^(J|k Missouri D /^vl- n i /^i o 1 Botanical Garden A Journal for Botanical Nomenclature 2012 Volume 22, Number 1 July 2012 Novon, A Journal for Botanical Nomenclature from the Missouri Botanical Garden The mission of the Missouri Botanical Garden is to discover and share knowledge about plants and their environment in order to preserve and enrich life. Novon publishes short articles whose primary purpose is the establishment of nomencla- ture in vascular plants and bryophytes. All articles are peer-reviewed by qualified, indepen- dent reviewers. Manuscripts must fully state and justify the reasons for proposing nova. These may include detailed comparisons with similar taxa, short keys to similar taxa, illustrations to similar taxa, and mechanical nomenclature reasons, among others. Manuscripts whose primary purpose is other than establishment of new nomenclature, which usually are longer manuscripts, cannot be accepted for review. These include reviews, revisions, monographs, or other papers that incidentally include nova. Manuscripts must follow the guidelines in the Checklist for Authors. The Checklist may be downloaded from the Garden’s web site, www.mbgpress.info, or authors may contact the editor at novon@mobot.org to request a copy. Novon will not knowingly accept manuscripts that have been simultaneously submitted to other journals for consideration or previously published in some form elsewhere. Victoria C. Hollowell Allison M. Brock Tammy Charron Cirri R. Moran Ihsan A. Al-Shehbaz Carmen Ulloa Ulloa George Yatskievych Kanchi N. Gandhi Nicholas J. Turland Roy E. Gereau Associate Editor Associate Editor Press Coordinator Consulting Editor Consulting Editor Consulting Editor Nomenclature Consultant Nomenclature Consultant Latin Editor Novon is included in the subscription price of the Annals of the Missouri Botanical Garden. Price for 2012: $180 per year U.S.A.; $190 Canada and Mexico; $215 all other countries. Four issues per volume. Postal address: Subscriptions: Annals of the Missouri Botanical Garden, P.0. Box 1897, Law- rence, KS 66044-8897; Manuscript submissions: Novon, Missouri Botanical Garden, P.0. Box 299, St. Louis, MO 63166-0299. Editorial queries: novon@mobot.org Orders for back issues: mbgpress@mobot.org Web site: http://www.mbgpress.info Missouri Botanical Garden 2012 © Novon is printed on paper that meets the re- quirements of ANSI/NISO Z39.48-1992. Novon (ISSN 1055-3177) is published quarterly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Periodicals postage paid at St. Louis, MO and additional mail- ing offices (USPS #006-777). POSTMASTER: Please send address changes to Annals of the Missouri Botanical Garden, P.0. Box 1897, Lawrence, KS 66044-8897. Information on the contents of Novon is present- ed in SciSearch®, Research Alert®, Current Con- tents®/Agriculture, Biology, and Environmental Sciences, APT Online, the CAB Abstract/Global Health database, and the ISI® database. The full-text of Novon is available online through BioOne™ (http://www.bioone.org). Volume 22 NO VON Number 1 W 2012 Passiflora cacao (Passifloraceae), a New Species from Southern Bahia, Brazil Luis Carlos Bemacci Centro de Recursos Geneticos Vegetais e Jardim Botanico, Instituto Agronomico de Campinas, Caixa Postal 28, 13001-970 Campinas, Sao Paulo, Brazil, bemacci@iac.sp.gov.br Margarete Magalhaes Souza Universidade Estadual de Santa Cruz, Departamento de Ciencias Biologicas, Pavilhao Jorge Amado, Rodovia Iheus-Itabuna, Km 16, Salobrinho 45662-900 Ilheus, Bahia, Brazil Abstract. Passiflora cacao Bemacci & M. M. Souza is described. The Passifloraceae has a limited known distribution area in southern Bahia, Brazil. This new species is not very abundant and occurs in a mountainous area associated with dense ombrophi- lous forest noteworthy for its great biodiversity, including many species of hummingbirds. Passiflora cacao is similar to P. edulis Sims, but it can easily be recognized by the cylindrical hypanthium and winged seeds. In addition, the flowers are usually smaller, with inner corona and outer corona series greater in Passiflora edulis Sims, the passion fruit, is the species in the Passifloraceae of greatest economic importance, being grown in many parts of the world (Meletti & Maia, 1999; Meletti et al., 2005; Bemacci et al., 2008). However, its close phylogenetic relatives are unclear (Killip, 1938; Muschner et al., 2003; Yockteng & Nadot, 2004). The Passifloraceae has ca. 500 to 600 described species (Escobar, 1988; Bemacci et al., 2005), but there could be 650 or even 750 species (Cronquist, 1981; Bemacci et al., 2005; Feuillet & MacDougal, 2007). The family’s distribution is concentrated in the tropical regions of America and Africa, and its largest genus, Passiflora L., is clearly American, with doi: 10.3417/2010043 a great number of species concentrated in South America, especially in the Andes (Ocampo et al., 2007) . Morphologically, Passiflora incamata L. is the species most similar to P. edulis, and there are even taxonomic treatments in which the taxa are consid- ered conspecific (Killip, 1938; Bernacci et al., 2008). While P. incamata is native to North America, occurring in the United States, P. edulis is native to South and Central America (Killip, 1938; Cervi, 1997). Possibly, because of the broad natural distribution of P. edulis, the species presents considerable morphological variation associated with several scientific names attributed at the species level (Killip, 1938; Cervi, 1997; Bemacci et al., 2008) . However, this morphological variation does not present any pattern that can be associated with geographic distribution, and several morphological characters vary independently (Killip, 1938; Cervi, 1997; Bemacci et al., 2008). To study the Passifloraceae of Minas Gerais, the first author, together with A. C. Cervi and collabo- many regions of that state. In a complementary study intended to better understand species not often collected, the authors of this article examined material from regions adjacent to the state of Bahia. Novon 22: 1-7. Published on 10 July 2012. Volume 22, Number 2012 Bernacci & Souza Passiflora (Passifloraceae) from Brazil Novon considered Least Concern (LC) for the risk of extinction and additionally because it was found in a conservation unit (RPPN). Attempts at cultivation have not been successful beyond the FI generation. Etymology. The epithet refers to cacao (or cacaueiro, as known in Brazil), a plant of interest and historical importance in the agricultural economy of the southern region of Bahia (the area where this Novon the hypanthium, as seen in P. coccinea Aubl. (Fischer & Leal, 2006), not because of its length. In the species of the series Passiflora, the length of between 0.17 and 0.48 (Bemacci et al., 2003), while as a rule (according to Killip [1938]) this relation is greater than 0.75 for species in section Tacsonioides DC. (subgenus Passiflora, supersection Stipulata Feuillet & J. M. MacDougal, which, however, includes only species with broad, sheetlike, and strongly asymmetric stipules [Feuillet & MacDougal, 2003] ). Passiflora luetzelburgii Harms, which has a tubular hypanthium and linear stipules (Vitta & Bemacci, 2004), was included in the subgenus Tacsonioides (DC.) Killip (Killip, 1938) but is included now in the series Setaceae Killip ex Cervi (which typically has species with few series in the corona) of the section Passiflora (Ulmer & MacDou- gal, 2004). Passiflora luetzelburgii clashed with the other species in section Tacsonioides by stipule characteristics. Passiflora trintae Sacco (1968) and P. fameyi Pessoa & Cervi (1992) are species accepted for the series Passiflora and were recently described. Passiflora caatingae L. K. Escobar (1989) was synonymized to P. trintae (Cervi, 1995). Including those species previously attributed to Passiflora ser. Serratifoliae Killip ex Cervi, a total of 13 species have been accepted within Passiflora ser. Passiflora (MacDougal & Feuillet, 2004; Ulmer & MacDougal, 2004) . Therefore, with the description of P. cacao, 14 species of the series Passiflora are now recognized, of which the majority, nine species, occur natively in Brazil. In all of these, with the exception of P. recurva Mast., the margin of the leaf blade varies from serrated to subentire and is only entire in P. recurva. The blade margins may sometimes vary within the same species, as in P. cincinnata Mast., in which it varies from serrate to crenate-serrate and subentire. Passiflora trintae has red flowers and an elongated hypanthium (cylindrical and campanulate in shape), which are uncommon characteristics for its series Passiflora but similar to P. luetzelburgii and P. racemosa Brot. (supersection Stipulata) in presenting reddish flowers (Killip, 1938; Vitta & Bemacci, pollination by hummingbirds, but these traits could have evolved independently. It is worth highlighting that where P. cacao occurs there are reports of many species of hummingbirds (cf. WikiAves, 2008). The elongated hypanthium of the new species could be an indication of its pollination by hummingbirds, although its flowers lack red coloration. Such species as P. cacao raise interesting questions about floral Paratypes. BRAZIL. Bahia: Itubera, Assentamento Lucas Dantas, litoral sul, mata ciliar, 13°43'S, 39°08'W, 3 Sep. 2001 (fr.), M. L. Guedes, D. L. Santana, D. M. Loureiro & L J. Alves 9448 (ALCB, CEPEC); Camacan, RPPN Serra Bonita, Mata Atlantica, 10 Dec. 2004 (A.), L Becker 3898 (IAC); 15°23'01.2"S, 39°34'06.9"W, 10 Dec. 2004 (fl.), L. C. Bemacci, M. M. Souza, N. E. Pereira, A. S. Maia & V. 0. Becker 3900 (IAC), 7 Mar. 2006 (immature fr.), R. Tsuji, L C. Bernacci & E. F. Souto 1188 (HPL, IAC). P> Tan 20(X. ill i. / ( P», > / u III)! \W( l l'( B) 50 Mar. 2007 (fl.), L. C. Bemacci 4405 (IAC); 10 Mar. 2008 (fl./fr.), L. C. Bemacci 4492 (IAC); (fl.), L. C. Bernacci 4494 (IAC, K); 18 Mar. 2008 (fl./fr.), L C. Bernacci 4500 (IAC, RB); 6 Mar. 2009 (fl.), L C. Bemacci 4677 (HUEFS, IAC). Acknowledgments. We thank Eduardo G. Gonsalves (Inhotim Cultural Institute, Bmmadinho, Minas Gerais) and Roy Gereau (MO) for their help with the Latin diagnosis; to Karolina Pastorek, illustrator and environmental manager, for her careful drawing of Passiflora cacao-, to the Universidade Estadual de Santa Cmz; to Instituto Plantamm, Sao Paulo, for funding that, among other things, allowed us to locate and observe the new species in nature and under cultivation; to IAC for enabling the collecting expeditions: and. finally, to Clemira Becker and Vitor 0. Becker, researchers at — and enthusiasts of — RPPN Serra Bonita, for maintaining the natural area and enabling our collection and observation of P. cacao. Thanks also to Andre Amorim, curator of CEPEC, for allowing us to examine the collection; to Victoria C. Hollowell, scientific editor; Stephen Tillett, reviewer; an anon- ymous reviewer; Jennifer Dropkin, Allison M. Brock, and Tammy Charron, editors, for relevant suggestions and assistance during the process of publication. Amorim, A. M. Flora da RPPN Serra Bonita, 1 , accessed 26 October 2011. Quesnelia (Bromeliaceae: Bromelioideae) from the At- lantic rain forest of Bahia, Brazil. Brittonia 61: 14-21. C. Clifton & A. M. V. Carvalho. 2005. The vascular 21: 1727-1752. Amorim, A. M.,J.G.Jardim,M.M.M. Lopes, P.Fiaschi,R. A. X. Borges, R. 0. Perdiz & W. W. Thomas. 2009. Angio- Bahia, Brasil. Biota Neotrop. 9(3). < http://www.biotaneotropica. Further Nomenclatural Action for the Cypresses (Cupressaceae) David J. de Laubenfels Department of Geography, Syracuse University, Syracuse, New York 13244, U.S.A. ddelaube@verizon.net Chad E. Husby* and M. Patrick Griffith Montgomery Botanical Center, 11901 Old Cutler Road, Miami, Florida 33156, U.S.A. * Authors for correspondence: chad@montgomerybotanical.org; patrick@montgomerybotanical.org Abstract. The relationships among cypress species of the closely related genera Cupressus L., Callitropsis Oerst., and the recently described Hesperocyparis Bartel & R. A. Price were examined using morpho- logical characters. Previous studies did not fully resolve New World and Old World cypresses, the status of Cupressus duclouxiana B. Hickel, Cupressus benthamii Endl., and Cupressus funebris Endl. being in question. Cupressus duclouxiana is sister to Hesperocyparis and Cupressus benthamii is sister to Callitropsis, while Cupressus funebris is considered a basal taxon to all other Cupressus investigated. The intrusion of Hesperocyparis lusitanica (Mill.) Bartel characters into the standard descriptions of Cupressus torulosa D. Don ex Lamb, is examined in order to clarify the differentiation of the respective genera. An emended description of Callitropsis constituting a grade of three species is provided. The new combination, Callitropsis funebris (Endl.) de Laub. & Husby, is provided for Cupressus funebris. Five names at new varietal rank are transferred to Hesperocyparis, as H. arizonica (Greene) Bartel var. nevadensis (Abrams) de Laub. [= Cupressus neva- densis Abrams], H. arizonica var. montana (Wiggins) de Laub. [= Cupressus montana Wiggins], H. goveniana (Gordon) Bartel var. abramsiana (C. B. Wolf) de Laub. [= Cupressus abramsiana C. B. Wolf], H. goveniana var. pygmaea (Lemmon) de Laub. [= Cupressus goveniana Gordon var. pygmaea Lemmon], and H. lusitanica var. lindleyi (Klotzsch ex Endl.) de Laub. [= Cupressus lindleyi Klotzch ex Endl.]. Key words: Callitropsis, Cupressaceae, Cupressus, Hesperocyparis. Three recent publications have divided the species commonly known as cypresses into separate genera. Supported by a large data set based on morphological data, Little (2006) decisively demonstrated that a division would be appropriate and assigned the New World cypresses to the genus Callitropsis Oerst. along with the Old World C. vietnamensis (Farjon & T. H. Novon 22: 8-15. Published on 10 July 2012. Nguyen) D. P. Little [= Xanthocyparis i Farjon & T. H. Nguyen]. Adams et al. (2009), after sampling and analyzing nuclear DNA of several species, concurred with the taxonomic separation but established the new genus Hesperocyparis Bartel & R. A. Price to include 16 species of the New World cypresses, with the exception of C. nootkatensis (D. Don) Oerst. ex D. P. Little, which was previously handled by Little (2006). Similarly, de Laubenfels (2009) also concurred with Little’s (2006) analysis and established the new genus Neocupressus de Laub. with eight species and seven varieties. The two synonymous genera, Hesperocyparis and Neocupres- sus, were separated in priority by five months, with Hesperocyparis the prior name. To finish the taxonomic segregation initiated by Little’s (2006) study, several nomenclatural transfers and taxonomic considerations remain for the genus Hesperocyparis and are addressed herein. Since the three genera Hesperocyparis, Cupressus L., and Callitropsis are closely related, relationships tion. Adams et al. (2009) treated every New World cypress taxon at the species rank. While de Laubenfels (2009) suggested that Cupressus bentha- mii Endl. may belong in Callitropsis, the analysis of Little (2006) resolved Cupressus benthamii with all the other New World species, placing it with them in the genus Callitropsis. Adams et al. (2009) did not later attempt to resolve the position of those several species of intermediate morphological characters, including Cupressus benthamii. Further, the feasibil- ity of distinguishing New and Old World cypresses cladistically with exclusively morphological charac- ters has not yet been demonstrated. Morphological data continue to provide useful and reliable phylogenetic resolution (Zander, 2003, 2007b; Lee, 2006) for organismal taxonomies. Morphological character states include important information about organisms that can augment molecular markers used in phylogenetic analyses doi: 10.3417/2010056 Novon wonder that Endlicher (1847: 57) indicated “Butan et Nepalia” as the habitat for C. torulosa. Wallich, who assembled the specimens cited by Don, also sent seed the same year that Cupressus torulosa was published “in 1824, and again in 1830” (Loudon, 1854: 2479). Soon young plants were under cultivation in Europe. In 1842, Loudon illustrated what is clearly the correct taxon for C. torulosa, calling it “[t]he Bhotan, or twisted, Cypress,” an error, for torulosa does not refer to a twisted form (cf. Loudon, 1854: 2479, figs. 2329-2331; Loudon, 1869: 1076, fig. 2000). More collections of diverse cypresses were rapidly being delivered from India, Nepal, and Bhutan, leading Lindley (1853: 168) to question whether there might be more than one cypress native to India based on differences among the specimens raised in England. He presented an oversized illustration (1853: 167, fig. 105) of a plant cultivated from Himalayan seeds described with glaucous foliage not at all flattened, which clearly identifies this as Hesperocyparis lusitanica, although Masters (1896) attributed this to C. macrocarpa Hartw. ex Gordon. Various compilations in the 19th century gave descriptions for Cupressus torulosa. Camere (1855: 118) referred to drooping branchlets, glaucous foliage, and glaucous seed cones, but never mentioned flattening, i.e., Hesperocyparis lusitanica. Gordon (1858: 69) also referred to drooping branchlets and glaucous leaves, which suggested H. lusitanica. Gordon further referred to C. torulosa as the “weeping cypress of travellers,” and he noted two cotyledons as well as the “Twisted or Bhotan Cypress.” In spite of Lindley’s admonition, all Himalayan cypresses were mistaken as similar. Even Camus (1914), who published a monograph on cypresses, repeated the same critical traits as Caniere. It may seem remarkable that Cupressus torulosa and Hesperocyparis lusitanica could be so unquestionably combined, but specimens of the two species are, in fact, quite similar. They both have scale leaves and seed cones of about the same size (C. torulosa 12-20 X 12-18 mm vs. H. lusitanica 12-20 X 12-20 mm). The dimorphism of C. torulosa i specimens is not obvious. Distichous branchlets when pressed require close observation: glaucousness generally disappears on dried specimens, and cotyle- dons are rarely available. Hill and de Fraine (1908) reported that the cotyledon number is three to five for C. torulosa, which is an error, but this was repeated by subsequent authors (Elwes & Henry, 1910; Camus, 1914; Ouden, 1949; Silba, 2005). Because H. lusitanica was commonly cultivated under the name C. torulosa, the confusion is not surprising. In 1968, Franco correctly specified the lectotype of Cupressus torulosa. He did not give a description of the species but did append a list of the specimens consulted, of which many were, in fact, Hesperocy- paris lusitanica. Only Shrestha (1974), among modem authors, restricted his brief description to characters appropriate to C. torulosa. In treating the Gymno- sperms of Nepal, Shrestha presumably inspected actual living wild trees. Farjon (2005: 223) continued the taxonomic confusion in his monograph of the Cupressaceae, describing Cupressus torulosa with the branchlets often with drooping ends, the leaves monomorphic, and the cotyledons (two)three or four(five). There is no mention of distichousness, and this description would encompass Hesperocyparis lusitanica. Even Eckenwalder (2009: 231), in his recent Conifers of the World, treated C. torulosa essentially the same as Farjon with characters preponderant for H. lusitanica. Finally, the monograph of Silba (2005) would go so far as to treat C. torulosa as a subspecies of H. lusitanica. However, C. torulosa is an important tree in India and Nepal, and the taxon needs careful distinction from H. lusitanica. la. Branchlets distichous; leaves dimorphic; cotyle- lb. Branchlets not distichous; leaves monomorphic; Relationship of Cupressus duclouxiana. The whole of the characters examined resolves Cupressus duclouxiana B. Hickel as sister to the eight species analyzed for Hesperocyparis. Cupressus duclouxiana differs from the Old World Cupressus primarily by shoots that are not distichous and foliage that is not dimorphic, in marked contrast to the rest of Cupressus (and a large part of Cupressaceae). On the other hand, C. duclouxiana does have cotyledon characters like Callitropsis that are not found in Hesperocyparis but Cupressus. This character combination places it with Juniperus, which is either sister to a larger cypress group or in a polytomy with cypresses (Little, 2006). This character suite is also shared by Widdringtonia Endl., and this genus would likely have a similar position as Juniperus had it been included in Little’s (2006) analysis. The seed cones of Juniperus are more or less fleshy and do not open when mature, and there is usually one wingless seed per fertile bract. The seed cones of Widdringtonia are reduced to four bracts, like many of those in Systematics and Ecology of a New Species of Seagrass ( Thalassodendron , Cymodoceaceae) from Southeast African Coasts Maria Cristina Duarte Tropical Botanical Garden, Tropical Research Institute, Travessa Conde da Ribeira 9, 1300-142 Lisbon, Portugal; and CIBIO, Research Center in Biodiversity and Genetic Resources, Campus Agrario de Vairao, Universidade do Porto, 4485-661 Vairao, Portugal. Author for correspondence: mcduarte@iict.pt Eduardo Mondlane University, Department of Biological Science, CP 257 Maputo 0100, Mozambique. sband@zebra.uem.mz Maria Manuel Romeiras Tropical Botanical Garden, Tropical Research Institute, Travessa Conde da Ribeira 9, 1300-142 Lisbon, Portugal; and