Molecular Phylogenetics and Evolution 77 (2014) 41–53 Contents lists available at ScienceDirect Molecular Phylogenetics and Evolution journal homepage: www.elsevier.com/locate/ympev Molecular systematics of subtribe Orchidinae and Asian taxa of Habenariinae (Orchideae, Orchidaceae) based on plastid matK, rbcL and nuclear ITS Wei-Tao Jin a, Xiao-Hua Jin a,⇑, André Schuiteman b, De-Zhu Li c, Xiao-Guo Xiang a, Wei-Chang Huang d, Jian-Wu Li e, Lu-Qi Huang f,⇑ a State Key Laboratory of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Beijing 10093, China Herbarium, Library, Art and Archives Directorate, Royal Botanical Gardens, Kew, Richmond, Surrey TW9 3AB, UK c Key Laboratory of Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan 650204, China d Shanghai Chenshan Botanical Garden, Chenhua Road 3888, Songjiang, Shanghai 201602, China e Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Menglun Township, Mengla County, Yunnan 666303, China f National Resource Centre for Chinese Material Medica, China Academy of Chinese Medical Science, Beijing 100700, China b a r t i c l e i n f o Article history: Received 24 October 2013 Revised 30 March 2014 Accepted 3 April 2014 Available online 16 April 2014 Keywords: Asia Generic delimitation Habenariinae Molecular phylogenetics Orchidinae a b s t r a c t The subtribe Orchidinae, distributed predominantly in Eastern Asia and the Mediterranean, presents some of the most intricate taxonomic problems in the family Orchidaceae with respect to generic delimitation. Based on three DNA markers (plastid matK, rbcL, and nuclear ITS), morphological characters, and a broad sampling of Orchidinae and selected Habenariinae mainly from Asia (a total of 153 accessions of 145 species in 31 genera), generic delimitation and phylogenetic relationships within the subtribe Orchidinae and Habenariinae from Asia were assessed. Orchidinae and Asian Habenariinae are monophyletic, and Orchidinae is divided into distinct superclades. Many genera, such as Amitostigma, Habenaria, Hemipilia, Herminium, Platanthera, Peristylus and Ponerorchis, are not monophyletic. Habenaria is subdivided into two distantly related groups, while Platanthera is subdivided into three even more disparate groups. Many previously undetected phylogenetic relationships, such as clades formed by the Amitostigma–Neottianthe–Ponerorchis complex, Platanthera latilabris group, Ponerorchis chrysea, Sirindhornia, and Tsaiorchis, are well supported by both molecular and morphological evidence. We propose to combine Hemipiliopsis with Hemipilia, Amitostigma and Neottianthe with Ponerorchis, Smithorchis with Platanthera, and Aceratorchis and Neolindleya with Galearis, and to establish a new genus to accommodate Ponerorchis chrysea. Tsaiorchis and Sirindhornia are two distinctive genera supported by both molecular data and morphological characters. A new genus, Hsenhsua, and 41 new combinations are proposed based on these findings. Ó 2014 Elsevier Inc. All rights reserved. 1. Introduction The subtribe Orchidinae consist of about 35 genera and 350–400 species, distributed mainly in Eastern Asia, the Mediterranean Region with a few species extending in Northern America, Southern America and Africa (Chen et al., 2009; Dressler, 1993; Kraenzlin, 1901; Lang, 1999; Pearce and Cribb, 2002; Pridgeon et al., 2001). Orchidinae is similar to Habenariinae in habitat preferences and many vegetative and floral characters, but these ⇑ Corresponding authors. E-mail addresses: jinweitao@ibcas.ac.cn (W.-T. Jin), orchid@ibcas.ac.cn, xiaohuajin@ibcas.ac.cn (X.-H. Jin), A.Schuiteman@kew.org (A. Schuiteman), dzl@mail.kib.ac. cn (D.-Z. Li), xiangxg2010@ibcas.ac.cn (X.-G. Xiang), hwc_zx@126.com (W.-C. Huang), ljw@xtbg.org.cn (J.-W. Li), huangluqi01@126.com (L.-Q. Huang). http://dx.doi.org/10.1016/j.ympev.2014.04.004 1055-7903/Ó 2014 Elsevier Inc. All rights reserved. two subtribes can be distinguished on the basis of the structures of stigma: Orchidinae usually has a concave and sessile stigma often with confluent lobes, whereas Habenariinae usually has stalked, convex and distinct stigma lobes (Dressler, 1993; Pridgeon et al., 2001). This morphological distinction has been supported by molecular evidence (Douzery et al., 1999). However, Inda et al. (2010, 2012) indicated that Habenariinae s.l. is paraphyletic, and several genera from Africa, such as Stenoglottis, Cynorkis, and Holothrix, were resolved as successive sister to Orchidinae + Habenaria and its alliance. Batista et al. (2013) showed that the Habenariinae clade (formed by Habenaria s.l. + Cynorkis + Stenoglottis) is sister to the Orchidinae clade formed by (Orchis + Platanthera). In practice, it can be difficult to distinguish between concave sessile stigma lobes and stalked convex ones in some alpine taxa, such as Androcorys, Herminium, Peristylus, Ponerorchis, and Smithorchis 42 W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 (Dressler, 1993; Lang, 1999; Pridgeon et al., 2001), while some true Platanthera species (Orchidinae) clearly have stalked stigma lobes. Kurzweil and Weber (1992), Pridgeon et al. (2001) and Chase et al. (2003) even suggested abandoning the recognition of Orchidinae and Habenariinae as distinct clades. Orchidinae is a medium-sized subtribe in Orchidaceae and one of many well-studied groups in Orchidaceae, however, the generic delimitation and classification within Orchidinae are still problematic and is complicated by the morphological diversity, wide distribution range, homoplasy of characters considered diagnostic at generic level, and the intergrading and overlapping morphological variation between genera (Aceto et al., 1999; Bateman et al., 2003, 2009; Box et al., 2008; Dressler, 1981, 1993; Hapeman and Inoue, 1997; Jin and Efimov, 2012; Luer, 1975; Pridgeon et al., 2001; Soliva et al., 2001; Tyteca and Klein, 2008). Based on morphological characters and/or analyses of molecular data, various generic delimitations and taxonomies have been proposed, especially on some systematically difficult genera, such as Dactylorhiza, Orchis s.l., Platanthera, Ponerorchis, and Tsaiorchis (Aceto et al., 1999; Bateman et al., 2003, 2009; Hapeman and Inoue, 1997; Hooker, 1890; King and Pantling, 1896, 1898; Lang, 1998, 1999; Soliva et al., 2001; Luer, 1975; Pridgeon et al., 2001; Tyteca and Klein, 2008). As previous molecular systematics of Orchidinae were largely based on sampling from the Mediterranean area (Bateman et al., 2003, 2009; Douzery et al., 1999; Soliva et al., 2001; Inda et al., 2012), and/or mainly utilized a single DNA marker (ITS) (Bateman et al., 2003, 2009; Douzery et al., 1999; Soliva et al., 2001), some conclusions and results were weakly supported or even without statistical support, and many taxonomic problems remain unresolved. Despite the high diversity of Orchidinae and Habenariinae in Eastern Asia, many taxa, especially those from monotypic/oligotypic genera, were not represented in previous molecular studies. Generic delimitation and systematic position of many genera in Orchidinae and Habenariinae from Asia, such as Aceratorchis, Amitostigma, Hemipilia, Hemipiliopsis, Neolindleya, Neottianthe, Platanthera, Ponerorchis, Smithorchis, and Tsaiorchis, are not or little known, and as a result many taxonomical suggestions and proposals remain to be tested (Bateman et al., 2003, 2009; Chen et al., 2009; Pridgeon et al., 2001). For a better understanding of the generic delimitation within Orchidinae and of the interrelationships among major clades within Orchidinae, it is desirable to base the analyses on multiple DNA markers and a denser sampling across many systematically difficult genera and their allies from Asia. In the present study, phylogenetic relationships were inferred using three DNA markers (plastid matK, rbcL and nuclear ITS sequence), with 153 samples representing 146 species of Orchidinae and selected Habenariinae, such as Androcorys, Habenaria, Herminium, and Peristylus, mainly from Asia, with the aims of (1) increasing our understanding of the generic delimitation within Orchidinae and Habenariinae; (2) reconstructing the phylogenetic interrelationships within Orchidinae. 