Meconopsis is an herbaceous genus native to the high altitude habitats across the Himalaya and adjacent plateau and mountain areas. Attractive Meconopsis flowers have spurred many European botanists to study the taxonomy of the genus resulting in numerous infrageneric classifications, dating from the first taxonomic revision in the late 19th century until the most recent monograph in 2014. All, however, were morphology-based treatments and largely inconsistent with one another. To investigate the incongruence among the previous taxonomic grouping strategies of the species in Meconopsis and settle the controversies, we employed a well-resolved molecular phylogeny built by analyzing four chloroplast markers (trnL-trnF intergenic spacer, matK, ndhF, and rbcL). We found that the evolutionary relationships revealed by our phylogeny disagreed to varying degrees with any infrageneric relationship suggested by previous authors. Therefore, we propose a revised classification based on our phylogenetic topology aswell as themorphological and cytological patterns reflected by the phylogenetic structure. To achieve a practical and approachable system, we have tried to retain as much as possible of phylogenetically meaningful components from previous taxonomies for the genus. As a result, we used the four major clades of our Meconopsis phylogeny as the bases for infrageneric sections (Meconopsis sect. Meconopsis, M. sect. Aculeatae, M. sect. Primulinae, and M. sect. Grandes). A key to the sections is provided, followed by a description and composition of each.
Meconopsis Vig., also known as Himalayan poppy or blue poppy, is an Old World genus in the subfamily Papaveroideae of Papaveraceae. Viguier (1814) established Meconopsis based on a single Western European species, Papaver cambricum L., that served as the type until recently. However, all the species later added to Meconopsis were discovered in South and East Asia. Kadereit et al. (2011) proposed returning Meconopsis cambrica (L.) Vig. to Papaver because geographical and molecular evidence showed that M. cambrica is not related to the rest of the species subsequently placed in Meconopsis, but embedded in the phylogeny of Papaver. Grey-Wilson (2014) later placed Meconopsis cambrica in a newly circumscribed monotypic genus Parameconopsis Grey-Wilson, but molecular evidence (Yuan 2002; Kadereit et al. 2011; Xiao 2013; Liu et al. 2014) clearly supported that this species should be returned to Papaver rather than treated as segregate monotypic genus.
The exclusion of M. cambrica as well as two other species, Meconopsis chelidonifolia Bureau & Franch. and Meconopsis oliveriana Franch. ex Prain, is in agreement with molecular work (Yuan 2002; Kadereit et al. 2011; Xiao 2013; Liu et al. 2014). The relationships among M. cambrica, Meconopsis, Cathcartia Hook. f., and Papaver are illustrated in Fig. 1 (Xiao 2013). Grey-Wilson (2014) officially transferred M. chelidonifolia and M. oliveriana from Meconopsis to Cathcartia. However, Cathcartia chelidonifolia (Bureau & Franch.) Grey-Wilson was typified by syntypes, and the lectotype of Cathcartia oliveriana (Franch. ex Prain) Grey-Wilson was not clearly indicated. We designate a lectotype for each of these species at the end of the Taxonomic Treatment below.
As Grey-Wilson moved the original type Meconopsis cambrica out of Meconopsis, he (2012) proposed conservation of the generic name Meconopsis for the Asiatic species with a new type: Meconopsis regia G. Taylor. Because the Nomenclature Committee for Vascular Plants recommended Grey-Wilson's proposal (in Taxon 62(6): 1318. 2013), we use the generic name “Meconopsis” for the Asian species following Article 14.16 of the International Code of Nomenclature (McNeill et al. 2012).
Meconopsis has traditionally been considered to consist of ca. 50–80 species. This large range of species numbers was mostly due to different species concepts implemented in previous taxonomic works. The genus exhibits high morphological (examples shown in Fig. 2) and ecological diversity: species range from a few centimeters to more than 2 m in height, are distributed from 3,000 to 5,800 m in elevation, and grow in distinctive habitats such as mountain woodland, alpine meadow, or rocky slopes. Moreover, various polyploids have been reported in Meconopsis (2n = 14, 22, 28, 56, 74, 76, 82, 84, and higher) (Ratter 1968; Ying et al. 2006; Kumar et al. 2013). However, there has not been any taxonomic scheme based on a comprehensive incorporation and evolutionary interpretation of the morphological, ecological, geographical, and cytological diversities in the genus. The lack of an integrated approach led to its ever-changing taxonomy over the last 200 yr. Previous taxonomic strategies for subdividing the genus (Prain 1895, 1906, 1915; Fedde 1909, 1936; Kingdon-Ward 1926, 1935; Taylor 1934;Wuand Chuang 1980; Chuang 1981; Grey-Wilson 2000, 2014) were based on different sets of morphological characters and growth habits that resulted in conflicting treatments. Here we discuss the two most influential and reasonably well-organized previous classifications of Meconopsis (Fedde 1909; Taylor 1934) as well as the most recently published monograph (Grey-Wilson 2014) to highlight the taxonomic inconsistency at the infrageneric level (Fig. 3B–D).
Fedde's (1909) classification was based on Prain's (1906) work, which divided Meconopsis into nine natural groups including Cambricae, Anomalae, Aculeatae, Primulinae, Bellae, Grandes, Torquatae, Robustae, and Chelidonifoliae. Prain (1906) also organized all his groups into two sections, M. sect. Eumeconopsis and M. sect. Polychaetia, based on leaf and stem trichome type. Fedde (1909) fully adopted Prain's (1906) system but assigned sectional rank to Prain's groups and elevated Prain's (1906) sections to subgenera (Fig. 3B).