BioFIG, University of Lisbon, Faculty of Science, Campo Grande, 1749-016 Lisbon, Portugal, mromeiras@iict.pt Abstract. Tropical seagrass communities are one of the most productive aquatic ecosystems on earth. A high diversity of seagrass species occurs in south- eastern Africa; however, these marine angiosperms are among the least studied in the world. To address this, we have revised Thalassodendron Hartog (Cymodoceaceae), one of the most representative seagrasses in these coastal waters. Morpho-anatom- ical analyses, complemented with field data, reveal that specimens from rocky habitats present a number of distinguishing characters (e.g., rhizome intemode lengths, leaf epidermal cells, and flower structures) that recommend their exclusion from the species T. ciliatum (Forssk.) Hartog. A new species from rocky habitats, T. leptocaule Maria C. Duarte, Bandeira & Romeiras, is thus described and illustrated, with the type from Mozambique, and an identification key for the investigated taxa is presented. Key words: Coastal waters, Cymodoceaceae, lUCJS Red List, seagrass, southeastern Africa, Thalassodendron. Seagrass species are marine angiosperms widely distributed in both tropical and temperate coastal wat< eal ig one of the most productive aquatic ecosystems on earth (Gullstrom et al., 2002; Spalding et al., 2003). Seagrass meadows function as shelters and nurseries for juvenile animals. Their canopy acts as a hydrodynamic barrier in near shore areas, and their roots and rhizomes stabilize bottom sediments, providing large benefits for fisheries (e.g., Bandeira, 1995; Gullstrom et al., 2002; Spalding et al., 2003; Novon 22: 16-24. Published on 10 July 2012. Eklof et al., 2005). Seagrass beds also play an important role in the development of carbonate sediment facies, promoting the settling and stabili- zation of fine sediment; moreover, seagrass stems and blades provide important habitats for epiphytic marine organisms, such as calcareous faunas and skeletal material to the substrate (Perry & Beaving- ton-Penney, 2005). Although seagrasses provide one of the most important coastal ecosystems, a decline in their distribution and abundance has been observed and is due to a growing coastal population and an over- exploitation of resources (e.g., Short & Wyllie- Echeverria, 1996; Barbera et al., 2005). Together with mangroves, coral reefs, and tropical rainforests, seagrass meadows are among the most threatened ecosyslems on earth (Waycott et al., 2009). Seagrass meadows and the ecosystems they support are also threatened by natural disturbances and the potential impacts of climate change (Bjork et al., 2008). The preservation of these ecosystems and their associated services — in particular, biodiversity, primary and secondary production, nursery habitat, and nutrient and sediment sequestration — should be a global priority (Orth et al., 2006). An ecosystem-based conservation and management strategy is essential to preserve not just seagrasses and their dependent communities, which include several threatened species, but also the entirety of coastal ecosystems (Orth et al., 2006; Hughes et al., 2009). doi: 10.3417/2010079 i (Cymodoceaceae) f Figure 1. Thalassodendron leptocaule Maria C. Duarte, Bandeira & Romeiras. — A. Community of T. leptocaule and seaweeds bract enclosing the female flower in fresh specimen. — D, " 3801, LISC). Scale bars: B = 15 mm; C, D = 5 20 Novon are distinctly different from T. leptocaule specimens from rocky habitats. Morpho-anatomical analyses indicated that leaf epidermal cells (either on adaxial or abaxial surfaces) are somewhat isodiametric or (Fig. 3A) and rectangular with sinuous anticlinal walls in T. leptocaule (Fig. 3B) (a character also present in Amphibolis ). The male flower in T. ciliatum is protected by four bracts rather than one, as is the case in T. leptocaule; anthers from T. ciliatum are considerably longer, 6-10 mm versus ca. 2.5 mm (Fig. 3C, D). The only male specimen known of T. i (Cymodoceaceae) f ,70 cm long, 1—2 mm w If F- A New Species of Pseudostellaria (Caryophyllaceae) from Korea Sangtae Lee, Kyeong-In Heo, and Seung-Chul Kim 10.341' 26 Novon Petal length (mm) Petal width (mm) Stem length (cm) Pedicel length (cm) Fruit pedicel length (cm) Distalmost subverticillate whole leaf length (L) and width (W) ratio (L/W) Mature distalmost subverticillate whole leaf length (cm) Mature distalmost subverticillate whole leaf length (L) and width (W) ratio (L/W) P. palibiniana 5-6.3 ± 0.63-7* 4.5- 5.9 ± 0.71-7.2* .1-1,5 ± 0.26-2* 3.5- 5.5 ± 1.16-7.8* 0.8-1 .2 ± 0.24-1.8* 2-3.6 ± 0.76-5* 7.3-13.5 ± 3.3-19.5 1.8- 2.4 ± 0.3-2. 8 1.8- 2.6 ± 0.5-3.7 1.9- 3.7 ± 0.?-5 0.6-1. 1 ± 0.41-2 2-3.7 ± 1.47-7.1 2.7- 5.5 ± 1.54-9.9 1.1-2.5 ± 0.72-4.6 1. 7- 2.2 ± 0.32-3.1 2.7— 4.4 ± 1 3.1-4.7 ± 1.21-8.6 0.6-2 ± 0.81—4.6 1.8- 2.6 ± 0.69—4.9 (SKIG199^04^001),W//. S5.n.W^KK^199^)4^8001),0 S. D. Jo s.n. (SI „ ~ j Qjh HKI050013 (SKK), 5. Tm ^ & K. Heo HKI050015 (SKK), S. T. Lee & K. . nous, 3.6-5.1 X 0.7-1.8 mm; petals 6 to 9, obovate, apex acute, emarginate, obtuse, or round, white, 2.9- 5.4 X 1-2 mm; stamens 12, filaments 3. 1-4.4 mm, anthers 0.5-0.6 mm; ovaries syncarpous, with 3 to 4 carpels, ovate, 1. 3-2.8 X 1.3-2. 6 mm, style distinct, 2-3.7 mm, stigma filiform to clavate. Fruit as capsule, reddish brown, septicidally dehiscent; seeds (4 to)9(to 15), reniform, 1.8-2.6 X 1. 7-2.2 mm, papillose, the papillae distributed evenly, roundly conical, their anticlinal cell walls branching radially and irregularly, tightly interwoven with other cell walls, the spinules of papilla slender, 0.02-0.03 mm. Phenology. Pseudostellaria longipedicellata was observed to flower in April with fruiting from May to trending belt running along the eastern margin of the Korean Peninsula. The population size is small (usually less than 100 individuals per population) each locality (or mountain). Populations are typically found in mid to lower elevations of mountains, especially along the hiking trails or park entrances. locality, have been threatened and have declined due to construction of some park structures, such as buildings and parking lots. It is urgent to take additional ecological and biological study for effec- tive conservation measures. We recommend an IUCN Red List category of Endangered, with small or declining populations (EN, C2) (IUCN, 2001). Distribution. Pseudostellaria longipedicellata is endemic to Mts. Taebaek, Daedeok, Deokhang, and Myeon in Gangwon Province, Korea. It was also collected from Mts. Worak and Joryeong in Chung- cheongbuk Province and Mt. Naeyeon in Gyeong- sangbuk Province and Mt. Biseul in Daegu Metro- politan City. IUCN Red List category. Pseudostellaria longi- pedicellata is native to three eastern provinces, i.e., Gangwon, Chungcheongbuk, and Gyeongsangbuk, as well as Daegu Metropolitan City on the Korean Peninsula. Additional populations likely exist along the Taebaek Mountain Range, a north- to south- Ecology and phenology. Pseudostellaria long- ipedicellata is sympatric with P. setulosa Ohwi, occurring along roadsides, and may be found on somewhat dry leaf mold in larch forests (Larix Mill.) near the south entrance of Mt. Taebaek National Park. Its flowering time overlaps with that of P. setulosa, beginning in April. Only a few individuals (approximately up to 50 with varying distances from a few centimeters to 1-2 m) of P. longipedicellata were In contrast, the sympatric species P. setulosa grew individuals of the new species were encountered at locations other than the type locality. Pseudostellaria Volume 22, Number 1 2012 Lee et al. 27 Pseudostellaria (Caryophyllaceae) from Korea longipedicellata was noted along roadsides on nearby Mt. Myeon in Gangwon; it was also seen at dry, sunny sites in rocky places in Mt. Worak National Park in Chungcheongbuk Province. Through examination of specimens in other herbaria, the distribution of the described species extended further south to Mt. Naeyeon as well as in Daegu Metropolitan City and Gyeongsangbuk Province. Etymology. The specific epithet longipedicellata highlights the dramatic elongation of the fertile pedicels post-anthesally. Common name. Tae-baek-gae-byeol-ggot (Kore- an). The vernacular name Taebaek was taken from the type locality, in Mt. Taebaek National Park. endemic to Mt. Jiri in Gyeongsangnam Province. However, P. okamotoi is easily distinguished from P. longipedicellata by having puberulent pedicels (Ohwi, 1937). Interestingly, both species share the intriguing elongation of the fruiting pedicels. Pedicels in P. okamotoi elongate (2.3-3.8 cm, sensu Choi, 1993) and bend downward (Choi, 1993), but not to the extent observed for P. longipedicellata (3. 8-6.9 Pseudostellaria heterophylla (Miq.) Pax, which occurs widely in the entire Korean Peninsula, is also similar to the new species, sharing four ± whorled distalmost leaves and their shapes. Differences in P. heterophylla include more flowers at stem apices (one to eight vs. one in P. longipedicellata ), only a single taproot (vs. one to four), and leaf blades that are basally ciliate/pubescent (vs. pubescent along the proximal margin in P. longipedicellata). Pedicels of P. heterophylla are shorter (5.8-28.5 mm) and lack the dramatic elongation seen in the new species (Choi, 1993). The surface morphology of seeds is variable and different among the species of Pseudostellaria in Korea (Choi, 1999). The seed surfaces of P. palibiniana, P. okamotoi, and P. heterophylla show roundly conical papillae distributed evenly, with the anticlinal cell walls branching radially and irregular- Slender spinules are seen at the top of papillae. In contrast, the branches of anticlinal cell walls are few and rough in P. heterophylla (Choi, 1999: 25, fig. 5), many and delicate in P. palibiniana (Choi, 1999: 26, fig. 9), or intermediate in P. okamotoi (Choi, 1999: 26, fig. 7). In particular, P. longipedicellata and P. palibiniana are quite distinct in several features of surface morphology (cf. Fig. 3). Branching of anticlinal cell walls of P. longipedicellata appear short and wide, but are long and narrow in P. palibiniana', spinules of the former are short, but elongate in the latter; ultra-reticular calcium carbon- ate debris on the papilla of the former is scarce or absent, but evenly distributed in the latter. The seed surface morphology of P. longipedicellata is quite distinctive from that of P. palibiniana, and more closely resembles that of P. okamotoi (Choi, 1999: 26, fig. 7). Cleistogamy is common within Pseudostellaria (Ohwi, 1937), and the cleistogamous capsules are borne at the axil of leaves typically at the proximal part of the stem. Based on the authors’ observations, cleistogamous capsules appear to be rare in P. longipedicellata, in contrast to P. palibiniana and P. heterophylla. This may suggest that P. longipedicel- lata is a facultative autogamous species, with this observation supported by the fact that in this species, seed sets approach 100% (S. T. Lee, unpublished). The color of fruits and pedicels is reddish brown. This color and the ground-level fruit might suggest that the fruits attract ants and that the seeds are dispersed by ants (Fig. 2D). Many wandering ants were observed under the clustered plants of Pseu- dostellaria longipedicellata, although ant dispersal of seeds was not directly observed. Among other taxa, ant-pollinated flowers bloom near the ground and/or exhibit similar reddish brown perianths (Proctor & Yeo, 1973; Beattie, 1985). This has been noted in Asarum L. (Aristolochiaceae; Smith et al., 1989), Leavenworthia Torr. (Brassicaceae; Wyatt, 1981), Lepidium L. (Brassicaceae; Gomes et al., 1986), Balanophora J. R. Forst. & G. Forst. (Balanophor- aceae; Kawakita & Kato, 2002), Microtis R. Br. (Orchidaceae; Peakall & Beattie, 1991), as well as other Caryophyllaceae, including Arenaria L. and Gypsophila L. (Gomes et al., 1986). Ants may disperse small seeds of many annual herbs (Lach, 2003), including Stellaria L., which is affined to Pseudostellaria (Kjellsson, 1985). The clustered habit of plants of P. longipedicellata may be caused not only by short rhizomes but also by the ants that collect the seeds for food reserve (Beattie, 1985; Lach, 2003). We did observe moist capsules that were partially foraged by insects (Fig. 2F), although in none of them did the dried capsule dehisce naturally. We did not observe any insect taxa other than ants in association with the fruits. Further studies are required to confirm the pollination and seed dispersal mechanisms in P. longipedicellata. Table 1). KOREA. Chungcheongbuk: Chungju, Seokmun- dong, 16 May 2006, *G. H. Nam NGH60317 (KH); Volume 22, Number 1 2012 30 Novon collections, to Soyoung Kwon for the illustration (Fig. 1), and Samjo Chung, a flower photographer and amateur ecologist, for the photographic series (Fig. 2) taken at the type locality of Pseudostellaria long- ipedicellata. We also thank the scientific editor and associate editors at NOVON, and two anonymous reviewers for their comments and suggestions on an earlier version of the manuscript. Identifications and Typifications of Rosa (Rosaceae) Taxa in North America Described or Used by E. W. Erlanson, 1925-1934 Walter H. Lewis Washington University in St. Louis, Department of Biology, 1 Brookings Drive, St. Louis, Missouri 63130-4899, U.S.A., and Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166-0299, U.S.A. lewis@biology.wustl.edu Anton A. Reznicek and Richard K. Rabeler University of Michigan Herbarium, 3600 Varsity Drive, Ann Arbor, Michigan 48108-2228, U.S.A. reznicek@umich.edu; rabeler@umich.edu Abstract. During her main publication period at the University of Michigan from 1926 to 1937, E. W. Erlanson described 18 new Rosa L. species and varieties (Rosaceae), most from the midwestem United States. The majority were grown and hybrid- ized at the University of Michigan Botanical Garden, where she collected many specimens for the University of Michigan Herbarium (MICH) over a period of years. Some living collections were hers, but most were from others who sent living plants usually without voucher specimens. For this reason and the inability until recently to use sheets as types from living collections under specific circumstances, type selections were at times inappropriate. Nevertheless, seven (almost half) of the cited holotypes here are in MICH, in addition to three lectotypes designated here, as well as isotypes and isolectotypes. A total of 40 taxa with either type data or in synonymy are identified in 10 categories of Rosa species, subspe- cies, and nothospecies following the Flora of North America treatment. Lectotypes are designated here for the following 19 names: R. acicularioides Schuette, R. acicularis Lindl. var. bourgeauiana Crep., R. arkan- sana Porter, R. blanda Aiton fo. carpohispida Schuette, R. blanda var. glandulosa Schuette, R. blanda var. hermannii Erlanson, R. blanda var. nuda Schuette, R. blanda var. subgeminata Schuette, R. califomica Cham. & Schltdl., R. Carolina L. var. aculeata (Schuette) Erlanson, R. Carolina var. litoralis Erlanson, R. deamii Erlanson, R. gemella Willd., R. nitida Willd., R. Carolina var. inermis Schuette, R. Carolina var. sepalorelevata Schuette, R. pensylvanica Michx., R. Xrudiuscula (Greene) Erlanson, and R. suffulta Greene var. valida Erlanson. One neotype is designated, R. palustris. In addition, holotypes of the following taxa are identified for the first time: R. blanda var. setigera Crep., R. blanda var. hispida Farw., and R. blanda fo. alba (Schuette ex Erlanson) a Femald. tl Novon 22: 32^2. Published on 10 July 2012. In her early research on Rosa L. (Rosaceae) at the University of Michigan, Eileen W. Erlanson de- scribed 18 new taxa, six species, and 12 varieties in two publications (Erlanson, 1926, 1928). Most are native to the Midwest, but a few occur in California, the Rocky Mountains, and New York State. One additional variety in California was recognized in a later publication (Erlanson, 1934). However, by the which provided experimental data for a revision of North American roses, the majority of taxa Erlanson had previously described were synonymized or excluded from consideration. It seems that Erlanson had initially followed E. L. Greene and P. A. Rydberg by describing new taxa routinely when variant characters were found, but she later dropped their approach when her research showed that inherent variations in the form of unreliable variation were common in Rosa species, a finding further confound- ed by both polyploidy and hybridization. Although Erlanson did not plan to produce a monographic treatment of North American roses, typifications were often difficult to provide since herbarium vouchers of wild collections were rarely made by those sending her live material. In a 1 July 1925 letter to C. C. Deam at Indiana University, Erlanson wrote, when discussing collected living material, “I wish I had herbarium specimens of all of them taken before they were sent here,” suggesting that many living collections destined for the University of Michigan Botanical Garden were not accompanied these potential types, we have used sheets taken from living material to designate lectotypes, thanks to the assistance of Kanchi Gandhi (GH), who reminded us that we could consider certain sheets from living doi: 10.3417/2011032 Volume 22, Number 1 2012 Lewis et al. 35 Identifications and Typifications of Rosa in North America floral information only available from flowering material in June or rarely to July, the cultivated material was used to describe the variety and constitutes uncited original material (Art. 9.10 [Mc- Neill et al., 2006]). Material of the second collection cited by Erlanson (1928) (W. Brotherton Jr. s.n., Oct. 1922) is also missing, but its living material Garden number 2668) has no priority over the first cited material. Erlanson did not include taxonomic consideration of the variety in 1934. Rosa blanda var. hispida was recognized by Erlanson (1934: 255), but except for fine prickles or aciculi on some branches, it is very similar to R. blanda. Progeny from a single wild collection of R. blanda included R. blanda var. hispida as well as R. blanda var. glandulosa and R. blanda var. acicular- ioides, both cited above, while the majority of progeny were typical of the species (Erlanson, 1934: 236- 237). See the R. blanda var. glandulosa discussion above. Surprisingly, Erlanson included the variety in her final key (1934, app. 3: 255). 3b. Rosa blanda Aiton fo. alba (Schuette ex Erlanson) Femald, Rhodora 50: 146. 1948. Basionym: Rosa blanda var. alba Schuette ex Erlanson, Pap. Michigan Acad. Sci. 5: 88. 1926. TYPE: U.S.A. Wisconsin: Brown Co., Albany, near Stado farm, 16 June 1895, ./. H. Schuette s.n. (holotype, F-379239). 3c. Rosa blanda Aiton fo. carpohispida (Schuette) W. H. Lewis, Rhodora 60: 240. 1958. Basio- nym: Rosa blanda var. carpohispida Schuette, Proc. Amer. Assoc. Advancem. Sci. 46: 279. 1898. TYPE: U.S.A. Wisconsin: Door Co., Little Sturgeon Bay, 17 Aug. 1891, J. H. Schuette s.n. (lectotype, designated here, US-752278). The collection selected as lectotype was cited in the exsiccatae of Rosa blanda fo. carpohispida, which illustrates its atypical stipitate-glandular hypanthia and pedicels also noted in the original description by Schuette. Erlanson recognized predominantly R. blanda var. carpohispida in a study of offspring of R. blanda from Quebec (1934: 237, tab. B), but R. blanda var. carpohispida was excluded from the final appendices 2 and 3. 4. Rosa bridgesii Crep. ex Rydb., Bull. Torrey Bot. Club 44(2): 83. 1917, nom. cons. (Enter, 2007). TYPE: U.S.A. California: 1857, Bridges 62 (holotype, K). related to R. gymnocarpa Nutt, and which was retained as an ecotype species of R. gymnocarpa (1934: 251). In addition, R. covillei Greene became a synonym of R. gymnocarpa (1934: 251). Erlanson (1934: 230, key 257) recognized R. yainacensis Greene (with synonyms R. myriadenia Greene and R. pinetorum A. Heller) as an ecotype species of R. durandii Crep. Rosa bridgesii nom. cons, is conserved against five other names (Ertter, 2007). The species is included in the Flora of North America treatment for Rosa (Lewis et al., 2012). 5. Rosa californica Cham. & Schltd., in D. F. L. von Schlechtendal, Linnaea 2: 35. 1827. TYPE: U.S.A. California: San Francisco, Chamisso & Eschschultz s.n. [Chamisso Herbarium, annotat- ed by Crepin] (lectotype, designated here by B. Ertter, LE). The lectotype label selected by B. Ertter in St. Petersburg includes San Francisco, “Nob.” and “Herb. Cham.,” making this the obvious candidate. The species was probably collected during the Kotzebue expedition by the ship Rurik at the Presidio, October 1816 (Eastwood, 1944). Rosa californica Cham. & Schltd. var. bidenticulata (Rydb.) Erlanson, Bot. Gaz. 96: 222. 1934. Basionym: Rosa bidenticulata Rydb., N. Amer. FI. 22(6): 518. 1918. TYPE: U.S.A. California: Shasta Co., Castella, 24 July 1912, A. Eastwood 1389 (holotype, A-32560; isotype, In his revision of the Rosa californica complex, Cole (1956) included R. californica var. bidenticulata discussion. However, B. Ertter (pers. comm., 2010) noted that R. californica var. bidenticulata “is part of the unresolved variation in R. californica, possibly bridgesii Crep. ex Rydb.” Clearly, the identification of this variety as R. pinetorum is very much in question and for now is retained within R. californica s.l. Erlanson (1934: app. 3, 257) recognized the variety in her final key. 6. Rosa Carolina L., Sp. PL 492, 1753, non Rosa Carolina L., 1762. TYPE: U.S.A. Carolina (holotype), designated by Heath (1992: 51, 80), confirmed by Jarvis (2007), J. Dillenius, Hort. Eltham. 