polyploidy is a common feature of some angiosperm genera, and changes in chromosome number within a genus have been linked to habitat variation and distributional patterns. Chromosome counts in Polystachya (Jones, 1966; Podzorski & Cribb, 1979) show some species to be tetraploid (with 390 TAXON 59 (2) • April 2010: 389–404 2n = 4x = 80 chromosomes), which tend to be more common away from the centre of diversity of the genus in the Neotropics and southern Africa. This study contributes additional ploidy information in the form of chromosome counts and genome size estimates to gain a clearer picture of the role of polyploidy in Polystachya diversification and whether changes in chromosome number or genome size can be related to biogeographical or ecological factors. In recent years, plastid DNA sequences have proved useful in orchid phylogenetics (e.g., Cameron, 2005; Van den Berg & al., 2005 Carlsward & al., 2006b; Fischer & al., 2007). Among plastid DNA regions, the trnK intron and matK gene in particular have been a focus in plant phylogenetics (e.g., Barfuss & al., 2005; Samuel & al., 2005; Hausner & al., 2006; Barthet & Hilu, 2007), including orchids (e.g., Gravendeel & al., 2001; Bytebier & al., 2007; Micheneau & al., 2008), and on that basis, this region was chosen for this study. Shaw & al. (2007) identified 13 highly variable regions of the plastid genome. Among these regions, initial results suggested the rps16 intron would be informative in Polystachya, and the further inclusion of the rps16-trnK spacer allowed us to use a contiguous 5-kb section of the plastid genome. During a further pilot study, the psbD-trnT spacer was easy to amplify and sequence and gave well-resolved trees of a subset of Polystachya samples, with high consistency and retention indices (Rupp, 2008). This region was also added to the genus-wide molecular dataset. MATERIALS AND METHODS Taxon sampling. — The DNA Bank at the Royal Botanic Gardens, Kew, provided many of our DNA samples (http:// data.kew.org/dnabank/homepage.html). Other material for DNA extraction came from plants growing at the Botanical Garden of the University of Vienna, including plants originating in cultivation and gathered in the wild; plants growing in the private collection of Isobyl la Croix in Ross-shire, Scotland; and from collections made by the authors in Cameroon. Voucher specimens are deposited in the Herbaria EA, K, WU and YA. Accessions are listed in the Appendix. For simplicity, we have used the broad definition of Polystachya concreta provided by Garay & Sweet (1974) and provide provenance information for our eleven samples of that species. This seems a more acceptable solution than the current convention of applying names to these plants based on their geographical location (P. concreta in the Neotropics and Asia, P. tessellata Lindl. in mainland Africa and P. mauritiana Spreng. on the Malagasy Islands). Each of these names encompasses plants with a range of ecologies and morphologies, as does P. concreta sensu lato. Outgroup taxa were chosen to represent the most closely related genera to Polystachya based on preliminary sequence data, and included Adrorhizon purpurascens, Sirhookera lanceolata and two species of Bromheadia. Among genera included in published studies of Epidendroideae, Phalaenopsis is closely related to Polystachya and the complete plastid genome of one species is available from GenBank (Chang & al., 2006). TAXON 59 (2) • April 2010: 389–404 Russell & al. • Phylogenetics and cytology of Polystachya DNA extraction. — DNA extractions were performed at the Department of Systematic and Evolutionary Botany, University of Vienna and the Jodrell Laboratory, Royal Botanic Gardens, Kew, using a modification of the protocols of Doyle & Doyle (1987), Li & al. (2007) and Tel-Zur & al. (for polysaccharide-rich samples; 1999). DNA was extracted from living material, silica gel–dried material (Chase & Hills, 1991), and tissue preserved in CTAB-salt solution (3% CTAB, 35% NaCl: Štorchová & al., 2000). We used approximately 100 mg of tissue, fresh weight. Silica gel–dried material was ground with glass beads in a 2-ml microcentrifuge tube. CTAB-preserved material was ground in cold sorbitol buffer (100 mM Tris, 0.35 M Sorbitol, 5 mM EDTA, pH 8.0, stored at 4°C, with 1% 2-mercaptoethanol and 1% PVP-40 added just before use) and a small amount of quartz powder using a mortar and pestle. Fresh material was ground either in liquid nitrogen with the rapid introduction of sorbitol buffer or directly in sorbitol buffer with quartz powder. To remove mucilaginous polysaccharides, we initially mixed the ground samples in sorbitol buffer in 12-ml centrifuge tubes, and centrifuged for ten minutes at 3000 rcf. We poured off the supernatant and added fresh sorbitol buffer, and repeated the sorbitol buffer cleaning until there was no visible mucilage layer in the sample pellet after centrifugation (usually three or four rounds, sometimes more). The samples were transferred back to 2-ml microcentrifuge tubes and incubated for one hour at 60°C with 700 μl high-salt 3× CTAB extraction buffer (100 mM Tris, 3 M NaCl, 3% CTAB, 20 mM EDTA, pH 8.0, preheated to 60°C and with 0.2% 2-mercaptoethanol and 1% PVP-40 added just before use) and 30 μl sarkosyl (30%). 700 μl chloroform/isoamyl alcohol (24 : 1) was added and the samples incubated at room temperature for 30 minutes, then centrifuged for ten minutes at 10,000 rcf. The upper aqueous phase was transferred to a microcentrifuge tube and DNA was precipitated by addition of 1/10 volume sodium acetate (3 M, pH 5.2) and 2/3 volume cold isopropanol, incubating the tubes at 4°C overnight. DNA was pelleted out by centrifugation at 14,000 rcf for 30 minutes, and washed twice with 500 μl 70% ethanol, before drying and resuspension in 50 μl TE buffer (10 mM Tris, 1 mM EDTA, pH 8.0). We removed RNA using 30 μg RNase A (Fermentas) incubating at 37°C for 30 minutes. PCR amplification and sequencing. — See Table 1 for primers used in this study. Most PCR reactions were 20 μl, containing 18.0 μl ABGene ReddyMix PCR Master Mix, 0.4 μl of each primer at 20 μM, 0.8 μl bovine serum albumin (20 μg/μl, Fermentas), and 0.4 μl template DNA. Reaction conditions were an initial denaturation at 80°C for 5 min, followed by 36 cycles of 94°C for 30 s, annealing temperature for 30 s and 72°C for 1 min. We used a final extension of 72°C for 5 min. Annealing temperature was either 55°C or 50°C depending on the amplification primers used. PCR products were cleaned by incubating at 37°C for 45 min, followed by denaturing at 80°C for 15 min, with one unit CIAP (calf intestinal alkaline phosphatase) and 10 units exonuclease I (both from Fermentas) to degrade single-stranded DNA fragments and dNTPs in the PCR product (Werle & al., 1994). Sequencing reactions were carried out with 1.0 μl ABI BigDye Terminators kit per reaction, 1.0 μl primer at 3.2 μM, and 8.0 μl cleaned-up PCR product, using 30 cycles of 96°C for 10 s, 50°C for 5 s, and 60°C for 4 min. Sequencing was performed on a 16-capillary sequencer, Applied Biosystems 3130xl Genetic Analyzer following the manufacturer’s protocols. Some samples stubbornly refused to yield readable sequences for certain loci, especially for the trnK intron upstream of matK due to poly-A/T sequences, and as a result the final matrix contains some areas of missing data. Sequence analysis. — Sequences were edited using FinchTV 1.4.0 (Geospiza Inc.) and assembled using either AutoAssembler 1.4.0 (Applied Biosystems, Perkin Elmer Corp.) or LaserGene 7.1 SeqMan (DNASTAR Inc.). Alignment was performed using MUSCLE (Edgar, 2004) and manually Table . Primers used in this study, in alphanumerical order. Primer namea Sequence Reference matK1200For 5′-GTATTGGGTCATCCTATTAGTAAACC-3′ This study matK1326R 5′-TCTAGCACACGAAAGTCGAAGT-3′ Cuenoud & al. (2002) matK-50Fmo 5′-GTTCTGACCATATTGCACTATGTATC-3′ This study matK50Ror 5′-TGAGCAAGTGAGTAAATAKACTCCTG-3′ This study matK550For 5′-TGRTTCAAATCCTTCAATGCTGGATC-3′ This study matK750Ror 5′-ATGTGTTCGCTCAAGAAAGACTCC-3′ This study psbD 5′-CTCCGTARCCAGTCATCCATA-3′ Shaw & al. (2007) rps16F2mo 5′-CTYGAGCCGTATGAGGARAAAACY-3′ This study rpsF 5′-GTGGTAGAAAGCAACGTGCGACTT-3′ Oxelman & al. (1997) rpsR2 5′-TCGGGATCGAACATCAATTGCAAC-3′ Oxelman & al. (1997) trnKR3an 5′-TCGAACCCGGAACTAGTCGG-3′ This study trnKF5an 5′-GTTGCTAACTCAAYGGTAGAGTACTC-3′ This study trnKR5an 5′-CCYTTSAGGATCAGTCGTGGTC-3′ This study trnT(GGU)-R 5′-CCCTTTTAACTCAGTGGTAG-3′ Shaw & al. (2007) a For primers designed for this study, the suffix ‘or’ denotes an orchid specific primer, ‘mo’ = monocotspecific and ‘an’ = angiosperm-specific. 