2. Materials and methods including Bonatea, Gennaria, Habenaria, Ophrys, Pseudorchis, and Serapias, were used to broaden the sampling of Orchideae in our analyses. Since previous results indicated that Orchideae is closely related to tribe Diseae (Bytebier et al., 2007; Douzery et al., 1999; Pridgeon et al., 2001), two species of Disa, Disa tripetaloides and Disa uniflora, were used as outgroups. The voucher information and the GenBank accession numbers used in this study are listed in Table A.1. 2.2. DNA extraction, amplification and sequencing Total genomic DNA was isolated from silica-gel-dried materials using a Plant Genomic DNA Kit (Beijing Biomed Co., LTD, Beijing, China). For this study, two plastid markers (the coding gene matK, rbcL) and the nuclear ribosomal DNA internal transcribed spacers (ITS) were used. The PCR and sequencing primers for matK, rbcL, and ITS are listed in Table A.2. The selected DNA regions were amplified by using a standard polymerase chain reaction (PCR). The sequencing reactions were performed by using the ABI Prism Bigdye Terminator Cycle Sequencing Kit (Applied Biosystems, ABI). 2.3. Phylogenetic analyses Sequences were aligned using the program Clustal X 1.83 (Thompson et al., 1997) and manually adjusted using BioEdit (Hall, 1999). The homogeneity between the ITS data and the combined plastid dataset (matK, rbcL) was tested using the incongruence length difference (ILD) (Farris et al., 1995), implemented in PAUP v4.0b10 (Swofford, 2002). Following Cunningham (1997), no cases of strongly supported incongruence were detected (P = 0.17), therefore, we combined ITS data and the plastid dataset (matK, rbcL) in SequenceMatrix v1.7.8 (Vaidya et al., 2011) to perform further phylogenetic analyses. The phylogenetic analyses for each matrix were performed using the maximum parsimony (MP) in PAUP v4.0b10 (Swofford, 2002) and Bayesian inference (BI) in MrBayes v3.1.2 (Ronquist and Huelsenbeck, 2003) on CIPRES Science Gateway Web server (Old MrBayes on XSEDE 3.1.2) (Miller et al., 2010). For the MP analyses, heuristic searches were conducted with 1000 replicates of random addition, one tree held at each step during the stepwise addition, tree-bisection–reconnection (TBR) branch swapping, MulTrees in effect, and steepest descent off. All of the characters were unordered and equally weighted, and the gaps were coded as missing data. To evaluate the node support, bootstrap analyses (Felsenstein, 1988) were performed using 1000 replicates, with 10 random taxon additions and heuristic search options. Prior to the Bayesian analysis, a model for sequence evolution for each matrix was determined by using ModelTest v3.7 (Posada and Crandall, 1998) under the Akaike information criterion. For the BI analyses, two separate four Markov chain Monte Carlo (MCMC) analyses were run, with 10,000,000 generations and sampling every 1000 generation. Majority rule (>50%) consensus trees were constructed after removing the ‘‘burn-in period’’ samples (the first 25% of the sampled trees). 2.1. Taxon sampling There are about 32 genera of Orchideae distributed in Asia, out of which about ten genera are endemic (Chen et al., 2009; Dressler, 1993; Jin et al., 2012; Pedersen et al., 2002; Pridgeon et al., 2001). In order to represent the taxonomic diversity of Orchideae in Asia, 153 accessions of 145 species in 31 genera, including 27 genera and 103 species from Asia (about 84% of Asian genera and 25% of Asian species), were included in this study. Additionally, several genera and some species from Africa, Europe, and South America, 3. Results 3.1. Sequences and alignment In this study, 215 new sequences were obtained. Sequence lengths were as follows: 825 bp for ITS region, 1254 bp for rbcL, 1870 bp for matK. The combined alignment of ITS and plastid regions comprised 3949 bp, 24% of which were 43 W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 Table 1 The statistics from the analyses of the chloroplast and nuclear data sets from the parsimony analysis. Information ITS matK rbcL Combined No. taxa Aligned length No. variable characters No. parsimony-informative characters Tree length (steps) Consistency index (CI) Retention index (RI) Model 144 825 507(61%) 445(54%) 3052 0.33 0.77 GTR + I + G 136 1870 642(34%) 419(22%) 1428 0.58 0.82 GTR + I + G 119 1254 181(14%) 103(8%) 334 0.62 0.83 GTR + I + G 153 3949 1330(34%) 967(24%) 4868 0.42 0.78 GTR + I + G parsimony-informative. Table 1 summarizes the properties of each aligned data partition. 3.2. Phylogenetic analyses The partition homogeneity test for plastid DNA + ITS shows there were no strongly supported incongruent results in the datasets (P = 0.17), therefore, we combined the datasets for simultaneous analyses. Phylogenetic relationships based on the ITS data had a better resolution than the two combined plastid DNA data (results not shown here). Based on the combined ITS and plastid DNA data, our findings are consistent in the overall topology of the trees produced with maximum parsimony (MP) and Bayesian inference (BI) methods, except for a few of the collapsed nodes. Bootstrap values (BS) were often lower than the Posterior Probability (PP) from the Bayesian analysis. The BI topology from the combined dataset is chosen as the primary tree for discussion of phylogenetic relationships (Figs. 1 and 2; the MP strict consensus tree is not shown). Our results indicate that the tribe Orchideae can be divided into two well supported sister clades, Habenariinae (PP = 100, BS = 67) (Fig. 1) and Orchidinae (PP = 100, BS = 97) (Fig. 2). Within the Orchidinae, six of nine major clades (Clade I to IX) are well supported along the backbone of the Orchidinae, and many interrelationships among them are well-resolved (Fig. 1). Within the Habenariinae, nine major clades, Clade X to XVIII, can be recognized (Fig. 2). Our results show that many morphological delimited genera of the Asian Orchideae in Pridgeon et al. (2001) and Chen et al. (2009) are not monophyletic, such as Hemipilia, Herminium, Platanthera, and Ponerorchis. Clade I (Fig. 1) comprises 42 species (PP = 89, BS < 50), most of which belong to Platanthera s.l., the remainders are from genera Herminium, Peristylus, and Smithorchis. The interrelationships within Clade I are not well resolved. Clade II comprises two species of Galearis, two monotypic genera, Aceratorchis and Neolindleya, and one species of Aorchis with weak support (PP = 86) (Fig. 1). Clade III includes one species of Pseudorchis (Fig. 1). Clade IV consists of four species from Dactylorhiza and Gymnadenia with strong support (PP = 100, BS = 85) (Fig. 1). Clade V includes three species of Orchis s.s. with robust support (PP = 100, BS = 99) (Fig. 1). Clade VI (PP = 100, BS = 93) consists of two species of Ophrys and two species of Serapias with strong support (PP = 100, BS = 99) (Fig. 1). Clade VII (Fig. 1) includes about 15 species from Amitostigma, Neottianthe, and Ponerorchis, and is subdivided into three well supported subclades: one subclade contains five species from Amitostigma and Ponerorchis; another subclade includes four species from Amitostigma and Neottianthe, including types of these two genera; and the third subclade includes four to six species of Amitostigma. Clade VIII (Fig. 1) comprises nine species, six from Hemipilia, two from two monotypic genera, Hemipiliopsis and Tsaiorchis, and one from Ponerorchis, with weak support (PP = 81, BS = 56). Clade VIII can be divided into two subclades, one includes Tsaiorchis, and the other includes the remainder with strong support (PP = 100, BS = 97). Clade IX includes a species from the oligotypic genus Sirindhornia (Fig. 1). Clade X includes four species, two of Androcorys and two of Porolabium with strong support (PP = 100, BS = 100) (Fig. 2). Clade XI consists of three species from Platanthera, i.e. Platanthera latilabris group, with strong support (PP = 100, BS = 100) (Fig. 2). Clade XII includes one species from Herminium, H. lanceum. Clade XIII (PP = 100, BS = 100) includes six species, two from Peristylus, and four from Herminium (Fig. 2). An unidentified species of Herminium is sister to clades X to XIII, with weak support for the sister group (Fig. 2). Clade XIV (PP = 100, BS = 100) includes two terminals from one species, Ponerorchis chrysea (Fig. 2). Clade XV includes nine species, eight from Asian Habenaria and one from Pecteilis, with strong support (PP = 100, BS = 99) (Fig. 2). Clade XVI includes seven species, six from Peristylus, and one from Platanthera, with substantial support (PP = 100, BS = 84) (Fig. 2). Clade XVII includes 24 species of Habenaria from Africa, Asia and Southern America, and two species of Botanea from Africa with some support (PP = 100, BS = 59) (Fig. 2). Clade XVIII includes two monotypic genera from Europe and Asia, Gennaria and Nujiangia, with strong support (PP = 100, BS = 100) (Fig. 2). 