Taylor (1934) also arranged the genus into two subgenera (Fig. 3C) which, however, were substantially different from those of Fedde's (1909). Taylor (1934) used the criterion of a stylar disc: the members of his Meconopsis subg. Discogyne were characterized by a style expanding into a flat disc surmounting the ovary; species of his M. subg. Eumeconopsis lacked this disc. His M. sect. Polychaetia (in M. subg. Eumeconopsis) that included the majority of the genus was divided into subsections and then series based on characters such as leaf persistence during winter, flower arrangement, and trichome type (Taylor 1934).
The most recent treatment of Meconopsis was by Grey-Wilson (2014), who excluded a few species (M. cambrica, M. chelidonifolia, M. oliveriana, Cathcartia villosa Hook.f. (syn: M. villosa), and Cathcartia smithiana Hand.-Mazz. (syn: M. smithiana)), relative to Taylor's (1934) treatment (Fig. 3 legend). Examined by the evidence from recent molecular phylogenetic studies (Yuan 2002; Carolan et al. 2006; Xiao 2013; Liu et al. 2014), the genus Meconopsis proposed by Grey-Wilson is a monophyletic taxon. Grey-Wilson (2014) divided the genus into four subgeneric divisions: subgenera Meconopsis, Grandes, Discogyne, and Cumminsia (shown in Fig. 3D). His M. subg. Meconopsis is similar to Fedde's (1909) M. sect. Robustae or Taylor's (1934) M. subsect. Eupolychaetia; his M. subg. Grandes is based on Fedde's M. sect. Grandes; and his M. subg. Discogyne corresponds to Fedde's M. sect. Torquatae or Taylor's M. subg. Discogyne. However, Grey-Wilson's (2014) concept of M. subg. Cumminsia had never been proposed before. Grey-Wilson's (2014) taxonomy is also much more complex than Fedde's (1909) or Taylor's (1934) by adding more subdivision levels, especially in his M. subgen. Cumminsia. For example, the majority of species in Taylor's (1934) M. ser. Aculeatae was divided to four sections which were further broken down into series by Grey-Wilson (2014).
Despite the removal of the outgroup species, the infrageneric classification of Meconopsis remains equivocal (Fig. 3B–D). Classifications based on selected morphological similarities have not reached a well agreed upon and stable taxonomy over the last 200 yr. Recent molecular phylogenetic studies (Yuan 2002; Liu et al. 2014) both employed trnL-trnF and ITS sequences. These two markers generated phylogenetic incongruences, and neither was sufficient to resolve the relationships within the genus (resulted in large basal polytomies). A low copy nuclear marker was utilized to investigate hybridization and polyploidization pattern in the genus (Xiao and Simpson 2014). However, because the species of Meconopsis range from diploid to dodecaploid and higher, estimating phylogenetic relationship using nuclear genes is problematic due to the difficulty and uncertainty of eliminating the effect of paralogy/orthology conflation and recombination. Therefore, we relied on single-copy chloroplast markers in this study to build a well-resolved molecular phylogeny for Meconopsis. Our resulting cpDNA tree guided the determination of taxonomic groups: each new section was based on a clade that matches, or shares close similarity in contained species with, previously published infrageneric groups, and is defined by the shared morphological and cytological characteristics of its included species.
Materials and Methods
Taxon Sampling—We sampled 40 species of Meconopsis (accessions) for this study that represent every section and series of Fedde (1909), Taylor (1934), and Grey-Wilson (2014). The first author made the determinations of the specimens. Although species delimitations are not in the scope of this study, it is worth noting that certain species have been defined very differently by authors. For example, we previously tested the species delimitation using phylogenetic methods and we found that “species” in Grey-Wilson's M. ser. Heterandrae together with most of the “species” in his M. ser. Racemosae formed a species complex called Meconopsis horridula complex (Xiao and Simpson 2015). Because these lacked clear species delimitations, we do not agree on specific ranks for most of the “species” in Grey-Wilson's (2014) M. sect. Racemosae. Thus, we did not highlight any species in Grey-Wilson's M. ser. Heterandrae in Fig. 3D. Nine outgroup species (accessions) were selected and sampled based on previous phylogenetic studies of Meconopsis (Yuan 2002) and Papaver (Carolan et al. 2006). Samples were collected from the wild, from the living collection in the Royal Botanical Garden at Edinburgh, and (with permission) from specimens in various herbaria. Species names, authorities, collection information, and sequence information are listed in Appendix 1. In addition, we included four Meconopsis accessions from Yuan's (2002) study, and downloaded their trnL-trnF spacer sequences from GenBank. Their vouchers and sequence information are also listed in Appendix 1. Genetic markers for our accessions that could not be successfully amplified, or for Yuan's (2002) accessions thatwere not available in GenBank,were coded as missing data (Appendix 1).