391 Russell & al. • Phylogenetics and cytology of Polystachya adjusted in MacClade 4.08 (Maddison & Maddison, 2005) following the guidelines of Kelchner (2000). The sequences were analysed together as a combined matrix: rps16 intron; rps16 exon 2; rps16-trnK spacer; trnK intron excluding matK; matK; psbD-trnT spacer. The trnK exons contained only invariant or (few) autapomorphic sites and were excluded from analyses; their inclusion would have complicated the partition models described below without contributing many data. Gap-rich portions of the matrix (≥ 50% missing data) were excluded from analyses, a total of 1433 sites. Of the remaining 5338 sites in the matrix, 794 were potentially parsimony-informative. Maximum parsimony (MP) analysis was performed in PAUP* 4.0b10 (Swofford, 2003), using tree bisection and reconnection (TBR) branch swapping, with multrees on and steepest descent off. To locate topographic islands in the dataset, an initial heuristic search was performed using 1000 random heuristic search replicates, saving the 10 most-parsimonious trees per replicate. The resulting 10,000 trees were used as the starting trees for a second heuristic search, with MaxTrees set to 50,000. Bootstrap percentages (BP) were obtained from 1000 replicates of TBR branch swapping, multrees on, and steepest descent off, saving 100 trees per replicate. Bayesian analyses were carried out in MrBayes 3.1.2 parallel version (Huelsenbeck & Ronquist, 2001; Altekar & al., 2004) using the resources of the Computational Biology Service Unit at Cornell University (http://cbsuapps.tc.cornell. edu). Three analyses were performed on the same data with different systems of partitioning the matrix (Nylander & al., 2004). We used MrModeltest v.2.3 (Nylander, 2004) to determine the most appropriate nucleotide substitution model in each partition and applied the model nominated by the Akaike information criterion (AIC) in each case. For the first analysis we used a single partition (i.e., the matrix was not partitioned) with the GTR + I + Γ model. For the second analysis we partitioned the data into coding and non-coding sites with the GTR + I + Γ model for both partitions. For the third analysis, we used six partitions to allow different character evolution in the following groups: rps16 intron; rps16 exon 2; rps16-trnK spacer; trnK intron excluding matK; matK; psbD-trnT spacer. We applied the GTR + Γ model to the rps16-trnK spacer and the psbD-trnT spacer, and the GTR + I + Γ model to the other four partitions. For the partitioned analyses, model parameters were unlinked between partitions, except for tree topology and branch length. For each analysis, two independent sets of four metropolis-coupled Monte Carlo Markov chains were run with default (Dirichlet) priors for 5,000,000 generations, sampling every 500 generations, with a burn-in of 25% and chains heated to 0.10 (increasing the frequency of data swapping between chains compared to the default value 0.20, at the cost of smaller topographic distances that can be negotiated between likelihood peaks). The program Tracer (Rambaut & Drummond, 2007b) was used to confirm that the chains had reached convergence, and the effective sample size (ESS) of each of the parameters was high. Genomic data. — Chromosome counts for Polystachya species were taken from Jones (1966), Fedorov (1969), Podzorski & Cribb (1979), Reich (2006), and our own observations. 392 TAXON 59 (2) • April 2010: 389–404 We collected freshly growing roots from plants in the living collections of the Botanical Gardens of the University of Vienna and the Royal Botanic Gardens, Kew, and prepared them first by making small longitudinal incisions in the tips down to the meristem using a scalpel blade and dissecting microscope, as described in Jones & Daker (1966). They were quickly transferred to 2 mM 8-hydroxyquinoline and incubated for two hours in darkness at room temperature, then two hours in darkness at 4°C. They were then fixed in 3 : 1 ethanol/acetic acid and stored at –20°C. Before staining, root tips were incubated in citrate buffer (4 mM citric acid, 6 mM sodium citrate, pH 4.8) for 10–15 min and digested for ca. 70 min in an enzyme mixture (1% pectolyase Y23, 1% drieselase, 1% cellulase R10, in citrate buffer) at 37°C. Digested roots were transferred to citrate buffer for 20 min. Root meristematic cells were isolated from root tips and spread on microscope slides with one drop of 60% acetic acid, and the slides were frozen and air-dried without their cover slips. Finally, chromosomes were stained with DAPI (2 μg/ml), mounted in Vectashield mounting solution, and viewed with a fluorescence microscope. Rupp (2008) showed that genome size estimates from flow cytometry could be used to infer ploidy in Polystachya. We have used data for Polystachya from Rupp (2008) and Rupp & al. (subm.) in addition to chromosome counts. Distribution data. — Ancestral distributions were inferred using Bayesian (Lagrange: Ree & Smith, 2008) and parsimony-based (DIVA: Ronquist, 1996) methods. Both programs require as input a single, fully resolved tree; the input tree for Lagrange should also include branch lengths. In both cases, we used the maximum clade credibility tree from the unpartitioned Bayesian analysis above, as calculated by TreeAnnotator v.1.4.7 (part of the BEAST package: Rambaut & Drummond, 2007a) after manually removing the burn-in trees from the two MrBayes runs and combining them in a single text file. DIVA was unable to process the large number of terminal taxa in a single analysis, so a series of analyses were performed using subtrees and summary trees to piece together the most parsimonious ancestral distribution. Species distributions were recorded from literature (Cribb, 1978, 1984; Geerinck, 1979, 1980; Podzorski & Cribb, 1979; La Croix & Cribb, 1998; Szlachetko & Olszewski, 2001; Szlachetko & al., 2004) and the World Monocot Checklist (Govaerts & al., 2009). We recorded the presence or absence of each species in five large areas covering the full range of the genus: (1) Asia, (2) Neotropics, (3) Malagasy Islands, (4) eastern Africa and (5) western Africa. For this study, ‘eastern Africa’ covered the African mainland east of, and including, the Albertine Rift montane forests and the dry forests of Zambia, southeastern Democratic Republic of Congo and southeastern Angola. The area extends north through Ethiopia and Eritrea and south through Malawi, Mozambique, Zimbabwe, Swaziland, and eastern South Africa, where Polystachya pubescens extends the total distribution of the genus along coastal forests as far as the eastern Cape. ‘Western Africa’ covered the African mainland west of, and including, the Congo basin and the islands in the Gulf of Guinea. TAXON 59 (2) • April 2010: 389–404 Russell & al. • Phylogenetics and cytology of Polystachya RESULTS lengths were similar relative to each other between the analyses but the branch length scale factor increased by more than an order of magnitude to make the branches unrealistically long (> 1 expected substitution per site from the base of the tree to the tips). This effect is discussed by Marshall & al. (2006) and might be due to not unlinking branch lengths between partitions (unlinking branch lengths produces a separate tree for each partition). Therefore, references here to Bayesian results will only relate to the unpartioned analysis, including the fully resolved maximum clade credibility tree (Fig. 2) used in the ancestral area analyses. Consensus trees from Bayesian inference and parsimony were similar, but the Bayesian tree (not shown) was more highly resolved. Polystachya adansoniae 1 and P. elegans were sister taxa; Polystachya laxiflora was sister to the Phylogenetic analysis. — The combined data matrix contained 794 potentially parsimony-informative sites. Maximum parsimony analysis found 50,000 equally most parsimonious trees before reaching the limit on the number of trees to retain in memory. Tree length was 2464, consistency index = 0.69, and retention index = 0.83. The strict consensus tree from parsimony analysis is presented in Fig. 1. For Bayesian analysis, partitioning of the data matrix resulted in phylogenetic trees identical in topology to that from the unpartitioned matrix, with only small differences in clade posterior probabilities between the three partitioning schemes. However, the partitioned analyses produced trees that were much ‘longer’ than the unpartitioned analysis; that is, branch 100 57 1.0 0.95 100 1.0 100 1.0 100 100 1.0 1.0 100 1.0 80 100 0.84 1.0 100 1.0 100 89 100 1.0 I. 1.0 1.0 73 72 1.0 0.99 87 1.0 89 1.0 85 83 71 0.98 0.84 1.0 84 100 1.0 1.0 100 98 1.0 II. 1.0 85 100 90 1.0 1.0 100 98 73 1.0 0.99 1.0 100 97 88 1.0 1.0 1.0 100 95 1.0 1.0 94 1.0 60 0.98 93 1.0 PŚĂůĂĞŶŽƉƐŝƐ ĂƉŚƌŽĚŝƚĞ subsp. ĨŽƌŵŽƐĂŶĂ ĚƌŽƌŚŝǌŽŶ ƉƵƌƉƵƌĂƐĐĞŶƐ ^ŝƌŚŽŽŬĞƌĂ ůĂŶĐĞŽůĂƚĂ ƌŽŵŚĞĂĚŝĂ ĮŶůĂǇƐŽŶŝĂŶĂ ƌŽŵŚĞĂĚŝĂ ƐƌŝůĂŶŬĞŶƐŝƐ PŽůǇƐƚĂĐŚǇĂĂĸŶŝƐ 2x P. ŽƩŽŶŝĂŶĂ 2x, 4x P. ůŽŶŐŝƐĐĂƉĂ 2x P. ŶĞŽďĞŶƚŚĂŵŝĂ 2x P. ĚĞŶĚƌŽďŝŝŇŽƌĂ 1 2x P. ĚĞŶĚƌŽďŝŝŇŽƌĂ 2 2x P. ŐŽĞƚǌŝĂŶĂ P. ǀĂŐŝŶĂƚĂ 1 P. ǀĂŐŝŶĂƚĂ 2 P. ƉŽůǇĐŚĂĞƚĞ 2x P. cf. ƌŚŽĚŽƉƚĞƌĂ 2x? P. ƐĞƟĐĂƵůŝƐ P.ĐŽƌŝƐĐĞŶƐŝƐ 2x? P. ƌĂŵƵůŽƐĂ 2x P. ƐĞƟĨĞƌĂ 2x P. ĂĚĂŶƐŽŶŝĂĞ 1 2x P. ĂĚĂŶƐŽŶŝĂĞ 2 2x P. ĞůĞŐĂŶƐ 2x W͘ĐĂůůƵŶŝŇŽƌĂ P. ĂůƉŝŶĂ 1 W͘ĂůƉŝŶĂ 2 P. ďŝĮĚĂ 2 2x P. ďŝĮĚĂ 1 2x P. ƚŚŽŵĞŶƐŝƐ P.ďŝĐĂůĐĂƌĂƚĂ P.ƚĞŶƵŝƐƐŝŵĂ 2x P. ĨƵůǀŝůĂďŝĂ 2x, 4x P. ŶǇĂŶǌĞŶƐŝƐ 3 2x P. ŐĂůĞĂƚĂ 1 2x P. ŐĂůĞĂƚĂ 2 2x P. cf. ŶǇĂŶǌĞŶƐŝƐ2 2x? P. cf. ŶǇĂŶǌĞŶƐŝƐ 1 2x? P. ƐƵƉĮĂŶĂ P. ůĂǆŝŇŽƌĂ 2x P.ƚƌĂŶƐǀĂĂůĞŶƐŝƐ 1 2x P. ĐĂůŽŐůŽƐƐĂ 1 2x P. ĐĂůŽŐůŽƐƐĂ 2 2x P. ƚƌĂŶƐǀĂĂůĞŶƐŝƐ 2 2x P. ĂůďĞƐĐĞŶƐ subsp. ŝŵďƌŝĐĂƚĂ 2x W͘ďĞŶŶĞƫĂŶĂ 1 P. ďĞŶŶĞƫĂŶĂ 2 Outgroup ĸŶĞƐ ,ƵŵŝůĞƐ ĞŶĚƌŽďŝĂŶƚŚĞ ĞŶĚƌŽďŝĂŶƚŚĞ /ƐŽĐŚŝůŽŝĚĞƐ PŽůǇĐŚĂĞƚĞ ĂƵůĞƐĐĞŶƚĞƐ? PŽůǇĐŚĂĞƚĞ ĂůůƵŶŝŇŽƌĂĞ ,ƵŵŝůĞƐ ĂƵůĞƐĐĞŶƚĞƐ ƵůƚƌŝĨŽƌŵĞƐ ĂƵůĞƐĐĞŶƚĞƐ CLADES III, IV and V, Fig. 1, part 2 (see overleaf) Fig. . Strict consensus tree from 50,000 equally most parsimonious trees after maximum parsimony analysis of the combined plastid matrix. Shortest trees had a length of 2464 steps, consistency index = 0.69 and retention index = 0.83. Numbers above branches are bootstrap percentages; numbers below branches are posterior probabilities from Bayesian analysis. Roman numerals I to V indicate the five main clades discussed in the text. Vertical bars to the right of the tree indicate the sectional classification of the genus. Ploidy is indicated next to species names, if known. 393 TAXON 59 (2) • April 2010: 389–404 Russell & al. • Phylogenetics and cytology of Polystachya 100 1.0 From Fig. 1, part 1 (see overleaf) 100 1.0 96 1.0 99 96 100 1.0 1.0 1.0 III. 91 100 1.0 99 1.0 1.0 100 100 1.0 1.0 100 1.0 70 1.0 100 1.0 98 98 1.0 1.0 93 1.0 100 94 97 1.0 1.0 100 1.0 100 1.0 IV. 1.0 100 91 99 1.0 1.0 1.0 100 1.0 80 1.0 100 99 96 1.0 1.0 1.0 100 81 1.0 <50 1.0 0.86 82 99 1.0 1.0 98 1.0 89 1.0 74 0.97 100 1.0 74 100 1.0 1.0 100 100 1.0 1.0 V. 100 99 1.0 100 1.0 1.0 79 0.99 92 1.0 100 1.0 93 63 1.0 65 1.0 0.97 77 90 55 1.0 1.0 0.98 95 1.0 65 0.92 394 89 1.0 P. eurychila P. paniculata 1 2x P. paniculata 2 2x P. concreta 1 (Madagascar) P. concreta 9 (Reunion) P. pinicola 2x P. golungensis 1 2x P. golungensis 2 2x P. cornigera P. tsinjoarivensis 1 P. tsinjoarivensis 2 P. humberƟi 1 2x P. humberƟi 2 2x P. oreocharis 1 P. oreocharis 2 P. steudneri P. dolichophylla 2x P. concreta 2 (Madagascar) 4x P. concreta 3 (Madagascar) P. concreta 4 (Madagascar) P. concreta 5 (Brazil) 4x P. concreta 6 (Cameroon) P. concreta 7 (Laos) 4x P. concretaϴ;DĂƵƌŝƟƵƐͿ4x P. concreta 10 (Comoros) 4x P. concreta 11 (Seychelles) 4x P. foliosa 1 4x P. foliosa 2 4x P. henrici P. modesta 2x P. odorata 1 2x P. odorata 2 2x P. melanantha P. lindblomii P. eurygnatha 2x P. fusiformis 2x, 4x P. kermisina P. spatella 1 P. spatella 2 P. confusa P. campyloglossa 1 2x, 4x? P. campyloglossa 2 2x, 4x? P. piersii 1 4x P. piersii 2 4x P. holsƟi P. Įscheri 1 P. Įscheri 2 P. pubescens 1 4x, 6x? P. pubescens 2 4x, 6x? P.