2: 325, pi. 245, fig. 316. 1732: “pi. Rosa Carolina fragrans, fobs mediotenus serratis.” Figures 1, 2. To our knowledge these are the first original figures of the Rosa Carolina holotype provided in a North American journal. This is significant since only three 36 Novon Volume 22, Number 1 2012 37 Lewis et al. Identifications and Typifications of Rosa in North America 6b. Rosa Carolina subsp. subserrulata (Rydb.) W. H. Lewis, Novon 18(2): 195. 2008. Basionym: Rosa subserrulata Rydb., N. Amer. FI. 22: 500. 1918. TYPE: U.S.A. Missouri: Taney Co., Swan [Township], 2 June 1899, B. F. Bush 42 (holotype, NY-429910; isotype, MO-5255457). Neuwiedia malipoensis , a New Species (Orchidaceae, Apostasioideae) from Yunnan, China Zhong-Jian Liu, 1,2,3 Li-Jun Chen,1 and Ke-Wei Liu 1,2 1 Shenzhen Key Laboratory for Orchid Conservation and Utilization, The National Orchid Conservation Center of China and The Orchid Conservation and Research Center of Shenzhen, Wangtong Road, Shenzhen 518114, People’s Republic of China. liuzj@sinicaorchid.org; chenlj@sinicaorchid.org 2 The Center for Biotechnology and BioMedicine, Graduate School at Shenzhen, Tsinghua University, Shenzhen University Town, Shenzhen 518055, People’s Republic of China. liukw@sinicaorchid.org 3 College of Forestry, South China Agricultural University, Wushan Road, Tianhe, Guangzhou 510642, People’s Republic of China Abstract. Neuwiedia malipoensis Z. J. Liu, L. J. Chen & Ke Wei Liu (Orchidaceae, Apostasioideae), from southeastern Yunnan Province in China, is described, illustrated, and photographed. The new species is characterized by a true corolla lip, which is widely obovate and concave, with a subterete claw at the base and a fleshy linear callus extending from its claw to the upper portion. By these features it can be easily distinguished from its allied N. veratrifolia Blume, N. balansae Baill. ex Gagnep., and other species of this genus. Key words: Apostasioideae, China, IUCN Red List, Neuwiedia, Orchidaceae, Yunnan. Neuwiedia Blume is a genus within the subfamily Apostasioideae (Orchidaceae) for which 17 names have been described. Approximately 10 species are distributed across southeastern Asia from southern China through Southeast Asia to New Guinea and the Pacific Islands. Only the one species, N. singapur- eana (Wall, ex Baker) Rolfe, has been reported for the Chinese flora (Chen, 1999). The flowers of this genus are nearly actinomorphic, with three fertile stamens and a simple lip similar to, but a little larger than, the remaining two petals; the lip’s midrib is usually raised on the upper surface. The new taxon was collected from southeastern Yunnan Province in China during fieldwork in 2008. It is similar in habit and floral morphology to other Neuwiedia, except that the corolla lip consists of a concave, widely obovate blade with a subterete claw at the base, and a fleshy linear callus that is both 1 mm wide and high and extends from its claw to the upper part of the corolla blade. This lip is quite distinct from the other two petals and the flowers are obviously zygomorphic in appearance, in contrast to other species in the genus. This undescribed species represents an advanced type in the genus. The lip structure easily distinguishes the new species from the allied N. veratrifolia Blume and N. balansae Baill. ex Gagnep., as well as other species of the The living plants and ecological environment were observed in the field in Malipo County (ca. 23°N) in southeast Yunnan, China, in August 2008. Fresh flowers were examined under a stereoscope (Gui- guang XTL-500, Guilin, China). Comparison of the vegetative and floral structures was made between this taxon and Neuwiedia veratrifolia and allies based on fresh material, herbarium specimens, and the available literature. Neuwiedia malipoensis Z. J. Liu, L. J. Chen & Ke Wei Liu, sp. nov. TYPE: China. Yunnan: Malipo, Ma An Shan, 1100 m, evergreen broad-leaved forest, limestone mountainous area, 1100 m, 11 Aug. 2008, Z. J. Liu 4058 (holotype, N0CC). Figures 1, 2. Species nova Neuwiediae veratrifoliae Blume et N. balansae Baill. ex Gagnep. similis, sed a priore floribus praeter callum flavum labelli albis atque labello late balansae ) longis, petiolo 12-20 cm (3-10 cm in N Terrestrial plants 65-70 cm tall; rhizome 13-15 cm, 1.5-1. 7 cm diam., with evident nodes, bearing stout roots. Leaves many, nearly tufted on the short stem; leaf blades oblong to lanceolate, 30-63 X 3.3- 4.8 cm, apex long-acuminate; petiole 12-20 cm, margin membranous, base slightly dilated and doi: 10.3417/2010093 Novon 22: 43-47. Published on 10 July 2012. Novon Volume 22, Number 1 2012 Liu et al. 45 Neuwiedia (Orchidaceae) from Yunnan, China up of inflorescence. — C. Flower (front view). — D. Flowers (side view). — E. Infructescence. — F. Capsule cross 46 Novon clasping. Scape to 35 cm; raceme ca. 11 cm, 20- flowered or more: rachis. with glandular trichomes; floral bracts ovate to lanceolate, 1-1.7 cm, abaxially with glandular trichomes. Flowers not resupinate, not opening widely; sepals narrowly elliptic, 1.6-1. 9 X 0.35-0.4 cm, hooded apically, shortly aristate toward abaxial apex, abaxially with glandular tri- chomes, dorsal sepal usually slightly smaller than lateral pair; petals broadly cuneate, obovate, 1.6-1 .8 X 0.5-0.6 cm, apex mucronate, abaxially sparsely glandular pubescent; labellum bipartite; blade ± rowing into a subterete claw, with a fleshy linear callus extending from the base of the claw distally for 2/3 of the blade length; column erect, 10-11 mm; stamens subequal to stigma; filaments 3.3-3. 7 X 0.6- 0.7 mm; anthers linear, 5.8-6.2 mm; ovary elliptic, 7-8 mm; style 6-6.5 mm, subterete; stigma 3-lobed. Immature capsule ellipsoid. Habitat and ecology. Neuwiedia malipoensis was observed to grow in wet places in a pristine evergreen broad-leaved forest on a limestone slope at the southern fringe of a subtropical zone of ca. 1100 m. The dominant plant associates include trees and shrubs such as Zanthoxylum armatum DC. (Rutaceae), Luculia pinceana Hook. (Rubiaceae), Brassaiopsis glomerulata (Blume) Regel (Araliaceae), Machilus rufipes H. W. Li (Lauraceae), Ficus hirta Vahl var. imberbis Gagnep. (Moraceae), Trevesia palmata (Roxb. ex Lindl.) Vis. (Araliaceae), Cornus oblonga Wall. (Comaceae), and Rubus paniculatus Sm. (Rosaceae). Herbaceous associates include Liparis delicatula Hook. f. and L. assamica King & Pantl. (Orchidaceae), Debregeasia orientalis C. J. Chen, Elatostema pachyceras W. T. Wang, Pilea glaberrima (Blume) Blume, and P. howelliana Hand.- Mazz. (Urticaceae), Impatiens arguta Hook. f. & Thomson (Balsaminaceae), Cynoglossum zeylanicum (Yahl ex Homem.) Thunb. ex Lehm. (Boraginaceae), Artemisia indica Willd. (Asteraceae), Galium asper- uloides Edgew. subsp. hojfmeisteri (Klotzsch) H. Hara (Rubiaceae), Microsorum superficial (Blume) Ching (Polypodiaceae), and Glaphyropteridopsis erubescens (Wall, ex Hook.) Ching (Thelypteridaceae). The type locality is in a subtropical highland in the of 17.6°C, ranging from 10.1°C to 23°C. The canopy density was estimated at 80%-85% coverage. The dry season extends from December to March, when there is no fog, and the rainy season is from April to November, when this area is often covered by dense fog in the night and morning (Liu et al., 2008a, 2008b). The soil types are mainly developed from limestone, and the soil layer is rather thin. The vegetation in this area consists of broad-leaved forests, coniferous forests, thickets, and grasslands (Liu et ah, 2008a, 2008b). IUCN Red List category. This species is very rare. There were altogether 97 individuals found exclusively in six populations. We failed to see more plants in the vicinity of the type locality at Malipo, activity. The conservation status of this new species is assessed as Endangered (EN), according to IUCN (2001) criteria. Discussion. In comparison to derived taxa in the other four subfamilies of the Orchidaceae, Neuwiedia of the Apostasioideae is a rather primitive genus, from whose allies or ancestors the Orchidaceae may have derived in a past paleogeologic period (Kocyan et al., 2004). The Apostasioideae remains a poorly known subfamily, with only the two genera Neuwiedia and Apostasia Blume. Neuwiedia is distinguished from Apostasia by its three fertile stamens with fairly long filaments, versus two stamens, with or without a style in Apostasia. The stigma is trilobed in N. malipoensis, but noted as capitate in Apostasia (Chen, 1999; Chen et al., 2009). The typical perianth of Neuwiedia is somewhat zygomorphic in appearance, with the lip slightly larger than the lateral two petals; the midvein of the lip is usually raised on the upper surface. The perianth of the new species shows a strong zygomorphic tendency and the central petal, or lip, is more or less differentiated into a short subterete claw and a widely obovate, concave blade. On its upper surface there is a fleshy linear callus extending from the base of the claw distally for two-thirds of the blade length. This makes the flowers more clearly zygomorphic in appearance, and there is no doubt they are most derived within the genus. This new species is quite different from Neuwiedia singapureana, the only species of this genus found in China, which has a lip similar in shape and size to the two lateral petals except for the midvein raised on the upper surface. Our new plant is somewhat similar to N. veratrifolia and N. balansae. Neuwiedia veratrifolia has yellow flowers, and the lip lacks a terete claw at the base, while N. balansae has smaller leaves, only 10-20 cm long, with shorter petioles 3- 10 cm long, and the fleshy linear callus on the lip Erythroxylum ayrtonianum (Erythroxylaceae): A New Species from Brazil Volume 22, Number 2012 Loiola & Sales Erythroxylum (Erythroxylaceae) from Brazil 0.7 mm, IPSl Psychotria kosraensis (Rubiaceae), a New Species from Kosrae, David H. Lorence* and Kenneth R. Wood Novon s 4.5-5 X 1.2-1 .3 r and E. Volume 22, Number 2012 Lorence & Wood Psychotria (Rubiaceae) from the Caroline Islands flower ( Wood et ll. 14132, PTBG). ’ & kusaiensis Kaneh. (Elaeocarpaceae), Flagellaria indica L. (Flagellariaceae), Cyrtandra kusaimontana Hosok. and C. urvillei C. B. Clarke (Gesneriaceae), Neuburgia celebica (Koord.) Leenh. (Loganiaceae), Astronidium se (Kaneh.) Markgr., Medinilla dwersifol and Melastoma malabathricum L. (Melast ;), Ficus virens Alton (Moraceae), Horsfield ;rtn.) Warb. (Myristicaceae), Pandanus kmc A New Species of Hymenoxys (Asteraceae, Helenieae, Tetraneuridinae) from Texas Jeffrey N. Mink Department of Biology, Baylor University, Waco, Texas 76798-7388, U.S.A. j_mink@baylor.edu Jason R. Singhurst Wildlife Diversity Program, Texas Parks and Wildlife Department, 4200 Smith School Road, Austin, Texas 78704, U.S.A. jason.singhurst@tpwd.state.tx.us Walter C. Holmes Department of Biology, Baylor University, Waco, Texas 76798-7388, U.S.A. walter_holmes@baylor.edu Abstract. Hymenoxys perpygmaea W. C. Holmes, Singhurst & Mink (Asteraceae, Helenieae, Tetraneur- idinae) is described as a new annual species. The species, which is ephemeral, is endemic to an alfisol prairie that contains mima mounds in Lamar County from northeastern Texas, U.S.A. It is similar to H. texana (J. M. Coult. & Rose) Cockerell, but is easily distinguished from that species by its diminutive rosette and fewer florets per capitulum. Key words: Asteraceae, Helenieae, Hymenoxys, IUCN Red List, Tetraneuridinae, Texas, U.S.A. Hymenoxys Cass, is a New World genus of approximately 25 species, 17 of which occur north of Mexico (Biemer, 2006). Two of the 17 species are annual, with one, H. texana (J. M. Coult. & Rose) Cockerell, being endemic to Texas (Cockerell, 1904; Mahler, 1983; Bridges, 1988). This latter species is of conservation concern, being listed as endangered (G2S2) by the United States Fish and Wildlife Service (Poole et al., 2007). Recent field study of the tall grass prairies of Lamar County in northeastern Texas has resulted in the discovery of another annual species of Hymenoxys, which is described below. Hymenoxys perpygmaea W. C. Holmes, Singhurst & Mink, sp. nov. TYPE: U.S.A. Texas: Lamar Co., Gambill Goose Prairie Refuge, 7.2 mi. W of Paris (+33.69617, -95.660550), 4 May 2009, W. C. Holmes, J. R. Singhurst & J. N. Mink 14504 (holotype, BAYLU; isotype, MO). Fig- Species nova Hymenoxydis texanae (J. M. Coult. & Rose) Cockerell similis, sed ab ea rosula 1. 3-2.4 cm in diametro (vs. 4-7+ cm) et flosculis disci in quoque capitulo 17 ad 32 (vs. 45 ad 75+) differt. Novon 22: 56-59. Published on 10 July 2012. Plants slender, erect, ephemeral, 5-6.9 cm tall; taproots slender, light brown, ca. 1.7 mm diam.; stems terete, glabrous, 0.3-0.7 mm diam., un- branched and leafless for the first 2-3 cm, divided distally into 1 or 2(4) head-bearing branches; rosettes 1.3-2. 4 cm diam. Leaves 3 to 6, elliptic to obovate, 6-12 X 3-4 mm, margins entire, surfaces glabrous, glandular, midvein whitish, conspicuous in the proximal 1/2 of blade, apices rounded, rosettes turning yellow and withering prior to or at time of flowering, leaving an enlarged node 1-2 mm diam., often with 1-3 mm leaf remnants persisting, cauline leaves alternate, 4 to 8 per branch, linear, 6.4-10.5 X 0.5-2 mm, sessile, margins mostly entire, the midstem leaves occasionally irregularly toothed, surfaces glabrous, glandular, apices obtuse. Pedun- cles quadrangular, obscurely winged to ridged, surfaces sparsely crisp-puberulent, capitula terminal, 1 or 2(4), semi-campanulate, 4.3-7.9 X 3. 1-5.4 mm; phyllaries in 2 series; outer 8, ovate-oblong, ca. 5 mm, glabrous, bases slightly connate, gibbous, margins entire, the distal 1/2 rather scabrous, apex acute; inner phyllaries 8, similar to outer, ca. 0.5 mm longer, apices rounded; receptacles hemispheric, ovoid, smooth, epaleate. Ray florets yellow, not exceeding phyllaries, ca. 7, pistillate, fertile, corollas ovate, ca. 2.8 mm, lobes (teeth) 3 or 4, rounded (ovate), ca. 0.4 mm. Disc florets yellow, bisexual, fertile, 17 to 32, glabrous, tube ca. 1.1 mm, throat cylindrical, ca. 1.4 mm, teeth deflate, ca. 0.2 mm; styles and style branches ca. 2.2 mm, undivided portion ca. 1.65 mm, style appendages ca. 0.55 mm; anthers ca. 0.8 mm, filaments ca. 0.3 m, appendages deflate. Cypselae obpyramidal, ca. 2 X 0.6 mm, densely velutinous; pappi possessing awned scales, the scales ovate, ca. 1 X 0.7 mm, margins entire, doi: 10.3417/2010065 Volume 22, Number 2012 Hymenoxys (Asteraceae) from Texas figure 1. Photograph oi Hymenoxys perpygmaea W. L. Holmes, Singhurst & Mink taken at the type location (scale applies to both the larger and inset images). Habit and capitula of the holotype Holmes, Singhurst & Mink 14504 (BAYLU). Photo by Darrell Vodopich, 4 May 2009. The inset image shows the basal leaves collected as the paratype Singhurst 18134 (BAYLU). Photo by J. R. Singhurst, 25 February 2010. (Asteraceae) from Texas Impatiens ecornuta , a New Name for Impatiens ecalcarata (Balsaminaceae), a Jewelweed from the United States and Canada Gerry Moore National Plant Data Team, Natural Resources Conservation Service, United States Department of Agriculture, 2901 East Lee Street, Greensboro, North Carolina 27401, U.S.A. gerry.moore@gnb.usda.gov Peter F. Zika WTU Herbarium, University of Washington, Box 355325, Seattle, Washington, 98195-5325, U.S.A. Catherine A. Rushworth Biology Department, Duke University, Box 90338, Durham, North Carolina 27708, U.S.A. Abstract. The new name Impatiens ecornuta Gerry Moore, Zika & Rushworth (Balsaminaceae) is proposed for the species of western North America currently recognized as I. ecalcarata Blank. The latter name is an illegitimate later homonym of I. ecalcarata Collett & Hemsl. Key words: Balsaminaceae, Impatiens, North America. In 1905, Joseph Blankinship described a new species of Impatie Pacific Northwest u; However, in 1890 Henry Collett and William Hemsley had previously described a new species of Impatiens from Burma using the name I. ecalcarata Collett & Hemsl. Thus, I. ecalcarata Blank, is an illegitimate later homonym that, barring formal conservation, must be replaced (McNeill el al., 2006, Art. 53.1). Formal conservation of the name I. ecalcarata Blank, does not seem warranted, since the range of this species is rather restricted and dissemination of a new name will not be difficult. It name I. ecalcarata Collett & Hemsl. (it is generally treated as a synonym) in Myanmar may warrant conservation of I. ecalcarata Blank. However, there is no guarantee that future revisions of Impatiens in Myanmar and adjacent areas would not lead to I. ecalcarata Collett & Hemsl. being held distinct. Thus, it seems best to leave I. ecalcarata Collett & Hemsl. available for use in pending revisions by not proposing I. ecalcarata Blank, for conservation over I. ecalcarata Collett & Hemsl. Therefore, the following name is provided as a replacement name for /. ecalcarata Blank. Impatiens ecornuta Gerry Moore, Zika & Rush- worth, nom. nov. Replaced name: Impatiens ecalcarata Blank., Sci. Stud. Montana Coll. Agric., Bot. 1: 84. 1905, nom. illeg., non Impatiens ecalcarata Collett & Hemsl., J. Linn. Soc., Bot. 28: 30, tab. 4. 1890. Impatiens biflora Walter var. ecalcarata M. E. Jones, Bull. Montana Univ., Biol. Ser. 15: 39. 1910. TYPE: U.S.A. Montana: Missoula Co. [now Sanders Co.], along a brook 1 m[ile] E of Plains, 9 Aug. 1901, J. W. Blankinship s.n. (holotype, MONT; isotypes, MONTU, WS, WTU). Unlike the other native Impatiens species in the United States and Canada (e.g., I. aurella Rydb., I. capensis Meerb., I. noli-tangere L., I. pallida Nutt.), and as the specific epithet indicates (as well as Blankinship’s specific epithet), the petaloid, lower saccate sepal in I. ecornuta is spurless. Impatiens ecornuta is known from wetlands in British Columbia, Idaho, Montana, Oregon, and Washington, at eleva- tions below 1200 m. West of the Cascade Range in Oregon and Washington, I. ecornuta hybridizes with I. capensis, which has been introduced into this area (/. capensis is a native elsewhere in Canada and the United States; Omduff, 1966, 1967; Zika, 2006b). The 1 sen I. capensis and 1. econ known as I. Xpacifica Zika, and this hybrid l genetically threatening the uncommon I. ei (Zika, 2006a, 2006b). Novon 22: 60-61. Published on 10 July 2012. doi: 10.3417/2011088 Nuevas Combinaciones en Adiantopsis (Pteridaceae, Cheilantheae) de America del Sur Novon Volume 22, Number 1 2012 Ponce & Scataglini Adiantopsis (Pteridaceae, Cheilantheae) de America del Sur con A. chlorophylla, pero en esta especie los ultimos segmentos son elfpticos o triangulares, a menudo Material examinado. ARGENTINA. Salta: Dpto. Cap- ital, Cerro Tunel, frente a Vaqueros, L. Novara 4979 (MCNS). BOLIVIA. La Paz: Inquisivi, Huanahawira, 3 km NW of Quime, 17°02'S, 67°15'W, cloud forest above Rio Khatu, L. J. Dorr et al. 6766 (US). PARAGUAY. Alto Paraguay: Cerro Cabrera, 19°38'S, 61°42'W, R. Fortunato et al. 3733 (SI). PERU. Cajamarca: Contumaza, La Pampa, Guzmango, A. Sagdstegui & R. Samamei 2937 (F, GH, US). 3. Adiantopsis recurvata (Baker) Ponce & Scatagli- ni, comb. nov. Basionimo: Cheilanthes recurvata Baker, J. Bot. 16(n.s. 7): 299-300. 