DNA Extraction, PCR, and Sequencing—Genomic DNA was extracted from silica-dried leaf materials or herbarium specimens using the DNeasy Plant Minikit (Qiagen, Valencia, California, USA). We chose the cpDNA marker trnL-trnF, which was shown to be phylogenetically informative in previous studies of Papaveroideae (Yuan 2002; Carolan et al. 2006).We also selected the cpDNA marker rbcL because it is commonly used for molecular dating in basal eudicot families (Wikström et al. 2001; Anderson et al. 2005; Bell et al. 2010) and also showed sequence variations in the species of Meconopsis we tested. Additionally, the cpDNA markers matK and ndhF were tested and selected because they were easy to amplify and significantly contributed to the resolution of the relationships at sectional level in Meconopsis. PCR amplification was carried out in 12 mL reaction volumes with 1–20 ng DNA, 1.0 unit of Taq polymerase (labmade, The University of Texas at Austin), 0.5X Failsafe Buffer B (Epicentre Biotechnologies, Madison, WI, USA), and 2.0 µmol/L primers. Forty-five PCR cycles were performed at 95°C for 30 sec, 50°C for 45 sec, and 72°C for 45 sec for each cycle. Internal primers were designed for amplifying herbarium samples. Primer pairs used are listed in Appendix 2. All of the PCR products were visualized on agarose gel containing Syber Safe DNA gel stain (Invitrogen, Eugene, Oregon, USA). Successfully amplified products were cleaned using ExoSap (Exonuclease I: New England Biolabs Beverly, MA, USA; Shrimp Alkaline Phosphatase: Progema, Madison, WI, USA) following the manufacturers' protocols. Cleaned PCR products were sequenced using an ABI 3730 DNA Analyzer at the Institute for Cell and Molecular Biology Core Facility at The University of Texas at Austin. Amplifying primers were used for sequencing. In addition, internal primers were also used for sequencing if the amplicon was longer than 900 base pairs (i.e. rbcL, matK, and ndhF).
Phylogenetic Analyses—Sequences were assembled in Geneious 5.5 ( http://www.geneious.com, Kearse et al. 2012), and aligned using Geneious Alignment (implemented in Geneious) with the default setting and 5 refinement iterations. Alignments were then reviewed and refined manually. Concatenated cpDNA data was analyzed using MrBayes v3.1.2 (Huelsenbeck and Ronquist 2005). Partition analysis was conducted for the combined cpDNA dataset with each cpDNA marker treated as a separate partition. The evolutionary models of nucleotide substitution were first selected by jModelTest (Posada 2008) under the Akaike information criterion (AIC), and we used the models most similar to the best fit models estimated by jModelTest and that were also available in MrBayes v3.1.2 for each gene partition: GTR + G for rbcL, GTR + I + G for ndhF, GTR + G for matK, and GTR + G for trnL-trnF dataset. Prior probability distributions on all parameters were set to the defaults. Twenty million generations were run using a Markov chain Monte Carlo method with four chains. Trees were collected every 100th generation. With 25% burn-in, a 50% majority rule consensus tree was calculated to generate a posterior probability (PP) for each node.
Results and Discussion
We obtained and analyzed 1756 nucleotide positions (with 358 variable sites) of matK, 1648 positions (with 231 variable sites) of ndhF, 1085 positions (with 588 variable sites) of trnL-trnF, and 1395 positions (with 55 variable sites) of rbcL sequences. The data are available from the Dryad Digital Repository: DOI:10.5061/dryad.1cr40 (Xiao and Simpson 2017). The recovered phylogenetic relationships within Meconopsis are illustrated in Fig. 3A, in which we show only the most closely related outgroup species, Papaver alpinum. The posterior probabilities were labeled above the branches on the resulting Bayesian consensus cpDNA tree. This phylogeny provided for the first time well resolved relationships among different subgroups of Meconopsis. Based on the resulting tree, we divided the genus into four monophyletic sections (Fig. 3A): Meconopsis sect. Meconopsis (PP = 1.00), M. sect. Aculeatae (PP = 0.95), M. sect. Primulinae (PP = 0.97), and M. sect. Grandes (PP = 0.82). Our new sections are color coded for easy examination of inconsistencies in earlier treatments (Fig. 3). Within each of our sections, we have not extensively studied species delimitations and reticulation patterns at the molecular level and are therefore reluctant to subdivide further each section before additional studies are conducted.
Our proposed Meconopsis sect. Meconopsis (e.g. M. paniculata, Fig. 2E) is significantly different from any previous treatment. Species in this section (highlighted in blue in Fig. 3B–D) were traditionally divided into two separate groups (i.e. Fedde's M. sect. Robustae and M. sect. Torquatae; Taylor's M. subsect. Eupolychaetia and M. subg. Discogyne; Grey-Wilson's M. subg. Meconopsis and M. subg. Discogyne). However, such treatments made Fedde's M. sect. Robustae, Taylor's M. subsect. Eupolychaetia, and Grey-Wilson's M. subg. Meconopsis all paraphyletic. Furthermore, none of the subdivisions in Taylor's (1934) M. subsect. Eupolychaetia or those in Grey-Wilson's (2014) M. subg. Meconopsis is monophyletic according to the phylogenetic result (Fig. 3A).
The grouping strategy of our Meconopsis sect. Aculeatae (e.g. M. sp and M. speciosa, Fig. 2C, D) is most similar to that of Taylor's (1934) M. ser. Aculeatae. Taylor's M. ser. Aculeatae is not monophyletic because it included M. sinuata and did not include M. delavayi (Fig. 3A, C), but most of its species share a common ancestor and the same chromosome number (2n = 56). Thus with a minor modification guided by the phylogenetic tree, we transformed Taylor's M. ser. Aculeatae to our monophyletic M. sect. Aculeatae.