Ăī͘heckmanniana P. johnstonii P. purpureobracteata 2x P. villosa 2x P. lawrenceana 2x P. zambesiaca 2x P. bella 4x P. laurenƟi P. fallax 2x P. virginea 2x P. vulcanica var. aconiƟŇora 2x P. vulcanica var. vulcanica 2x P. pachychila 2x P. undulata 2x P. tenella 1 P. tenella 2 P. poikilantha 1 P. poikilantha 2 P. poikilantha var. leucorhoda 2x P. melliodora P. caespiƟĮca subsp. laƟlabris 1 2x P. caespiƟĮca subsp. laƟlabris 2 2x P. maculata 2x P. bicarinata P. cultriformis 1 2x P. cultriformis 2 2x P. virescens 4x P. cf. rosea 2 4x P. monophylla 4x P. cf. rosea 1 4x P. clareae 4x P. anceps 4x P. tsaratananae 1 4x P. tsaratananae 2 4x Eurychilae Caulescentes Polystachya (unplaced) Superpositae Polychaete Polystachya Humiles Superpositae Kermisinae Superpositae Humiles ĸnes Humiles ĸnes ĸnes Cultriformes (unplaced) TAXON 59 (2) • April 2010: 389–404 Russell & al. • Phylogenetics and cytology of Polystachya P. transvaalensis–P. bennettiana clade; the backbone of clade III was better resolved with P. eurychila sister to a clade containing all of the P. concreta accessions; P. odorata was sister to an unresolved clade of P. concreta, P. foliosa, P. modesta and P. henrici accessions. Polystachya forms a monophyletic group with good support and high sequence divergence from Adrorhizon, Bromheadia, Sirhookera and Phalaenopsis. Among the outgroup genera, Phalaenopsis has the most divergent sequence; the other three genera appear more closely related to each other than any is to Phalaenopsis or Polystachya. Figure 1 shows the sectional classification of species included in the study. The species representing currently recognised infrageneric taxa do not form monophyletic groups. In particular, the large sections Cultriformes, Caulescentes Kraenzl., Affines Kraenzl., Humiles Summerh. and Polychaete are all polyphyletic with regard to the sequence data. The tree in Fig. 1 can be divided into six parts for discussion, corresponding to five monophyletic clades with good support (BP 96%–100%), numbered I to V, and a grade of species-poor lineages. Polystachya affinis represents the earliest extant lineage to diverge from the rest of the genus, followed consecutively by lineages represented here by P. ottoniana, P. longiscapa, a P. neobenthamia–P. dendrobiiflora clade and a clade represented by members of sect. Isochiloides Summerh., which is sister to the single large clade containing clades I to V (BP 100%). Branches separating these clades are short, with just one to twelve substitutions (from the equally most parsimonious trees from PAUP*, not shown), but have good support (BP > 90%), except for those combining clades I and II (BP 71%) and clades IV and V (BP 81%) as sister groups. Clades III, IV and V form a clade, as do clades IV plus V. Two accessions of P. concreta (1 and 9, from Madagascar and Réunion) form a clade separate from the other nine accessions of P. concreta, which fall into a polytomy shared with P. foliosa, P. henrici, P. odorata and P. modesta. Despite the wide range and morphological variation in this group, there is little sequence divergence, with no more than 15 base substitutions between any two accessions (0.003%). ƌŽŵŚĞĂĚŝĂĮŶůĂǇƐŽŶŝĂŶĂ ƌŽŵŚĞĂĚŝĂƐƌŝůĂŶŬĞŶƐŝƐ Fig. . Maximum clade credibility tree from Bayesian analysis of the combined plastid DNA data matrix. Branch lengths are proportional to the expected number of substitutions per site (scale bar 0.005). Roman numerals I to V indicate the five main clades discussed in the text. The five main distribution areas are represented by colours: green = eastern Africa; blue = western Africa; yellow = Malagasy Islands; red = Neotropics; light blue = Asia. Coloured circles next to species names indicate extant distribution areas for that species. Coloured circles at internal nodes represent the ancestral distribution results from DIVA analysis. The absence of a coloured circle at a node indicates ambiguous results from DIVA. Coloured branches indicate historical distributions from Lagrange analysis when the descendants of a node could be assigned to distribution areas with a relative probability > 0.90. Black branches indicate a lack of statistical support for any particular distribution split. Outgroup species (grey) were not included in the biogeographical analyses. Phalaenopsis aphrodite subsp. ĨŽƌŵŽƐĂŶĂ Adrorhizon purpurascens ^ŝƌŚŽŽŬĞƌĂůĂŶĐĞŽůĂƚĂ WŽůǇƐƚĂĐŚǇĂĂĸŶŝƐ ͻͻͻ W͘ŽƩŽŶŝĂŶĂ ͻ P. longiscapa ͻ W͘ŶĞŽďĞŶƚŚĂŵŝĂ ͻ W͘ĚĞŶĚƌŽďŝŝŇŽƌĂ 1 ͻ W͘ĚĞŶĚƌŽďŝŝŇŽƌĂ 2 ͻ P. goetziana ͻ P. vaginata 1 ͻ P. vaginata 2 ͻ W͘ƉŽůǇĐŚĂĞƚĞ ͻͻ P. cf. rhodoptera (São Tomé) ͻ W͘ƐĞƟĐĂƵůŝƐ ͻ W͘ƐĞƟĨĞƌĂ ͻ I. P. coriscensis ͻ W͘ƌĂŵƵůŽƐĂ ͻͻ W͘ĐĂůůƵŶŝŇŽƌĂ ͻͻ P. alpina 2 ͻ P. alpina 1 ͻ P. adansoniae 2 ͻͻ P. elegans ͻ P. adansoniae 1 ͻͻ W͘ďŝĮĚĂ 2 ͻ W͘ďŝĮĚĂ 1 ͻ W͘ƚŚŽŵĞŶƐŝƐ ͻ W͘ŶǇĂŶǌĞŶƐŝƐ 3 ͻͻ W͘ĨƵůǀŝůĂďŝĂ ͻ P. galeata 1 ͻͻ II. P. galeata 2 ͻͻ P. cf. ŶǇĂŶǌĞŶƐŝƐ 2 ͻͻ P. cf. ŶǇĂŶǌĞŶƐŝƐ 1 ͻͻ W͘ƐƵƉĮĂŶĂ ͻ P. bicalcarata ͻ W͘ƚĞŶƵŝƐƐŝŵĂ ͻͻ W͘ůĂǆŝŇŽƌĂ ͻ P. transvaalensis 1 ͻͻ P. caloglossa 2 ͻͻ P. caloglossa 1 ͻͻ P. transvaalensis 2 ͻͻ W͘ďĞŶŶĞƫĂŶĂ 2 ͻͻ W͘ďĞŶŶĞƫĂŶĂ 1 ͻͻ P. albescens subsp. ŝŵďƌŝĐĂƚĂ ͻͻ 0.005 CLADES III, IV and V, Fig. 2, part 2 (see overleaf) 395 Russell & al. • Phylogenetics and cytology of Polystachya III. (see overleaf) From Fig. 2, part 1 IV. V. 0.005 396 TAXON 59 (2) • April 2010: 389–404 P. paniculata 1 ͻͻ Distribution. — Figure 2 shows distribution data, P. paniculata 2 ͻͻ both extant and ancestral as inferred from DIVA and P. cornigera ͻ P. tsinjoarivensis 1 ͻ Lagrange, mapped onto the maximum clade credP. tsinjoravensis 2 ͻ ibility tree from Bayesian analysis. P. oreocharis 1 ͻ P. oreocharis 2 ͻ There are two main clades of Malagasy endemics, P. humberƟi 2 ͻ one in clade III and the other in clade V. Neotropical P. humďerƟi 1 ͻ species appear in two separate subclades in clade III. P. eurychila ͻͻ P. concreta 9 ͻͻͻͻͻ Specimens of P. concreta, P. foliosa, P. henrici, and P. concreta 1 ͻͻͻͻͻ P. modesta form a polytomy in clade III with low P. pinicola ͻ P. golungensis 1 ͻͻ levels of sequence divergence between them. These P. golungensis 2 ͻͻ specimens originate from places as diverse as Brazil, P. steudneri ͻͻ P. dolichophylla ͻ the Dominican Republic, Cameroon, Madagascar, the P. odorata 2 ͻͻ Comoros, Mauritius, the Seychelles, and Laos. P. odorata 1 ͻͻ P. foliosa 2 ͻ Considering just the division of the range of P. concreta 6 ͻͻͻͻͻ Poly stachya in Africa into eastern and western arP. modesta ͻͻ P. concreta 5 ͻͻͻͻͻ eas, the early-divergent lineages are mostly endemic P. concreta 8 ͻͻͻͻͻ to the eastern part, except for P. affinis, which is P. concreta 10 ͻͻͻͻͻ P. foliosa 1 ͻ widespread in tropical Africa. Other species with P. concreta 2 ͻͻͻͻͻ an eastern distribution are all from clades IV and P. concreta 4 ͻͻͻͻͻ P. henrici ͻ V. Species with a western distribution are clustered P. concreta 3 ͻͻͻͻͻ in clades I and II; the species that are widespread in P. concreta 7 ͻͻͻͻͻ Africa (found in both east and west) are mostly found P. concreta 11 ͻͻͻͻͻ P. lindblomii ͻ in clades I, II and III. P. fusiformis ͻͻͻ Lagrange was unable to suggest an ancestral disP. eurygnatha ͻ P. kermisina ͻ tribution for the root of the genus; the wide extant disP. spatella 2 ͻ tribution of P. affinis led DIVA to suggest an equally P. spatella 1 ͻ P. melanantha ͻ wide ancestral distribution for its common ancestor P. confusa ͻ with the rest of the genus as the most parsimonious reP. campyloglossa 1 ͻ P. campyloglossa 2 ͻ sult. However, both methods supported purely eastern P. piersii 2 ͻ African distributions for the next most-basal nodes P. piersii 1 ͻ P. holsƟi ͻ and internal branches. P. Įscheri 2 ͻ Despite the large number of widespread species in P. Įscheri 1 ͻ P. pubescens 2 ͻ clades I and II, both DIVA and Lagrange suggested P. pubescens 1 ͻ western African distributions for the ancestral species P. aī. heckmanniana ͻ of those two clades. Similarly, both programs showed P. johnstonii ͻ P. purpureobracteata ͻ clades IV and V to be strongly eastern African, with at P. villosa ͻ one point a species distributed in both eastern Africa P. lawrenceana ͻ P. zambesiaca ͻ and the Malagasy Islands giving rise to endemic eastP. bella ͻ ern African and Malagasy daughter species. P. laurenƟi ͻ P. virginea ͻ Except for the monophyletic clade of Malagasy P. fallax ͻ endemics, DIVA favoured western African ancestral P. vulcanica var. aconiƟŇora ͻ P. vulcanica var. vulcanica ͻ distributions in clade III, but many internal nodes P. undulata ͻ have several equally parsimonious solutions. LaP. pachychila ͻ P. tenella 1 ͻ grange gave no support to any particular set of geoP. tenella 2 ͻ graphical splits except again for the P. poikilantha 2 ͻ P. poikilantha 1 ͻ Malagasy species. P. poikilantha var. leucorhoda ͻ Genomic data. — Ploidy for PolyP. caespiƟĮca subsp. laƟlabris 2 ͻ P. caespiƟĮca subsp. laƟlabris 1 ͻ stachya species included in this study P. melliodora ͻ is given in Table 2, from published P. maculata ͻ chromosome counts and genome size P. cultriformis 2 ͻͻͻ P. cultriformis 1 ͻͻͻ estimates, and our own data (e.g., Fig. P. bicarinata ͻ 3). The basic chromosome number P. cf. rosea 2 ͻ P. cf. rosea 1 ͻ for the genus is x = 20. Genome size P. monophylla ͻ (1C) values are in the range 0.60–0.92 P. virescens ͻ P. clareae ͻ (–0.99) pg for diploids, and 1.10–1.80 pg P. anceps ͻ for tetraploids. The data are also shown P. tsaratananae 2 ͻ P. tsaratananae 1 ͻ with species names on Fig. 1. Species Chomosome counts, 2n Genome size, 1C [pg] Ploidy level Species Polystachya adansoniae 40a,b 0.60–0.85f,g 2x P. longiscapa 0.64g 2x f,h 2x P. maculata 0.61g 2x 2x P. modesta 1.47–1.53f 4x 0.64g 2x a P. affinis 40 P. albescens subsp. imbricata 40b P. anceps 80a 1.47g 4x P. monophylla P. bifida 40a 0.61g 2x P. neobenthamia P. bella 80 0.81–0.83 a f P. caespitifica subsp. latilabris P. caloglossa 40 40a a,c f,g P. nyanzensis 40 0.79–0.82 2x P. odorata 40b 0.81–0.92f,g f,g P. campyloglossa 40 ; 80? 