1878. TIPO: Paraguay. Asuncion, Co. Lambare, “Pontes rocailleuses du Cerro-Lambare,” 1 Nov. 1876, B. Balansa 358 (holotipo, K [barcode] 000633276; isotipos, BM, G, P). Figura ID. Distribucion. Adiantopsis recurvata se distribuye casi exclusivamente en Paraguay (Chaco, Guaira, Paraguari), siendo muy rara en Argentina (Misiones), de donde existe una coleccion muy antigua del area con cerrado en esta provincia (Santa Ana y San Ignacio). Su habito es cespitoso, creciendo en vegetacion de chaco y cerrado o en montes riberenos, entre rocas o piedras. Observaciones. En la revision de las especies del grupo Cheilanthes dichotoma (Ponce & Morbelli, 1989) se incluyo a C. recurvata en la sinonimia de C. orbignyana. Un analisis mas amplio de este grupo permitio encontrar diferencias entre ambas especies. Cheilanthes recurvata se distingue de C. orbignyana hialino mucho mas amplio y por los ultimos Aunque la especie no ha sido muestreada para los estudios moleculares, se la trata bajo este genera por sus similitudes morfologicas con el resto de las rizomaticas bicoloras, pecfolos y ejes atropurpureos, laminas herbaceas a cartaceas, ultimos segmentos peciolulados, presencia de pelos glandulares 2- celulares, de cabezuela blancuzca o amarillenta, esporas crestado-reticuladas. Material examinado. ARGENTINA. Misiones: Santa Ana, 26 Nov. 1896, N. Alboffs.n. (SI). PARAGUAY. Alto Paraguay: Cerro Galvan, Puerto Casado, T. Rojas 1829 (SI). Guaira: Cordillera de Ybytyruzu, rd. to Polilla, Zardini & Velazquez 13575 (CTES). Paraguari: Cerro Acahay, T. Rojas 3006 (P, SI, US). Cheilanthes tweediana Hook., Sp. Fil. 2: 84, tab. 96B. 1852. TIPO: Brasil. “S. Brazil, banks of the river Parana,” s.d., J. Tweedie s.n. (holotipo, K [barcode] 000633277; isotipo, US- 01470293 [K fragm.]). Figura IE. Distribucion. Adiantopsis tweediana se distribuye en Bolivia (Cochabamba, Santa Cruz), Brasil (Mato Grosso), Paraguay (Caaguazu, Cordillera) y Argentina (Chaco, Cordoba, Corrientes, Formosa, Santa Fe). A esta especie se la encuentra tanto en los valles secos interandinos y matorrales xeroffticos (chaquenos), como en bosques en galena o en planicies con suelos arcillosos erosionados. Observaciones. Adiantopsis tweediana se incluye en el grupo Cheilanthes dichotoma, por sus pecfolos castano oscuros con escamas bicoloras, laminas 2- pinnadas, raquis acanalados con aristas laterales, ultimos segmentos peciolulados, herbaceos, lobula- dos; pseudoindusios lobulados, membranaceos. Se diferencia del resto de las especies del grupo por sus rizomas rastreros, horizon tales y sus hojas de contomo lineal, remotas o subremotas. En el analisis filogenetico (Ponce & Scataglini, inedito) se encuen- tra como especie hermana de C. dichotoma, los (2011) en su redefinition del genera Adiantopsis. Material examinado. ARGENTINA. Chaco: Colonia Benitez, Schultz 30/2218 (US). Cordoba: San Javier, La Vina, H. H. Bartlett 20618 (SI, US). Corrientes: Dpto. Lavalle, Canada de los Perros, ruta 27, al S de Santa Lucia, A. Schinini 5149 (CTES, F). Formosa: Pilcomayo, Isla Leona, I. Morel 7686 (G). Santa Fe: Lanteri, Est. Bonazzola, M. M. Job 1245 (GH). BOLIVIA. Cochabamba: Cercado, Cochabamba, Cerro San Pedro, 17°22'32"S, 66°08'06"W, N. Ritter 2752 (LPB). Santa Cruz: Andres Ibanez, 12 km E of Santa Cruz on rd. to Cotoca, 17°46'S, 63°04'W, M. Nee 36295 (G, NY, P, SI). BRASIL. Mato Grosso: 20 km E Porto Murtinho, no chao, J. M. Pires & P. P. Furtado 1 7294 (CTES, NY). PARAGUAY. Caaguazu: Arroyo Yakare’f, along S side from rte. 2, E. Zardini & Velazquez 10747 (G). Cordillera: San Bernardino, E. Hassler 208 (SI). Agradecimientos. Deseamos agradecer a los co- legas del Instituto Darwinion y a Gonzalo Marquez (LP) por colectar parte de las muestras de helechos utilizadas en este trabajo, a los curadores de los herbarios visitados y consultados, por su atencion y prestamos. Las tareas y viajes de coleccion se financiaron con subsidios para Proyectos de Inves- tigacion Plurianuales (PIP 5262/2005-2008) del Consejo Nacional de Investigaciones Cientfficas y Tecnicas (CONICET, Argentina). Cuatrecasas Award (2007) permitio el estudio del material depositado en el Herbario de Smithsonian Institution. A New Combination in Phemeranthus (Montiaceae) and Notes on the Circumscription of Phemeranthus and Talinum (Talinaceae) from the Southwestern United States and Northern Mexico Price & Gentianella weigendii (Gentianaceae), an Unusual New Gentian from Peru Volume 22, Number 1 2012 Pringle Gentianella (Gentianaceae) from Peru Habitat and phenology. Gentianella weigendii is known only from the type collection. It was found in well-developed jalca vegetation, among numerous shrubs and perennial herbs, at 4000 m ( Weigend et al. 5083, in sched.). The March date of this collection provides the only information available on its IUCN Red List category. Data are deficient for recommended listing (DD) (IUCN, 2001). With the species being known only from a single locality in northern Ancash Department, it seems likely that its total geographic range is not extensive, although according to Maximilian Weigend (pers. comm., 23 June 2009), Gentianella weigendii was locally common at the type locality. Etymology. Gentianella weigendii is named for Maximilian Weigend, of the Institut fur Biologie- Systematische Botanik und Pflanzengeographie-der Freien Universitat der Berlin, who collected the type specimen and thought that it represented a new Volume 22, Number 1 2012 Pringle Gentianella (Gentianaceae) from Peru coralline trichomes are absent; the corollas are green and scarcely longer than the calyces; and the corolla lobes are ovate-triangular, shorter than the tube. The flowers of G. weigendii and G. thyrsoidea are similar in aspect, but in G. thyrsoidea the filaments are equal, 5-8 mm, all inserted at the same level about 0.5X the height of the corolla tube. At 10-60X magnification I detected no floral glands in G. thyrsoidea ( Weigend et al. 5125 and 7712, HAM), but observations with fresh material and with SEM would be desirable. Vegetatively, Gentianella weigendii and G. thyrso- idea differ distinctly. Plants of G. thyrsoidea are 0.2- 3.5 dm; those of G. weigendii are 3-6 dm. In G. thrysoidea the basal and near-basal leaves are very numerous, often over 100, and the cauline leaves are closely spaced the entire length of the stem (distal intemodes commonly 2-8 mm, occasionally to 15 mm, rarely to 40 mm on the tallest plants). The leaves are ascending and distally incurved, and in combi- nation with the closely spaced axillary cymules, often conceal much of the stem. All of the leaves are of the abruptly expanded portion just above the basal sheath, reminding Ruiz and Pavon of pine needles (as indicated by the old epithet pinifolia ). The leaves are about eight per whorl at the mid-cauline nodes, more numerous at the proximal nodes, with the bases united into a sheath, which is 8-12 mm on the larger plants. The entire plant above ground forms a dense cylinder 2-6 cm in diam. In G. weigendii the basal leaves are fewer and wider, 7-12 cm X 4-13 mm; the distal cauline leaves are lance-triangular, 2.5-7 cm X 7-13 mm; and the leaves are not expanded near the and the leaves spread widely, so that much of the stem is visible. The cauline leaves are in whorls of four, separate to the base. From Gilg’s description of Pitygentias Gilg, which he segregated from South American Gentiana s.l. in 1916, it might be suspected that Gentianella weigendii had already been described, but no earlier specific epithet is available for this species. Gilg treated Pitygentias as comprising two species, both of which are now included in Gentianella thyrsoidea (Fabris, 1958). Although in 1916 Gilg described the nectaries of Pitygentias as five calluses between the adnate portions of the stamens, he included in P. thyrsoidea (Hook.) Gilg plants that he (Gilg, 1896) had previously described as Gentiana tupa Gilg, in which he had said that the nectaries were scarcely conspicuous. The material that he identified as P. pinifolia (Ruiz & Pav. ex G. Don) Gilg in 1916 included flowers preserved in alcohol, received from August Weberbauer, as well as herbarium specimens. If Weberbauer’s botanizing had extended to the range of Gentianella weigendii, it might be that Gilg’s description of the nectaries of Pitygentias was based on alcohol-preserved flowers of G. weigendii, whereas the herbarium specimens that he identified as P. pinifolia and P. thyrsoidea, and earlier as Gentiana tupa, represented Gentianella thyrsoidea. The name Gentiana tupa was based on plants from Diezmo, Peru. A Ruiz and Pavon specimen from Diezmo at BC (according to Ibanez, 2006), and a Ruiz and Pavon collection (now divided, F and MA) interpreted by Fabris (1958) and Macbride (1959) as an isotype of the name Gentiana pinifolia, were both designated by the same unpublished name by Ruiz and Pavon in herb., and both were annotated as P. pinifolia by Gilg. These specimens are probably from a single collection, of which I identified a component as Gentianella thyrsoidea. The type of the basionym Gentiana thyrsoidea Hook. ( Cruikshanks s.n., BM) likewise represents the species to which the name Gentianella thyrsoidea is currently applied. Although the similarities of Gentianella weigendu to G. thyrsoidea and G. weberbaueri have been noted above, little evidence, morphological or molecular, exists to indicate the phylogenetic relationships among the South American species of Gentianella or between the South American species and those of other continents. Divisions of Gentianella (historical- ly included in Gentiana ) into sections have included all species, or all except G. magellanica (Gaudich.) Fabris ex D. M. Moore, in a single section, and divisions of Gentianella into segregate genera subsequent to Gilg’s recognition of Pitygentias have not discussed South American species. Some char- acter states occur in diverse combinations rather than consistently occurring in syndromes, suggesting the frequent occurrence of homeoplasy. Molecular phy- logenetic studies have indicated that Gentianella, with paired nectaries, is monophyletic (Hagen & Kadereit, 2001), but few South American species have been included in such studies to date. Weigend et al. ( Weigend et al. 5083, in sched.) suggested that the flowers of Gentianella weigendii might be cleistogamous. Self-pollination has been demonstrated in some Gentianaceae, including some high-altitude grassland species in South America (Freitas & Sazima, 2009), but studies of pollination in South American Gentianella have remained specula- tive, despite the interest engendered by the diverse floral morphology among these species. Gilg (1916) reported that the flowers of G. thyrsoidea (as Pitygentias spp.) appeared to be cleistogamous, as A New Species of Pyrrosia (Polypodiaceae) from Made 22: 75-77. 76 Taxonomic Notes Group of Astragalus sect. Novon — C. Standard. — D. Keel. — E. Wings. — F. Androecium. — G. Pistil. — H. Pod. — I. Telophase I in A. mishodaghmontanus. Drawn from the holotype Ranjbar 19161 (BASU). Scale bar: A = 2 cm; B-G = 1 cm; H = 1 cm; I = 6 pm. Staelia catolensis (Rubiaceae), una Nueva Especie de Catoles, Bahia, Brasil Cc Novon Volume 22, Number 1 2012 Salas & Cabral Staelia (Rubiaceae) de Bahia, Brasil o determinado, mientras que difiere en las hojas aciculares con apfculo punzante, levemente crasas, de seccion hemicircular, de 3-6 X O. 2-0.5 mm (vs. hojas lineares con apice no punzante, papiraceas y de seccion aplanada, de 5- 20 X 1-2.5), corola extemamente papilosa (vs. glabra reticulada (vs. tectado-perforada). Por otro lado, S. thymoides es una especie que habita en el noreste de Argentina, noroeste de Rio Grande do Sul, Brasil, sur de Paraguay oriental y noroeste de Uruguay en areas de campos o pastizales primarios o modificados. Staelia catolensis presenta una irrflorescencia con crecimiento determinado, por lo que todas sus ramas florfferas tienen sus meristemas apicales consumidos en glomerulos terminales. Esta caracterfstica es compartida no solo por S. thymoides, sino tambien por S. hatschbachii J. H. Kirkbr. (Endemica de Grao Mogol, Minas Gerais). De acuerdo a Weberling (2007), este tipo de inflorescencia esta relacionado al habito anual, lo cual se pudo comprobar en invernaculo en S. thymoides. Los especimenes examinados se encuentran iden- tificados en los tres herbarios como Staelia catecho- sperma K. Schum. [= Anthospermopsis catechosperma (K. Schum.) J. H. Kirkbr.], de la cual puede ser diferenciada por ser un sufrutice ramificado desde la base (vs. arbusto con eje simple ramificado apical- mente), por las hojas aciculares de seccion hemi- circular y punzante (vs. hojas angostamente elipticas o elipticas, seccion plana y apice no punzante); capsula con dehiscencia longitudinal y oblicua, valvas separadas de las semillas (vs. capsula con dehiscencia longitudinal, valvas con semillas inclu- sas despues de la dehiscencia y septo intercarpelar cordado). Distribucion y habitat. Esta especie esta restrin- gida a Bahia, Brasil, dentro del complejo de la Chapada Diamantina en Serra de Catoles. En esta cas por investigadores brasilenos y britanicos desde 1983 hasta el presente. De esta coleccion, se conoce al unico especimen de Staelia catolensis. El Sr. Wilson Ganev (colector del holotipo), residente local y gufa experto del distrito de Catoles, junto a otros coleccionistas (R. M. Harley y D. J. N Hind de K y L. P. Queiroz de HUEFS, entre otros) realizo en esa region innumerables viajes de coleccion entre 1983 y 1995; dichos especimenes fueron depositados prin- cipalmente en los herbarios HUEFS, K y SPF. El analisis de esas colecciones, permite sugerir que el material tipo representa probablemente la unica . catolensis. Esta especie proviene de areas de campos rupestres, vegetacion que en el distrito de Catoles se encuentran bien representada entre 1200 y 1700 m. El area se caracteriza por suelos arenoso- pedregosos, ocasionalmente anegadizos, con numer- osos afloramientos rocosos. La vegetacion predom- inante esta representada por plantas lenosas, de porte pequeno y perennifolias, las hojas, tallos y rai'ces (hojas esclerificadas, coriaceas, suculentas o con indumentos particulares), lo que les permite sobre- vivir a los grandes cambios termicos, a la gran variabilidad hidrica y a los efectos del fuego (Zappi et al., 2003). La region de Catoles, se caracteriza ademas por mas de 30 especies endemicas, la mayorfa recientemente descritas y entre las cuales se incluye un genero denominado Catolesia D. J. N. Hind. En cuanto a la tribu Spermacoceae, fue descrita recientemente Borreria catolensis E. L. Cabral & L. M. Miguel basada en colecciones de los municipios Abaira y Piata (Cabral et al., 2011). IUCN Red List category. Debido a que se conoce una unica coleccion en los ultimos 19 anos en un area relativamente bien colectada, se le asigna una categorfa de amenazada de acuerdo a los criterios de IUCN (2001), por lo tanto consideramos a Staelia catolensis en Peligro Crftico: CR B2a,b(iv). Moifologia polmica. Los granos de polen fueron analizados bajo microscopio optico y electronico de barrido, la descripcion se basa en la observacion de 20 granos. Descripcion bajo MO: granos (7 a)8 zonocolporados, esferoidales, isopolares y radio- simetricos. Tamano mediano (P = 24-26.7 y E = 24.6-25.7 pm). Colpos largos, de 17.5-18.3 pm (ECA/P: < 0.46). Endoabertura en endocmgulo, 4- 7 pm de largo. Exina de 2.2-2. 7 pm de espesor, sexina igual o mas gruesa que la nexina. MEB: exina microreticulada, heterobrocada, lumenes 0.1-1. 2 pm de diametro, circulares, alargados o apocolpio, muros 0.4-0. 6 pm de ancho, con espmulas 0. 1-0.2 pm, uniformemente distribuidas en todo el grano (Fig. 2A-D). El polen de Staelia thymoides fue descrito por Salas y Cabral (2010a) como de tamano mediano (P = 29 pm y E = 27.5 pm), (— ' 7)8-colporado y colpos de 16-21 pm. Los autores tambien senalan que las especies por ellos analizadas presentan el polen morfologicamente muy similar. Las principales diferencias en cuanto al habito, hojas, polen, distribucion y ambientes de S. thymoides y S. catolensis se resumen en la Tabla 1. P = 24-26.7 y E = 24.^25.7 P = 26.6-35.5 y E = 27.3-31.8 New Combinations in North American Schoenoplectiella (Cyperaceae) Derek R. Shiels and Anna K. Monfils Central Michigan University Herbarium, Department of Biology, Central Michigan University, Mount Pleasant, Michigan 48858, U.S.A. Author for correspondence: d.r.shiels@gmail.com Abstract. An investigation of taxonomic diversity in the North American members of Schoenoplectus (Rchb.) Palla and Schoenoplectiella Lye indicates the need to relocate species and varieties from Schoeno- plectus sect. Actaeogeton (Rchb.) J. Raynal to Schoenoplectiella. A transfer of four taxa is proposed here: Schoenoplectiella purshiana (Femald) Lye var. williamsii (Femald) Shiels & Monfils, S. smithii (A. Gray) Shiels & Monfils var. smithii, S. smithii var. leviseta (Fassett) Shiels & Monfils, and S. smithii var. setosa (Femald) Shiels & Monfils. Key words: bulmsh, Cyperaceae, Schoenoplectiel- la, Schoenoplectus, systematics, taxonomy. Schoenoplectus (Rchb.) Palla s.l. has four sections designated by morphological and ecological data (Raynal, 1976; Smith & Hayasaka, 2001): Schoeno- plectus sect. Schoenoplectus, Schoenoplectus sect. Mal- acogeton (Ohwi) S. G. Sm. & Hayasaka, Schoenoplectus sect. Actaeogeton (Rchb.) J. Raynal, and Schoenoplectus sect. Supini (Cherm.) J. Raynal. Species within Schoenoplectus sect. Schoenoplectus and Schoenoplectus sect. Malacogeton are perennials and have yellow to dark brown smooth achenes. Species within Schoeno- plectus sect. Actaeogeton and Schoenoplectus sect. Supini are annuals or perennials and have light brown to black striate achenes. Species in Schoenoplectus sect. Schoenoplectus have stout rhizomes and floral scales with distinctly emarginate apices, while species in Schoenoplectus sect. Malacogeton have slender rhi- zomes often terminating in fleshy tubers and floral scales with entire or minimally emarginate apices. Species in Schoenoplectus sect. Supini often display amphicarpy and all have one or two nodes with cauline leaves, while species in Schoenoplectus sect. Actaeogeton never display amphicarpy and have nodeless culms with basal leaves. According to Govaerts et al. (2012), there are 76 species in Schoenoplectus s.l.; ca. 25 of these are not assigned to a section in the sectional accounts of Raynal (1976) and Smith and Hayasaka (2001). Lye (2003) recognized the polyphyly of Schoeno- plectus and proposed a revision resulting in a more narrowly defined Schoenoplectus and a newly erected genus, Schoenoplectiella Lye. This revision relied Muasya et al. (1998, 2000), which, combined, included two species of Schoenoplectus sect. Schoe- noplectus (Schoenoplectus lacustris (L.) Palla and Schoenoplectus triqueter (L.) Palla) and two species from Schoenoplectus sect. Supini (Schoenoplectiella juncea (Willd.) Lye and Schoenoplectiella articulata (L.) Lye). Muasya et al. (2000) found Schoenoplectus lacustris (the type species for Schoenoplectus) more closely related to Actinoscirpus (Ohwi) R. W. Haines & Lye and Bolboschoenus (Asch.) Palla than to the sampled congeners from Schoenoplectus sect. Supini. plectus s.l. into Schoenoplectiella: 23 species from Schoenoplectus sect. Supini and three from Schoeno- plectus sect. Actaeogeton (Schoenoplectiella juncoides (Roxb.) Lye, Schoenoplectiella purshiana (Femald) Lye, and Schoenoplectiella wallichii (Nees) Lye). Lye did not recognize sections in his treatment. Lye’s (2003) generic description of Schoenoplectiella relies on the presence of amphicarpy, tropical to subtropical distribution, and the “smaller” annual habit. Amphicarpy is present in Schoenoplectus sect. Supini but not in the three transferred species from Schoenoplectus sect. Actaeogeton. In addition, three species that Lye (2003) moved to Schoenoplectiella are restricted to North America and can occur in a temperate climate (Schoeonoplectiella hallii (A. Gray) Lye, Schoenoplectiella purshiana, and Schoenoplectiella saximontana (Femald) Lye). Schoenoplectiella jun- coides was transferred and is a perennial species with individuals that can reach 70 cm in height (Koyama, 1958). As currently described, not all species in Schoenoplectiella fit Lye’s (2003) description. Since Lye established Schoenoplectiella in 2003, five phylogenetic studies have been published that include modest sampling of species from Schoeno- plectus s.l. (Yano & Hoshino, 2005; Simpson et al., 2007; Muasya et al., 2009a, 2009b; Jung & Choi, 2010). These studies include up to 13 species, utilize additional genetic markers, and include sampling from several close relatives in the Fuireneae. In all analyses, species in Schoenoplectus sect. Actaeogeton and Schoenoplectus sect. Supini are in a clade separate from species in Schoenoplectus sect. Schoe- noplectus and Schoenoplectus sect. Malacogeton. Citing molecular and morphological evidence, Jung and Choi (2010) transferred five species from doi: 10.3417/2011097 Novon 22: 87-90. Published on 10 July 2012. A New Combination in Senna (Leguminosae, Caesalpinioideae) from the Department of Valle del Cauca, Colombia Philip A. Silverstone-Sopkin Departamento de Biologfa, Universidad del Valle, A.A. 25360, Cali, Valle, Colombia pasilverstone_sopkin@yahoo.com Abstract. A new combination, Senna tocotana (Rose ex Britton & Killip) Silverst., is proposed for Chamaefistula tocotana Rose ex Britton & Killip, the latter name previously placed in synonymy of S. viminea (L.) H. S. Irwin & Barneby. The Colombian species S. tocotana is not conspecific with the Jamaican species S. viminea. They differ notably in the size and shape of the fruits and in the arrangement of seeds; the fruits of S. viminea are Resumen. Una nueva combinacion, Senna tocotana (Rose ex Britton & Killip) Silverst., se propone para Chamaefistula tocotana Rose ex Britton & Killip, la cual estaba colocada en la sinonimia de S. viminea (L.) H. S. Irwin & Bameby. La especie colombiana S. tocotana no pertenece a la misma especie como la especie jamaicana S. viminea. Ellas difieren nota- ordenacion de las semillas; los frutos de S. viminea semdlas son bisenadas (en dos filas), versus The genus Senna Mill. (Leguminosae, Caesalpi- nioideae) includes 350 species, of which 80% are found in the New World (Marazzi et al., 2006). The New World species were revised by Irwin and Bameby (1982), who divided the genus into six sections and 35 series. The largest of the sections is Senna sect. Chamaefistula (DC. ex Collad.) H. S. Irwin & Bameby. It is characterized by simple trichomes, ebracteolate pedicels, usually zygomor- phic flowers, palmately veined sepals, and not more than seven fertile stamens, with the longest abaxial stamens parallel to the floral axis. Within the section Chamaefistula, the series Bacillares (Benth.) H. S. Irwin & Barneby is characterized by a woody, caulescent, evergreen habit, leaves with two pairs of leaflets and one or two glands on the rachis, dimorphic fertile stamens with beaked anthers, and a turgid, often pulpy fruit, with at least 50 seeds arranged transversely to the longitudinal axis of the fruit. Marazzi et al. (2006) showed that, based on three chloroplast genes, the section Chamaefistula was paraphyletic, although the series Bacillares was supported as monophyletic. In 1986, Jorge E. Ramos-Perez and I began a floristic inventory of the few remaining small patches One of these remnants is an island of 12.5 ha of forest surrounded by open fields of sugar cane at the Hacienda El Medio, near La Paila in the Municipio de Zarzal. This forest is the last vestige of a large cacao plantation that was abandoned in the 1930s. Here, we encountered an unidentified species of Senna (sect. Chamaefistula, ser. Bacillares ), which seemed at first to be an undescribed taxon. Later, I examined specimens at COL and VALLE that were identified by E. P. Killip (probably in the 1940s) as Chamaefistula tocotana Rose ex Britton & Killip; these specimens were collected between 1944 and 1946 by Jose Cuatrecasas-Arurm (/. Cuatrecasas 18393, 19565, 22019 ) from the eastern slope of the Cordillera Occidental of the Andes in the department of Valle del Cauca, Colombia, between 1680 and 2250 m. The collections had been filed under S. viminea (L.) H. S. Irwin & Bameby, following the synoptic treatment by Irwin and Bameby (1982), who placed C. tocotana in the synonymy of S. viminea without making the new combination in Senna. The lectotype of Senna viminea (P. Browne s.n., LINN-528.6, collected in Jamaica) was designated by Lawcett and Rendle (1920: 102). I have seen images of the lectotype and five non-type specimens of S. viminea from Jamaica, and of the type collection of Chamaefistula tocotana (H. Pittier 720, holotype, US; isotype, NY), collected in Colombia. Data on labels of some of the specimens examined indicate that S. viminea is a vine, whereas C. tocotana is a shrub or small tree. In the material examined, the fruits of the specimens of Senna viminea from Jamaica differ to such a degree from those of the type collection of doi: 10.3417/2007117 Novon 22: 91-95. Published on 10 July 2012. Novon Isotype specimen of S. tocotana (Rose ex Britton & Killip) Silverst. (H. Pittier 720, NY). Chamaefistula tocotana that they cannot belong to the same species (Fig. 1). The largest fruits of S. viminea are relatively short and broad. Measurements of eight fruits from four Jamaican specimens of S. viminea (C. L. G. Bertero s.n., MO; N. L. Button 2718, NY; A. S. Hitchcock s.n., MO; J. W. Walker 297, US) are 1.9- 5.3 X 0.4-2. 