The Meconopsis sect. Primulinae (e.g. M. bella, Fig. 2A) we proposed is somewhat similar to Grey-Wilson's M. sect. Cumminsia. Our M. sect. Primulinae includes both Grey-Wilson's M. sect. Cumminsia and M. sect. Bellae because of their phylogenetic relatedness as well as morphological consistency. It is notable that species in our M. sect. Primulinae (highlighted by pink in Fig. 3B–D) and M. sect. Aculeatae (highlighted by orange in Fig. 3B–D) were never clearly separated from each other in any previous classification. For example, in Fedde's (1909) M. sect. Primulinae, thetype species M. primulina is not related to the rest of the sectionwhich actually belong to our M. sect. Aculeatae. Taylor's (1934) M. ser. Aculeatae is polyphyletic due to the inclusion of Meconopsis sinuata, a species in our M. sect. Primulinae. Grey-Wilson's (2014) M. subg. Cumminsia, a polyphyletic group, combined both taxa, our M. sects. Aculeatae and Primulinae. However, our reconstructed phylogeny indicated that each of these two taxa has its own distinct evolutionary history, which was only reflected by our proposed classification. We will further discuss their morphological similarities and differentiation in the Taxonomic Treatment below.
Our Meconopsis sect. Grandes (e.g. M. grandis, Fig. 2B) closely corresponds to Feddes's M. sect. Grandes or Grey-Wilson's (2014) M. subg. Grandes. Grey-Wilson (2014) further divided this group into M. sects. Grandes and Simplicifoliae, a method similar to that of Taylor's (1934). However, the position of M. simplicifolia on our cpDNA phylogeny does not support their treatments.
In summary, each of the new sections we proposed is a monophyletic group based on the reconstructed cpDNA phylogeny. Traditional Meconopsis classifications employed selected morphological similarities and were found largely inconsistent with the monophyly of the grouping in light of our resulting tree. Monophyly, an objective standard in our method, can minimize artificial preference and can be used to fairly evaluate the incongruence among traditional taxonomies. The value of the previous treatments was also incorporated into our system by retaining the previous taxonomic units that are supported by the phylogenetic result. We defined each section by the shared morphological characteristics and cytological patterns of its contained species, which will be described and discussed below along with the key to each section. As mentioned earlier, species delimitations have not been systematically investigated. We have applied phylogenetic species concept consistently throughout our study and provided an “Included Species” list for each section based on our observation, analyses, and best estimation. However, these tentative lists should be updated when future investigations are available.
Herbs, monocarpic or polycarpic, with yellow or white latex; roots taproots or fibrous, or both. Leaves cauline and in basal rosettes or only in basal rosettes with the rosette leaves evergreen or senescing in winter, petiolated or sessile; lamina ovate, obovate, elliptic, oblong, oblanceolate, nearly linear, pinnatifid, pinnatisect, rarely bipinnatifid or bipinnate; margin entire, serrate, sinuate, lobed, or deeply divided. Flowers solitary, borne on basal scapes, or flowers arranged in raceme-like or panicle-like cymes; bracts leafy or reduced. Calyx caducous; sepals normally 2, occasionally 3 or 4 (particularly in terminal flowers). Corolla large and showy, often saucer- to bowl-shaped; petals commonly 4–8, rarely more than 12. Stamens numerous; filaments filiform, occasionally dilated; anthers commonly yellow to orange, but changing colors with age. Ovary superior, subspherical, ovate, or obovate to narrowly subcylindric; locule normally 1 with 3–6 fused carpels, ovules numerous; style often distinct but short, usually less than 1/2 of the ovary length, occasionally inconspicuous, sometimes basally expanding into a disk covering the top of ovary; stigma usually capitate or clavate, occasionally star-shaped with 3–9 stigmatic rays variously decurrent on style. Fruit a subspherical, ovate, obovate, or elliptic to subcylindrical capsule, dehiscing septicidally by 3–9 valves from apex to base usually for a short distance or occasionally to near the base. Seeds many, reniform, falcate-oblong, or elliptic-oblong; testamost commonly reticulated or corrugated, sometimes papillose. Chromosome number: 2n = 14, 22, 28, 56, 74, 76, 82, 84, 118, 120, 164; x = 7.
Key to Sections
1. Monocarpic perennials. Basal rosette persistent through winter 1. M. sect. Meconopsis
1. Polycarpic perennials, or monocarpic biennials or perennials. Leaves deciduous during the winter 2
2. Plants usually densely hirsute with barbellate trichomes, often bearing a dense tuft of persistent leaf bases interspersed with dense barbellate bristles. Root-system fibrous or with slender taproot, or with a combination of the two 4. M. sect. Grandes
2. Plants glabrous to densely bristly with non-barbellate trichomes, often with simple sharp bristles, lacking persistent leaf bases or, if leaf-bases present, plants never densely bristly at the base. Root consisting of a taproot only 3
3. Plants armed with dense to sparse sharp bristles, or rarely subglabrous. Taproot usually stout and elongated (exceeding 7 cm in length). Upper cauline leaves noticeably reduced in size relative to lower cauline leaves and basal leaves. Petals normally blue or purple-violet, rarely red, white, or yellow; when blue, usually more than 6 flowers per plant 4
4. Basal leaves usually lanceolate to elliptic-oblanceolate or oblanceolate; entire, pinnatilobate, or rarely pinnatisect or pinnaticompound at the margin; basally attenuating into petiole 2. M. sect. Aculeatae
4. Basal leaves narrowly oblong to narrowly lanceolate; more or less sinuate at the margin; tapering at the base into a long flattened Petiole 3. Meconopsis sinuata (M. sect. Primulinae)
3. Plants usually sparsely vestitured with weak trichomes or subglabrous. Taproot usually slender (less than 7 cmin length). Upper cauline leaves usually similar in size to the lower leaves. Petals usually pale blue to pale purple-blue, sometimes yellow or white; but never bright purple-violet; when blue, usually fewer than 5 flowers per plant 3. M. sect. Primulinae
1. Meconopsis sect. Meconopsis. Meconopsis subg. Meconopsis, Gen. Meconopsis: 43. 2014.—TYPE: M. regia G. Taylor, J. Bot. 67: 259. 1929.