0.69–0.74 2x, 4x? P. pachychila 80a 1.45g 4x P. paniculata 40a P. concreta (Africa) 40b; 80a 0.73–0.75f,g, 1.33f 2x, 4x P. piersii 80a g 2x 2x f 0.70–0.77 ; 1.41–1.42 P. ottoniana P. aff. clareae a,b,e 2x 4x g Ploidy level 40b 0.71f 2x b Genome size, 1C [pg] 1.10, 1.75–1.80 b a Chomosome counts, 2n 2x, 4x f 0.64 2x 0.64g 2x 4x g P. concreta (Neotropics) 80 1.39–1.43 4x P. pinicola P. concreta (Malagasy Is.) 80a 1.33–1.54f,g,h 4x P. poikilantha var. leucorhoda 40c P. concreta (Laos) 1.40f 4x P. polychaete 40e 0.60f 2x P. coriscensis 0.93–0.99g 2x? P. pubescens 80a; 120?a,b 1.38–1.80f,g 4x, 6x? P. cultriformis e a,c 38–39 ; 40 0.62–0.65 g,h 2x 0.74 0.66 P. purpureobracteata 0.58 2x P. ramulosa 40 P. dolichophylla 0.74h 2x P. rhodoptera 38e 0.76 P. elegans P. eurygnatha 40 P. fallax 40a P. foliosa 80 2x b 0.70–0.71f a 1.40–1.49 b f f,g 2x P. setifera P. tenuissima 4x P. transvaalensis P. fulvilabia 40 1.62 2x, 4x P. tsaratananae P. fusiformis 40a,b 0.67–0.70g; 1.45–1.75f 2x, 4x P. undulata P. galeata 40a,b,d 0.73–0.83f,h 2x P. villosa 0.64g 2x P. virescens P. golungensis 0.75 P. humbertii g 2x P. virginea P. isochiloides 40 b 2x P. vulcanica P. lawrenceana 40a 2x P. zambesiaca P. laxiflora 40a,b a 0.72g b 2x P. cf. rosea 2x 2x b 0.63f 2x g 4x f 2x 1.48 0.81 40c 2x b 40 2x g 1.58 4x 40c 2x 0.66–0.73f,g 2x 80a 1.55g 4x a g 40 0.67 c a,c c 35? ; 40 ; 62? 2x g 0.74–0.76 2x 0.68g 2x 2x c 397 Data sources for chromosome counts: this study; Jones (1966); Podzorski & Cribb (1979); d Reich (2006); e Fedorov (1969). Data sources for genome size ranges: f Rupp & al. (subm.); g Rupp (2008); h Reich (2006). Russell & al. • Phylogenetics and cytology of Polystachya P. dendrobiiflora f 2x 2x g f TAXON 59 (2) • April 2010: 389–404 Table . Chromosome numbers and genome size ranges of Polystachya species included in this study. Russell & al. • Phylogenetics and cytology of Polystachya TAXON 59 (2) • April 2010: 389–404 Fig. . DAPI-stained chromosome spreads of Polystachya. Scale bars are 5 μm. A, P. fusiformis (2n = 40); B, P. virginea (2n = 40); C, P. concreta from Réunion (2n = 80); D, P. concreta from Madagascar (2n = 80). Polyploidy (2n = 80) has occurred in 17 of the study species, scattered throughout the genus but especially in sect. Polystachya Kraenzl. and in the Malagasy species in clade V. For Polystachya concreta we have included more accessions in the study to cover the geographical range, and both diploids and tetraploids have been recorded, so for this species chromosome numbers are only provided in Fig. 1 if they are known for that particular accession. DISCUSSION Polystachya phylogeny. — This study confirms the inclusion of Neobenthamia within Polystachya (Schlechter in Warburg, 1903; Reich, 2006). It is sister to P. dendrobiiflora, with which it has many morphological similarities, such as slender stems with a narrow, conical pseudobulb and distichous, linear leaves. The flowers of both are showy, thin-textured, white to pink in colour and borne on a slender pedicel. From the infrageneric classification represented on Fig. 1, the only section with multiple representatives to appear monophyletic based on current taxon sampling is sect. Isochiloides. Sections Polystachya and Dendrobianthe Schltr. could be considered paraphyletic; all the other sections are polyphyletic. Morphological analysis will be needed to refine the sectional classification and identify morphological synapomorphies for infrageneric groups; many of the characters (e.g., unifoliate shoots; setose bracts; superposed stems) previously used for sectional delimitation are homoplasious. The sister to the rest of the genus is Polystachya affinis. Polystachya bancoensis Burg is not included in this analysis but has similar morphology. The two species’ appearance is different from other plants in the genus, with atypical large 398 subspherical or flattened pseudobulbs, a pendulous inflorescence with an indumentum of brown hairs, and large bracts relative to the flowers. These plants belong to the large sect. Affines, but are not closely related to other members of the section sampled here, which are mainly found in clade IV. The next consecutively diverging lineages are morphologically diverse and represented here by P. ottoniana, P. longiscapa, a clade containing P. neobenthamia and P. dendrobiiflora, and a clade containing the two sampled members of sect. Isochiloides. This study does not include other members of sections Dendrobianthe or Isochiloides: from sect. Dendrobianthe we lack Polystachya zuluensis L. Bolus from South Africa and Swaziland. Section Isochiloides contains a further nine species with distributions in eastern Africa. Six species are restricted to the Eastern Arc mountain ranges of Tanzania, and other species are found in Malawi and western Tanzania. Polystachya vaginata is more widely distributed and occurs from Kenya south to Zambia and Zimbabwe. The well-supported (BP 100%) clade sister to sect. Isochiloides contains the majority of the genus. Topographically, it consists of: (1) a backbone of early divergences, which are fully resolved but not all well supported (BP 71%–100%); (2) several large clades (our clades I to V) with good support (BP 96%–100%). Clade I reunites members of sect. Calluniflorae Kraenzl. with sect. Polychaete (except for P. steudneri), following their separation by Cribb (1978). The two western African species P. alpina and P. rhodoptera also appear here. Section Polychaete contains ca. 15 species with more or less dense inflorescences of small flowers with setose bracts. Polystachya steudneri is morphologically dissimilar to the rest of the section, with deciduous leaves, a secund inflorescence with a sheathed peduncle and rachis, and bracts acutely triangular rather than setose. TAXON 59 (2) • April 2010: 389–404 Clade II contains species belonging to sections Caulescentes (leafy-stemmed plants lacking pseudobulbs) and Cultriformes, a large section (> 40 spp.) defined by having only a single apical leaf, but which is otherwise heterogeneous. Members of this section in clade II include the widespread and variable P. galeata and species with similar morphology— generally robust plants with coriaceous leaves and fleshy flowers with a large mentum. Species boundaries are unclear in this group, especially between P. galeata, P. nyanzensis and P. fulvilabia. Our study indicates the two “P. aff. nyanzensis” specimens from Cameroon are more closely related to P. supfiana, also from Cameroon, than to a third P. nyanzensis plant grown in cultivation; this last P. nyanzensis is more closely related to P. galeata and P. fulvilabia from Sierra Leone and Congo, respectively. Also in clade II are P. tenuissima and P. bicalcarata, two species in sect. Cultriformes with grasslike leaves and small flowers borne on slender shoots. Clade III contains several distinct subclades of morphologically diverse plants, but it is unresolved at the base. One comprises a group of diminutive plants endemic to the Malagasy Islands. They can be characterized by their small stature and inflorescences of small flowers opening widely, with labella with undulate margins, short side-lobes and a horn-shaped central callus. There are also two clades of plants belonging to sect. Polystachya, characterised by short, leafy stems with an obscure pseudobulb, oblanceolate to obovate leaves, and either racemose or paniculate inflorescences, commonly secund, with a sheathed peduncle. Some members have moniliform labellar food hairs, as opposed to clavate or setose in other sections (Davies & al., 2002). This is the only section to have spread to the Neotropics and Asia. There are ca. 20 species in the Neotropics, including P. concreta, for which the range also extends through Africa, the Indian Ocean, and Southeast Asia. Species delimitations in this section are difficult; for example the definition of P. concreta applied by Garay & Sweet (1974) is not widely accepted, with the synonym P. tessellata still generally used for plants on mainland Africa, either P. mauritiana, P. tessellata, or P. concreta for plants on Madagascar and the Indian Ocean islands, and P. concreta used in the Neotropics and Asia. Phenotypes represented by the synonym P. estrellensis Rchb. f., for example, are sometimes maintained as a separate species. On mainland Africa, there is variation in flower colour, including plants with concolor yellow or purple flowers. Similar variation exists in other areas, with orange, white, and pink forms occurring in Madagascar and other Indian Ocean islands. Nine samples out of eleven in this study, from Africa, Madagascar, Indian Ocean islands and the Neotropics, fall into a polytomy shared with P. foliosa, P. henrici, P. odorata, and P. modesta. Another two P. concreta accessions, from Madagascar and Réunion, belong to a different clade with P. pinicola (Neotropical) and P. golungensis. Assuming correct identification of the specimens and reliable data collection and analysis, reasons for P. concreta specimens falling into two different clades include morphological homoplasy to the extent that unrelated plants can be assigned to the same morphological species; introgression and plastid capture (Rieseberg & Soltis, 1991) between Russell & al. • Phylogenetics and cytology of Polystachya P. concreta and plants in the P. pinicola/P. golungensis clade; or multiple hybrid origins (Soltis & Soltis, 1999) of P. concreta tetraploids (we have so far found only tetraploid P. concreta outside mainland Africa and the Gulf of Guinea islands, but we cannot yet say that they are allotetraploids), with one parent species contributing the plastid genome in some populations and the other parent contributing the plastid genome in others. At present we cannot rule out any of these possibilities, although future