4 cm (1.5^.8X longer than wide). In contrast, the largest fruit (immature) of the holotype ( H . Pittier 720, US) of C. tocotana is elongate and narrow, 9.5 X 0.5 cm (19X longer than wide). Some Venezuelan specimens were provisionally determined by Irwin and Bameby (1982), following Novon Studies in Andean 0 cotea (Lauraceae). I. Species with Hermaphrodite Flowers and Fistulose Twigs Occurring Above 1000 m Altitude aceae) Volume 22, Number 2012 100 Novon Volume 22, Number 2012 slightly ridged, to 5 mm d buds densely appressed pubescent. Leaves 15-30 X Volume 22, Number 1 2012 107 van der Werff Studies in Andean Ocotea (Lauraceae) Figure 6. Holotype of Ocotea obovatifolia van der Werff ( van der Werff et al. 12490, MO-6327502). Fruits are from van der Werff etal. 12503 (MO). 6-13 cm, alternate, obovate or obovate-elliptic, thinly surface; venation slightly raised on the lower surface; chartaceous, the base attenuate or acute, flat or domatia absent; secondary veins 6 to 9 on each side slightly inrolled, the apex obtuse or rounded, of the leaf, curved upward near the margin, but not glabrous or very sparsely appressed pubescent on loop-connected; petioles 12-19 mm, flat, not winged, the lower surface; venation immersed on the upper with a similar indument as the twigs. Inflorescences Two New Infraspecific Taxa of Orychophragmus violaceus (Brassicaceae) in Northeast China Ling Wang, Xi-Juan Ma, and Chuan-Ping Yang* Northeast Forestry University, Harbin 150040, Heilongjiang, People’s Republic of China. wanglingyaoyao@yahoo.com.cn; 1241964981@qq.com * Author for correspondence: yangcp@nefu.edu.cn Abstract. Two new varieties of Orychophragmus violaceus (L.) 0. E. Schultz (Brassicaceae) are described from Heilongjiang Province in northeastern talus Ling Wang & Chuan P. Yang, differs from the typical variety of the species in that the petal has five or more obtuse teeth (vs. smooth petals) in the upper third of the expanded petal. The clawed petals have an expanded apical section that is 10-13 mm long (vs. 7-13 mm), and the stems and pedicels are puberulent (vs. glabrous). A second new variety is 0. violaceus var. variegatus Ling Wang & Chuan P. Yang, which differs from the typical variety by the marked petal variegation of white and purple (vs. no petal variegation). The uppermost leaves are often undivided, with sharply serrate margins (vs. irregu- larly toothed leaves). These distinctions have persist- ed and been stable after eight years of cultivation in field experiments in Heilongjiang. A key to the infraspecific taxa of 0. violaceus is included. Both new varieties are cataloged as of Least Concern (LC), based on IUCN Red List categories and criteria. Key words: Brassicaceae, China, Cruciferae, Heilongjiang, IUCN Red List, Orychophragmus. The genus Orychophragmus Bunge (Brassicaceae) was established in 1833 (Lu, 1980), based on the one species 0. sonchifolia Bunge. A second species, 0. violaceus (L.) 0. E. Schulz, was added in 1916 and it was long accepted that only two species were known in East and Central Asia, with the one species in China as 0. violaceus [= Brassica violacea L.]. This species is widely distributed in Heilongjiang, Jilin, Liaoning, Hebei, Shanxi, Henan, Shandong, Anhui, Jiangsu, Zhejiang, Jiangxi, Shaanxi, Gansu, and Sichuan provinces as wild and cultivated popula- tions. Lrom such broad geographic distribution and wide ranges of variation in leaf structure, shape, size, and indumentum, at least five additional varieties of 0. violaceus have been described, including variety homaeophyllus (Hance) 0. E. Schulz, variety hupe- hensis (Pamp.) 0. E. Schulz (1923), variety interme- dius (Pamp.) 0. E. Schulz (1923), variety subinte- grifolius (Pamp.) 0. E. Schulz, and variety lasiocarpus Migo. New species considered to be closely related to 0. violaceus were also described from China and include 0. taibaiensis Z. M. Tan & B. X. Zhao (from Shanxi Province), 0. diffusus Z. M. Tan & J. M. Xu, 0. limprichtianus (Pax) Al-Shehbaz & G. Yang (both from Zhejiang), and 0. ziguiensis Z. E. Zhao & J. Q. Wu. Recent fieldwork and herbarium research by the authors support the discovery of two additional varieties of the species Orychophragmus violaceus. Both belong to this species because they share the same height, flowering time, and number of flowers per stem. The new variety 0. violaceus var. odontopetalus Ling Wang & Chuan P. Yang differs from the typical variety in that the petals have five or more obtuse teeth (vs. ovate petals) on the upper third of the expanded blade. The clawed petals have an expanded apical section that is 10-13 mm long (vs. 7-13 mm), and the stems and pedicels are puberulent (vs. glabrous). The second new variety, 0. violaceus var. variegatus Ling Wang & Chuan P. Yang, differs from the typical variety in that the petals are strikingly white and purple variegated (vs. no variegation). Its uppermost leaves are often undivided, with sharply serrate margins (vs. irregu- larly toothed). The two new taxa are described, particularly regarding the variable characters of the petals, which are taxonomically significant. These petal features remained stable after eight years of cultivation in field plots at the Northeast Forestry University campus at Harbin in Heilongjiang Prov- 1. Orychophragmus violaceus (L.) 0. E. Schulz, Bot. Jahrb. Syst. 54(3, Beibl. 119): 56. 1916. Basionym: Brassica violacea L., Sp. PI. 2: 667. 1753. TYPE: China. Henan: Neixang Xian, Baotianman Nature Reserve Yinghu Gou, 20 May 1994, D. Boufford, Liu, Ying, Zhang & Zhu 26131 (neotype, designated by Al-Shehbaz in Cafferty & Jarvis [Taxon 51: 532. 2002], A not seen; isoneotypes, E not seen, MO not seen). doi: 10.3417/2009133 Novon 22: 109-113. Published on 10 July 2012. Volume 22, Number 2012 Wang et al. Ill Orychophragmus (Brassicaceae) in Northeast China Wang & C. RYang 09009 (NEFU). — B. Close-up of flower Showing the obtuse teeth on petals; taken from specimen It NEFU 112 Novon Figure 2. Orychophragmus violaceus var. variegatus Ling Wang & Chuan P. Yang. — A. Holotype specimen, L. Wang & X. J. Ma 8015 (NEFU). — B. Close-up of flower, showing the purple and white variegation of the petals, taken from specimen at NEFU by the authors. Ligustrum guangdongense (Oleaceae), a New Species from China Rui-jiang Wang Volume 22, Number 1 2012 Wen & Wang Ligustrum (Oleaceae) from China 115 distribution of punctate glands in L. punctifolium M. C. Chang. Scale bars: A, B = 1 cm; C, E = 2 mm; D, F, G, J = 1 mm; H, I = 2012 ’ Ligustrum (Oleaceae) from China Carex orestera (Cyperaceae), a New Sedge from the Mountains of California Peter F. Zika WTU Herbarium, University of Washington, Box 355325, Seattle, Washington 98195-5325, U.S.A. zikap@comcast.net Abstract. Carex orestera Zika (Cyperaceae) is described from western North America and is endemic to five counties in central and southern California in the Sierra Nevada range and the adjacent White Mountains. The taxon is separable from C. albonigra Mack, in its more lanceolate to oblong scales with a conspicuous pale midvein and relatively narrower perigynia. The new species is sectional key is provided for these sedges in Key words: California, Carex, Cyperaceae, IUCN Red List. Carex L. sect. Racemosae G. Don (Cyperaceae) is a northern hemisphere group of ca. 60 species with 11 indigenous species reported from California (Murray, 2002). One widespread member of the section is C. albonigra Mack., with a range from Alaska to Arizona in western North America. Specimens called C. albonigra from California have puzzled caricologists, because the specimens do not agree with typical Rocky Mountain plants (Murray, 1969, 2002; Wilson et al., 2008; Taylor, 2011; Zika et al., 2012). The Californian populations are described here as a novelty. The new sedge was first collected in 1930 by F. W. Peirson. Carex orestera Zika, sp. nov. TYPE: U.S.A. California: Mono Co., subalpine talus, above NW end of Gardisky Lake, 3250 m, 11 Aug. 2010, P. F. Zika 25338 & S. Matson (holotype, WTU; isotypes, CAN, CAS, CHSC, GH, JEPS, MICH, MO, NY, OSC, RSA, US). Figures 1, 2A, C. ratione angustioribus differt. Plants perennial, densely to loosely caespitose; rhizomes short or slightly elongated, 1-30 X 1.5-2. 5 mm, clothed in light brown to mid brown scales, sometimes slightly fibrous; fertile stems (culms) sharply triangular, scabrous distally, longer than leaves, 3.5-39 cm tall, bractlike leaf with blade 1-9.5 cm bladeless scalelike basal leaves. Stem bases light brown, sometimes suffused with red-purple. Leaves blue-green, glabrous, papillose adaxially and abax- ially, blades folded in V- or W-shape when fresh, sometimes flattening with age, abaxial surface with midvein more pronounced than lateral veins, margins scabrous, blades 20-110 X 1.3-3. 5 mm; sheath faces smooth, white to translucent or pale brown, truncate or shallowly notched, not fibrillose, often overlapping and obscured. Inflorescences dark purple, erect, 14- 60 X 4^12 mm; 2 to 4(5) ascending spikes, in elliptic to ovate (rarely oblong) cluster, rarely basal spikes separated 15-33 mm; inflorescence bract green or more usually dark purple with green midvein, sheath 0-4 mm, blade green or purple and green, 0-65 mm; basal spike subsessile, rarely on erect peduncle 3-21 mm; lateral spikes pistillate, narrowly oblong to lance-oblong, 4-12(— 19) X 2.5-5 mm; terminal spike 7-20 X 4-8.5 mm, mostly mixed sex, perigynia usually terminal and spike 9/6 (gynecandrous) or 9/6/9, rarely 9/6, 6/9 (androgynous), or 6, staminate portion of spike usually less than 70% of length; scales from lower half of lateral pistillate spikes usually lanceolate to oblong or elliptic, narrowly ovate, sometimes narrowly elliptic, elliptic- ovate, or ovate, dark purple, smooth or papillose to scabrous distally, midvein usually pale and conspic- uous, 2. 7-3. 7 mm, margins with narrow hyaline zone 0.05-0.5 mm wide, hyaline margin broadest on staminate scales, base clasping rachis, apex blunt, mucro 0. 1-0.3 mm, scabrous or not. Perigynia mostly oblanceolate to narrowly elliptic, less common variants are elliptic to flattened but bulging over achene, dark purple or largely so on a green background, the green aging to golden brown, not scabrous, papillose at 20X, 2.4- 3.6 X 0.9-1. 4(— 1.6) mm, faces nerveless or 1 to 3 faint adaxial nerves, 1 to 6 faint abaxial nerves, base Novon 22: 118-124. Published on 10 July 2012. doi: 10.3417/2011039 Volume 22, Number 2012 Volume 22, Number 1 2012 Zika Carex (Cyperaceae) from California 121 caespitose habit, dark and small, erect inflorescences terminal spikes, and dark purple scales. In addition, at 20X magnification, they share papillose perigynia, foliage, and stems, and sometimes the scales are papillose as well, demarcating them both from C. heteroneura W. Boott, which has essentially no i of Carex albonigra papillae on the perigynia, scales, and stems, and often grows at elevations below 3000 m. The similar C. helleri is found near treeline growing with C. orestera, but C. helleri has essentially no papillae on the perigynia, has occasional papillae on some scale midveins, and differs in the lower pistillate scales with their distinctive awns, often more than 0.5 mm. The corresponding scale tips of C. orestera are 0.3 mm. In the field, C. helleri usually looks quite denser headlike cluster. Table 1 shows a comparison of critical characters for Carex orestera and C. albonigra. The base of the stem is often different. Carefully made collections of C. albonigra, showing the basal scales on fertile stems and new shoots, are usually purple, and at 10X show no more than a few fibers as the scales age. In contrast, roughly a third of the collections of C. : the 1 : the r brown. Carex orestera stem bases can be slightly or moderately fibrous. In the inflorescence, C. albonigra typically has dark purple scales with an inconspic- uous dark midvein, while C. orestera has a prominently contrasting pale or whitish midvein. Mature scale shape (when perigynia are mature) can be useful in Carex sect. Racemosae, although many species are curved and slightly clasping at the base, making it difficult to see their outline even with species of this section tend to show much more variation in scale and perigynium shape than usually credited in floras and is one reason that existing identification keys can be challenging. Both C. albonigra and C. orestera are found near timberline and tolerate a broad range of soil moisture, bedrock, snow cover, and wind. Some of their variation in scale and perigynium shape may be a phenotypic response . Scales and pengym 3 taken from the lo\ Common 2 I hape Perigynium length (mm) broadly elliptic to broadly ovate, elliptic-oblong, oblong, elliptic, ovat scale body obovate, broadly ovate lanceolate to oblong or elliptic, with scale body oblanceolate to narrowly elliptic Validation of the Name Veratrum hybridum (Liliales, Melanthiaceae): The Correct Name for Crisped Bunch-Flower 126 Novon 157. MELANTHIUM. Cor ; 6-p*tala perfif. tentia, color tm mutant ia , Capful* apice divaricate. # Petalis ungulculatts imprimis albis demum ob- fcuro-rubris fcminibus femi evtitis* virginicum petalis planis.'fnaculis duabus flavis 1. notatis, floribus plerumque herma- phroditis. hybridum 2. petalis plicato-undulatis itoroaculatls, floribus mafcuHset fcemineismixtis. monoicum 3. petalis plants* maculis 2 luteis ; flo- t _ ^Uc/LnJkiitr* toijttu -nbns inferioribus 'raafculis majo-^f/7 • ribus, panicuiis lateralibus ; fupe- rioribus foemineis racemo terminali. Figure 1. Description of Melanthium hybridum in the facsimile edition of Flora Caroliniana (Walter, 1946: 125), showing with glands) and not the adjective modified by owners) of an original copy of Flora Caroliniana. The establishment of Melanthium hybridum as the correct name for the crisped bunch-flower generates an unresolved nomenclatural issue, relative to the circumscription of Melanthium L. (two to four species) versus Veratrum L. (17 to 45 species). The two genera have a complex taxonomic and nomen- clatural history, and Veratrum has been variously defined with Melanthium submerged totally or in part (cf. summaries in Zimmerman, 1958; Bodkin, 1978; Zomlefer et al., 2003). Recent molecular analyses show that recognition of the traditional Melanthium makes Veratrum paraphyletic, supporting a recom- mendation to submerge Melanthium within Veratrum s.l. (Zomlefer et al., 2003). However, although these two Linnaean generic names have equal priority (Linnaeus, 1753), Thunberg (1797) first cited Veratrum as the synonym of Melanthium, thereby establishing the priority of Melanthium over Veratrum (Art. 11.5; McNeill et al., 2012). Therefore, combi- nations witnin Veratrum based on Melanthium are valid and legitimate, but are incorrect. A proposal (Zomlefer et al., 2010) to continue the widespread usage of Veratrum, which would best serve stability of nomenclature (Art. 14.2, McNeill et al., 2012), was approved by the Nomenclature Committee for Vascular Plants (Barrie, 2011; Brummitt, 2011; McNeill et al., 2011). The voted acceptance of this proposal at the Melbourne International Botanical Congress establishes the name Veratrum as now conserved over Melanthium (Barrie, 2011; Brummitt, 2011; McNeill et al., 2012). This necessitates the transfer of the name Melanthium hybridum to Veratrum, as previously attempted by the monographer Zimmerman (1958), who accepted Walter’s name and circumscribed a broadly defined, monophyletic Vera- trum. However, his recombination (Kupchan et al., 1961: 11) was not validly published (Gandhi, 2007, online), because the reference to the basionym is incomplete, lacking a citation of publication, page, and date (Art. 33.4, McNeill et al., 2012). The complete reference for this new combination is provided below, giving James Hall Zimmerman (1924-1992) credit for his studies on Veratrum and including recognition of the priority of Walter’s name for the crisped bunch- flower (Zimmerman, 1958; Kupchan et al., 1961). Veratrum hybridum (Walter) J. H. Zimmerman ex Zomlefer, comb. nov. Basionym: Melanthium hybridum Walter, FI. Carol. 125. 1788. Lei- 127 Alpinia rugosa (Zingiberaceae), a New Species from Hainan, China Pu Zou, Yu-Shi Ye, Sen-Jen Chen, f Zhong-Yi Chen, and Jing-Ping Liao Volume 22, Number 2012 I Acknowledgment of Reviewers Novon 22: 131-132. Ji-Pei Yue Sir www.mbgpress.info CONTENTS -j4J|k Missouri Botanical Garden A Journal for Botanical Nomenclature ME 2 2 V 0 L U NUMBER 2 2012 Volume 22, Number 2 November 2012 Novon, A Journal for Botanical Nomenclature from the Missouri Botanical Garden The mission of the Missouri Botanical Garden is to discover and share knowledge about plants and their environment in order to preserve and enrich life. Novon publishes short articles whose primary purpose is the establishment of nomencla- ture in vascular plants and bryophytes. All articles are peer-reviewed by qualified, indepen- dent reviewers. Manuscripts must fully state and justify the reasons for proposing nova. These may include detailed comparisons with similar taxa, short keys to similar taxa, illustrations to similar taxa, and mechanical nomenclature reasons, among others. Manuscripts whose primary purpose is other than establishment of new nomenclature, which usually are longer manuscripts, cannot be accepted for review. These include reviews, revisions, monographs, or other papers that incidentally include nova. Manuscripts must follow the guidelines in the Checklist for Authors. The Checklist may be downloaded from the Garden’s web site, www.mbgpress.info, or authors may contact the editor at novon@mobot.org to request a copy. Novon will not knowingly accept manuscripts that have been simultaneously submitted to other journals for consideration or previously published in some form elsewhere. Victoria C. Hollowell Editor Allison M. Brock Associate Editor Tammy Charron Associate Editor Cirri R. Moran Press Coordinator Ihsan A. Al-Shehbaz Consulting Editor Carmen Ulloa Ulloa Consulting Editor George Yatskievych Consulting Editor Kanchi N. Gandhi Nomenclature Consultant Nicholas J. Turland Nomenclature Consultant Roy E. Gereau Latin Editor Novon is included in the subscription price of the Annals of the Missouri Botanical Garden. Price for 2012: $180 per year U.S.A.; $190 Canada and Mexico; $215 all other countries. Four issues per Postal address: Subscriptions: Annals of the Missouri Botanical Garden, P.0. Box 1897, Law- rence, KS 66044-8897; Manuscript submissions: Novon, Missouri Botanical Garden, P.0. Box 299, St. Louis, MO 63166-0299. Editorial queries: novon@mobot.org Orders for back issues: mbgpress@mobot.org Web site: http://www.mbgpress.info © Missouri Botanical Garden 2012 0 Novon is printed on paper that meets the re- quirements of ANSI/NISO Z39.48-1992. Novon (ISSN 1055-3177) is published quarterly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Periodicals postage paid at St. Louis, MO and additional mail- ing offices (USPS #006-777). POSTMASTER: Please send address changes to Annals of the Missouri Botanical Garden, P.0. Box 1897, Lawrence, KS 66044-8897. Information on the contents of Novon is present- ed in SciSearch®, Research Alert®, Current Con- tents®/Agriculture, Biology, and Environmental Sciences, APT Online, the CAB Abstract/Global Health database, and the ISI® database. The full-text of Novon is available online through BioOne™ (http://www.bioone.org). Volume 22 NO VON Number 2 W 2012 A New Species of Amaranthus (Amaranthaceae) from Salta, Argentina 134 Novon 1.5-4 X 0.5-0.7 Jacquemontia macrocalyx (Convolvulc the , a New Species Endemic to Brazil 22: 137-140. 