4. Discussion 4.1. An overview of phylogenetics of Orchideae With broader sampling and more DNA markers, our results confirmed the earlier findings that Orchidinae (PP = 100, BS = 97) (Fig. 1) and Asian Habenariinae (PP = 100, BS = 67) (Fig. 2) are two well resolved sister groups. However, our results indicated that the morphological distinctions, such as the types of stigma lobes, between these two subtribes are problematic (see discussion of each clade). Inda et al. (2012) indicated that Habenariinae is paraphyletic, and several genera from Africa, such as Stenoglottis, Satyrium, Cynorkis, and Holothrix, were resolved as successive sister to Orchidinae + Asian Habenariinae. Batista et al. (2013) showed that the Habenariinae clade (formed by Habenaria s.l. + Cynorkis + Stenoglottis) is sister to the Orchidinae clade formed by (Orchis + Platanthera). Given the complex taxonomy of Orchideae, it would be premature to make any firm conclusion about these two subtribes. More importantly, our results showed that many previous findings of interrelationships within Orchideae based on molecular phylogenetics were strongly supported, and many unknown or overlooked phylogenetic relationships have been detected. Asian Orchidinae is subdivided into two superclades (Fig. 1). Superclade A includes Dactylorhiza, Galearis, Gymnadenia, Orchis s.s., and Platanthera s.l. (PP = 87, BS = 58). Superclade B includes Amitostigma, Hemipilia, Neottianthe and Sirindhornia (PP = 100). Clade VI, formed by Ophrys and Serapias, is suggested as a sister to Superclade A with strong support, and Sirindhornia is resolved as sister to the rest of Superclade B with weak support (PP = 100) (Fig. 2). 44 W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 Fig. 1. Phylogenetic tree obtained by Bayesian analysis of the combination of ITS and plastid regions, showing the detailed relationships of subtribe Orchidinae. Numbers above the branches indicate posterior probabilities (PP) and bootstrap percentages (BS). ‘‘–’’ indicates node is not supported in the analysis. ‘‘’’ indicates node is with support value 100%. W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 45 Fig. 2. Phylogenetic tree obtained by Bayesian analysis of the combination of ITS and plastid regions, showing the detailed relationships of subtribe Habenariinae. Numbers above the branches indicate posterior probabilities (PP) and bootstrap percentages (BS). ‘‘–’’ indicates node is not supported in the analysis. ‘‘’’ indicates node is with support value 100%. Asian Habenariinae is subdivided into three moderately to strongly supported superclades (Fig. 2). Superclade C (PP = 98), consisting of Androcorys, Porolabium, Herminium, Ponerorchis chrysea, Platanthera latilabris group, and some species of Habenaria, occur mostly in montane to alpine regions in Asia. Superclade D (PP = 100, BS = 84), comprising Peristylus, occur mainly in tropical and subtropical Asia with a few species extending into alpine regions. Superclade E (PP = 100), including Habenaria from Asia, Africa and the Neotropics, African Bonatea, and two monotypic genera from Europe and Asia, Gennaria and Nujiangia, mainly occurs in tropical regions with a few species extending into subtropical mountains. The clade formed by Gennaria and Nujiangia is resolved as sister to the remainder in Superclade E. As Orchidinae is well-represented in Europe, the generic delimitation and classification of many genera around the Mediterranean have been thoroughly studied, for example, Orchis and its alliance by Aceto et al. (1999), Ophrys by Soliva et al. (2001), Serapias by Bellusci et al. (2008), Dactylorhiza by Devos et al. (2006) and 46 W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 Box et al. (2008). Recently, molecular phylogenetics of Habenaria from Neotropics has been investigated based on comprehensive sampling from Neotropics (Batista et al., 2013). Hence, in our analyses, sampling of genera or species from the Africa, Europe, and Neotropics is mainly used as placeholders for these taxa. Their phylogeny will not be discussed here because our focus is on phylogenetic analyses on Asian taxa. 4.2. Phylogeny and generic delimitation 4.2.1. Platanthera (Clade I, and XVI) Platanthera consists of about 100–150 species and is among the larger genera in Orchideae, however, generic delimitation of Platanthera is unclear. Luer (1975) suggested that several previous recognized genera, such as Blephariglottis, Lacera, Limnorchis, and Tulotis, should be treated as sections within Platanthera, and Dressler (1993) suggested that Platanthera is close to Dactylorhiza–Galearis alliance instead of Habenaria. These suggestions have been supported by recent results of molecular studies (Bateman et al., 2003, 2009; Hapeman and Inoue, 1997). Bateman et al. (2009) broadened Platanthera to include more genera, such as Diphylax, Piperia, and Tsaiorchis. Based on seed micromorphology, however, Gamarra et al. (2008) stated that Limnorchis should be separated from Platanthera as a distinct genus. Our analyses do not only support most of these conclusions, but also indicate that Platanthera s.l. is polyphyletic and subdivided into three distant related groups that belong to two subtribes, Platanthera clade (Clade I, Orchidinae), Platanthera latilabris group (Clade XI, Habenariinae), and Platanthera biermanniana (included in Clade XVI, Habenariinae). The Platanthera clade (Clade I, Orchidinae) is not strongly supported and the interrelationships within the genus are not well resolved. However, the Platanthera clade is well characterized by morphological characters, such as fleshy rootstock (no tuber), leaves basal to cauline and grading into foliaceous bracts, sheathing at base, lip spurred, column short and truncate, anther broad, loculi more or less separated by connective, stigmas two and more or less divergent at the base of the entrance of spur. Platanthera latilabris group, consisting of three closely related species, P. clavigera, P. edgeworthii, P. latilabris (Fig. 3a), form an unambiguous clade deeply nested within Habenariinae. Species of this clade are characterized by ovoid to globose tubers, densely flowered inflorescence, linear and fleshy lip with a callus at base, stigma lobes more or less stalked, extending at base of lip (Fig. 3a). Hooker (1890) treated these species as members of Habenaria, Szlachetko and Kras (2006) transferred these species into the Neotropical genus, Habenella. Our results indicated that Platanthera latilabris group and Habenella (included in Clade XVII) are two quite distantly related groups. Instead, Platanthera latilabris group are resolved as sister to Androcorys plus Porolabium with weak support (PP = 92, BS = 56). Platanthera biermanniana resembles species of Peristylus in gross morphology. Our analyses show that P. biermanniana is deeply nested within Peristylus with robust support, which is congruent with morphological characters, such as cylindric-ovoid tuber, 3lobed lip, short spur, anther with very narrow connective. Several species of Herminium (H. carnosilabre and H. angustilabre), Peristylus (P. nematocaulon), and the monotypic Smithorchis (S. calceoliformis) are deeply nested within Platanthera s.l. These species are restricted in alpine region with elevation ranging from 3500 to 4300 m. The taxonomy has been complicated by these species having minute flower (the smallest in Orchidinae) and a greatly reduced column which is occupied almost entirely by the anther. Most genera of Asian