analysis of nuclear DNA sequences might provide additional evidence. In clade IV, members of sect. Superpositae Kraenzl. (excluding P. oreocharis) and sect. Kermisinae P.J. Cribb together form a subclade; they are characterised by their fusiform pseudobulbs, superposed in branching structures. Inflorescences are racemose or paniculate, lax, bearing medium-sized flowers with a prominent cylindrical mentum. The other members of clade IV belong to sects. Affines and Humiles. They are morphologically consistent within the clade but not necessarily within their respective polyphyletic sections; the accessions in clade IV are small plants with conical, clustered pseudobulbs and relatively large, fleshy flowers in a simple raceme. Species ranges are confined to an arc across eastern Africa, from Kenya and Uganda, south along the Rift Valley and neighbouring mountain ranges into Zambia, Zimbabwe, Mozambique, and South Africa. Species ranges in the sect. Superpositae/sect. Kermisinae clade also conform to this distribution, with two exceptions: P. fusiformis is more widely distributed across tropical Africa and the Indian Ocean; P. superposita Rchb. f. (not included here) is restricted to western Africa, from Cameroon to Gabon. Clade V includes the remaining species of sect. Cultriformes and also two species from sect. Affines and a separate clade of Malagasy endemics characterized by a robust habit, ligulate or oblanceolate leaves and fleshy, often brightly coloured flowers. Except for the Malagasy endemics and the widespread P. cultriformis, all species in clade V are restricted to eastern or east-central Africa. The members of sect. Cultriformes in this clade are heterogeneous, from the robust P. cultriformis to the diminutive P. vulcanica and P. caespitifica. The section as a whole is shown here to be polyphyletic, with the defining characteristic of a single leaf below the inflorescence probably having evolved in two separate clades. The clade V Madagascan endemics appear to have arisen from this group but subsequently lost the single-leaved habit. Distribution. — There have been at least two separate dispersals of Polystachya to the Neotropics. Increased taxon sampling of Neotropical species may reveal more dispersal events. Neotropical P. pinicola is sister to the African P. golungensis, suggesting a transatlantic dispersal from Africa to South America. The other Neotropical species, P. foliosa and P. concreta, are closely related to each other and to African, Malagasy and Asian plants, with no resolution in the tree and little sequence divergence. The group is nested within mainland-African species such as P. odorata and P. dolichophylla. This suggests a recent dispersal from Africa to tropical forests across the Indian Ocean, southern and southeastern Asia, and the Neotropics. More resolution in this group might 399 Russell & al. • Phylogenetics and cytology of Polystachya be obtained using other data sources such as nuclear DNA sequences or AFLP markers, but the dispersal routes and possible biological factors allowing such a rapid sequence of long distance dispersals are well worth further study. Examples of biological factors could include improved wind or other vector dispersal (Nathan & al., 2002; Nathan, 2006), less specific substrate or mycorrhizal requirements for germination and survival of seedlings (Bidartondo & Read, 2008), and/or the ability of pioneering individuals to found colonies through vegetative reproduction or selfing or apomixis (Mogie & al., 2007; Hörandl & al., 2008). The phenomenon of long-distance dispersal is complex, but absolutely important in plant evolution (Cain & al., 2000). There are two main clades of Malagasy endemics, one in clade III consisting of miniature species with white or pale pink flowers; another in clade V consisting of larger plants with coriaceous and often glaucous foliage and usually brightly coloured flowers. From branch lengths (Fig. 2) and phylogenetic position, this species group in clade V is a more recent dispersal and derived from sect. Cultriformes. Considering the plants on mainland Africa, two centres of diversity were apparent from the literature studied: (1) eastcentral Africa associated with the Eastern Arc Mountains and the Albertine Rift Highlands of the Democratic Republic of Congo, Uganda, Rwanda and Burundi south into Malawi; (2) west-central Africa associated with the highlands of southwest Cameroon and the coastal and moist lowland forests surrounding the Gulf of Guinea. The former corresponds to montane and sub-montane rainforests that have long been recognized as bearing ancient and highly endemic floras (e.g., Fjeldså & Lovett, 1997; Burgess & al., 2007). The western area corresponds well with the inferred areas of lowland tropical forest during Pleistocene dry periods and glacial maxima (Nichol, 1999). The different species compositions of eastern and western Africa reflect an ancient phylogenetic division between the two areas, followed by more recent dispersal. The early– divergent lineages are mostly endemic to the eastern area (especially the Eastern Arc Mountains) as are clades IV and V. Few species in these groups can be found in both western and eastern areas, exceptions being P. affinis and P. cultriformis, although clade V also contains a clade of Malagasy endemics as discussed above. By contrast, species endemic to the western area are mostly restricted to clades I and II. They are associated with a large number of species with wide distributions across tropical Africa. Many of these species are common in anthropogenic habitats such as coffee plantations and villages, as well as open woodland and disturbed forest. The results from DIVA show at least 18 dispersal events from western to eastern Africa, compared to just three from eastern to western. However, it is possible that DIVA provides more concrete ancestral area results than are merited by the data. For example, several of the internal nodes in clade III are given western African distributions, based purely on the extant distribution of P. dolichophylla in western Africa. If this one species had an eastern African distribution, the relationship between eastern and western Africa in this clade according to DIVA would be reversed, with ancestral nodes 400 TAXON 59 (2) • April 2010: 389–404 being assigned to only eastern Africa. Our results from the likelihood-based Lagrange are more conservative, with fewer strongly supported inferences of ancestral area. This more accurately reflects the inherent uncertainty of ancestral area analysis (e.g., Xiang & Thomas, 2008), especially in a scenario such as this with incomplete taxon sampling, an inability to represent phylogenetic uncertainty in the analysis, and no fossil evidence to assign areas to internal nodes a priori. However, where Lagrange does give well-supported results, they agree with the results from DIVA, and both programs support the conclusion of greater dispersal from western to eastern Africa from ancestral lineages that were endemic to those areas for long periods of time. It is possible that the restriction of African wet forest plants to refugia during Pleistocene dry periods (Coetzee, 1964; Nichol, 1999; Cohen & al., 2007) has had a big impact on the patterns of diversity that we see today. For example, during these periods there may have been a strong selection in western Africa for plants that were better dispersers or more catholic in their ecological preferences. Cohen & al. (2007) thought it likely that stable African refugia would be difficult to locate, and biodiversity was instead concentrated successively in different areas during cycles of extreme climate change. The eastern African Polystachya species tend to occur more in montane and submontane forests, areas with high ecological heterogeneity, and might have been better able to adjust to changing climatic conditions through elevational range fluctuation over short geographical distances (Kiage & Liu, 2006; Ehrich & al., 2007) than the more lowland, western African plants. Eastern African montane vegetation may have been more extensive during the Pleistocene dry periods than they are today (Ehrich & al., 2007; Wronski & Hausdorf, 2008). In spite of more frequent dispersal from western to eastern Africa in recent periods, the presence of several early diverging clades of eastern African species suggests an early radiation of the genus in eastern Africa and subsequent spread. Genome evolution. — There have been at least eight instances of tetraploidy in Polystachya. In some cases speciation has occurred subsequently giving rise to tetraploid clades; in other cases two ploidy states occur within a species. This frequency and scattered distribution of polyploidy is unusual in Orchidaceae. It may reflect hybrid speciation and may have had a significant impact on the evolution of some lineages. Chromosome counts and genome size measurements from a greater range of species and individuals would give a fuller picture of the extent of polyploidy; in this study we have been limited by the availability of living plants. Notably, all members of the P. concreta species group that have dispersed outside Africa, for which data are available, are tetraploid; mainland-African plants can be diploid or tetraploid, and their African sister species are diploid. Correlation between polyploidy and greater dispersal has been observed in some plant groups (Brochmann & al., 2004; Lowry & Lester, 2006; Hijmans & al., 2007) and could be due to the development of uniparental reproduction, increasing the ability of pioneer plants to establish new populations; or it could be due polyploids being better able to establish in suboptimal habitats TAXON 59 (2) • April 2010: 389–404 because they express a novel combination of genes. This is a complex phenomenon reviewed for example by Mogie & al. (2007). However, it is not a universal trend (Stebbins & Dawe, 1987; Hörandl, 2006; Barringer, 2007). Further research would be desirable into breeding systems, mycorrhizal ecology and