Meconopsis sect. Robustae Fedde, Pflanzenr. (Engler) 40 (IV.104): 267. 1909.—TYPE: M. robusta Hook. f. & Thomson.
Meconopsis sect. Torquatae Fedde, Pflanzenr. (Engler) 40 (IV.104): 265. 1909. Meconopsis subg. Discogyne G. Taylor, Account Gen. Meconopsis: 107. 1934.—TYPE: M. torquata Prain.
Meconopsis subsect. Eupolychaetia G. Taylor, Account Gen. Meconopsis: 30. 1934.—TYPE: M. paniculata Prain.
Meconopsis subg. Discogyne (G. Taylor) Grey-Wilson, Gen. Meconopsis: 44. 2014.—TYPE: M. discigera Prain.
Monocarpic perennials with taproots; 0.3–2.5 m tall at anthesis, frequently more than 1 m tall. Stems and leaves hirsute or pubescent, commonly with barbellate or branched trichomes. Leaves retained in an evergreen dense basal rosette for a few years before flowering. Leaf blades oblanceolate or elliptic to oblong, pinnatifid or pinnatisect, with serrate, lobed or divided margins, up to 60 cm long. Inflorescence a raceme-like or panicle-like cyme most commonly with 1–5 (up to 15) flowered cymules; bracts leafy or reduced. Petals usually 4; commonly yellow, red, blue to violet. Ovary ellipsoid to oblong, usually setose, rarely glabrous; style distinct and short, occasionally expanding at the base into a disk surmounting the ovary; stigma normally capitate. Capsules oblong to ellipsoid, or ovoid to ellipsoid. Chromosome number 2n = 56, rarely 2n = 28.
Included Species—Meconopsis autumnalis P. A. Egan; M. chankheliensis Grey-Wilson; M. dhwojii G. Taylor; M. discigera Prain; M. ganeshensis Grey-Wilson; M. gracilipes G. Taylor; M. manasluensis P. A. Egan; M. napaulensis DC.; M. paniculata Prain; M. pinnatifolia C. Y. Wu & H. Chuang ex L. H. Zhou; M. regia G. Taylor; M. robusta Hook. f. & Thomson; M. simikotensis Grey-Wilson; M. staintonii Grey-Wilson; M. superba King ex Prain; M. taylorii L. H. J. Williams; M. tibetica Grey-Wilson; M. torquata Prain; M. violacea Kingdon-Ward; M. wallichii Hook.; M. wilsonii Grey-Wilson.
Species in this section share characters of a perennial monocarpic habit and retention of a dense evergreen rosette of leaves for a few years before flowering, and the latter is absent in other sections in the genus. Most species in this section are tall plants (usually more than 50 cm and up to 2.5 m tall when mature). However, a subgroup (highlighted by red branch in Fig. 3A) in the section contains species usually less than 50 cm tall and characterized by the style expanding into a flat disc at the base. This unique disc structure was emphasized by all of the previous classifications and species with the disc structure had always been grouped into a distinct unit (i.e. Fedde's M. sect. Torquatae, Taylor's M. subg. Discogyne, or Grey-Wilson's M. subg. Discogyne). Our M. sect. Meconopsis for the first time recognized and put emphasis on the phylogenetic relatedness instead of relying on one morphological character to perform infrageneric division.
2. Meconopsis sect. Aculeatae Fedde, Pflanzenr. (Engler) 40 (IV.104): 255. 1909. Meconopsis ser. Aculeatae (Fedde) G. Taylor, Account Gen. Meconopsis: 78. 1934.—TYPE: M. aculeata Royle, Ill. Bot. Himal. Mts. [Royle] 1: 67. 1839.
Meconopsis ser. Delavayanae G. Taylor, Account Gen. Meconopsis: 76. 1934.—TYPE: M. delavayi Franch.
Meconopsis sect. Racemosae C. Y. Wu & H. Chuang, Acta Bot. Yunnan. 2(4): 374. 1980.—TYPE: M. racemosa Maxim.
Meconopsis sect. Forrestii C. Y. Wu & H. Chuang, Acta Bot. Yunnan. 2(4): 376. 1980. Meconopsis sect. Forrestianae Grey-Wilson, Gen. Meconopsis: 46. 2014.—TYPE: M. forrestii Prain.
Meconopsis sect. Impediatae Grey-Wilson, Gen. Meconopsis: 46. 2014.—TYPE: M. impedita Prain.
Monocarpic biennials, or perennials with taproots; up to 1 m tall at anthesis. Stems and leaves aculeate with simple non-barbellate trichomes, or occasionally subglabrous. Leaves senescing and deciduous during the winter. Leaf lamina ovate, oblanceolate to oblong, elliptic to oblong, pinnatifid, or pinnatisect with margins normally entire, lobed or divided, up to 25 cm long. Flowers borne on basal scapes, or in bracteate or ebracteate raceme-like cymes, or both. Petals 4–12, commonly blue or violet, rarely white, yellow, red or dark red. Ovary subspherical, or ellipsoidal to narrowly subcylindric, densely covered by sharp bristles to glabrous; style distinct; stigma capitate or clavate. Capsules oblong, ovoid, obvoid to narrowly subcylindrical. Chromosome number 2n = 56, rarely 2n = 14.