Orchideae were distinguished on the basis of floral characters, especially of the column (Lang, 1999; Pridgeon et al., 2001). Our observations and morphological comparison established that these species have oblong to fusiform rootstock (which are near horizontal in S. calceoliformis), entire and spurred lip, naked viscidium, and a relative obvious anther connective, which supports the transfer of these species into Platanthera. Dozens of Platanthera species from the Himalayas, such as P. bakeriana, P. leptocaulon, P. roseotincta, and others, have two projecting stigmatic lobes (stalked stigma lobes) and an entire lip (Fig. 3b). Duthie (1906), Hooker (1890), King and Pantling (1896, 1898), and Tuyama (1966, 1971, 1975) treated these species as Habenaria, while Kraenzlin (1901), Lindley (1835), Lang (1998, 1999), and Soó (1929) placed them in Platanthera. Lang (1998) even proposed a subgenus, subgen. Stigmatosae, to contain these species. Our results indicate that stalked stigma lobes have evolved at least twice in Platanthera, one in P. yadongensis (Jin et al., 2013), the other in P. exelliana and its relative species. This renders the morphological distinction between Habenariinae and Orchidinae, which is mainly based on the morphology of the stigma, problematic. 4.2.2. Aceratorchis, Aorchis, Galearis, and Neolindleya (Clade II) Both Aceratorchis (Fig. 3c) and Neolindleya are little-known monotypic genera. Bateman et al. (2009) considered Aceratorchis to be a peloric form of Galearis. Neolindleya was previously considered as Gymnadenia (Bateman et al., 2003), then it was separated as a distinctive genus and sister to Galearis (Bateman et al., 2009; Efimov et al., 2009). Pridgeon et al. (2001) considered Aorchis as congeneric with Galearis. In our phylogenetic analyses, Aceratorchis, Aorchis, Galearis and Neolindleya form a weak supported clade (PP = 86) (Fig. 1), while Aceratorchis and Neolindleya were resolved as successive sister to Galearis and Aorchis. These four genera typically share a stolon-like rhizome, anther loculi parallel with distinct rostellum between them, concave stigma, and viscidium enclosed in bursicle. Considering the weak support of the clade, we tentatively proposed to subsume all other three genera into Galearis. However, this needs to be tested by further studies. 4.2.3. Amitostigma, Neottianthe, and Ponerorchis (Clade VII) The generic borders between Ponerorchis and Amitostigma are unclear due to the difficulty of distinguishing the supposed diagnostic characters, such as appendages of column and the bursicle, and several species have been transferred back and forth between these genera. However, it is nevertheless somewhat unexpected that Amitostigma, Neottianthe and Ponerorchis are nested together and form a moderately supported group (Fig. 1). This clade mainly occurs in Eastern Asia, with one or two species extending into the northern temperate zone. Members of this clade share an ovoid tuber, slender plants with 1–3 slightly fleshy leaves, flowers secund along rachis, dorsal sepal and petals forming a hood, the base of lip more or less connecting the column, column short, lateral appendages obvious, parallel anther cells, stigma lobes situated and confluent under rostellum. The subclade formed by three species of Neottianthe (including the type, Neottianthe cucullata) and Amitostigma gracile (type of Amitostigma) is strongly supported and well characterized by morphological characters, such as leaves basal, elliptic to ovate, petals and three sepals forming a hood, lip papillose. Ponerorchis is said to differ from Amitostigma and Neottianthe by having a viscidium enveloped in a bursicle (see Pridgeon et al., 2001). Our results indicate that a bursicle has independently evolved four times in Orchideae (Clade II, VII, VIII, XIV). Moreover, our fieldwork established that the presence of a bursicle is difficult to determine even in living specimens, and some species of Amitostigma, such as A. monanthum (Fig. 3g), A. yuanum (Fig. 3h), do have bursicles. It is almost impossible to assign some species, such as Amitostigma yuanum (Fig. 3h), Ponerorchis nana (Fig. 3i), and A. farreri, to Amitostigma or Ponerorchis, W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 47 Fig. 3. Diversity of Orchideae from Asia. (a) flowers of Platanthera latilabris, front view; (b) flowers of Platanthera roseotincta, front view; (c) Plants of Aceratorchis tschiliensis, showing the stolon-like rhizome; (d) plant of Hemipilia purpureopunctata; (e) flowers of Hemipilia purpureopunctata; (f) plants of Tsaiorchis neottianthoides; (g) flower of Ponerorchis monantha; (h) flowers of Amitostigma yuana; (i) flower of Ponerorchis nana; (j) flowers of Sirindhornia pulchella; (k) Hsenhsua chrysea. (Photos taken by Xiaohua Jin.) based on morphological characters. On these grounds, we propose to unite Amitostigma and Neottianthe into Ponerorchis. 4.2.4. Hemipilia, Hemipiliopsis,and Tsaiorchis (Clade VIII) Both Hemipiliopsis (Fig. 3d, e) and Tsaiorchis (Fig. 3f) are littleknown and narrowly distributed monotypic genera. Garay and Kettredge (1985) placed Tsaiorchis neottianthoides (= Habenaria keiskeoides) in Amitostigma, Pridgeon et al. (2001) treated Tsaiorchis as Diphylax. Bateman et al. (2009) positioned Tsaiorchis neottianthoides in Platanthera. Bateman et al. (2003) stated that the monotypic Hemipiliopsis is closely related to Ponerorchis brevicalcarata, and suggested these two species should be assigned to Hemipilia. In our analyses, Hemipiliopsis is nested within Hemipilia with robust support (PP = 100, BS = 97), while Tsaiorchis is resolved as sister to Hemipiliopsis plus Hemipilia with weak support. These results agree with morphological characters. On the one hand, members of this clade (Clade VIII) typically share a distinctly elongate and erect rostellum. On the other hand, Tsaiorchis differs greatly from Hemipiliopsis and Hemipilia both in vegetative and floral characters. Hemipiliopsis and Hemipilia are characterized by cylindrical to ovoid-globose tubers, leaf solitary, basal, prostrate fleshy, more or less distinctly spotted, inflorescence with purple spots (Fig. 3d, e), stigma lobes confluent. In contrast to these, Tsaiorchis (Fig. 3f) is characterized by hairy and horizontally extending rhizomes, one to two chartaceous and green leaves, inflorescence green, two lateral appendages longer than anther, rostellum elongate, stigma lobes separate, linear and extending at the base of lip. Although Luo and Chen (2003) established Hemipiliopsis on the basis of its less developed rostellum and the shape of spur, Hemipiliopsis is characterized by the well-developed and erect rostellum as long as half of anther locus (Fig. 3e). The sister taxon of Hemipiliopsis, Ponerorchis brevicalcarata, fits well in Hemipilia. In addition, Hemipiliopsis is sympatric with Hemipilia. Based on these findings, we propose to broaden Hemipilia to include Hemipiliopsis, and maintain Tsaiorchis as a distinctive genus. 4.2.5. Sirindhornia (Clade IX) Sirindhornia is a recently established and morphologically distinctive genus. There are three species and they are mainly distributed in border regions in Thailand, Myanmar, and China. However, 48 W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 the systematic position of Sirindhornia has been disputed. Pedersen et al. (2002) stated that Sirindhornia is close to Hemipilia, Ponerorchis and reminiscent of Orchis. Instead, Chen et al. (2009) included Sirindhornia monophylla (type of Sirindhornia) in Ponerorchis. In our phylogenetic analyses, Sirindhornia is unambiguously resolved as sister to the remaining genera, including Hemipilia, Tsaiorchis, in Superclade B, which is supported by morphological characters of Sirindhornia. Sirindhornia is characterized by the elongate tuber, solitary and fleshy leaf, leaf convolute and not spreading, stem and ovary papillose–pubescent, ciliate bract, lip spurred, stigma lobes concave, and viscidium enveloped in a bursicle (Fig. 3j). In addition, species of Sirindhornia typically grow in limestone regions, and flower from April to June, which is one to two months earlier than other Orchidinae from Eastern Asia. Based on these findings, we recognize Sirindhornia as a distinct genus. 