physiological tolerances of this species group compared to, for example, diploid P. pinicola, which is also found in South America, and related African Polystachya species. The origins of the tetraploids are of particular interest, and our understanding of this and other aspects of Polystachya evolution could be improved, for example, by adding data from the nuclear genome (e.g., van den Hof & al., 2008). This work is ongoing, and future analysis of nuclear data may reveal the nature of reticulation that might have occurred and give a fuller picture of the evolution of the genus. ACKNOWLEDGEMENTS Thanks to Tod Stuessy for his support and providing the facilities in Vienna for this work. The expertise and collections of Manfred Speckmaier, Suranjan Fernando, Anton Sieder and Gunter Fischer have been invaluable to the project. Isobyl la Croix supplied important material including living specimens from her private collection. We give thanks to Philip Cribb for discussions and for helping with some specimen identifications. Elfriede Grasserbauer, Gudrun Kohl and especially Verena Klejna contributed laboratory work in Vienna. Hanna Weiss-Schneeweiss, Eva Temsch, Johann Greilhuber and Ilia Leitch provided help with laboratory space, materials, training and ideas. Jeff Wood and Clare Drinkell were more than helpful during visits to the herbarium at Kew. Martin Cheek, George Goslin, JeanMichel Onana, Laura Pearce, Olivier Sene and Valery Noiha Naomi made fieldwork possible. Thanks go to two anonymous reviewers who helped improve the manuscript. This project is funded by the Austrian Science Fund (FWF) grant number AP19108. LITERATURE CITED Altekar, G., Dwarkadas, S., Huelsenbeck, J.P. & Ronquist, F. 2004. Parallel metropolis coupled Markov chain Monte Carlo for Bayesian phylogenetic inference. Bioinformatics 20: 407–415. Barfuss, M.H.J., Samuel, R., Till, W. & Stuessy, T.F. 2005. Phylogenetic relationships in subfamily Tillandsioideae (Bromeliaceae) based on DNA sequence data from seven plastid regions. Amer. J. Bot. 92: 337–351. Barringer, B.C. 2007. Polyploidy and self-fertilization in flowering plants. Amer. J. Bot. 94: 1527–1533. Barthet, M.M. & Hilu, K.W. 2007. Expression of matK : Functional and evolutionary implications. Amer. J. Bot. 94: 1402–1412. Bidartondo, M.I. & Read, D.J. 2008. Fungal specificity bottlenecks during orchid germination and development. Molec. Ecol. 17: 3707–3716. Brenan, J.P.M. (ed.). 1954. Plants collected by the Vernay Nyasaland expedition of 1946. Mem. New York Bot. Gard. 9: 1–132. Brochmann, C., Brysting, A.K., Alsos, I.G., Borgen, L., Grundt, H.H., Scheen, A.C. & Elven, R. 2004. Polyploidy in arctic plants. Biol. J. Linn. Soc. 82: 521–536. Burgess, N.D., Butynski, T.M., Cordeiro, N.J., Doggart, N.H., Fjeldså, J., Howell, K.M., Kilahama, F.B., Loader, S.P., Lovett, Russell & al. • Phylogenetics and cytology of Polystachya J.C., Mbilinyi, B., Menegon, M., Moyer, D.C., Nashanda, E., Perkin, A., Rovero, F., Stanley, W.T. & Stuart, S.N. 2007. The biological importance of the Eastern Arc Mountains of Tanzania and Kenya. Biol. Cons. 134: 209–231. Bytebier, B., Bellstedt, D.U. & Linder, H.P. 2007. A molecular phylogeny for the large African orchid genus Disa. Molec. Phylog. Evol. 43: 75–90. Cain, M.L., Milligan, B.G. & Strand, A.E. 2000. Long-distance seed dispersal in plant populations. Amer. J. Bot. 87: 1217–1227. Cameron, K.M. 2001. An expanded phylogenetic analysis of Orchidaceae using three plastid genes: rbcL, atpB, and psaB. P. 104 in: Botany 2001 Abstracts. St. Louis: Botanical Society of America. Cameron, K.M. 2005. Leave it to the leaves: A molecular phylogenetic study of Malaxideae (Epidendroideae, Orchidaceae). Amer. J. Bot. 92: 1025–1032. Carlsward, B.S., Stern, W.L. & Bytebier, B. 2006a. Comparative vegetative anatomy and systematics of the angraecoids (Vandeae, Orchidaceae) with an emphasis on the leafless habit. Bot. J. Linn. Soc. 151: 165–218. Carlsward, B.S., Whitten, W.M., Williams, N.H. & Bytebier, B. 2006b. Molecular phylogenetics of Vandeae (Orchidaceae) and the evolution of leaflessness. Amer. J. Bot. 93: 770–786. Chang, C.C., Lin, H.C., Lin, I.P., Chow, T.Y., Chen, H.H., Chen, W.H., Cheng, C.H., Lin, C.Y., Liu, S.M., Chang, C.C. & Chaw, S.M. 2006. The chloroplast genome of Phalaenopsis aphrodite (Orchidaceae): Comparative analysis of evolutionary rate with that of grasses and its phylogenetic implications. Molec. Biol. Evol. 23: 279–291. Chase, M.W., Cameron, K.M., Barrett, R.L. & Freudenstein, J.V. 2003. DNA data and Orchidaceae systematics: a new phylogenetic classification. Pp. 69–89 in: Dixon, K.W., Kell, S.P., Barrett, R.L. & Cribb, P.J., (eds.), Orchid conservation. Kota Kinabalu: Natural History Publications (Borneo). Chase, M.W. & Hills, H.H. 1991. Silica-gel: An ideal material for field preservation of leaf samples for DNA studies. Taxon 40: 215–220. Coetzee, J.A. 1964. Evidence for a considerable depression of the vegetation belts during the Upper Pleistocene on East African mountains. Nature 204: 564–566. Cohen, A.S., Stone, J.R., Beuning, K.R.M., Park, L.E., Reinthal, P.N., Dettmar, D., Scholz, C.A., Johnsor, T.C., King, J.W., Talbot, M.R., Brown, E.T. & Ivory, S.J. 2007. Ecological consequences of early Late Pleistocene megadroughts in tropical Africa. Proc. Natl. Acad. Sci. U.S.A. 104: 16422–16427. Cribb, P.J. 1978. Studies in the genus Polystachya (Orchidaceae) in Africa. Kew Bull. 32: 743–766. Cribb, P.J. 1984. Polystachya. Pp. 333–402 in: Polhill, R.M. (ed.), Flora of tropical East Africa: Orchidaceae, part 2. Rotterdam: Balkema. Cuenoud, P., Savolainen, V., Chatrou, L.W., Powell, M., Grayer, R.J. & Chase, M.W. 2002. Molecular phylogenetics of Caryophyllales based on nuclear 18S rDNA and plastid rbcL, atpB, and matK DNA sequences. Amer. J. Bot. 89: 132–144. Davies, K.L., Roberts, D.L. & Turner, M.P. 2002. Pseudopollen and food-hair diversity in Polystachya Hook. (Orchidaceae). Ann. Bot. 90: 477–484. Doyle, J.J. & Doyle, J.L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19: 11–15. Dressler, R.L. 1993. Phylogeny and classification of the orchid family. Cambridge: Cambridge Univ. Press. Edgar, R.C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32: 1792–1797. Ehrich, D., Gaudeul, M., Assefa, A., Koch, M.A., Mummenhoff, K., Nemomissa, S., Consortium, I. & Brochmann, C. 2007. Genetic consequences of Pleistocene range shifts: Contrast between the Arctic, the Alps and the East African mountains. Molec. Ecol. 16: 2542–2559. Fedorov, A.A. (ed.). 1969. Chromosome numbers of flowering plants. Leningrad: Academy of Sciences of the USSR. 401 Russell & al. • Phylogenetics and cytology of Polystachya Fischer, G.A., Gravendeel, B., Sieder, A., Andriantiana, J., Heiselmayer, P., Cribb, P.J., Smidt, E.D.C., Samuel, R. & Kiehn, M. 2007. Evolution of resupination in malagasy species of Bulbophyllum (Orchidaceae). Molec. Phylog. Evol. 45: 358–376. Fjeldså, J. & Lovett, J.C. 1997. Geographical patterns of old and young species in African forest biota: The significance of specific montane areas as evolutionary centres. Biodiv. Cons. 6: 325–346. Freudenstein, J. & Rasmussen, F. 1999. What does morphology tell us about orchid relationships? A cladistic analysis. Amer. J. Bot. 86: 225–248. Garay, L.J. & Sweet, H.R. 1974. Terminilogía de orchidearum Jacquinii. Orquideologia 9: 200–210. Geerinck, D. 1979. Notes taxonomiques sur des Orchidacées d’Afrique centrale VI: Polystachya Hook. sect. Cultriformes Kraenzl. Bull. Jard. Bot. Natl. Belg. 49: 409–420. Geerinck, D. 1980. Répartition géographic d’Orchidacées africaines I: Polystachya Hook. Bull. Jard. Bot. Natl. Belg. 50: 99–116. Govaerts, R., Campacci, M.A., Holland Baptista, D., Cribb, P.J., George, A., Kreuz, K. & Wood, J. 2009. World checklist of Orchidaceae. The Board of Trustees of the Royal Botanic Gardens, Kew. http://apps.kew.org/wcsp/ (accessed 31 January 2009). Gravendeel, B., Chase, M.W., De Vogel, E.F., Roos, M.C., Mes, T.H.M. & Bachmann, K. 2001. Molecular phylogeny of Coelogyne (Epidendroideae; Orchidaceae) based on plastid RFLPs, matK, and nuclear ribosomal ITS sequences: Evidence for polyphyly. Amer. J. Bot. 88: 1915–1927. Hausner, G., Olson, R., Simon, D., Johnson, I., Sanders, E.R., Karol, K.G., McCourt, R.M. & Zimmerly, S. 2006. Origin and evolution of the chloroplast trnK (matK) intron: A model for evolution of group II intron RNA structures. Molec. Biol. Evol. 23: 380–391. Hijmans, R.J., Gavrilenko, T., Stephenson, S., Bamberg, J., Salas, A. & Spooner, D.M. 2007. Geographical and environmental range expansion through polyploidy in wild potatoes (Solanum section Petota). Global Ecol. Biogeogr. 16: 485–495. Hörandl, E. 2006. The complex causality of geographical parthenogenesis. New Phytol. 171: 525–538. Hörandl, E., Cosendai, A.-C. & Temsch, E.M. 2008. Understanding the geographic distributions of apomictic plants: A case for a pluralistic approach. Pl. Ecol. Divers. 1: 309–320. Huelsenbeck, J.P. & Ronquist, F. 2001. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755. Jones, K. 1966. The chromosomes of orchids: I. Polystachya Hook. Kew Bull. 20: 357–359. Jones, K. & Daker, M.G. 1966. The chromosomes of orchids: III. Catasetinae Schltr. Kew Bull. 22: 421–427. Kelchner, S.A. 2000. The evolution of non-coding chloroplast DNA and its application in plant systematics. Ann. Missouri Bot. Gard. 87: 482–498. Kiage, L.M. & Liu, K.B. 2006. Late Quaternary paleoenvironmental changes in East Africa: A review of multiproxy evidence from palynology, take sediments, and associated records. Progr. Phys. Geogr. 30: 633–658. Kraenzlin, F. 1926. Monographie der Gattung Polystachya Hook. Repert. Spec. Nov. Regni Veg. Beih. 39: 1–136. La Croix, I. & Cribb, P.J. 1998. Polystachya. Pp. 321–363 in: Pope, G.V. (ed.), Flora Zambesiaca, vol. 11, part 2. Kew: Royal Botanic Gardens. Li, J.T., Yang, J., Chen, D.C., Zhang, X.L. & Tang, Z.S. 2007. An optimized mini-preparation technique to obtain high-quality genomic DNA from mature leaves of sunflower. Genet. Molec. Res. 6: 1064–1071. Lowry, E. & Lester, S.E. 2006. The biogeography of plant reproduction: potential determinants of species’ range sizes. J. Biogeogr. 