138 Novon Volume 22, Number 2 2012 Buril & Alves 139 Jacquemontia (Convolvulaceae) from Brazil Outer sepal 0.45-1.3 X 0.35-0.8 c 1 .2-1.8 X 0.8-1. 5 c: Brazil (Bahia) dromous, with 5 pairs of secondary veins; petiole 4-6 mm. Inflorescences as monochasial or dichasial cymes, 2- or 3-flowered; peduncles 0.5-4.5 cm, axillary, when terminal very short, velutinous; bracts 2, ca. 5 mm, linear, velutinous; sepals membranous, entire, unequal, the outer two 1.2-1 .8 X 0.8-1. 5 cm, ovate to deltoid, rarely lanceolate, base cordate, apex acuminate to 1-1.4 X 0.3-0. 4 cm, asymmetric, inner two 0.3-0.5 X 0.2-0.25 cm, entire, ovate to lanceolate, base rounded, apex acuminate, pubescent; corolla 1.4-1. 5 cm long, funnelform, apex of midpetaline line ciliate, lilac with the tube light pink; stamens ca. 8 mm, filaments pilose on the base, anthers ovoid, white; nectary absent; ovary ca. 0.5 mm, oblong, glabrous, 2-locular, 2 ovules per locule; style 0.5-1 cm, stigmatic lobes ca. 0.5 mm, Two New Species of Stereospermum (Bignoniaceae) from Madagascar Martin W. Callmander Missouri Botanical Garden, P.0. Box 299, St. Louis, Missouri 63166-0299, U.S.A, and Conservatoire et Jardin botaniques de la ville de Geneve, ch. de Flmperatrice 1, case postale 60, CH-1292 Chambesy, Switzerland, martin.callmander@mobot.org Peter B. Phillipson Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166-0299, U.S.A. , and Museum national d’Histoire naturelle, Departement de Systematique et Evolution, UMR 7205, Case Postale 39, 57 rue Cuvier, F-75231 Paris cedex 05, France, peter.phillipson@mobot.org George E. Schatz Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166-0299, U.S.A. george.schatz@mobot.org Abstract. A recent review of the Malagasy Bigno- niaceae for the Catalogue of the Vascular Plants of Madagascar has led to the discovery of two new species in the genus Stereospermum Cham.: S. gentryi Callm., Phillipson & G. E. Schatz and S. randria- naivoi Callm., Phillipson & G. E. Schatz. Stereo- with red glands that give a reddish color when dry, and a long, lax, paniculate inflorescence bearing flowers with large corollas that are red-purple inside the mouth and throat. Stereospermum randrianaivoi can be distinguished by its usually 3-foliolate leaves i thick, coriaceous leaflets that are discolorous, upper surface a dull gray-green, and shiny in vivo. Both of the new species are provided with illustra- tions, a discussion of their morphological affinities, and a conservation threat analysis based on the IUCN Red List Criteria. Resume. Une revision recente de la famille des Bignoniaceae pour le Catalogue des plantes de Madagascar nous permets de decouvrir deux nou- velles especes dans le genre Stereospermum Cham.: S. gentryi Callm., Phillipson & G. E. Schatz and S. randrianaivoi Callm., Phillipson & G. E. Schatz. Stereospermum gentryi est unique possedant des folioles a limbe tachete de glandes rouges un aspect rougeatre sur le sec et une longue et diffuse inflorescence portant des fleurs a large corolles dont Pinterieur de la gorge et de la bouche est rouge pourpre. Stereospermum randrianaivoi peut etre distingue par ses feuilles generalement a 3-folioles, son limbe coriace, epais, discolore qui seche de fagon rosatre (brillant in vivo). Les deux nouvelles especes Madagascar is the second largest center of Bignoniaceae diversity in the world after South America, with ca. 80 currently accepted species (and numerous varieties) in nine genera (Perrier de la Bathie, 1938a, 1938b; Capuron, 1960, 1970; Gentry, 1977; Zjhra, 2006). A recent review of the Malagasy Bignoniaceae for the Catalogue of the Vascular Plants of Madagascar () has enabled us to refine taxon delimitations and has led to the discovery of new Bojer ex Meisn. (Callmander & Phillipson, 2012), Ophiocolea H. Perrier (Callmander et al., 2011), and Rhodocolea Baill. (Callmander & Phillipson, 2011). Stereospermum Cham, is distributed from Africa (four species) and Madagascar (12 spp., of which two are described in this article), to southern and southeastern Asia (four spp.). The genus can be recognized easily by its flowers with a narrowly cylindric corolla tube and its cylindric, podlike fruit with a bony seed chamber possessing alternating deep cavities in which the seeds are lodged (Schatz, 2001). The high level of species diversity in Madagascar corresponds to considerable morpholog- much of the country and occurring in varied habitats and on different substrates. Certain species are quite doi: 10.3417/2010115 Novon 22: 141-147. Published on 30 November 2012. Figure 1. Stereospermum gentryi Callm., Phillipson & G. E. Schatz. — A. Fertile branching with inflorescence. — B. Detail of a Rafamantanantsoa 3479 (TAN). Scale bars: A = 2 cm; B-D = 1 cm. Observations. Stereospermum gentryi is distin- color when dry, and with a reddish brown indument guished from other Malagasy Stereospermum by its on both blade surfaces. Also distinctive are the large leaves with 21 leaflets or more, with the lamina trumpet-shaped corollas (70-80 mm long) with the adaxially spotted with red glands giving a reddish throat and mouth red-purple inside, the tube Novon 44° 46“ 48° 50“ expanding gradually to the mouth, and the long, lax paniculate inflorescence (ca. 20-30 cm long), quite unlike those of any other species in the genus, which are mostly white, smaller (under 60 mm long), narrowly tubular and expanding abruptly at the mouth, and borne in a smaller inflorescence. In Madagascar, only a single species of Stereospermum has larger flowers: S. longiflorum Capuron. It is endemic to dry forest of the extreme north, the flowers are white, and the inflorescence is highly reduced. Paratypes. MADAGASCAR. Toliara Prov.: Andohahe- la RN, E of Ifarantsa-Enakara Ambony Rd., 24°46'S, 46°51'E, 26 Apr. 1988 (11.), A. J. M. Leeuwenberg, J. Floret, P. P. Lowry & R. Rajemisa 13962 (MO, P, WAG); Anosy, Col du Maningotry, 24°44'S, 46°51'E, 30 Jan. 1990 (£.), R. Rabevohitra 2253 (MO, P, TAN). i Callm., Phillipson & G. E. Schatz, sp. nov. TYPE: Madagascar. Antsiranana Prov.: foret de Sahafary, [12°34'S, 049°26'E], [50 m], 10 Dec. 1952 (fl. & fr.), Service Forestier 6285 (holotype, P [P00730636]; isotypes, G [G00303500], MO [MO-097472], MO [MO-2160847], P [P00730638], TEF). Figure 3. et al. 147 Bauhinia tuichiensis (Fabaceae, Cercideae), una Especie Nueva del Bosque Seco de la Region Madidi, Bolivia Novon 22: 148-151. doi: 10.3417/2009062 Volume 22, Number 2 2012 Cayola & marrones hacia el apice, (8-)10 X 2(-3.5) mm, i 0.8-1. 1 X ca. 1 mm, 1 Nomenclatural Notes on Eriotheca (Bombacoideae, Malvaceae) Marilia Cristina Duarte Programa de Pos-Graduagao em Biodiversidade Vegetal e Meio Ambiente, Instituto de Botanica, Secretaria do Meio Ambiente, Cx. Postal 3005, 01061-970, Sao Paulo, SP, Brazil. mcdbot@hotmail .com Gerleni Lopes Esteves Instituto de Botanica, Secretaria do Meio Ambiente, Cx. Postal 3005, 01061-970, Sao Paulo, SP, Brazil, gerleniibot@yahoo.com.br Abstract. As part of the taxonomic treatment of Eriotheca Schott & Endl. in Brazil (Bombacoideae, Malvaceae), a lectotype is designated for Bombax crenulatum K. Schum. [= E. gracilipes (K. Schum.) A. Robyns], and an epitype is selected for E. globosa (Aubl.) A. Robyns. The change of status is proposed for two names in Eriotheca. Eriotheca candolleana (K. Schum.) A. Robyns var. longipes A. Robyns is recognized at the species level as E. longipes (A. Robyns) M. C. Duarte & G. L. Esteves; E. macro- phylla (K. Schum.) A. Robyns subsp. sclerophylla (Ducke) A. Robyns is also recognized at species rank as E. sclerophylla (Ducke) M. C. Duarte & G. L. Esteves. Resumo. Durante o estudo taxonomico de Eriotheca Schott ex Endl. no Brasil (Bombacoideae, Malva- ceae), verificou-se que mudangas eram necessarias. Sao designados um lectotipo para Bombax crenula- tum K. Schum. [= E. gracilipes (K. Schum.) A. Robyns] e um epftipo para E. globosa (Aubl.) A. Robyns. Duas mudangas de status sao propostas: E. candolleana (K. Schum.) A. Robyns var. longipes A. Robyns para E. longipes (A. Robyns) M. C. Duarte & G. L. Esteves e E. macrophylla (K. Schum.) A. Robyns subsp. sclerophylla (Ducke) A. Robyns para E. sclerophylla (Ducke) M. C. Duarte & G. L. Esteves. with the free filaments that emerge from the apex of the staminal tube (Robyns, 1963; Duarte, 2010). Eriotheca has been historically interpreted differ- ently in the taxonomic literature. The genus was established by Schott and Endlicher in 1832 on the wo species originally described within Bombax L. Schumann (1886) and van den Brink (1924) would later reject the use of Eriotheca, either synonymizing it under Bombax or recognizing it within Bombax sect, y Eriotheca (Schott & Endl.) Bakh. It was not until 1963 that Andre Robyns would again, and his revision of Bombax s.l. is the most significant recent study of these taxa. The current authors recognize 19 species in Eriotheca, divided in four subgenera, three of which have Brazilian representatives, as Eriotheca subg. Eriotheca Schott & Endl., Eriotheca subg. Tartagalia (Capurro) A. Robyns, and Eriotheca subg. Macrosiphon A. Robyns (Duarte, 2010). During the preparation of a taxonomic revision of Eriotheca in Brazil, nomenclature within Eriotheca was clarified to facilitate future taxonomic otheca Schott & Endl., Melet. Bot. 35. 1832. TYPE: Eriotheca pubescens (Mart. & Zucc.) Schott & Endl. (lectotype, designated by Robyns [1963: 125]). Eriotheca Schott & Endl. (Bombacoideae, Malva- ceae s.l.) comprises about 24 species distributed in South America. In Brazil, the genus is represented by 16 species, from northern Brazil to the state of Sao Paulo, which represents its southern extent. The genus is known to occur in the Atlantic rainforest, as well as Amazon forest and cerrado. Eriotheca is characterized by its small flowers that reach to 6.5 cm long and by the 18 to 170 stamens partially connate, Discussion. The history of the species type of the genus Eriotheca began in 1825, when Martius and Zuccarini described Bombax parvifolium Mart. & Zucc. (Martius, 1825). In 1826, the same authors committed an orthographic error, writing B. parvi- florum instead of B. parvifolium (Martius, 1826). In 1832, Schott and Endlicher described the genus Eriotheca, characterized by the stamens partially Novon 22: 152-154. Published on 30 November 2012. doi: 10.3417/2010066 A New Species of Pelexia (Orchidaceae, Spiranthinae) from Sao Paulo, Brazil Alessandro W. C. Ferreira Universidade Federal do Maranhao, Campus V. Rua Raimundo Jose Pimenta s.n., 65200-000, Pinheiro, MA, Rrazil. alessandrowcf@yahoo.com.br Maria Ines S. Lima Departamento de Botanica, Universidade Federal de Sao Carlos, Rodovia Washington Luis, km 235, Caixa Postal 676, 13565-905, Sao Carlos, SP, Brazil, ines@ufscar.br Emerson R. Pansarin Departamento de Biologia, Faculdade de Filosofia, Ciencias e Letras de Ribeirao Preto, Universidade de Sao Paulo, Av. Bandeirantes 3900, 14040-901, Ribeirao Preto, SP, Brazil. epansarin@ffclrp.usp.br Abstract. A new species of Pelexia Poit. ex Lindl. (Orchidaceae, Spiranthinae) occurring in central Sao Paulo, southeastern Brazil, is described and illus- trated as P. vinosa A. W. C. Ferreira, M. I. S. Lima & Pansarin. Pelexia vinosa is recognized by its leaves that are present at flowering and its dark purple leaf blades with reddish margins. Inflorescences are sparsely pubescent and reddish. The red sepals contrast with the white hyaline petals and labellum. The species is notable for its spurlike nectary that is parallel and adnate to the ovary. The new species is morphologically similar to P. laxa (Poepp. & Endl.) Lindl. In addition, the need to preserve native areas of the interior of Sao Paulo State (habitat of P. vinosa ) is discussed. Resumo. Uma nova especie de Pelexia Poit. ex Lindl. (Orchidaceae, Spiranthinae) que ocorre na regiao central do estado de Sao Paulo, e descrita e ilustrada como P. vinosa A. W. C. Ferreira, M. I. S. Lima & Pansarin. Pelexia vinosa e reconhecida pelas folhas que estao sempre presentes durante o perfodo de floragao, e pelas laminas foliares roxo-escuras com margens avermelhadas. As inflorescencias sao aver- melhadas e esparsamente pubescentes. As petalas avermelhadas contrastam com as petalas e labelo, que sao brancos, e pelo nectario paralelamente disposto e adnato ao ovario. Essa nova especie e morfologicamente semelhante a P. laxa (Poepp. & nativas no interior do estado de Sao Paulo (habitat de P. vinosa ), tambem e discutida. Within the Orchidaceae, the tribe Cranichideae is cosmopolitan in distribution and comprises ca. 90 genera and 1600 species that are further recognized among six subtribes, one of which is the Spiranthinae (Salazar, 2003). The Spiranthinae encompass 39 genera, of which Pelexia Poit. ex Lindl. contains 76 species distributed from Mexico to Argentina (Hoehne, 1945; Pabst & Dungs, 1975; Salazar, 2003). Pelexia consists of terrestrial, lithophytic, or more rarely epiphytic herbs (Salazar, 2003). In Brazil, 38 species of Pelexia are found (Barros et ah, 2012), occurring in several vegetation physiognomies, in- cluding ombrophilous (e.g., Miller & Warren, 1996; Romanini & Barros, 2007), semideciduous meso- phytic (e.g., Pansarin & Pansarin, 2008, 2010), Amazonian forests (Ribeiro, 1992), as well as cerrado vegetation (e.g., Batista & Bianchetti, 2003). Fur- thermore, Brazilian species of Pelexia can be found in diverse habitats that include marshy areas, rocky outcrops, and both dry and wet grasslands (Cogniaux, 1893-1896; Hoehne, 1945; Pabst & Dungs, 1975; Sprunger et ah, 1996). Seven species of Pelexia are known to occur in the central region of the state of Sao Paulo (Ferreira et ah, 2010). This is an area Brazilian biomes, namely the cerrado and Atlantic forest (Kronka et ah, 1993; Soares et ah, 2003). The climate of the region is classified as “Cwa,” namely mesothermic with a dry winter season (Koppen, 1948). The new species described herein was discovered while the authors carried out a floristic survey of the Orchidaceae in central Sao Paulo in southeastern Brazil. The genus Pelexia is characterized by the presence of fleshy and fasciculate roots, and the leaves are doi: 10.3417/2010102 Novon 22: 155-159. Published on 30 November 2012. hyaline petals and labellum, and with the spur parallel and completely adnate to the ovary. The new species is closely related to P. laxa. However, P. vinosa differs from P. laxa by the absence of maculae on the leaf blades, the more congested inflorescence, the lateral sepals not decurrent, the dorsal sepal an apical lobe that is transversally oval and triangular and with lateral lobes that are falcate. In P. laxa, the leaves are maculate, the inflorescences are more elongate and lax, the lateral sepals are decurrent, the dorsal sepal is elliptic, and the labellum is yellow in the middle with an apical lobe that is cordiform and with lateral lobes that are triangular. Although Barros et al. (2012) reported the occurrence of P. laxa from northern Brazil, this species was recorded from the central region of the state of Sao Paulo, flowering from September to November (Ferreira et al., 2010). Because species of Pelexia are cultivated as ornamental orchids (Salazar, 2003), we believe that the new species also has ornamental potential, based on its beautiful dark purple leaf blades and the reddish inflorescences and sepals that markedly contrast with the white petals and labellum. In Brazil, native areas of the cerrado vegetation, the Atlantic rainforests, and ecotonal areas between both biomes (the natural habitat of Pelexia vinosa ) have been considered crucial for preservation and conservation as rich in species diversity and endemism. Currently, these Brazilian biomes are 158 Novon Polystichum oblanceolatum , a New Species in Section Haplopolystichum (Dryopteridaceae) from Guangxi, China HaiHe College of Life Sciences, Chongqing Normal University, Shapingba, Chongqing 400047, People’s Republic of China, hehaicq@gmail.com Li-Bing Zhang* Missouri Botanical Garden, P.0. Box 299, St. Louis, Missouri 63166-0299, U.S.A., and Chengdu Institute of Biology, Chinese Academy of Sciences, P.0. Box 416, Chengdu, Sichuan 610041, People’s Republic of China. * Corresponding author: Libing.Zhang@mobot.org Abstract. A new fern species, Polystichum oblan- ceolatum H. He & Li Bing Zhang (Dryopteridaceae), is described and illustrated from a karst cave in Guangxi, southern China. As a member of Poly- stichum Roth sect. Haplopolystichum Tagawa, P. oblanceolatum is morphologically most similar to P. dielsii Christ and P. liui Ching, but is distinguishable from the latter two by its often oblanceolate lamina with acute apex, its repand or broadly dentate pinnae with non-spinulose toothed distal margins, and the less developed, more or less truncate, but not sharply acute, basal acroscopic auricles of the pinnae. The new taxon appears to be endemic to a single karst cave in Tian’e County in northern Guangxi and is considered to be Critically Endangered (CR), based on IUCN Red List criteria. A key is provided to facilitate the identification of P. oblanceolatum and morphologically similar species. Key words: China, Dryopteridaceae, Guangxi, IUCN Red List, Polystichum. Species of Polystichum Roth (Dryopteridaceae) are widely distributed in the mountains of subtropical and warm- temperate areas in the Northern Hemi- the superficial similarity of many taxa, and the many genus is estimated to contain (180 to)225 to 400 species worldwide (Kramer, 1990; Fraser- Jenkins, 1991; Kung et al., 2001; Wang & Wang, 2001; Lu, 2005), with about 250 or more species estimated in Asia (Benniamin et ah, 2008). In China, Polystichum is the most species-rich fern genus, with most known species being found in southwestern China and adjacent regions (Kung et ah, 2001). Since the publication of the Chinese-language Flora Reipubli- cae Popularis Sinicae, which recognized 168 species in Polystichum (Kung et al., 2001), more than 13 new species from southwestern China have been de- scribed (cf. Zhang & He, 2012). Our extended fieldwork with respect to the fern flora in Chongqing, Guangxi, Guizhou, Sichuan, and Yunnan in recent years has so far led to the discovery of several new species from limestone areas, most of which are endemic to a single cave or sinkhole (Zhang & He, 2009a, 2009b, 2009c, 2010, 2011, 2012; He & Zhang, 2010, 2011; Zhang et al, 2010). However, the high diversity of Polystichum in China, especially in less explored southwestern China, is still far from being well understood. During fieldwork in 2009, we found a species of Polystichum with mostly oblanceolate leaves from a karst cave in Tian’e County, northern Guangxi, China. Herein we describe this as another new cave species of Polystichum sect. Haplopolystichum Taga- Polystichum oblanceolatum H. He & Li Bing Zhang, sp. nov. TYPE: China. Guangxi: Tian’e Co., Bala (Laopeng) Yaozu Autonomous Town- ship, Madong Village, Zhichang, on weathered crust of limestone rocks at bottom of one of two small caves inside a larger karst cave, 24°48.59'N, 107°03.05'E, ca. 890 m, 13 Nov. 2009, L. B. Zhang & H. He 5179 (holotype, CDBI; isotypes, CDBI, CTC, MO). Figures 1, 2. Species Polysticho liui Ching et P. dielsii Chnst affims, spinuloso-dentatis atque auriculis basalibus acroscopicis Plants perennial, evergreen, (2.8-)6.5-9.5 cm tall; rhizome erect, ca. 4.3 X 2. 7-4.3 mm with bases of remnant old petioles, sparsely scaly; scales lanceo- late, ca. 0.7 X 0.1 mm, dark brown, margins entire; Novon 22: 160-165. Published on 30 November 2012. doi: 10.3417/2011010 Novon He & , College of Life Li-Bing Zhang * doi: 10.341' Volume 22, Number 2 2012 Heetal. 167 Polystichum loratum (Dryopteridaceae) from Guizhou, China when dry, densely covered with scales abaxially, to)25 to 55 pairs, alternate, 1-2 n sparsely scaly (especially at base of petiolules) adaxially; abaxial rachis scales ovate to lanceolate, similar to those of petiole but smaller, lighter in color and thinner in texture, adaxial rachis scales deltoid and subulate, dark brown, to 2 X 0.4 mm, hairlike outgrowths only at broadest base, margins nearly entire, apex caudate; pinnae nearly leathery, in (16 apart even if other pinnae are imbricate), asymmetri- cally rectangular to oblong (or slightly falcate toward apex), 3. 