4.2.6. Androcorys, Herminium, and Porolabium (Clade X, XI, XII, and XIII) Morphologically, it is difficult to determine the systematic positions of Androcorys, Herminium, and Porolabium in Orchideae due to the relative small flowers, short column occupied almost by anther, greatly reduced stigma and rostellum (Pridgeon et al., 2001). Pridgeon et al. (2001) proposed to broaden Androcorys to include some members of Herminium with spurred lip; Bateman et al. (2003) suggested that some temperate members of Peristylus should be transferred to Herminium. In our analyses, Androcorys, Herminium, and Porolabium are deeply embedded within Habenariinae, and Herminium is not monophyletic. Some misplaced members have to be excluded and assigned to other genera, while some species currently residing in other genera need to be transferred to it. Herminium lanceum is resolved as immediate sister to groups formed by clade X plus XI with moderate support (PP = 99, BS < 50). Our observations indicate that H. lanceum differs from other species of Herminium by having parallel anther loculi, one pulvinate stigma swelling at the base of the column, and transverse rostellum, while most Herminium (including the type, H. monorchis) have more or less divergent anther loculi, two stalked stigma lobes extending along the base of column, and a cylindrical rostellum. A little known species of Herminium is resolved as immediate sister to clade X, XI, XII and XIII with weak support. Our morphological studies show that this entity is similar to Herminium and shares many morphological characters, such as ovoid tubers, viscidium involute and hornlike, two stalked stigma lobes extending at the base of column. This entity differs from Herminium by having parallel anther loculi with a long (as long as half of anther locus) and slender appendage between them. More molecular data and a better understanding of the morphological characters are needed to elucidate the phylogeny of this little known entity. Tang and Wang (1940) separated Porolabium from Herminium on the basis of having a lip with two pores. Pridgeon et al. (2001) stated that the rostellum of Porolabium is identical to Herminium; Pearce and Cribb (2002) suggested that Androcorys ophioglossoides resembles Porolabium biporosum in morphological characters. Our results indicate that Porolabium is nested within Androcorys with strong support, and these two genera are sympatric. We found that Androcorys and Porolabium both have a pulvinate stigma at the base of the column. There are several divergent taxonomic options to classify Herminium and its alliance into one or more monophyletic groups. One is to broaden Herminium to include Androcorys and its alliance (Clade X, XI, XII, XIII). Another is to recognize each clade in this alliance (Clade X, XI, XII, XIII) as a distinctive genus. However, with the currently available sampling and weak supported interrelationships among these clades, it would be premature to make any firm decision. 4.2.7. Ponerorchis chrysea (Habenaria chrysea, Orchis chrysea, Clade XIV) Ponerorchis chrysea (Fig. 3h) may have the largest flowers in Orchidinae, however, its systematic position is poorly understood. Schlechter (1924) treated Habenaria chrysea as a member of Orchis s.l., Soó (1966) transferred it to Ponerorchis, Hunt (1971) included it in Chusua. In our analyses, Ponerorchis chrysea is resolved as sister to the entity comprised of Clades X–XIII, with moderate support (PP = 100, BS = 69) (Fig. 2). Most members of these clades have basal and green leaf (or leaves), yellow, yellowish or green flowers, which strikingly differ from Ponerorchis. Our detailed morphological examination indicate that Ponerochis chrysea is a quite out of place in Ponerorchis on account of several morphological characters, such as leaves basal, sheaths long and clasping peduncle, floral bracts similar to leaf and enclosing the spur, pedicel longer than ovary, anther connective elongate and drawn out, three slightly convex stigma lobes confluent under rostellum. Some of these characters, such as pedicel longer than ovary, connective elongate and drawn out, are even unique in Orchideae in Eastern Asia. On these grounds, we propose to establish a new genus, Hsenhsua, to accommodate this taxon. Hsenhsua X.H. Jin, Schuit. et W.T. Jin, gen. nov. Figs. 3k and 4. Type species: Hsenhsua chrysea (W.W. Sm.) X.H. Jin, Schuit., W.T. Jin et L.Q. Huang. Included taxa: Hsenhsua chrysea (W.W. Sm.) X.H. Jin, Schuit.,W.T. Jin et L.Q. Huang, comb. nov. (Basionym Habenaria chrysea W.W. Sm., Notes from the Royal Botanic Garden, Edinburgh 13(63–64): 204–205. 1921.) Diagnosis: Tuber subglobose. Stem 2 leaved. Leaves green, basal. Inflorescence 1-flowered. Floral bracts foliaceous, sheathing, enclosing the spur. Pedicel longer than ovary; anther connective elongate and drawn out; rostellum protruding with two arms; three slightly convex stigma lobes confluent under rostellum. Morphological evidence: Morphologically, Hsenhsua is similar to Ponerorchis s.l. by sharing subglobose tuber, three-lobed lip with a distinct spur, rostellum protruding with two arms, but differs from the latter by having foliaceous bracts sheathing and enclosing the spur, pedicel longer than ovary, anther connective elongate and drawn out. Hsenhsua can be readily distinguished from Herminium s.l. by its foliaceous bracts, elongate anther connective, three slightly convex stigma lobes confluent under rostellum. Molecular evidence: In the current study, the combined analyses of the datasets of three markers (two chloroplast, one nuclear) shows moderate support for a sister group relationships of Hsenhsua with Herminium s.l. (including Androcorys, Herminium, Porolabium and other species) (PP = 100, BS = 69). Habit and Distribution: Growing in humid alpine grassland and shrub between 3600 and 4200 m in the Eastern Himalayas, including China (NW Yunnan, SE to S Xizang), and Bhutan. Etymology: Hsenhsua is named in honor of Prof. Hsen-Hsu Hu, the renowned Chinese scientist who was a pioneer and founder of modern botany in China. 4.2.8. Habenaria and its alliance (Clade XV, XVII, and XVIII) Habenaria is one of the largest genera of orchid family and widespread across the tropical and subtropical regions of the world (Pridgeon et al., 2001). There is much debate about the generic delimitation and infrageneric systems (see Batista et al., 2013; Pridgeon et al., 2001). Bateman et al. (2003) stated that Habenaria is polyphyletic and needs to be dismantled into many monophyletic genera. On the other hand, Batista et al. (2013) indicated that Neotropical Habenaria is monophyletic and closely related to some African congeners, and suggested that a revision of the infrageneric system rather than an extensive generic fragmentation is most appropriate. In our phylogenetic analyses with 36 sampled species from Asia, Africa and the Neotropics, Habenaria is resolved into two W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 49 Fig. 4. Plant of Hsenhsua chrysea. (a) habit of Hsenhsua chrysea; (b) front view of flower; (c) bract; (d) dorsal sepal, lateral sepal, petal and lip; (e) lateral view of flower (bract removed); (f) lateral view of longitudinal section of column and upper part of ovary. (Drawn by Yunxi Zhu.) unambiguously supported groups: Clade XV and clade XVII (Fig. 2). Clade XV includes nine Asian species mainly distributed in subtropical region, such as H. dentata, H. rhodocheila and Pecteilis gigantea, and is sister to the predominantly alpine Herminium s.l. clade (Clades X–XIII) and Ponerorchis chrysea. Clade XVII, mainly including African and South American Habenaria, and Bonatea, is resolved as sister to two monotypic genera, Nujiangia and Gennaria. It is interesting to note that five Asian Habenaria are nested within Clade XVII. Asian tropical Habenaria stenopetala is resolved as sister to the African-Neotropical H. heyneanna plus H. foliosa with strong support (PP = 100, BS = 100). Asian alpine H. intermedia is resolved as sister to African montane H. praestans with strong support (PP = 100, BS = 100); both belong