Included Species—Meconopsis aculeata Royle; M. bikramii Aswal (a rare species collected from Himalaya Pradesh in India, placed in this section because the original author suggested it is allied to M. aculeate; no material was available for examination and its palmately lobed lower cauline leaves cast doubt on its affinity); M. concinna Prain; M. delavayi Franch. Ex Prain; M. forrestii Prain; M. georgei G. Taylor; M. henrici Bureau & Franch.; M. horridula Hook. f. & Thomson; M. impedita Prain; M. lancifolia Franch.; M. latifolia Prain; M. muscicola Tosh. Yoshida, H. Sun & Boufford; M. neglecta G. Taylor; M. pseudovenusta G. Taylor; M. pulchela Tosh. Yoshida, H. Sun & Bouford; M. venusta Prain; M. yaoshanensis Tosh. Yoshida, H. Sun & Boufford.
As indicated by its name, Meconopsis sect. Aculeatae is characterized by sharp-pointed bristles on leaf and stem surface. Species in this section most commonly bear blue flowers (e.g. Fig. 2D) or purple-violet flowers (e.g. Fig. 2C). The flower colors of this section are indicated by branch color in Fig. 3A. The blue-flowered species, form a basal grade to the species with purple-violet flower which suggests that purpleviolet is a derived characteristic in this section (Fig. 3A). Species with purple-violet flowers tend to be less robust with shorter stature and less dense bristles than the blue-flowered species, and were once believed to resemble the also short-statured species Meconopsis primulina (the type species of our M. sect. Primulinae). There is overlap at the characters of indumentum, stature, and leaf shape between the species with purple-violet flowers in ourM. sect. Aculeatae and those in our M. sect. Primulinae, but the two sections can be easily separated by the petal color (Fig. 2A, C). Species in our M. sect. Primulinae do not have the deep purple-violet color of those in M. sect. Aculeatae. Additionally, our M. sect. Primulinae species are distributed mainly in the east Himalaya while the purpleviolet flowered species in M. sect. Aculeatae are distributed mainly in the Hengduan Mountains. The phylogenetic evidence also strongly supports the separation of two genetically distant clades of M. sects. Aculeatae and Primulinae.
3. Meconopsis sect. Primulinae Fedde, Pflanzenr. (Engler) 40 (IV.104): 259. 1909. Meconopsis ser. Primulinae (Fedde) G. Taylor, Account Gen. Meconopsis: 71. 1934. Meconopsis sect. Cumminsia (Prain) Grey-Wilson, Gen. Meconopsis: 45. 2014.—TYPE: M. primulina Prain, J. Asiat. Soc. Bengal, Pt. 2, Nat. Hist. 64: 319. 1895.
Meconopsis sect. Bellae Fedde, Pflanzenr. (Engler) 40 (IV.104): 261. 1909. Meconopsis ser. Bellae (Fedde) G. Taylor, Account Gen. Meconopsis: 103. 1934.—TYPE: M. bella Prain.
Monocarpic or polycarpic perennials with taproots; frequently less than 25 cm tall, rarely exceeding 50 cm at anthesis (except in Meconopsis sinuata that ranges from 30–65 cm in height). Leaves and stems most frequently sparsely vestitured with weak non-barbellate trichomes or subglabrous, rarely aculeate with sharp bristles. Leaves senescing during the winter; lamina variable in shape and margin type, frequently less than 7 cm long (rarely exceeding 15 cm). Flowers born on basal scapes or arranged in simple cyme with 2–8 flowers in axils of upper cauline leaves. Petals 4–8, commonly pale blue to pale purple-blue, sometimes pink or yellow. Ovary usually ellipsoid to oblong, or narrowly ellipsoidal to narrowly subsylindric sometimes subspherical, usually subglabrous or with sparse bristles; style distinct, usually short, but sometimes longer than the ovary; stigma capitate. Capsules obovoid or narrowly obovoid to narrowly subcylindrical. The only known chromosome number (M. bella) is 2n = 22.
Included Species—Meconopsis argemonantha Prain; M. bella Prain; M. florindae Kingdon-Ward; M. lyrata (H. A. Cummins & Prain) Fedde; M. primulina Prain; M. sinuata Prain; M. wumungensis K. M. Feng; M. zangnanensis L. H. Zhou.
Species in this section tend to have a dwarf and slender aspect with short root and weak stem as well as a brittle and sparse indumentum. Although blue flowers are common in this section, most species tend to be more pale or faded than the bright blue color present in other sections. Meconopsis sinuata in this section is morphologically distinct being taller than other species and armed with dense spines. Unsurprisingly, M. sinuata used to be grouped with species in our M. sect. Aculeatae (e.g. in M. sect. Aculeatae in Fedde 1909, or M. ser. Aculeatae in Taylor 1934). However, it is easy to distinguish living plants of M. sinuata from species ofM. sect. Aculeatae by the shape of ovary and style and especially by the leaf morphology (see the Key to Sections).
Meconopsis bella, another species in our M. sect. Primulinae, was traditionally placed in its own section (i.e.M. sect. Bellae in Fedde 1909 and Grey-Wilson 2014) or series (i.e. M. ser. Bellae in Taylor 1934), all based on its unique characteristic of a bellshaped ovary. However, the general morphology (e.g. height, leaf shape, flower arrangement, petal color) of M. bella matches that of our M. sect. Primulinae and phylogenetic result supports the inclusion of M. bella in M. sect. Primulinae, which indicate that the unique feature of having a bell-shaped ovary does not warrant special status.