5-6.5 X 1.8-3 mm, lowest pairs usually 1/2 to 2/3 as long as middle pinnae, asymmetrically deltoid or widely so, shortly petiolulate (petiolules to ca. 1 mm, attached to rachis at an acroscopic angle of ca. Hoy a daimenglongensis (Apocynaceae, Asclepiadoideae), a New Species from Yunnan, China Passifiora unipetala , a New Bat-pollinated Species of Passiflora supersect. Tacsonia (Passifloraceae) Missouri Botanical Garden, P.0. Box 299, St Louis, Missouri 63166-0299, U.S.A. Volume 22, Number 2 2012 176 Novon doi: 10.341' Volume 22, Number 2 2012 Li et al. 181 Meconopsis (Papaveraceae) from Sichuan, China ovate, ca. 1 mm, yellow; ovary ellipsoid-oblong, ca. 7 glabrous; style slender, 2-3 mm, stigma capitate, cream colored. Fruit not seen. Distribution and habitat. Meconopsis xiang- chengensis is currently known only from two sites in the southern Hengduan Mountains, in alpine mead- ows, at altitudes of 4385-4700 m. The type locality is in Xiangcheng County of southwestern Sichuan Province and the second site is to the south, in Shangri-La County of northwestern Yunnan Province, China. The new species may be considered endemic to these Hengduan Mountain localities. IUCN Red List category. Meconopsis xiangchen- gensis was collected only from two sites and is therefore assumed to be rare. It is considered Endangered (EN), according to IUCN Red List criteria (2001), since its known area of occupancy is less than 500 km2 (criterion B2) and the total population size is estimated to be smaller than 2500 mature individuals (criterion C). Phenology. Meconopsis xiangchengensis was ob- served with flowers in July. Etymology. The specific epithet of the new species refers to the type locality, Xiangcheng County in Sichuan Province, China. Discussion. According to Wu and Zhuang (1980), Wu et al. (1999), and Taylor’s (1934) monographic to Meconopsis sect. Simplicifoliae (Tayl.) C. Y. Wu & H. Chuang in Meconopsis ser. Delavayanae Tayl. The Dryopteris jiucaipingensis (Dryopteridaceae), a New Species in Dryopteris sect. Hirtipedes from Guizhou, China zaipingensis P. S. Wang, Q. Luo & Li Bing Zhang. —A. Habit. — B. Portion of (Dryopteridaceae) Polystichum tiankengicola (Dryopteridaceae), a New Species from Karst Sinkhole from Guizhou, China Qiang Luo Luo&; 188 Novon Woodsia guizhouensis (Woodsiaceae), a N Area in Guizhou, China Li-Bing Zhang* Chengdu 192 Novon Volume 22, Number 2 2012 Luo & Zhang 1 Woodsia (Woodsiaceae) from Guizhou, Chir Taxonomic Revision of the Casearia ulmifolia Complex (Salicaceae) Ronaldo Marquete Instituto Brasileiro de Geografia e Estatfstica, Av. Republica do Chile 500, 15° andar, Rio de Janeiro, RJ, CEP 20031-170, Brazil, and Jardim Botanico do Rio de Janeiro, Programa de Pos Graduagao em Botanica, Rua Pacheco Leao, 2040, Rio de Janeiro, RJ, 22460-030, Brazil. marquete@ibge.gov.br; rmarquet@jbrj.gov.br Vidal de F. Mansano i Jardim Botanico do Rio de Janeiro, Rua Pacheco Leao 915, Rio de Janeiro, RJ, 22460-030, Brazil, vidalmansano@gmail.com Abstract. Field studies and examination of herbar- ium specimens have led to a taxonomic revision of the Casearia ulmifolia Vahl ex Vent, complex (Salica- ceae). A new species circumscription is presented for C. ulmifolia, C. cotticensis Uittien is reestablished, and one new species, C. murceana R. Marquete & Mansano, is described from Para, Brazil. Morpholog- ical descriptions, a taxonomic key, geographical distributions, diagnostic characteristics of the spe- cies, and illustrations are provided. Key words: Brazil, Casearia, Flacourtiaceae, IUCN Red List, Para, Salicaceae, South America. Casearia Jacq. (Salicaceae, which now includes the non-cyanogenic Flacourtiaceae, Chase et al., 2002; Angiosperm Phylogeny Group, 2009) is a pantropical genus that includes ca. 180 species occurring in many different vegetation types ranging from dense ombrophilous forests, seasonal semi- deciduous forests, savannas, and coastal vegetation. Sleumer (1980) cited 80 species of Casearia in the Neotropics and subdivided this genus into six sidered as six informal groups. Casearia ulmifolia Vahl ex Vent, belongs to the Arboreae group of the Casearia sect. Casearia, which comprises 10 species. While reviewing Casearia species in Brazil for taxonomic study (Marquete, unpublished), we noted that Sleumer (1980) broadly circumscribed C. ulmifolia, synonymizing 10 names in recognition of this species. This resulted in a taxon with broad morphological variation. Casearia ulmifolia, for instance, has flowers with oblong anthers, an barbate to glabrous, longitudinal anther dehiscence with lateral dehiscent zones, and stigmas with of C. cotticensis Uittien present deltoid anthers, a developed connective with glabrous or subglabrous Novon 22: 196-206. Published on 30 November 2012. apical glands, transverse anther dehiscence with dehiscent zones oriented downward, and stigmas that are tomentose. Despite these distinctions, Sleumer (1980) considered these taxa conspecific. The complex Casearia ulmifolia studied here is characterized by its stipule shape (subulate or linear- lanceolate) and indument (villous or tomentose), leaf blade with acuminate to cuspidate apex, oblong or deltoid anthers, barbate, glabrous or tomentose apical gland, elongated, ovate ovary, and slightly sulcate or fimbriate seed testa. Species of the Casearia ulmifolia complex occur in the Amazon region bordered by the states of Mato Grosso to the south and the states of Roraima and Amapa to the north. This distribution reaches the Atlantic coast of southeastern Brazil, since the complex extends from Maranhao and Piauf southward to Bahia, Espirito Santo, and Minas Gerais. These species also occur in other South American countries such as Colombia, Guyana, Peru, Surinam, Trinidad, and Venezuela. These plants were noted to grow at altitudes from 50 to 950 m. The species of this complex occur in dense ombrophilous forests and its variations, according to Veloso et al. (1991). Despite the fact that Casearia cotticensis and C. ulmifolia are distributed in close localities, they are morphologically distinct. They differ by leaf texture, trichomes, venation, and especially anther dehis- cence. Casearia murceana R. Marquete & Mansano, found in Para State in eastern Amazonia, is more similar to C. ulmifolia, differing by the inflorescence structure, absence of glands, and flower indument. Sleumer (1954, 1980), who studied the Malaysian (eight species) and Neotropical species of Flacourtia- ceae, respectively, and Breteler (2008), in his work for Casearia from Africa, mentioned that the flowers of the genus do not offer characters to distinguish the species. We verified that floral characters in the doi: 10.3417/2011011 198 Novon 200 Novon 869334, P [B fragm.] not seen, L [barcode] 00010777 [Bfragm.] not seen, L photo). Lozania nunkui (Lacistemataceae), a New Species from the Sandstone Plateaus of the Cordillera del Condor in Ecuador and Peru David, A. Neill and Mercedes Asanza Universidad Estatal Amazonica, Facultad de Ingeniena Ambiental, km 2 V2 via a Napo, Puyo, Pastaza, Ecuador, davidneill53@gmail.com; gman2260@gmail.com Abstract. Lozania nunkui D. A. Nei (Lacistemataceae), a new species from the Cordillera del Condor region of southeastern Ecuador and adjacent northern Peru, is described and illustrated. It is distinguished from other species in the genus by the glabrous, entire, relatively thick and coriaceous, sclerophyllous leaves, and uniquely, by the five or all other known species. The new species is restricted to nutrient-poor, highly acidic sandstone plateaus within its range, and the thick sclerophyllous leaves appear to be an ecological adaptation to that edaphic condition. Resumen. Se describe y se ilustra Lozania nunkui D. A. Neill & Asanza (Lacistemataceae), una nueva especie de la region de la Cordillera del Condor en el sureste del Ecuador y la zona colindante del norte de Lozania nunkui various genera and families (Neill, 2005; Daly et al., 2012). Most of these undescribed taxa are evidently restricted to the outcrops of Cretaceous sandstone of the Hollrn Formation, which constitute about 20% of the land area in the Cordillera del Condor region. Some of these substrate-restricted species exhibit morphological features that appear to reflect an ecological adaptation to the highly acidic, nutrient- poor soils that are derived from the sandstone substrate, frequently including glabrous, thick, coriaceous or sclerophyllous leaves. Among these novelties that are evidently local endemics on the sandstone plateaus of the Cordillera del Condor, is a species of Lozania that we and our colleagues have found on several sandstone outcrops in the region. Peru. ! por las hojas glabras, enteras, relativamente gruesas, coriaceas y esclerofilas, y por los cinco u ocasional- mente seis sepalos, en vez de los cuatro sepalos que presentan todas las demas especies. Dentro de su rango geografico, la especie nueva esta restringida a nutrientes y muy acidos, y las hojas gruesas y esclerofilas parecen ser una adaptation ecologica a esa condition edafica. D. A. Neill & Asanza, sp. nov. TYPE: Ecuador. Morona-Santiago: Limon-In- danza, Cordillera del Condor, Centro Shuar Kuankus, summit of Cerro Chuank Naint (“Vulture Mountain” in Shuar), a sandstone massif, 1270 m, 3°3'36"S, 78°14'25"W, 18 Sep. 2005 (fl., fr.), D. Neill, M. Asanza, W. Quizhpe & C. Morales 14657 (holotype, QCNE; isotypes, AAU, GB, HUT, K, LOJA, MO, NY, US). Lozania Mutis ex Caldas I small Neotropical genus of just four accepted species at present (Sleumer, 1980; Gentry, 1988), with a distribution including southern Central America (southeast Nicaragua to Panama), the tropical Andes (Venezuela to Bolivia), and western Amazonia, including Amazonas, Brazil. All known species are small trees, sometimes shrubby, occurring in the understory of mature-phase moist and wet forests up to ca. 2400 m elevation. In the Cordillera del Condor region of southeastern Ecuador and northern Peru, floristic inventory work in recent years has resulted in collections of a Tree attaining 10 m in height with trunk ca. 10 cm diam.; branches glabrous, mostly borne horizontally. Leaves elliptic to elliptic-obovate, broadly obtuse to ceous, rather thick, stiff and sclerophyllous, com- pletely glabrous above and below, or glabrescent with a few scattered simple unicellular hairs on the lower surface, with entire margins, the blade (4-)5-10(— 11) X (2.5-)3-5 cm, petiole 7-9 mm, glabrous or with a few strigose hairs; leaf surface lustrous, conspicu- ously glossy on upper surface, somewhat duller beneath, costa impressed above, prominent below; lateral nerves 4 to 7, curved ascending, secondary doi: 10.3417/2009021 Novon 22: 207-211. Published on 30 November 2012. 208 Novon Clethra concordia (Clethraceae), a Shrubby New Species from the Crest of the Cordillera del Condor on the Peru-Ecuador Border David A. Neill Universidad Estatal Amazonica, Facultad de Ingenierfa Ambiental, km 2 V2 via a Napo, Puyo, Pastaza, Ecuador, davidneill53@gmail.com Hamilton Beltran Herbario USM, Museo de Historia Natural, Universidad Mayor Nacional de San Marcos, Casilla 14-0434, Lima 14, Peru, wilmersantiago@hotmail.com Wilson Quizhpe Herbario LOJA, Universidad Nacional de Loja, Facultad de Ciencias Agrfcolas, Casilla 11-01-249, Loja, Ecuador, wwquizhpe@yahoo.es Abstract. Clethra concordia D. A. Neill, H. Beltran & Quizhpe (Clethraceae), a thin-stemmed shrub or treelet from the sandstone Machinaza plateau in the Cordillera del Condor region on the Peru-Ecuador border, is described and illustrated. The new species is distinct from other species of Clethra L. in its small leaves, which are evidently an ecological adaptation to the highly acidic, nutrient-poor sandstone sub- strate where it occurs. Resumen. Se describe y se ilustra Clethra concor- dia, D. A. Neill, H. Beltran & Quizhpe (Clethraceae), un arbusto o arbolito con tallos delgados, procedente de la meseta de Machinaza, de roca arenisca, en la region de la Cordillera del Condor en la frontera entre Peru y Ecuador. La especie nueva se distingue de las demas especies de Clethra L. en su estatura pequena y en las hojas muy pequenas, gruesas y esclerofilas, las que evidentemente constituyen una adaptacion al substrato de arenisca muy acido y pobre en nutrientes, en donde se encuentra esta especie. Clethra L. is a widespread genus with ca. 70 to 80 species distributed in warm temperate areas and tropics of the Western Hemisphere from the southeastern United States to southern Brazil, in warm temperate and tropical eastern Asia from southern China to Malaysia, and on the Atlantic island of Madeira (Sleumer, 1967; Hamilton, 1985; Fior et al., 2003). Clethra has traditionally been considered the sole genus in Clethraceae, but recent evidence (Anderberg & Zhang, 2002) suggests that the Neotropical genus Purdiaea Planch, should be placed in Clethraceae rather than its traditional position in Cyrillaceae. Within Clethra, about 45 species are found in the Neotropics, from Mexico south to southern Brazil and northern Argentina, mostly in montane forests of the Mesoamerican isthmus and in the tropical Andes. Nine species have been reported for Ecuador (Gustafsson, 1992) and 11 species for Peru (Brako & Zarucchi, 1993). All of the Neotropical species are placed in Clethra sect. Cuellaria (Ruiz & Pav.) DC., characterized by the evergreen, persistent leaves and the flowers with stamens and styles included. Most of the Neotropical Clethra species are mid-sized canopy or subcanopy trees, attaining at least 10 m in height, although some in height, particularly those species that grow at high elevations or at forest edges. The Cordillera del Condor is an eastern outlier of from north to south and forms part of the international border between southeastern Ecuador and northern Peru. The geology of the Condor range is complex; one of the sedimentary strata that make up the Condor is the crystalline sandstone Hollm formation of mid-Cretaceous age, which has been uplifted and exposed with the rise of the Andes in the last 3 to 4 million years (Gregory-Wodzicki, 2000) to form a discontinuous series of flat-topped or inclined plateaus, sandstone fragments that vary in size from a few hectares to 20-30 km2 in size, at elevations from 300 m to ca. 2900 m. The soil derived from the crystalline white quartzsite bedrock of the Hollm formation is highly acidic and very poor in N, P, K, and other nutrients essential for plant growth, so as in other areas with oligotrophic white sand or sandstone- Novon 22: 212-216. Published on 30 November 2012. doi: 10.3417/2009038 213 5 mm 214 Novon Volume 22, Number 2 2012 Neill et al. Clethra concordia (Clethraceae) 215 Distribution < ecology. Clethra concordia known from just two localities, one in Ecuador and one in Peru, about 7 km apart on the summit of the nearly level Machinaza plateau. This is one of the largest and highest of the sandstone plateaus in the Cordillera del Condor, and it sits astride the international border. The western part of the plateau, on the Ecuadorian side, is in the watershed of the Rio Machinaza, a tributary of the Rio Zamora; the eastern portion, on the Peruvian side, is in the watershed of the Rio Comainas, a tributary of the Rio Cenepa that drains into the Rio Maranon; the watershed boundary is hardly recognizable on this flat plateau, but it constitutes the international border. The entire Machinaza plateau, mostly at 2300-2400 m, but dipping to about 2150 m at the eastern margin, covers an area of ca. 15 km2, and comprises the only known area of occurrence and suitable habitat for C. concordia. The vegetation on this plateau is a low, rather sparse scrub, with shrubs mostly less than 1.5 C. concordia is Weinmannia condorensis Z. S. Rogers, another example of a low, shrubby plant with very small, sclerophyllous leaves whose congeners are mostly Andean trees with larger, thinner, more mesophyllous leaves. Several other shrubs in various plant families at this site are believed to be locally endemic species not yet known to science (Neill, pers. obs.). Fragments of charred stems indicate that the area was burned over in the past, probably in the 1980s or early 1990s. There are other sandstone plateaus in the Ecuadorian portion of the Cordillera del Condor, mostly below 2000 m elevation, with a concordia has not been found at any other locality. The adaptive advantage of thick, stiff, sclerophyl- lous leaves, which Clethra concordia shares with many other species that are edaphically restricted to oligotrophic white-sand or sandstone-derived soils, appears to be related to protection from herbivory, since the loss of hard-won leaf tissue would threaten the energetic cost of sclerophylly (Turner, 1994). This hypothesis has been tested experimentally in lowland 2006). Several other recently described species from the sandstone plateaus of the Cordillera del Condor, such as Dacryodes uruts-kunchae Daly, M. C. Martinez & D. A. Neill (Rurseraceae), also have small, thick sclerophyllous leaves. There is a more thorough discussion of the adaptive advantage and IUCN Red List category. Clethra concordia, although it occurs in two countries, is known from a single population with an area of occupancy of less than 20 km2, as far as is known. Although the population is not imminently threatened, it may be fire in a year of dry weather, or rapid climate change. Under these criteria (IUCN, 2001), this species should be listed in the IUCN category of Vulnerable (VU). Etymology. The specific epithet “concordia” is a compound word from the Latin, translated literally as “with hearts” or “with hearts (working) together,” and denoting harmony and peace. The epithet commemorates the peace accords and treaty signed by the presidents of Ecuador and Peru in 1998, and ratified by the respective legislatures of those countries, in the culmination of the diplomatic negotiations following the 1995 border dispute : War of Cenepa. Some of the heaviest fighting took place at this site in January to February 1995, and the armies of both Ecuador and Peru planted lethal land mines in the Clethra concordia is locally common. After the 1998 treaty that established the international border once and for all, a path was cleared through the mine field in order to place an obelisk-shaped boundary marker, a hito in Spanish. Two of the authors of C. concordia represent the community of botanists of Peru and Ecuador, and we hope the name will serve as a reminder and hito of our peaceful bi-national scientific endeavors. The epithet is anagram of “condorica,” which denotes the Discussion. The reduced stature and small, thick, coriaceous leaves of Clethra concordia are extremes within the genus, such that, based on vegetative characters, specimens of this taxon are not readily recognizable as Clethra, and for that reason, the original material collected in Peru in 1994 lan- guished in the family indet cabinets in the herbarium of the Field Museum, Chicago, for a decade, until a look at the 3-valved capsule triggered a flash of recognition in the mind of the senior author. Among its Andean congeners, C. concordia is mostly closely approached by C. ovalifolia Turcz., a species that ranges from Andean Venezuela to northern Peru (Sleumer, 1967; Gustafsson, 1992) in its shrubby stature and small, glabrous or glabrate leaves, but the Two Taxonomic Innovations in South American Gentianella (Gentianaceae) James S. Pringle Royal Botanical Gardens, P.0. Box 399, Hamilton, Ontario, Canada L8N 3H8. jpringle@rbg.ca Abstract. Gentianella undulatisepala J. S. Pringle, a new species from Bolivia, is described for the Gentianaceae. It is distinguished from other solitary- flowered species in La Paz Department by its long pedicels and the undulate margins of its calyx lobes. Gentianella arenarioides (Gilg) J. S. Pringle is restored to accepted taxonomic status, with the name neotypified, and the new nomenclatural combination is published. Resumen. Se describe por las Gencianaceas Gen- tianella undulatisepala J. S. Pringle, una nueva especie de Bolivia. Se diferencia de las otras especies de flores solitarias en el Departamento de La Paz por los pedicelos largos y los margenes ondulados de los lobulos del caliz. Gentianella arenarioides (Gilg) J. S. Pringle se restaura al estado taxonomico aceptado, con el nombre neotypificado, y la combinacion de nomenclatura nueva se publica. New species of Gentianella Moench continue to be discovered in the Andes of South America, and additional collections of previously described taxa permit improvements in the classification of Gentia- nella in this region of its greatest diversity. In the present study, one new species is described and another is restored to taxonomic acceptance and nomenclaturally transferred. 1. Gentianella arenarioides (Gilg) J. S. Pringle, comb. nov. Basionym: Gentiana arenarioides Gilg, Repert. Spec. Nov. Regni Veg. 2: 39. 1906. TYPE: Peru. Cajamarca: Prov. Cajamarca, Cajamarca-Bambamarca rd., 55 km N of Cajamarca, 78°35'W, 6°55'S, natural grassland, glaciated terrain, 3750 m, 2 June 1984, D. N. Smith & I. Sanchez Vega 7433 (neotype, designated here, MO; isotype, HAM). Distribution. As no specimens of Gentianella arenarioides were previously known to be extant, three other collections are cited here, two from IUCN Red List category. Data remain deficient for any recommended status in the IUCN Red List (Data Deficient [DD], IUCN, 2001). Discussion. Gentiana arenarioides (1906) was described from a single specimen formerly at Berlin (B), no longer extant, A. Weberbauer 3995 (B images at F and MO), from Cajamarca, Peru. No other specimens so identified were subsequently cited by Macbride (1959). Fabris (1958) included this name at species rank in the synonymy of Gentianella graminea (Kunth) Fabris, noting that the depth of calyx lobing, used by Gilg (1916) as a key character to distinguish these species, is variable and does not differ significantly between the two. In Brako and Zarucchi (1993), I accepted this synonymy, but from the study of specimens seen more recently I conclude that these species are distinct. Gentianella arena- rioides differs from G. graminea in that its flowers are generally solitary, rarely in twos or threes, on peduncles or pedicels 2-10 cm long. In G. graminea, even the shorter stems generally bear a corymboid inflorescence of three to eight or more flowers, and the central pedicels are generally less than 2.5 cm, although the lowest are sometimes longer. The corolla lobes of G. arenarioides are widely spatulate to obovate, 1.4-1.8X as long as wide, widest at 0.65- 0.8X their length, while those of G. graminea are oblanceolate, 2-3X as long as wide, widest at 0.55- 0.6X their length. The transition from the crowded more abrupt in G. arenarioides than in G. graminea. Gilg (1906) and Macbride (1959) noted the similarity of Gentiana arenarioides to G. tristicha Gilg (now Gentianella tristicha (Gilg) J. S. Pringle, within which I here include G. lilacina (Gilg) Zarucchi and G. mesembryanthemoides (Gilg) T. N. Ho & S. W. Liu) of Ancash Department. Plants of G. tristicha are larger than those of G. arenarioides, with stems usually over 20 cm tall, mostly whorled leaves l-2.5(— 3) mm wide, and three to many flowers per stem. The stems of G. arenarioides are generally less doi: 10.3417/2011034 Novon 22: 217-219. Published on 30 November 2012. 