to Habenaria sect. Mutipartitae. Two monotypic and morphologically isolated genera in Habenariinae, Gennaria and Nujiangia, were uncovered as sister to African and American Habenaria with strong support. These two genera share several morphological characters, such as two alternate leaves, densely flowered inflorescence, 3-lobed lip with a short spur, auricles longer than anther, and a convex stigma (Jin et al., 2012). Although with limited sampling of Habenaria and its alliance from Africa and Neotropics, our phylogenetic analyses indicate that the taxonomy of Habenaria is even more complicated than previously thought. It is clear that as currently defined Habenaria is patently polyphyletic. There have been several highly divergent taxonomic proposals to delimit Habenaria to be monophyletic. One is the generic fragmentation of Habenaria into many genera, such as Bonatea, Habenella, and many others; another is to broaden Habenaria to include nearly all genera, such as Androcorys, Herminium, and Peristylus, in Habenariinae. The third is to redefine clade XVII as Habenaria s.s., with typification of Habenaria macroceratitis, while clade XV is segregated as a distinct genus. With the currently 50 W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 available sampling it is premature to make any firm decision. Future work with more comprehensive sampling and more molecular data will undoubtedly provide new insights into the taxonomy of Habenaria. 4.2.9. Peristylus (Clade XVI) Peristylus is a widespread genus in Asia with about 70 species (Pridgeon et al., 2001), however, it is almost neglected in previous studies of Orchideae. There is some debate on its distinction from Habenaria. Seidenfaden (1977) proposed to define Peristylus on the basis of a set of characters, such as size of floral parts, the position of stigma lobes and caudicles. Pridgeon et al. (2001) stated that Peristylus typically have pulvinate stigmatic swellings at the base of the column and adnate to the base of the lip. Comber (2001) also suggested a set of morphological characters to define Peristylus. In our phylogenetic analyses, Peristylus is resolved as an unambiguously supported clade within Habenariinae, and immediate sister to the group formed by Clades X to XV (Fig. 2). Some misplaced species of Peristylus, such as P. nematocaulon and P. coeloceras, are nested within Platanthera or Herminium, whereas at least one misplaced Platanthera with trilobed lip and ovoid tubers, namely P. biermanniana, is nested within Peristylus. Geographically, Peristylus is mainly distributed in Asia in tropical to subtropical regions, while Platanthera and Herminium mainly occur in alpine regions. In our phylogenetic analyses, the currently delimited Peristylus s.s. is strongly supported by molecular evidence and is more uniform in morphological characters, characterized by having ovoid to globose tubers, flowers small, dorsal sepal and petals forming a hood, lip trilobed and spurred, spur shorter than ovary, and pulvinate stigmas adnate to the base of lip. Therefore, we suggest to make some intergeneric transfers to retain Peristylus as a monophyletic genus which mainly includes members from tropical and subtropical regions. 4.3. Biogeography Orchidinae are mostly distributed in northern temperate or alpine regions, however, recent studies indicate that the widespread tropical genus Brachycorythis is sister to the remaining Orchidinae (Bateman et al., 2003; Inda et al., 2010). There are two centers of biodiversity of Orchidinae. One is the Mediterranean Region, the other is the Pan-Himalayas (including Himalayas and the Hengduan Mountains) (Pridgeon et al., 2001; Lang, 1998). Inda et al. (2012) stated that Orchidinae have diversified in the Mediterranean Region in the last 15 Million years. Many taxa of Superclade A, such as Himantoglossum, Ophrys, Pseudorchis and Serapias, are predominantly Mediterranean (Pridgeon et al., 2001). On the other hand, many genera in Superclade A, such as Platanthera and its alliance, have their biodiversity centers in Pan-Himalayas (Chen et al., 2009; Inda et al., 2012; King and Pantling, 1898; Lang, 1998; Pridgeon et al., 2001). Bateman et al. (2003) and Inda et al. (2012) showed that Platanthera and its alliance are deeply nested within the Mediterranean Region group. Our results indicate that Clade III, IV, V, and VI (the Mediterranean Region group) are resolved as successive sister to Platanthera and its alliance. Members of Superclade B, including Hemipilia, Ponerochis, Sirindhornia and Tsaiorchis, are mostly restricted to mountain regions in Eastern Asia, especially the Pan-Himalayas. Tropical/subtropical genera, such as Sirindhornia and Tsaiorchis, are resolved as successive sister to alpine taxa (Fig. 1). In contrast to Orchidinae, Habenariinae are widespread in tropical/subtropical regions around the world and only a few species occur in alpine regions. Recent studies indicate that tropical African genera, such as Cynorkis and, Stenoglottis, are basal groups in the subtribe (Bateman et al., 2003; Batista et al., 2013; Inda et al., 2010, 2012). Our results indicate that alpine groups, such as Herminium and its alliance (including Clade X, XI, XII, XIII, and XIV, most endemic in Pan-Himalayas), are deeply nested within Habenariinae (Fig. 2). Taken together, it seems that the tropical ancestral group of Orchideae radiated in the Mediterranean region in the last 15 million years and in the Pan-Himalayas in the last 8 million years (Geographical history of Pan-Himalayas see Shi et al., 1998; Biogeography of Pan-Himalayas see Sun, 2002a, 2002b). 5. Main conclusions Based on three DNA markers (plastid matK, rbcL, and nuclear ITS), morphological characters and comprehensive sampling, our current study greatly advances our understanding of the phylogeny of Orchidinae and Asian Habenariinae. Orchidinae are subdivided into two sister groups: one is Superclade A + Clade VI (including Gymnadenia, Ophrys, Orchis and its alliance, and Platanthera and its alliance) with diversity centers in the Mediterranean region and Pan-Himalayas, the other is Superclade B (Hemipilia, Ponerorchis, Sirindhornia, and Tsaiorchis) and is mainly distributed in Eastern Asia. Habenaria is subdivided into two distant related groups: one group mainly distributed in tropical and subtropical Asia, the other group (including type of Habenaria) is widespread in both Old and New Tropics. Elements of alpine regions are deeply nested within each tropical/subtropical group. Many genera, such as Herminium, Platanthera, Peristylus and Ponerorchis, are not monophyletic. Many previously undetected phylogenetic relationships, such as clades formed by Ponerorchis chrysea, the Platanthera latilabris group, Sirindhornia, and Tsaiorchis, are well supported by both molecular and morphological evidence. We propose to combine Hemipiliopsis with Hemipilia, to broaden Ponerorchis to include Amitostigma and Neottianthe, to combine Aorchis, Aceratorchis, and Neolindleya with Galearis, and to establish a new genus to accommodate Ponerorchis chrysea. Tsaiorchis and Sirindhornia are recognized as two distinctive genera. A nomenclatural revision is provided in the Appendix A. At the same time, the relationships of many groups, such as Herminium and its alliance (including Androcorys, Herminium, Platanthera latilabris group, and Porolabium) and Galearis and its alliance (including Aceratorchis, Aorchis, Galearis, and Neolindleya) remain unresolved. Further work based on broader sampling and more markers is needed. Acknowledgements Funds were provided by grants from the National Natural Science Foundation of China (Grant Nos. 31107176, 31311120061) and the Chinese Special Fund for Medicine Research in the Public Interest (201407003). Appendix A A.1. Nomenclatural revision 1. Galearis Raf. (1) Galearis camtschatica (Cham.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Orchis camtschatica Cham., Linnaea 3: 27 (1828). 2. Hemipilia Lindl. Synonym. Hemipiliopsis Y.B. Luo et S.C. Chen, syn. nov. (2) Hemipilia purpureopunctata (K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Habenaria purpureopunctata K.Y. Lang, Acta Phytotax. Sin. 16(4): 127 (1978). 