33: 1975–1982. Maddison, D.R. & Maddison, W.P. 2005. MacClade 4: Analysis of phylogeny and character evolution. Sunderland, Massachusetts: Sinauer. 402 TAXON 59 (2) • April 2010: 389–404 Marshall, D.C., Simon, C. & Buckley, T.R. 2006. Accurate branch length estimation in partitioned Bayesian analyses requires accommodation of among-partition rate variation and attention to branch length priors. Syst. Biol. 55: 993–1003. Micheneau, C., Carlsward, B.S., Fay, M.F., Bytebier, B., Pailler, T. & Chase, M.W. 2008. Phylogenetics and biogeography of Mascarene angraecoid orchids (Vandeae, Orchidaceae). Molec. Phylog. Evol. 46: 908–922. Mogie, M., Britton, N.F. & Stewart-Cox, J.A. 2007. Asexuality, polyploidy and the male function. Pp. 195–214 in: Hörandl, E., Grossniklaus, U., Van Dijk, P. & Sharbel, T. (eds.), Apomixis: Evolution, mechanisms and perspectives. Ruggell: Gantner. Mytnik-Ejsmont, J. 2007. Contribution à la révision de Polystachyinae Schlechter (Orchidaceae). 1. Szlachetkoella Mytnik. Richardiana 7: 55–60. Mytnik-Ejsmont, J. & Szlachetko, D.L. 2007a. Contribution à la révision de Polystachyinae Schlechter (Orchidaceae). 2. Geerinckia Mytnik & Szlachetko. Richardiana 7: 61–63. Mytnik-Ejsmont, J. & Szlachetko, D.L. 2007b. Contribution à la révision de Polystachyinae Schlechter (Orchidaceae). 3. Disperanthoceros Mytnik & Szlachetko. Richardiana 7: 64–66. Mytnik-Ejsmont, J. & Szlachetko, D.L. 2008a. Contribution à la révision de Polystachyinae Schlechter (Orchidaceae). 4. Epiphorella Mytnik & Szlachetko. Richardiana 8: 12–17. Mytnik-Ejsmont, J. & Szlachetko, D.L. 2008b. Contribution à la révision de Polystachyinae Schlechter (Orchidaceae). 5. Unguiculabia Mytnik & Szlachetko. Richardiana 8: 18–22. Mytnik-Ejsmont, J. & Szlachetko, D.L. 2008c. Contribution à la révision de Polystachyinae Schlechter (Orchidaceae). 6. Dendrobianthe Mytnik & Szlachetko. Richardiana 8: 23–27. Nathan, R. 2006. Long-distance dispersal of plants. Science 313: 786–788. Nathan, R., Katul, G.G., Horn, H.S., Thomas, S.M., Oren, R., Avissar, R., Pacala, S.W. & Levin, S.A. 2002. Mechanisms of long-distance dispersal of seeds by wind. Nature 418: 409–413. Nichol, J.E. 1999. Geomorphological evidence and Pleistocene refugia in Africa. Geogr. J. 165: 79–89. Nylander, J.A.A. 2004. MrModeltest (version 2). Available from http:// www.abc.se/~nylander/. Nylander, J.A.A., Ronquist, F., Huelsenbeck, J.P. & Nieves-Aldrey, J.L. 2004. Bayesian phylogenetic analysis of combined data. Syst. Biol. 53: 47–67. Oxelman, B., Liden, M. & Berglund, D. 1997. Chloroplast rps16 intron phylogeny of the tribe Sileneae (Caryophyllaceae). Pl. Syst. Evol. 206: 393–410. Pettersson, B. & Nilsson, O. 1993. Floral variation and deceit pollination in Polystachya rosea (Orchidaceae) on an inselberg in Madagascar. Opera Bot. 121: 237–245. Podzorski, A.C. & Cribb, P.J. 1979. A revision of Polystachya sect. Cultriformes (Orchidaceae). Kew Bull. 34: 147–186. Pridgeon, A.M., Cribb, P.J., Chase, M.W. & Rasmussen, F.N. (eds.) 2005. Genera Orchidacearum, vol. 4, Epidendroideae, part 1. Oxford: Oxford Univ. Press. Pridgeon, A.M., Cribb, P.J., Chase, M.W. & Rasmussen, F.N. (eds.) In press. Genera Orchidacearum, vol. 5, Epidendroideae, part 2. Oxford: Oxford Univ. Press. Rambaut, A. & Drummond, A.J. 2007a. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol. Biol. 7: 214. Rambaut, A. & Drummond, A.J. 2007b. Tracer (version 1.4). Available from http://tree.bio.ed.ac.uk/software/Tracer. Ree, R.H. & Smith, S.A. 2008. Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Syst. Biol. 57: 4–14. Reich, D. 2006. Molecular phylogenetics of Polystachya. Diploma thesis, University of Vienna. Rieseberg, L.H. & Soltis, D.E. 1991. Phylogenetic consequences of cytoplasmic gene flow in plants. Evol. Trends Pl. 5: 65–84. TAXON 59 (2) • April 2010: 389–404 Ronquist, F. 1996. DIVA (version 1.1). Available from http://www.ebc .uu.se/systzoo/research/diva/diva.html. Rupp, B. 2008. Genome size and molecular phylogeny of selected Polystachya species (Orchidaceae). Diploma thesis, University of Vienna. Samuel, R., Kathriarachchi, H., Hoffmann, P., Barfuss, M.H.J., Wurdack, K.J., Davis, C.C. & Chase, M.W. 2005. Molecular phylogenetics of Phyllanthaceae: Evidence from plastid matK and nuclear PHYC sequences. Amer. J. Bot. 92: 132–141. Shaw, J., Lickey, E.B., Schilling, E.E. & Small, R.L. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. Amer. J. Bot. 94: 275–288. Soltis, D.E. & Soltis, P.S. 1999. Polyploidy: Recurrent formation and genome evolution. Trends Ecol. Evol. 14: 348–352. Stebbins, G.L. & Dawe, J.C. 1987. Polyploidy and distribution in the European flora: A reappraisal. Bot. Jahrb. Syst. 108: 343–354. Stévart, T. & Nguema, N. 2004. Trois espèces et trois combinaisons nouvelles de Polystachya (Orchidaceae) du Cameroun, de Guinée Équatoriale et du Gabon. Adansonia 26: 217–233. Štorchová, H., Hrdličková, R., Chrtek, J., Tetera, M., Fitze, D. & Fehrer, J. 2000. An improved method of DNA isolation from plants collected in the field and conserved in saturated NaCl/CTAB solution. Taxon 49: 79–84. Summerhayes, V.S. 1942. African orchids: XII. Bot. Mus. Leafl. 10: 257–299. Summerhayes, V.S. 1947. African orchids: XVII. Kew Bull. 2: 123–133. Swofford, D.L. 2003. PAUP*: Phylogenetic analysis using parsimony (and other methods), version 4. Sunderland, Massachusetts: Sinauer. Szlachetko, D.L. & Olszewski, S. 2001. Polystachya. Pp. 493–591 in: Achoundong, G. & Morat, P. (eds.), Flore du Cameroun, vol. 35, Russell & al. • Phylogenetics and cytology of Polystachya Orchidacées II. Yaoundé: Ministere de la Recherche Scientifique et Technique (MINREST). Szlachetko, D.L., Sawicka, M. & Kras-Łapińska, M. 2004. Polystachya. Pp. 244–302 in: Morat, P. (ed.), Flore du Gabon, vol. 37, Orchidaceae II. Paris: Múseum National d’Histoire Naturelle. Tel-Zur, N., Abbo, S., Myslabodski, D. & Mizrahi, Y. 1999. Modified CTAB procedure for DNA isolation from epiphytic cacti of the genera Hylocereus and Selenicereus (Cactaceae). Pl. Molec. Biol. Reporter 17: 249–254. Van den Berg, C., Goldman, D.H., Freudenstein, J.V., Pridgeon, A.M., Cameron, K.M. & Chase, M.W. 2005. An overview of the phylogenetic relationships within Epidendroideae inferred from multiple DNA regions and recircumscription of Epidendreae and Arethuseae (Orchidaceae). Amer. J. Bot. 92: 613–624. Van den Hof, K., van den Berg, R.G. & Gravendeel, B. 2008. Chalcone synthase gene lineage diversification confirms allopolyploid evolutionary relationships of European rostrate violets. Molec. Biol. Evol. 25: 2099–2108. Warburg, O. (ed.). 1903. Kunene-Sambesi-Expedition. Berlin: Verlag des Kolonial-Wirtschaftlichen Komitees. Waterman, R.J. & Bidartondo, M.I. 2008. Deception above, deception below: Linking pollination and mycorrhizal biology of orchids. J. Exp. Bot. 59: 1085–1096. Werle, E., Schneider, C., Renner, M., Völker, M. & Fiehn, W. 1994. Convenient single-step, one tube purification of PCR products for direct sequencing. Nucleic Acids Res. 22: 4354–4355. Wronski, T. & Hausdorf, B. 2008. Distribution patterns of land snails in Ugandan rain forests support the existence of Pleistocene forest refugia. J. Biogeogr. 35: 1759–1768. Xiang, Q.Y. & Thomas, D.T. 2008. Tracking character evolution and biogeographic history through time in Cornaceae: Does choice of methods matter? J. Syst. Evol. 46: 349–374. Appendix. Accession list for DNA sequences. Taxon: number: provenance, Kew DNA [Kew DNA bank accession number], Kew living collection number, HBV (University of Vienna botanical garden) living collection number, voucher specimen details, GenBank accession numbers for rps16-trnK/psbD-trnT sequences. Outgroup: Adrorhizon purpurascens Hook. f.: Sri Lanka, Kew DNA 15745, Chase 15745 (K), GQ145084/GQ144965. Bromheadia finlaysoniana (Lindl.) Miq.: Brunei, Kew DNA 21766, Duangjai 039 (BRUN, K), GQ145085/GQ144966. Bromheadia srilankensis Kruiz & de Vogel: Sri Lanka, Kew DNA 15746, Chase 15746 (K), GQ145086/–. Phalaenopsis aphrodite Rchb. f. subsp. formosana Christenson ‘Taisugar TS-97’: complete plastid genome AY916449. Sirhookera lanceolata Kuntze: Sri Lanka, Kew DNA 15748, Chase 15748 (K), GQ145211/GQ145083. Polystachya adansoniae Rchb. f.: 1: Nigeria, Kew DNA 17957, Bytebier 429/94/469 (EA), GQ145088/GQ144968. 2: Cameroon, A. Russell 92 (YA), GQ145089/–. Polystachya affinis Lindl.: Nigeria, Kew DNA 21165, Chase 21165 (K), GQ145090/GQ144969. Polystachya albescens Ridl. subsp. imbricata (Rolfe) Summerh.: HBV ORCH06240, A Russell ORCH06240 (WU), GQ145091/GQ144970. Polystachya alpina Lindl.: 1: Cameroon, A. Russell 67 (YA), GQ145092/ GQ144971. 2: Cameroon, A. Russell 83 (YA), GQ145093/GQ144972. Polystachya anceps Ridl.: Madagascar, Fischer & Sieder FS4068 (WU), GQ145094/ GQ144973. Polystachya bella Summerh.: Kenya, Kew DNA 17950, Bytebier 783 (EA), GQ145095/GQ144974. Polystachya bennettiana Rchb. f.: 1: Kenya, Kew DNA 17958, Bytebier 338/94/418 (EA), GQ145096/GQ144975. 2: Kew DNA 19186, Mugambi & Odhiambo 81/01 (EA), GQ145097/GQ144976. Polystachya bicalcarata Kraenzl.: Cameroon, A. Russell 81 (YA), GQ145098/GQ144977. Polystachya bicarinata Rendle: Kenya, Kew DNA 17959, Bytebier 621/95/1226 (EA), GQ145099/GQ144978. Polystachya bifida Lindl.: 1: São Tomé, Kew DNA 25885, Kew living collection: 2001-3989, GQ145100/GQ144979. 2: Cameroon, A. Russell 71 (YA), GQ145101/GQ144980. Polystachya caespitifica Kraenz. subsp. latilabris (Summerh.) P.J. Cribb & Podz.: 1: Kenya, Kew DNA 17960, PCP (Plant Conservation Program of the East African National Museum) 187/95/718 (EA), GQ145102/GQ144981. 2: HBV ORCH06423, A. Russell ORCH06423 (WU), GQ145103/GQ144982. Polystachya calluniflora Kraenzl.: Cameroon, A. Russell 63 (YA), GQ145104/GQ144983. Polystachya caloglossa Rchb. f.: 1: Cameroon, A. Russell 41 (YA), GQ145105/–. 2: Cameroon, A. Russell 104 (YA), GQ145106/GQ144984. Polystachya campyloglossa Rolfe: 1: HBV ORCH07312, A. Russell ORCH07312 (WU), GQ145107/GQ144985. 