4. Meconopsis sect. Grandes Fedde, Pflanzenr. (Engler) 40 (IV.104): 262.1909. Meconopsis ser. Grandes (Fedde) G. Taylor, Account Gen. Meconopsis: 56. 1934. Meconopsis subg. Grandes (Fedde) Grey-Wilson, Gen. Meconopsis: 44. 2014.—TYPE: M. grandis Prain, J. Asiat. Soc. Bengal, Pt. 2, Nat. Hist. 64: 320. 1895.
Meconopsis ser. Simplicifoliae G. Taylor, Account Gen. Meconopsis: 49. 1934. Meconopsis sect. Simplicifoliae (G. Taylor) C. Y. Wu & H. Chuang, Acta Bot. Yunnan. 2(4): 375. 1980.—TYPE: M. simplicifolia (D. Don) Walp.
Monocarpic or polycarpic perennials with taproots or a fibrous root system or the combination; up to 1.5 m tall at anthesis. Leaves and stems hirsute with barbellate trichomes. Leaves senescing and deciduous during the winter; lamina frequently ellipic to narrowly oblanceolate, narrowly elliptic or narrowly lanceolate, longitudinally nerved and entire at the margin, sometimes oblanceolate to lanceolate- or ellipticoblong with entire, serrate or lobed margins and pinnate venation; up to 30 cm long; uppermost cauline leaves sometimes aggregated in a false whorl and bearing flowers in their axils. Flowers normally fewer than 8, solitary on basal scapes, or in the axil of cauline leaves. Petals most commonly 4, or up to 10; blue, violet, yellow or red. Ovary usually ellipsoid to oblong, pubescent to hispid; styles commonly distinct, sometimes inconspicuous; stigma capitate or subclavate or star-shaped with 3–9 stigmatic rays variously decurrent relative to the style (in this condition, the style more or less resembling a star-shaped column). Capsules oblong to ellipsoid. Chromosome numbers from 2n = 74 to 164, most frequently 84.
Included Species—Meconopsis betonicifolia Franch.; M. biloba L. Z. An, Shu Y. Chen & Y. S. Lian; M. grandis Prain; M. integrifolia (Maxim.) Franch.; M. punicea Maxim.; M. quintuplinervia Regel; M. simplicifolia (D. Don) Walp.; M. sherriffii G. Taylor.
The members of this section are easy to identify by their long and dense barbellate trichomes and the presence of a fibrous root system. Species in this section are more popular than those of other sections in Scottish gardens, not only for the brilliant colors of their large and showy flowers (Fig. 2B), but also for their easy cultivation and (frequent) polycarpic habit. This section also has the highest chromosome numbers among the genus with 2n = 74, 76, 82, 84, 118, 120, 164.
Cathcartia Chelidonifolia (Bureau & Franch.) Grey-Wilson, Gen. Meconopsis: 374. 2014.—LECTOTYPE (designated here): CHINA. Sichuan: Ta-Tsien-Lou [Kangding], P. G. E. Bonvalot & Prince Henri d'Orléans s. n. (P, barcode 00739028; isolectotype: P, barcode 00739029).
Cathcartia Oliveriana (Franch. ex Prain) Grey-Wilson, Gen. Meconopsis: 376. 2014.—LECTOTYPE (designated here): CHINA. Hubei, Henry 6863 (K, barcode K000653215; isolectotypes: BM, K).
We thank the reviewers and the editors for suggestions that helped to significantly strengthen the manuscript. We also acknowledge the herbaria and curators of Royal Botanical Garden at Edinburgh, Harvard University, University of Texas, and Kunming Institute of Botany, Chinese Academy of Science for their hospitality and help. Dr. David Boufford (Harvard Herbaria), Dr. Alan Elliott (RBGE), and Dr. Paul Egan (Trinity College Dublin, the University of Dublin) have kindly shared experimental materials and photos, we greatly appreciate their support.
- Anderson, C. L., K. R. Bremer, and E. M. Friis. 2005. Dating phylogenetically basal eudicots using rbcL sequences and multiple fossil reference points. American Journal of Botany 92: 1737–1748. Google Scholar
- Bell, C.D., D. E. Soltis, and P. S. Soltis. 2010. The age and diversification of the angiosperms re-revisited. American Journal of Botany 97: 1296–1303. Google Scholar
- Carolan, J. C., I. L. Hook, M. W. Chase, J. W. Kadereit, and T. R. Hodkinson. 2006. Phylogenetics of Papaver and related genera based on DNA sequences from ITS nuclear ribosomal DNA and plastid trnL intron and trnL-F intergenic spacers. Annals of Botany (Oxford) 98: 141–155. Google Scholar
- Chuang, H. 1981. The systematic evolution and the geographical distribution of Meconopsis Vig. Acta Botanica Yunnanica, Kunming 3: 139–146. Google Scholar
- Fedde, F. 1909. Papaveraceae-Hypecoideae et Papaveraceae-Papaveroideae. Pp. 244–271 in Das Pflanzenreich (Engler) iv vol. 104, ed. A. Engler. Leipzig: Wilhelm Englemmann. Google Scholar
- Fedde, F. 1936. Papaveraceae. Pp. 98–105 in DieNatürlichen Pflanzenfamilien 2 vol. 17b, ed. A. Engler et al. Berlin: Duncker and Humblot. Google Scholar
- Grey-Wilson, C. 2000. Poppies: A guide to the poppy family in the wild and in cultivation. Portland, Oregon : Timber Press. Google Scholar
- Grey-Wilson, C. 2012. Proposal to conserve the name Meconopsis (Papaveraceae) with a conserved type. Taxon 61: 473–474. Google Scholar