3. Peristylus Blume. W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 (3) Peristylus biermannianus (King et Pantl.) X.H. Jin, Schuit. & W.T. Jin, comb. nov. Basionym Habenaria biermanniana King et Pantl., J. Asiat. Soc. Bengal. 64: 343 (1895). 4. Platanthera Rich. Synonym. Smithorchis Tang et F.T. Wang, syn. nov. (4) Platanthera angustilabris (King et Pantl.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Herminium angustilabre King et Pantl. J. Asiat. Soc. Bengal, 65(2): 131. 1895 (1896). (5) Platanthera calceoliformis (W.W. Sm.) X.H. Jin, Schuit. et W.T. Jin, comb.nov. Basionym Herminium calceoliforme W.W. Sm., Notes Roy. Bot. Gard. Edinburgh. 13: 211 (1921). (6) Platanthera carnosilabris (Tang et F.T. Wang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Herminium carnosilabre Tang et F.T. Wang, Bull. Fan Mem. Inst. Biol. Bot. 10: 32 (1940). 5. Ponerorchis Rchb. f. Synonyms. Amitostigma Schltr., syn. nov. Neottianthe (Rchb.) Schltr., syn. nov. (7) Ponerorchis alpestris (Fukuy.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma alpestre Fukuy., Bot. Mag. Tokyo 49: 664 (1935). (8) Ponerorchis amplexifolia (Tang et F.T. Wang) X.H. Jin, Schuit. et W.T. Jin, comb. nov.Basionym Amitostigma amplexifolium Tang et F.T. Wang, Bull. Fan Mem. Inst. Biol. Bot. 7: 3 (1936). (9) Ponerorchis basifoliata (Finet) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Peristylus tetralobus f. basifoliatus Finet, Rev. Gén. Bot. 13: 525, pl. 13(C). (1901). (10) Ponerorchis bidupensis (Aver.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Hemipilia bidupensis Aver., Lindleyana 14: 222 (1999). (11) Ponerorchis bifoliata (Tang et F.T. Wang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma bifoliatum Tang et F.T. Wang, Bull. Fan Mem. Inst. Biol. Bot. 7: 127 (1936). (12) Ponerorchis capitata (Tang et F.T. Wang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma capitatum Tang et F.T. Wang, Bull. Fan Mem. Inst. Biol. Bot. 7: 4–5 (1936). (13) Ponerorchis camptoceras (Rolfe) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Habenaria camptoceras Rolfe, J. Linn. Soc., Bot., 29: 319 (1892). (14) Ponerochis compacta (Schltr.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Neottianthe compacta Schltr., Acta Horti Gothob. 1: 136 (1924). (15) Ponerorchis cucullata (L.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Orchis cucullata L., Sp. Pl.: 939 (1753). (16) Ponerorchis cucullata var. calcicola (W.W. Sm.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Gymnadenia calcicola W.W. Sm., Notes Roy. Bot. Gard. Edinburgh. 8: 188 (1914). (17) Ponerorchis dolichocentra (Tang, F.T. Wang et K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma dolichocentrum Tang, F.T. Wang et K.Y. Lang, Acta Phytotax. Sin. 20(1): 84 (1982). 51 (18) Ponerorchis faberi (Rolfe) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Habenaria faberi Rolfe, Kew Bull. 1896: 201 (1896). (19) Ponerorchis farreri (Schltr.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma farreri Schltr., Repert. Spec. Nov. Regni Veg. 20: 378 (1924). (20) Ponerorchis gonggashanica (K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma gonggashanicum K.Y. Lang, Acta Phytotax. Sin. 22(4): 315 (1984). (21) Ponerorchis gracilis (Blume) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Mitostigma gracile Blume, Mus. Bot. Lugd.-Bat. 2: 190 (1856). (22) Ponerorchis keiskei (Finet) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Gymnadenia gracilis var. keiskei Finet, Bull. Soc. Bot. France 47: 280 (1900). (23) Ponerorchis kinoshitae (Makino) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Gymnadenia kinoshitae Makino, Bot. Mag. (Tokyo) 23: 137 (1909). (24) Ponerorchis lepida (Rchb. f.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Gymnadenia lepida Rchb. f., Otia Bot. Hamburg.: 51 (1878). (25) Ponerorchis luteola (K.Y. Lang et S.C. Chen) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Neottianthe luteola K.Y. Lang et S.C. Chen, Acta Phytotax. Sin. 35(6): 545 (1996). (26) Ponerorchis monantha (Finet) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Peristylus monanthus Finet, Rev. Gén. Bot. 13: 323 (1901). (27) Ponerorchis oblonga (K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Neottianthe oblonga K.Y. Lang, Acta Phytotax. Sin. 35(6): 544 (1997). (28) Ponerorchis ovata (K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Neottianthe ovata K.Y. Lang, Acta Phytotax. Sin. 35(6): 542 (1997). (29) Ponerochis papilionacea (Tang, F.T. Wang et K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma papilionaceum Tang, F.T. Wang et K.Y. Lang, Acta Phytotax. Sin. 20(1): 83 (1982). (30) Ponerorchis parciflora (Finet) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Peristylus tetralobus f. parciflorus Finet, Rev. Gén. Bot. 13: 525 (1902) (as ‘parceflorus’). (31) Ponerorchis physoceras (Schltr.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma physoceras Schltr., Acta Horti Gothob. 1: 133 (1924). (32) Ponerorchis pinguicula (Rchb. f. et S. Moore) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Gymnadenia pinguicula Rchb. f. et S. Moore., J. Bot. 16: 135 (1878). (33) Ponerochis simplex (Tang et F.T. Wang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma simplex Tang et F.T. Wang, Bull. Fan Mem. Inst. Biol. Bot. Ser. 10: 25 (1940). (34) Ponerorchis secundiflora (Hook. f.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. 52 W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 Basionym Peristylus secundiflorus Kraenzl., Orchid. Gen. Sp. 1: 518 (1898). (35) Ponerorchis tetraloba (Finet) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Peristylus tetralobus Finet, Rev. Gén. Bot. 13: 524 (1901). (36) Ponerorchis thailandica (Seidenf. et Thaithong) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma thailandicum Seidenf. et Thaithong in G. Seidenfaden, Contr. Orchid Fl. Thailand 13: 8 (1997). (37) Ponerorchis tibetica (Schltr.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma tibeticum Schltr., Repert. Spec. Nov. Regni Veg. 20: 379 (1924). (38) Ponerorchis trifurcata (Tang, F.T. Wang et K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma trifurcatum Tang, F.T. Wang et K.Y. Lang, Acta Phytotax. Sin. 20(1): 80, pl. 1(5–8) (1982). (39) Ponerorchis wenshanensis (W. H. Chen, Y. M. Shui et K.Y. Lang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma wenshanense W. H. Chen, Y. M. Shui et K.Y. Lang, Acta Bot. Yunnan. 25(5): 521 (2003). (40) Ponerorchis yuana (Tang et F.T. Wang) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Amitostigma yuanum Tang et F.T. Wang, Bull. Fan Mem. Inst. Biol. Bot. 10: 26 (1940). 6. Tsaiorchis Tang et F.T. Wang. (41) Tsaiorchis keiskeoides (Gagnep.) X.H. Jin, Schuit. et W.T. Jin, comb. nov. Basionym Habenaria keiskeoides Gagnep., Bull. Soc. Bot. Fr. 78:71. (1931). Synonym Tsaiorchis neottianthoides Tang et F.T. Wang, Bull. Fan Mem. Inst. Biol. Bot. 7: 133 (1936). Appendix B. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ympev.2014. 04.004. References Aceto, S., Caputo, P., Cozzolino, S., Gaudio, L., Moretti, A., 1999. Phylogeny and evolution of Orchis and allied genera based on ITS DNA variation: morphological gaps and molecular continuity. Mol. Phylogenet. Evol. 13, 67–76. Bateman, R.M., Hollingsworth, P.M., Preston, J., Luo, Y.B., Pridgeon, A.M., Chase, M.W., 2003. Molecular phylogenetics and evolution of Orchidinae and selected Habenariinae (Orchidaceae). Bot. J. Linn. Soc. 142, 1–40. Bateman, R.M., James, K.E., Luo, Y.B., Lauri, R.K., Fulcher, T., Cribb, P.J., Chase, M.W., 2009. Molecular phylogenetics and morphological reappraisal of the Platanthera clade (Orchidaceae: Orchidinae) prompts expansion of the generic limits of Galearis and Platanthera. Ann. Bot.