2: HBV ORCH06244, photo voucher–contact author, GQ145108/GQ144986. Polystachya clareae Hermans: Madagascar, Fischer & Sieder s.n., 27 Jan. 2007 (WU), GQ145109/GQ144987. Polystachya concreta (Jacq.) Garay & H.R. Sweet: 1: Madagascar, Kew DNA 17854, Chase 17854 (K), GQ145110/GQ144988. 2: Madagascar, Kew DNA 17859, Chase 17859 (K), GQ145111/GQ144989. 3: Madagascar, Kew DNA 17860, Chase 17860 (K), GQ145112/GQ144990. 4: Madagascar, Fischer & Sieder FS3210 (WU), GQ145113/GQ144991. 5: Brazil, HBV ORCH06604, A. Russell ORCH06604, GQ145114/–. 6: Cameroon, A. Russell 40 (YA), GQ145115/GQ144992. 7: Laos, HBV ORCH07344, photo voucher–contact author, GQ145116/ GQ144993. 8: Mauritius, HBV ORCH07278, GQ145118/GQ144994. 9: Réunion, HBV “Chase & Samuel Reunion 1”, GQ145117/GQ144995. 10: Comoros, HBV ORCH07417, photo voucher–contact author, GQ145119/GQ144996. 11: Seychelles, Kew DNA 25884, Kew living collection 2003-406, A. Russell Kew2003-406 (WU), GQ145120/GQ144997. Polystachya confusa Rolfe: Kenya, Kew DNA 17947, Bytebier & al. 122 (EA), GQ145121/GQ144998. Polystachya coriscensis Rchb. f.: HBV ORCH07314, A. Russell ORCH07314 (WU) GQ145122/GQ144999. Polystachya cornigera Schltr.: Madagascar, Fischer & Sieder FS3208 (WU), GQ145123/–. Polystachya cultriformis (Thouars) Lindl. ex Spreng.: 1: Kew DNA 19182, Mugambi & Odhiambo 054/98/1607 (EA), GQ145124/ GQ145000. 2: Madagascar, Fischer & Sieder FS1045 (WU), GQ145125/GQ145001. Polystachya dendrobiiflora Rchb. f.: 1: Kenya, Kew DNA 17962, PCP (Plant Conservation Program of the East African National Museum) 063/98/1621 (EA), GQ145126/GQ145002. 2: Kew DNA 19184, Mugambi & Odhiambo 064/98/1622 (EA), GQ145127/GQ145003. Polystachya dolichophylla Schltr.: Cameroon, Kew DNA 25886, Chase 25886 (K), GQ145128/GQ145004. Polystachya elegans Rchb. f.: Cameroon, A. Russell 74 (YA), GQ145129/GQ145005. Polystachya eurychila Summerh.: Kenya, Kew DNA 17963, Bytebier 337/94/417 (EA), GQ145130/GQ145006. Polystachya eurygnatha Summerh.: photo voucher–contact author, GQ145131/GQ145007. Polystachya fallax Kraenzl.: Uganda, 403 Russell & al. • Phylogenetics and cytology of Polystachya TAXON 59 (2) • April 2010: 389–404 Appendix. Continued. Kew DNA 17922, Chase 17922 (K), GQ145132/GQ145008. Polystachya fischeri Rchb. f. ex Kraenzl.: 1: Kenya, Kew DNA 17964, Pearce 616/94/607 (EA), GQ145133/GQ145009. 2: Kenya, Kew DNA 17965, Bytebier 674/95/1280 (EA), GQ145134/GQ145010. Polystachya foliosa (Hook.) Rchb. f.: 1: Dominica, Kew DNA 25887, Kew living collection 2001-3986, GQ145135/GQ145011. 2: Venezuela, HBV ORCH07028, GQ145136/–. Polystachya fulvilabia Schltr.: “Congo”, Kew DNA 17855, Chase 17855 (K), GQ145137/GQ145012. Polystachya fusiformis (Thouars) Lindl.: Madagascar, Fischer & Sieder FS621 (WU), GQ145138/ GQ145013. Polystachya galeata (Sw.) Rchb. f. 1: Kew DNA O-1496, Chase O-1496 (K), GQ145139/GQ145014. 2: Sierra Leone, Kew 14650, Chase 14650 (K), GQ145140/GQ145015. Polystachya goetziana Kraenzl.: Kenya, Kew DNA 17955, Bytebier 1772 (EA), GQ145141/GQ145016. Polystachya golungensis Rchb. f. 1: Kenya, Kew DNA 17966, Bytebier 582/95/1156 (EA), GQ145142/GQ145017. 2: HBV ORCH05170, A. Russell ORCH05170 (WU), GQ145143/GQ145018. Polystachya aff. heckmanniana Kraenzl.: Malawi, photo voucher–contact author, GQ145144/GQ145019. Polystachya henrici Schltr.: Madagascar, Kew DNA 17856, Chase 17856 (K), GQ145145/GQ145020. Polystachya holstii Kraenzl.: Kenya, Kew DNA 19260, Bytebier 071/98/1629 (EA), GQ145146/GQ145021. Polystachya humbertii H. Perrier: 1: Madagascar, Fischer & Sieder FS2079 (WU), GQ145147/GQ145022. 2: Madagascar, Fischer & Sieder FS3017 (WU), GQ145148/GQ145023. Polystachya johnstonii Rolfe: HBV ORCH06241, photo voucher–contact author, GQ145149/GQ145024. Polystachya kermisina Kraenzl.: Rwanda, HBV ORCH07240, photo voucher–contact author, GQ145150/GQ145025. Polystachya laurentii De Wild.: photo voucher–contact author, GQ145151/GQ145026. Polystachya lawrenceana Kraenzl.: Malawi, HBV ORCH06412, photo voucher–contact author, GQ145152/GQ145027. Polystachya laxiflora Lindl.: HBV ORCH07315, A. Russell ORCH07315 (WU), GQ145153/GQ145028. Polystachya lindblomii Schltr.: Kenya, Kew DNA 17967, Bytebier 1142/98/1695 (EA), GQ145154/GQ145029. Polystachya longiscapa Summerh.: Tanzania, HBV ORCH06411, GQ145155/GQ145030. Polystachya maculata P.J. Cribb: Burundi, HBV ORCH07263, photo voucher–contact author, GQ145156/GQ145031. Polystachya melanantha Schltr.: Kenya, Kew DNA 17954, Bytebier 1783 (EA), GQ145157/GQ145032. Polystachya melliodora P.J. Cribb: Tanzania, Kew DNA 17923, Chase 17923 (K), GQ145158/GQ145033. Polystachya modesta Rchb. f.: HBV ORCH05165, GQ145159/GQ145034. Polystachya monophylla Schltr.: Madagascar, Fischer & Sieder FS3042 (WU), GQ145160/ GQ145035. Polystachya neobenthamia Schltr.: HBV ORCH07214, photo voucher–contact author, GQ145087/GQ144967. Polystachya cf. nyanzensis Rendle: 1: Cameroon, A. Russell 99 (YA), GQ145161/GQ145036. 2: Cameroon, A. Russell 100 (YA), GQ145162/GQ145037. 3: HBV ORCH06425, photo voucher–contact author, GQ145163/GQ145038. Polystachya odorata Lindl.: 1: Nigeria, Kew DNA 17857, Chase 17857 (K), GQ145164/GQ145039. 2: Cameroon, A. Russell 42 (YA), GQ145165/GQ145040. Polystachya oreocharis Schltr.: 1: Madagascar, Fischer & Sieder FS2082 (WU), GQ145166/GQ145041. 2: Madagascar, Fischer & Sieder FS3152 (WU), GQ145167/–. Polystachya ottoniana Rchb. f.: Kew DNA 25888, Kew living collection 2005-964, GQ145168/GQ145042. Polystachya pachychila Summerh.: HBV ORCH07310, A. Russell ORCH07310, GQ145169/GQ145043. Polystachya paniculata (Sw.) Rolfe: 1: Ethiopia, Kew DNA 25889, Kew living collection 1984-4977, GQ145170/GQ145044. 2: Cameroon, L. Pearce 27 (YA), GQ145171/–. Polystachya piersii P.J. Cribb: 1: Kenya, Kew DNA 17948, Bytebier & al. 101/95/1186 (EA), GQ145172/GQ145045. 2: Kew DNA 19185, Bytebier & al. 136/95/1207 (EA), GQ145173/GQ145046. Polystachya pinicola Barb.Rodr.: Brazil, HBV ORCH06606, GQ145174/GQ145047. Polystachya poikilantha Kraenzl.: 1: Kenya, Kew DNA 17953, Bytebier 956/97 (EA), GQ145175/GQ145048. 2: Kenya, Kew DNA 19261, Bytebier 956/97/524 (EA), GQ145176/GQ145049. Polystachya poikilantha Kraenzl. var. leucorhoda (Kraenzl.) P.J. Cribb & Podz.: HBV ORCH06272, photo voucher–contact author, GQ145177/GQ145050. Polystachya polychaete Kraenzl.: “Congo”, Kew DNA 25890: Kew living collection 2001-3987, GQ145178/GQ145051. Polystachya pubescens (Lindl.) Rchb. f.: 1: Kew DNA O-700, Kurzweil 1849, GQ145179/GQ145052. 2: HBV ORCH05171, A. Russell ORCH05171 (WU) GQ145180/GQ145053. Polystachya purpureobracteata P.J. Cribb & la Croix: Malawi, HBV ORCH07384, photo voucher–contact author, GQ145181/GQ145054. Polystachya ramulosa Lindl.: Cameroon, A. Russell 55 (YA), GQ145182/ GQ145055. Polystachya cf. rhodoptera Rchb. f.: São Tomé, Kew DNA 25891, Kew living collection 2001-3973, GQ145183/GQ145056. Polystachya cf. rosea Ridl.: 1: Madagascar, Fischer & Sieder FS3025 (WU), GQ145184/GQ145057. 2: Madagascar, Fischer & Sieder FS796 (WU) GQ145185/GQ145058. Polystachya seticaulis Rendle: “Congo”, Kew DNA 17924, Chase 17924 (K), GQ145186/GQ145059. Polystachya setifera Lindl.: Kew DNA O-1493, Chase O-1493 (K), GQ145187/GQ145060. Polystachya spatella Kraenzl.: 1: Kenya, Kew DNA 17951, Bytebier 949 (EA), GQ145188/GQ145061. 2: Kenya, Kew DNA 19263, Khayota 381 (EA), GQ145189/GQ145062. Polystachya steudneri Rchb. f.: Kenya, Kew DNA bank 17956, Bytebier 712/95/1305 (EA), GQ145190/GQ145063. Polystachya supfiana Schltr.: Cameroon, A. Russell 51 (YA), GQ145191/GQ145064. Polystachya tenella Summerh.: 1: Kenya, Kew DNA 17952, Bytebier 955/97/1524 (EA), GQ145193/GQ145066. 2: Kenya, Kew DNA 19262, Bytebier 955/97/1523 (EA), GQ145194/GQ145067. Polystachya tenuissima Kraenzl.: Kenya, Kew DNA 17968, Bytebier 428/94/468 (EA), GQ145195/GQ145068. Polystachya thomensis Summerh.: Sao Tome, Kew DNA 17858, Chase 17858 (K), GQ145196/GQ145069. Polystachya transvaalensis Schltr.: 1: Kenya, Kew DNA 17969, Bytebier 951/97/1519 (EA), GQ145197/GQ145070. 2: Kenya, Kew DNA 19264, Bytebier & Luke 1774 (EA), GQ145198/–. Polystachya tsaratananae H. Perrier: 1: Madagascar, Kew DNA 17861, Chase 17861 (K), GQ145199/ GQ145071. 2: Madagascar, Fischer & Sieder FS642 (WU), GQ145200/GQ145072. Polystachya tsinjoarivensis H. Perrier: 1: Madagascar, Fischer & Sieder FS3209 (WU), GQ145201/GQ145073. 2: Madagascar, HBV FS4182, photo voucher–contact author, GQ145202/GQ145074. Polystachya undulata P.J. Cribb & Podz: Kew DNA 17862, Chase 17862 (K), GQ145203/GQ145075. Polystachya vaginata Summerh.: 1: Kenya, Kew DNA 17949, Bytebier 566/95/1140 (EA), GQ145204/GQ145076. 2: Kenya, Kew DNA 19265, Bytebier 452/97/1587 (EA), GQ145205/GQ145077. Polystachya villosa Rolfe: HBV ORCH07216, A. Russell ORCH07216 (WU), GQ145192/GQ145065. Polystachya virescens Ridl.: Madagascar, Fischer & Sieder FS1002 (WU), GQ145206/GQ145078. Polystachya virginea Summerh.: Tanzania, HBV ORCH06422, GQ145207/GQ145079. Polystachya vulcanica Kraenzl. var. aconitiflora (Summerh.) P.J. Cribb & Podz.: Kew DNA 17863, Chase 17863 (K), GQ145208/GQ145080. Polystachya vulcanica Kraenzl. var. vulcanica: Kenya, Kew DNA 19266, Bytebier 954/97/1522 (EA), GQ145209/GQ145081. Polystachya zambesiaca Rolfe: Malawi, photo voucher–contact author, GQ145210/GQ145082. Sirhookera lanceolata Kuntze: Sri Lanka, Kew DNA 15748, Chase 15748 (K), GQ145211/GQ145083. 404

RELATED PAPERS

RELATED TOPICS

Log In

Phylogenetics and cytology of a pantropical orchid genus Polystachya (Polystachyinae, Vandeae, Orchidaceae): Evidence from plastid DNA sequence data

Phylogenetics and cytology of a pantropical orchid genus Polystachya (Polystachyinae, Vandeae, Orchidaceae): Evidence from plastid DNA sequence data

Phylogenetics and cytology of a pantropical orchid genus Polystachya (Polystachyinae, Vandeae, Orchidaceae): Evidence from plastid DNA sequence data

Phylogenetics and cytology of a pantropical orchid genus Polystachya (Polystachyinae, Vandeae, Orchidaceae): Evidence from plastid DNA sequence data

Phylogenetics and cytology of a pantropical orchid genus Polystachya (Polystachyinae, Vandeae, Orchidaceae): Evidence from plastid DNA sequence data

Related Papers

RELATED PAPERS

RELATED TOPICS