- Grey-Wilson, C. 2014. The genus Meconopsis blue poppies and their relatives. London: Royal Botanical Garden, Kew. Google Scholar
- Huelsenbeck, J. and F. Ronquist. 2005. MrBayes 3.1 Manual. Florida State University: School of Computational Science. Google Scholar
- Kadereit, J. W., C. D. Preston, and F. J. Valtueña. 2011. Is Welsh poppy, Meconopsis cambrica (L.) Vig. (Papaveraceae), truly a Meconopsis? New Journal of Botany 1: 80–88. Google Scholar
- Kearse, M., R. Moir, A. Wilson, S. Stones-Havas, M. Cheung, S. Sturrock, S. Buxton, A. Cooper, S. Markowitz, C. Duran, T. Thierer, B. Ashton, P. Mentjies, and A. Drummond. 2012. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28: 1647–1649. Google Scholar
- Kingdon-Ward, F. 1926. Notes on the genus Meconopsis, with some additional species from Tibet. Annals of Botany (Oxford) 40: 535–546. Google Scholar
- Kingdon-Ward, F. 1935. A sketch of the geography and botany of Tibet, being materials for a flora of that country. Journal of the Linnean Society of London. Botany 50: 239–265. Google Scholar
- Kumar, S., S. M. Jeelani, S. Rani, R. C. Gupta, and S. Kumari. 2013. Cytology of five species of subfamily Papaveroideae from the Western Himalayas. Protoplasma 250: 307–316. Google Scholar
- Liu, Y., Y. Liu, F. Yang, and X. Wang. 2014. Molecular phylogeny of Asian Meconopsis based on nuclear ribosomal and chloroplast DNA sequence data. PLoS One 9: e104823. Google Scholar
- McNeill, J., F. R. Barrie, W. R. Buck, V. Demoulin, W. Greuter, D. L. Hawksworth, P. S. Herendeen, S. Knapp, K. Marhold, J. Prado, W. F. Prud'homme Van Reine, G. F. Smith, J. H. Wiersema, and N. J. Turland. 2012. International Code of Nomenclature for algae, fungi, and plants (MelbourneCode). Regnum Vegetabile 154. Königstein: Koeltz Scientific Books. Google Scholar
- Posada, D. 2008. jModelTest: phylogenetic model averaging. Molecular Biology and Evolution 25: 1253–1256. Google Scholar
- Prain, D. 1895. Some additional Papaveraceae. Journal of the Asiatic Society of Bengal. Part 2. Natural History 64: 303–327. Google Scholar
- Prain, D. 1906. A review of the genera Meconopsis and Cathcartia. Annals of Botany (Oxford): 323–370. Google Scholar
- Prain, D. 1915. Some additional species of Meconopsis. Bulletin of Miscellaneous Information, Royal Gardens , Kew : 129–177. Google Scholar
- Ratter, J. 1968. Cytological studies in Meconopsis. Notes from the Royal Botanic Garden Edinburgh 28: 191–200. Google Scholar
- Sang, T., D. Crawford, and T. Stuessy. 1997. Chloroplast DNA phylogeny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae). American Journal of Botany 84: 1120–1136. Google Scholar
- Taberlet, P., L. Gielly, G. Pautou, and J. Bouvet. 1991. Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology 17: 1105–1109. Google Scholar
- Taylor, G. 1934. An account of the genus Meconopsis. London: New Flora and Silva. Google Scholar
- Viguier, L. A. 1814. Histoire naturelle, médicale et économique, des pavots et des argémones. Montpellier: J. Martelainé. Google Scholar
- Wikström, N., V. Savolainen, and M. W. Chase. 2001. Evolution of the angiosperms: calibrating the family tree. Proceedings of the Royal Society of London. Series B, Biological Sciences 268: 2211–2220. Google Scholar
- Wu, Z. Y. and H. Chuang. 1980. A study on the taxonomic system of the genus Meconopsis. Acta Botanica Yunnanica, Kunming 2: 371–381. Google Scholar
- Xiao, W. 2013. Molecular systematics ofMeconopsisVig. (Papaveraceae): taxonomy, polyploidy evolution, and historical biogeography from a phylogenetic insight. Ph.D. dissertation. Austin: The University of Texas at Austin. Google Scholar
- Xiao, W. and B. Simpson. 2014. The role of the triploid hybrid in the evolution of Meconopsis (Papaveraceae): a preliminary study on ancient polyploid and hybrid speciation. Lundellia 17: 5–17. Google Scholar
- Xiao, W. and B. Simpson. 2015. Phylogenetic analysis of Meconopsis (Papaveraceae) and evaluation of two controversial taxonomic species. Lundelia 18: 14–18. Google Scholar
- Ying, M., H. Xie, Z. Nie, Z. Gu, and Y. Yang. 2006. A karyomorphological study on four species of Meconopsis Vig. (Papaveraceae) from the Hengduan Mountains, SW China. Caryologia 59: 1–6. Google Scholar
- Yuan, C. 2002. The phylogeny, systematics and biogeography of Meconopsis Vig. (Papaveraceae) and Craigia W. W. Sm. & W. E. Evans (Tiliaceae). Ph.D. dissertation. Guangzhou: Zhong Shan University. Google Scholar
Appendix 1. Voucher and sequence information (Accession number; species name; (Collecting) country: Subdivision; voucher (herbarium); GenBank ID for matK, ndhF, trnL-trnF, rbcL. Accessions beginning with “X” were analyzed and sequenced in this study, and accessions beginning with “Y” were published by Yuan (2002). “-”, denotes a missing sequence).