-London 104, 431–445. Batista, J.A.N., Borges, K.S., de Faria, M.W.F., Proite, K., Ramalho, A.J., Salazar, G.A., van den Berg, C., 2013. Molecular phylogenetics of the species-rich genus Habenaria (Orchidaceae) in the new world based on nuclear and plastid DNA sequences. Mol. Phylogenet. Evol. 67, 95–109. Bellusci, F., Pellegrino, G., Palermo, A.M., Musacchio, A., 2008. Phylogenetic relationships in the orchid genus Serapias L. based on noncoding regions of the chloroplast genome. Mol. Phylogenet. Evol. 47, 986–991. Box, M.S., Bateman, R.M., Glover, B.J., Rudall, P.J., 2008. Floral ontogenetic evidence of repeated speciation via paedomorphosis in subtribe Orchidinae (Orchidaceae). Bot. J. Linn. Soc. 157, 429–454. Bytebier, B., Bellstedt, D.U., Linder, H.P., 2007. A molecular phylogeny for the large African orchid genus Disa. Mol. Phylogenet. Evol. 43, 75–90. Chase, M.W., Cameron, K.M., Barret, R.L., Freudenstein, J.V., 2003. DNA data and Orchidaceae systematics: a new phylogenetic classification. In: Dixon, K.W., Kell, S.P., Barret, R.L., Cribb, P.J. (Eds.), Orchid Conservation. Natural History Publications, Kota Kinabalu, pp. 69–89. Chen, S.C., Lang, K.Y., Stephan, W.G., Phillip, J.C., Paul, O., 2009. Subfam. Orchidoideae. In: Wu, Z.Y., Raven, P.H., Hong D.Y. (Eds.), Flora of China. vol. 25. Science Press, Beijing, Missouri Botanical Garden, St. Louis, pp. 45–166. Comber, J.B., 2001. Orchids of Sumatra. The Royal Botanic Gardens Kew, London. Cunningham, C.W., 1997. Can three incongruence tests predict when data should be combined? Mol. Biol. Evol. 14, 733–740. Devos, N., Raspé, O., Jacquemart, A., Tyteca, D., 2006. On the monophyly of Dactylorhiza Necker ex Nevski (Orchidaceae): is Coeloglossum viride (L.) Hartman a Dactylorhiza? Bot. J. Linn. Soc. 152, 261–269. Douzery, E.J.P., Pridgeon, A.M., Kores, P., Linder, H.P., Kurzweil, H., Chase, M.W., 1999. Molecular phylogenetics of Diseae (Orchidaceae): a contribution from nuclear ribosomal ITS sequences. Am. J. Bot. 86, 887–899. Dressler, R.L., 1981. The Orchids: Natural History and Classification. Harvard University Press, Cambridge. Dressler, R.L., 1993. Phylogeny and classification of the orchid family. Dioscorides Press, Portland. Duthie, J.F., 1906. The orchids of the North-Western Himalaya. Ann. Roy. Bot. Gard. (Calcutta) 9, 85–211. Efimov, P.G., Lauri, R.K., Bateman, R.M., 2009. Neolindleya Kraenzl. (Orchidaceae), an enigmatic and largely overlooked autogamous genus from temperate East Asia. Kew. Bull. 64, 661–671. Farris, J.S., Kälersjö, M., Kluge, A.G., Bult, C., 1995. Constructing a significance test for incongruence. Syst. Biol. 44, 570–572. Felsenstein, J., 1988. Phylogenies from molecular sequences: inference and reliability. Annu. Rev. Genet. 22, 521–565. Gamarra, R., Galán, P., Herrera, I., Ortúňez, E., 2008. Seed micromophology supports the splitting of Limnorchis from Platanthera (Orchidaceae). Nordic J. Bot. 26, 61– 65. Garay, A., Kettredge, W., 1985. Notes from the Ames Orchid Herbarium. Bot. Mus. Leaf. Harv. Univ. 30, 47–59. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98. Hapeman, J.R., Inoue, K., 1997. Plant-pollinator interactions and floral radiation in Platanthera (Orchidaceae). In: Givnish, T.J., Sytsma, K.J. (Eds.), Molecular Evolution and Adaptive Radiation. Cambridge University Press, Cambridge, pp. 433–454. Hooker, J.D., 1890. The flora of British India 6: Orchideae. Reeve, London. Hunt, P.F., 1971. Notes on Asiatic Orchids: VI. Kew. Bull. 26, 171–185. Inda, L.A., Pimentel, M., Chase, M.W., 2010. Contribution of mitochondrial cox1 intron sequences to the phylogenetics of tribe Orchideae (Orchidaceae): do the distribution and sequence of this intron tell us something about its evolution? Taxon 59, 1053–1064. Inda, L.A., Pimentel, M., Chase, M.W., 2012. Phylogenetics of tribe Orchideae (Orchidaceae: Orchidoideae) based on combined DNA matrices: inferences regarding timing of diversification and evolution of pollination syndromes. Ann. Bot. London 110, 71–90. Jin, X.H., Efimov, P.G., 2012. Platanthera ovatilabris and P. dulongensis spp. nov. and new records of Platanthera (Orchidaceae, Orchidoideae) for Yunnan and Tibet, China. Nordic J. Bot. 30, 291–298. Jin, X.H., Li, D.Z., Xiang, X.G., Lai, Y.J., Shi, X.C., 2012. Nujiangia (Orchidaceae: Orchideae): a new genus from the Himalayas. J. Syst. Evol. 50, 64–71. Jin, W.T., Zhou, H.L., Jin, X.H., 2013. Platanthera yadongensis (Orchidaceae, Orchideae), a new species from Tibet, China. Syst. Bot. 38 (4), 982–986. King, G., Pantling, R., 1896. A second series of new orchids from Sikkim. J. Asiat. Soc. Bengal, Pt. 2, Nat. Hist., 118–134. King, G., Pantling, R., 1898. The orchids of Sikkim-Himalaya. Ann. Roy. Bot. Gard. (Calcutta) 8, 1–342. Kraenzlin, F., 1901. Orchidacearum Genera et Species, vol. 1. Mayer and Müller, Berlin. Kurzweil, H., Weber, A., 1992. Floral morphology of southern African Orchideae. II. Habenariinae. Nordic J. Bot. 12, 39–61. Lang, K.Y., 1998. A new subgenus of Platanthera (Orchidaceae). Acta Phytotaxon. Sin. 36, 449–458. Lang, K.Y., 1999. Flora Republicae Popularis Sinicae, vol. 17. Science Press, Beijing. Lindley, J., 1835. The Genera and Species of Orchidaceous Plants. Ridgways Piccadilly, London. Luer, C.A., 1975. The Native Orchids of the United States and Canada Excluding Florida. New York Botanical Garden, New York. Luo, Y.B., Chen, S.C., 2003. Hemipiliopsis, a new genus of Orchidaceae. Novon 13, 450–453. Miller, M.A., Pfeiffer, W., Schwartz, T., 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: Proceedings of the Gateway Computing Environments Workshop (GCE), 14 November 2010, New Orleans, LA, pp. 1–8. Pearce, N.R., Cribb, P.J., 2002. Orchideae. In: Pearce, N.R., Cribb, P.J. (Eds.),The Orchids of Bhutan. Royal Botanic Garden Edinburgh, Edinburgh, Royal Government of Bhutan, Bhutan, pp. 122–192. Pedersen, H.A., Suksathan, P., Indhamusika, S., 2002. Sirindhornia, a new orchid genus from Southeast Asia. Nordic J. Bot. 22, 391–404. Posada, D., Crandall, K.A., 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14, 817–818. Pridgeon, A.M., Cribb, P.J., Chase, M.W., Rasmussen, F.N., 2001. Genera Orchidacearum, vol. 2. Oxford University Press, Oxford. Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19, 1572–1574. Schlechter, R., 1924. Additamenta ad Orchideologiam chinensem. (Orchidaceae novae et criticae) Repert. Spec. Nov. Regni Veg. 19, 372–383. Seidenfaden, G., 1977. Orchid genera in Thailand V: Orchidoideae. Dansk Bot. Ark. 31, 1–149. W.-T. Jin et al. / Molecular Phylogenetics and Evolution 77 (2014) 41–53 Shi, Y.F., Li, J.J., Li, B.Y., 1998. Uplift and Environmental Changes of Qinghai-Tibetan Plateau in the Late Cenozoic. Guangdong Science and Technology Press, Guangzhou. Soliva, M., Kocyan, A., Widmer, A., 2001. Molecular phylogenetics of the sexually deceptive orchid genus Ophrys (Orchidaceae) based on nuclear and chloroplast DNA sequences. Mol. Phylogenet. Evol. 20, 78–88. Soó, R., 1929. Revision der Orchideae-Ophrydineae von Ostasien und dem Himalaya. Annls Hist.-Nat. Mus. Nat. Hung. 26, 339–384. Soó, R., 1966. Die sog. Orchis Arten der Ostasiatisch-Nordamerikanischen Flora. Acta Bot. Acad. Sci. Hung. 12, 351–354. Sun, H., 2002a. Tethys retreat and Himalayas-Hengduanshan Mountains uplift and their significance on the origin and development of the Sino-Himalayan elements and alpine flora. Acta Bot. Yunnan. 24, 273–288. Sun, H., 2002b. Evolution of Arctic-Tertiary Flora in Himalayan-Hengduan Mountains. Acta Bot. Yunnan. 24, 671–688. Swofford, D.L., 2002. PAUP⁄: phylogenetic analysis using parsimony (⁄ and Other Methods), version 4.0b10. Sinauer, Sunderland. 53 Szlachetko, D.L., Kras, M., 2006. Notes sur le genre Habenella. Richardiana 6, 33–39. Tang, T., Wang, F.T., 1940. Contributions to the knowledge of eastern Asiatic Orchidaceae I. Bull. Fan Mem. Inst. Biol. Bot. Ser. 10, 23–46. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G., 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882. Tuyama, T., 1966. Orchidaceae. In: Hara, H. (Ed.), The Flora of Eastern Himalaya. University of Tokyo, Tokyo, pp. 424–452. Tuyama, T., 1971. Orchidaceae. In: H. Hara (Ed.), The Flora of Eastern Himalaya, Second Report. University of Tokyo, Tokyo, pp. 176–196. Tuyama, T., 1975. Orchidaceae. In: H. Ohashi (Ed.), The Flora of Eastern Himalaya, Third Report. University of Tokyo Press, Tokyo, pp. 137–164. Tyteca, D., Klein, E., 2008. Genes, morphology and biology – the systematics of Orchidinae revisited. J. Eur. Orch. 40, 501–544. Vaidya, G., Lohman, D.J., Meier, R., 2011. SequenceMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics 27, 171–180.