218 Novon Nectandra ovatocaudata , a New Species of Lauraceae from Cajamarca, Peru Jens G. Rohwer (Rohwer, 1993), and so he asked for my opinion about doi: 10.341' Validation of Names of Fossil Fungi from Tertiary Sediments of India R. K. Saxena Birbal Sahni Institute of Palaeobotany, 53 University Road, Lucknow-226 007, India rksaxena2207 @y ahoo.com Abstract. The names of 14 species of fossil fungi are validated, two recognized by Kar, one by Rao, and 11 by Nandi et al. The repositories of their holotypes and this validating information is provided herein. All names and their validating descriptions/diagnoses are ascribed to their original authors. These new species names are: Diporisporites giganticus R. K. Kar, Hypoxylonites bhubanensis B. Nandi & S. Banerjee, H. elsikii B. Nandi & S. Banerjee, H. minimus B. Nandi & S. Banerjee, H. neogenicus B. Nandi & S. Banerjee, H. subrotundus B. Nandi & S. Banerjee, H. thindii B. Nandi & A. Sinha, Kalviwadithyrites saxenae M. R. Rao, Lirasporis elongatus R. K. Kar, Spirotremesporites ellipticus B. Nandi & S. Banerjee, S. longiletus B. Nandi & S. Banerjee, S. miocenicus B. Nandi & S. Banerjee, S. reniformis B. Nandi & S. Banerjee, and S. tertiarus B. Nandi, S. Banerjee & A. Sinha. The genus name Kalviwadithyrites M. R. Rao is also here validated, together with the name of its srngle specres, K saxenae M. R. Rao. The species mentioned above belong to Xylariaceae, Microthyr- iaceae, and Fungi Imperfecti (Amerosporae and Dictyosporae). Key words: Diporisporites, Fungi Imperfecti, Hy- poxylonites, India, Kalviwadithyrites, Lirasporis, Mi- crothyriaceae, Spirotremesporites, Xylariaceae. Encountering dispersed fungal fruiting bodies, spores, and mycelia is a common aspect of palynological investigations. Records of fossil fungi from the Indian Tertiary sediments have been published in the form of two catalogues (Saxena, 1991, 2006). During the preparation of the latter catalogue (Saxena, 2006), the author noticed the names of one genus and 14 species of fossil fungi (belonging to Xylariaceae, Microthyriaceae, and Fungi Imperfecti) that were not validly published, because the authors of these names did not specify the single herbarium or collection or institution in which the types were conserved (cf. ICBN, Art. 37.7, McNeill et al., 2006). Two of these species were described by Kar (1990) from the subsurface Miocene and Mio-Pliocene sediments of Tripura in northeast- ern India. The one fungal genus, viz. Kalviwadithy- rites M. R. Rao, and its proposed type, K saxenae M. R. Rao, were described by Rao (2003) from the Sindhudurg Formation (Miocene) of Maharashtra, western India. Further, Nandi et al. (2003) recorded several species of Hypoxylonites Elsik (1990) and Spirotremesporites Duenas-Jimenez (1979) from the Cretaceous and Tertiary sediments of northeastern India. Of these, six species of Hypoxylonites and five species of Spirotremesporites were described as new but were not validly published. All of the above-mentioned species are based on sound morphological features and are well recogniz- able. This is evident because their original authors compared these species with other species of the respective genus and demonstrated that each of the proposed species was different from them. These fossil fungal names are validly published below and in each case are ascribed to the original author(s), with a full and direct reference to the previously and effectively published descriptions/diagnoses and illustrations as provided by the same original author(s). The holotype of each species name, as designated by the original author(s), along with its repository, is also cited. The slides of Kar (1990) and Rao (2003) are stored in the museum of the Birbal Sahni Institute of Palaeobotany, in Lucknow, India. Bharati Nandi (pers. comm.) informed the author that her slides containing holotypes are stored in the Pteridology and Palynology Laboratory, Department of Botany, Kalyani University, Kalyani, India. I. Diporisporites Hammen (Fungi Imperfecti, Amero- The diagnostic characters of Diporisporites are as opposite ends. 1. Diporisporites giganticus R. K. Kar, sp. nov. Validating diagnosis and illustrations in R. K. Kar, J. Palynol. 26: 196, pi. 8, figs. 114, 115. 1990. India. NE India, Tripura, subsurface Miocene sediment (holotype, designated by Kar [1990: pi. 8, fig. 114], based on slide no. 9681/1/9, Q44/3, deposited at Museum, Birbal Sahni Institute of Palaeobotany, Lucknow, doi: 10.3417/2009095 Novon 22: 223-226. Published on 30 November 2012. 224 Novon II. Hypoxylonites Elsik (Xylariaceae). The diagnostic characters of Hypoxylonites a follows: oval to elongate, bilateral, psilate, elongate scar, slit or furrow. 1. Hypoxylonites bhubanensis B. Nandi & S. Banerjee, sp. nov. Validating description and illustrations in Nandi et al., Acta Palaeontol. Sin. 42(1): 58, pi. I, figs. 1, 2, text fig. 3, 1. 2003. TYPE: India. NE India, Mizoram, Renkte Kawn-Sherlui Rd., Late Tertiary sediment (holotype, designated by Nandi et al. [2003: pi. I, fig. 2], based on slide no. RKSL 50/IV, coordinates 107.5/72.3, Pteridology & Palynol- ogy Laboratory, Department of Botany, Kalyani University, Kalyani, India). 2. Hypoxylonites elsikii B. Nandi & S. Banerjee, sp. Nandi et al., Acta Palaeontol. Sin. 42(1): 58, pi. I, figs. 5, 9, 16, 17, text fig. 3, 3. 2003. TYPE: India. NE India, Mizoram, Renkte Kawn- Sherlui Rd., Late Tertiary sediment (holotype, designated by Nandi et al. [2003: pi. I, fig. 16], based on slide no. RKSL 50/IV, coordinates 105.3/14.0, deposited at Pteridology & Palynol- ogy Laboratory, Department of Botany, Kalyani University, Kalyani, India). 3. Hypoxylonites minimus B. Nandi & S. Baneijee, sp. nov. Validating description and illustrations in Nandi et al., Acta Palaeontol. Sin. 42(1): 59, pi. I, figs. 12, 21, text fig. 3, 5. 2003. TYPE: India. NE India, Mizoram, Renkte Kawn- Sherlui Rd., Late Tertiary sediment (holotype, designated by Nandi et al. [2003: pi. I, fig. 12], based on slide no. RKSL 50/IV, coordinates 96.4/62.7, deposited at Pteridology & Palynol- ogy Laboratory, Department of Botany, Kalyani University, Kalyani, India). 4. Hypoxylonites neogenicus B. Nandi & S. Baner- jee, sp. nov. Validating description and illustra- tions in Nandi et al., Acta Palaeontol. Sin. 42(1): 59-60, pi. I, figs. 13, 14, text fig. 3, 2. 2003. TYPE: India. NE India, Mizoram, Renkte Kawn-Sherlui Rd., Late Tertiary sediment (holotype, designated by Nandi et al. [2003: pi. I, fig. 14], based on slide no. RKSL 50/V, coordinates 101.8/53.6, deposited at Pteridology & Palynology Laboratory, Department of Botany, Kalyani University, Kalyani, India). 5. Hypoxylonites subrotundus B. Nandi & S. Baneijee, sp. nov. Validating description and illustration in Nandi et al., Acta Palaeontol. Sin. 42(1): 60, pi. I, fig. 19, text fig. 3, 4. 2003. TYPE: India. NE India, Mizoram, Renkte Kawn-Sherlui Rd., Late Tertiary sediment (holotype, designated by Nandi et al. [2003: pi. I, fig. 19], based on slide no. RKSL 50/V, coordinates 101.8/53.6, deposited at Pteridology & Palynology Laboratory, Department of Botany, Kalyani University, Kalyani, India). 6. Hypoxylonites thindii B. Nandi & A. Sinha, sp. nov. Validating description and illustrations in Nandi et al., Acta Palaeontol. Sin. 42(1): 60, pi. II, figs. 3, 14, 15, text fig. 3, 6. 2003. TYPE: India. NE India, Mizoram, Renkte Kawn- Sherlui Rd., Late Tertiary sediment (holotype, designated by Nandi et al. [2003: pi. II, fig. 3], based on slide no. RKSL 50/IV, coordinates 97.5/57.2, deposited at Pteridology & Palynol- ogy Laboratory, Department of Botany, Kalyani University, Kalyani, India). III. Kalviwadithyrites M. R. Rao (Microthyriaceae). The diagnostic characters of Kalviwadithyrites are as follows: cleistothecium subcircular to circular in shape, dimidiate, non-ostiolate. Two types of cells present, pores absent. No hyphae present. Marginal cells rectangular to polygonal in shape, larger in size, covering outer part; central cells’ thickness 2- or 3- layered, squarish and isodiametric. Rao (2003) compared Kalviwadithyrites with Phragmothyrites Edwards, Microthyriacites Cookson, Parmathyrites Jain & Gupta, Siwalikiathyrites Saxena & Singh, and Ratnagiriathyrites Saxena & Misra, and demonstrated that Kalviwadithyrites can be differen- tiated from the other genera in having two types of nonporate cells. Kalviwadithyrites M. R. Rao, gen. nov. Validating diagnosis in Rao, Palaeobotanist 52(1- 3): 118. 2003. TYPE: Kalviwadithyrites saxenae M. R. Rao, sp. nov., monotypic. The generic name Kalviwadithyrites M. R. Rao was not validly published because the species name indicating its type was not validly published (McNeill et al., 2006: Art. 10.1 and 37.7). The type, K saxenae, is validated by providing the missing information on the repository of the holotype. . Kalviwadithyrites saxenae M. R. Rao, gen. Abt. B, A New Species of Leymus (Poaceae, Triticeae) from China En-Hua Tang, Ying Hu, Jian-Dong Li, and Hong-Xing Xiao* School of Life 130024, I A liiiiifiiuiijtHBMHB iiiiiuiiiiiiiitllitllitliil ill! 1 1 Ssaaili Jiialla unsmiwu nviiiiiiiifi.il ,wmm •\msm a » JJJJlIlllllJllJ uiumumum^^ 11 iii lyillli 230 Novon Volume 22, Number 2 2012 Tang et al. Leymus (Poaceae, Triticeae) from China 231 Leymus chinensis 4(g) Leymus chinensis 15 (y) Leymus chinensis 7 (y) Leymus chinensis 1 (y) Leymus chinensis 12 (g) Leymus chinensis 13 (y) Leymus chinensis 11 (y) Leymus chinensis 10 (y) Leymus chinensis 2(g) Leymus chinensis 6 (g) Leymus chinensis 8 (g) Leymus chinensis 14 (g) I Leymus pubens 5 Leymus pubens 3 ' Leymus pubens 9 Leymus arenarius 26 Leymus secalinus 18 Leymus pseudoracemosus 25 Leymus mollis 27 C Leymus karelinii 22 leymus angustus 17 nnus akmolinensis 16 ^'-Le i Leymus ambiguus 19 Leymus erianthus 24 Clade II ] Clade IV Group A Group B 0.05 Figure 2. Neighbor-joining tree based on the sequences of nuclear ITS and chloroplast tmL-F genic region for Leymus pubens H. X. Xiao and related Leymus species. Numbers at nodes are bootstrap percentages greater than or equal to 50%. The letter “g” 232 Novon Agave temacapulinensis (Agavaceae), a New Ditepalous Species from the Los Altos Region, Jalisco, Mexico Jose Antonio Vazquez- Garcia Instituto de Botanica, Departamento de Botanica y Zoologfa, Universidad de Guadalajara-CUCBA, Km. 15.5 carr. Guadalajara a Nogales, Las Aguas, Nextipac, Zapopan, 45110, Jalisco, Mexico. jvazquez@cucba.udg.mx Miguel de Jesus Chazaro-Basdnez Laboratorio de Biogeografia, Departamento de Geograffa, Universidad de Guadalajara-CUCSH, Avenida de los Maestros y Mariano Barcenas, Guadalajara, 44260, Jalisco, Mexico. chazaro55@hotmail.com Miguel Angel Muniz-Castro and Gregorio Nieves-Herndndez Instituto de Botanica, Departamento de Botanica y Zoologfa, Universidad de Guadalajara-CUCBA, Km. 15.5 carr. Guadalajara a Nogales, Las Aguas, Nextipac, Zapopan, 45110, Jalisco, Mexico. miguelmunizcastro@gmail.com; gnieves@cucba.udg.mx Abstract. Agave temacapulinensis A. Vazquez & Chazaro (Agavaceae) is described and illustrated from Jalisco, Mexico. The new species belongs to Agave L. subg. Agave and is a member of the Ditepalae group, sensu Gentry. It is related to A. wocomahi Gentry in terms of its rosette size, solitary habit in most cases, the smooth texture of the leaves, the red-tipped dimorphic tepals, and the deep floral mahi in having light gray, glaucous rosettes (vs. dark green to glaucous green); short acuminate leaves (vs. long acuminate); crenate margins (vs. straight to undulate) with smaller marginal teeth (8-10 mm vs. 10-20 mm); larger panicles (5.5-6.5 m vs. 3-5 m) with more lateral branches (16 to 18 vs. 8 to 15); and smaller seeds (5-6 X 3-4 mm vs. 7 X 4.5-5 mm). Resumen. Una nueva especie, Agave temacapuli- nensis A. Vazquez & Chazaro (Agavaceae), es descrita e ilustrada de Jalisco, Mexico. Pertenece a Agave L. subg. Agave y corresponde al grupo Ditepalae sensu Gentry. La especie esta relacionada con A. wocomahi Gentry en terminos de tamano de roseta, habito no surculoso en la mayorfa de los casos, textura suave de las hojas, tepalos dimorficos con puntas rojizas y tubo floral profundo, pero difiere de esta ultima por poseer rosetas de color glauco-gris claro (vs. verde oscuro-verde glauco); hojas corta- mente acuminadas (vs. largamente acuminadas); margenes crenados (vs. rectos a ondulados); con dientes mas pequenos (8-10 mm vs. 10-20 mm); pamculas mas grandes (5.5-6.5 m vs. 3-5 m) con mas ramas laterales (16 a 18 vs. 8 a 15); y semillas mas pequenas (5-6 X 3-A mm vs. 7 X 4.5-5 mm). Key words: Agave, Ditepalae, IUCN Red List, Jalisco, maguey bianco, Temacapulm, western Mex- Westem Mexico is the richest center of diversity of the genus Agave L., with 37 species (Vazquez-Garcfa et ah, 2007). In North America the most species-rich regions are in Mexico, specifically in the states of Oaxaca with ca. 30 species (Garcfa-Mendoza, 2004), Jalisco with 28 species (Vazquez-Garcfa et al., 2007), and Durango with 27 species (Gonzalez-Elizondo et al., 2009). Two species of Agave are native to the Rfo Verde canyon at the El Zapotillo dam project area, both from the Canadas de Obregon and Mexticacan municipalities in the Los Altos region of Jalisco. Agave angustifolia Haw. is a wide-ranging century plant from Mexico to Costa Rica, confined to volcanic, basaltic rocks. The second, an undescribed species, is confined to calcareous outcrops at the ecotones between the Juniperus L. forest, thorn forest, and Taxodium Rich, gallery forest. Initial observa- tions on the vegetative populations of this unde- scribed Agave were recorded by Ingeniero Oscar Valencia at Acasico, Mpio. Mexticacan, in 2009 (Miguel Chazaro, pers. comm.). On 5 May 2010, flowering specimens collected in the vicinity of Palmarejo, Mpio. Canadas de Obregon (Vdzquez- Garcia & Chazaro 9070, IBUG, MEXU), revealed its medium-sized panicles with dimorphic tepals, allow- doi: 10.3417/2011065 Novon 22: 235-239. Published on 30 November 2012. Volume 22, Number 2 2012 Leaf texture * Data from Gonzalez and Galvan (1992). July-Sep. On the Type of Pseudosporochnus (Fossil Cladoxylopsida) Wang Qi State Key Laboratory of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Beijing 100093, People’s Republic of China, happyking@ibcas.ac.cn Xue Jin-Zhuang Key Laboratory of Orogenic Belts and Crustal Evolution, School of Earth and Space Sciences, Peking University, Beijing 100871, People’s Republic of China, pkuxue@pku.edu.cn Abstract. Pseudosporochnus Potonie & Bernard (1904; fossil Cladoxylopsida) has been viewed as one of the iconic vascular plants in the Middle Devonian, but the type of the generic name remains problematic. Although P. verticillatus (Krejcf) Obrhel has been previously regarded as the type of the generic name, its so-called basionym Chondrites verticillatus Krejcf was not validly published and thereby has no nomenclatural status. Thus, the specific epithet verticillatus has no priority over krejcii, which Potonie and Bernard first used (1904) for the species of Pseudosporochnus, and the combination P. verticillatus was not validly pub- lished. The name P. verticillatus that Obrhel intended to combine in 1961 is rather a new species name and should be cited as P. verticillatus Krejcf ex Obrhel. However, P. verticillatus, when published (Obrhel, 1961), was an illegitimate superfluous name for P. krejcii (Stur) Potonie & Bernard. Both names attach to a lectotype designated by Obrhel (i.e., D-364, National Museum at Prague) and were originally used under the name Sporochnus krejcii Stur. Pseudosporochnus krejcii (Stur) Potonie & Bernard should be the type of Pseudosporochnus Potonie & Bernard. Key words: Cladoxylopsida, Middle Devonian, Pseudosporochnus. Pseudosporochnus Potonie & Bernard, vascular plant, has received extensive attention and been reported from the Middle Devonian of Scotland, the Czech Republic, Belgium, Germany, Estonia, Romania, Kazakhstan, South China, Norway, and the United States (Potonie & Bernard, 1904; Arber, 1921; Hirmer, 1927; Lang, 1927; H0eg, 1931, 1942; Krausel & Weyland, 1933; Leclercq, 1940, 1970; Obrhel, 1959, 1960, 1961; Leclercq & Banks, 1962; Raleanu & Semaka, 1966; Stockmans, 1968; Jurina, 1969; Schweitzer & Cai, 1987; Kalamees, 1988; Stein & Hueber, 1989; Berry & Fairon-Demaret, 1997, 2002; Schweitzer, 1999; Berry, 2008; Xue & Hao, 2008). Pseudosporochnus is characterized by its digitate branching pattern, terminal sporangia, and dissected xylem strands. The fossil taxon was assigned to its own family (Pseudosporochnaceae), order (Pseudosporochnales), and in turn to the Cladoxylopsida (Pichi-Sermolli, 1958, 1959; Le- clercq & Banks, 1962; Berry & Stein, 2000; Taylor et ah, 2009; Xue et al., 2010), which was the first plant group to attain an arborescent habit, forming one of the earliest known forests on the Earth (Dilcher et al., 2004; Meyer-Berthaud & Decombeix, 2007, 2009; Stein et ah, 2007; Meyer-Berthaud et al., 2010). Phylogenetically, Pseudosporochnus is of significance in that it has been considered one of the stem groups or candidate ancestors of the ferns and/or sphenopsids (Stein et al., 1984; Kenrick & Crane, 1997; Wang & Guo, 2009). To date, eight species names have been used in Pseudosporochnus (Potonie & Bernard, 1904; Obrhel, 1959, 1960, 1961; Leclercq & Banks, 1962; Leclercq, 1970; Mustafa, 1978; Schweitzer & Cai, 1987; Kalamees, 1988; Stein & Hueber, 1989). However, the nomenclature of some of the species names and in particular the generitype of Pseudosporochnus re- mains problematic. In this paper, we intend to clarify the type of Pseudosporochnus. Krejcf (1882) briefly described three kinds of fossil fucoids, which he described as Haliserites zonarioides Krejcf, H. spinosus Krejcf, and Chordrites verticillatus Krejcf, from the Middle Devonian of Bohemia, Czech Republic. Neither the cited specimens nor the illustrations were presented by Krejcf. It can be inferred that Krejcf utilized Sternberg’s (1833) system of classification on fossil fucoids, as he mentioned two of his generic names, i.e., Zonarites Sternberg and Haliserites Sternberg. Meanwhile, the spelling as Chordrites was thought to be a typographical error or a lapsus calami (i.e., a slip of the pen) for the fossil fucoid generic name Chondrites Sternberg, 1833 (Stur, 1882; Potonie & Bernard, 1904; Obrhel, 1961), because there is no other internal evidence that Novon 22: 240-243. Published on 30 November 2012. doi: 10.3417/2011006 Polystichum yaanense (Dryopteridaceae), a Remarkable New Species from Sichuan, China Liang Zhang College of Forestry, Sichuan Agricultural University, Ya’an, Sichuan 625014, People’s Republic of China, and Chengdu Institute of Biology, Chinese Academy of Sciences, P.0. Box 416, Chengdu, Sichuan 610041, People’s Republic of China Li-Bing Zhang * Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166-0299, U.S.A., and Chengdu Institute of Biology, Chinese Academy of Sciences, P.O. Box 416, Chengdu, Sichuan 610041, People’s Republic of China Jun Liu* College of Forestry, Sichuan Agricultural University, Ya’an, Sichuan 625014, People’s Republic of China ^Corresponding authors: Libing.Zhang@mobot.org; JLiu2502@yahoo.com.cn Abstract. A new fern species, Polystichum yaa- nense Liang Zhang & Li Bing Zhang (Dryopterida- ceae), is described from Bifengxia, or the Bluish Green Peak Canyon, in Ya’an Prefecture in Sichuan Province, China. The new species is morphologically unique in having yellowish green leaves, a thin but firm leaf texture, veins visible on the adaxial leaf surface, and narrowly ovate to lanceolate rachis scales. Ecologically, the new taxon grows in acidic soils, in contrast to other species in section Haplopolystichum Tagawa. Polystichum yaanense is considered to be Critically Endangered (CR), based on IUCN Red List criteria. Ten other species of Polystichum Roth occur in the type locality within the same canyon; all 11 species are distinguished from one another by a dichotomous key based on morphology. Key words : China, IUCN Red List, Polystichum, Sichuan. In December 2008, Hai He, a pteridologist based at the CTC Herbarium, affiliated with Chongqing Normal University, noticed images of fem species on the Internet posted by the first author (L.Z.). The images of a noteworthy species of Polystichum Roth (Dryopteridaceae) caught Hai He’s attention and were later forwarded to the second author (L.-B.Z.), who species. Subsequent fieldwork was conducted on 15 March 2009, and the new species is described here. The type locality is in Bifengxia, or the Bluish Green Peak Canyon, which is located in the Ya’an Prefecture directly north of Ya’an City in Sichuan Province. The locale is a scenic area known for its well-protected vegetation and for the Giant Panda Reserve. The canyon itself is about 13 km long and is composed of two segments that form a V shape. Streams and waterfalls run perennially in the canyon. The second author (L.-B.Z.) has also identified two more localities of the new species that were collected by Wen-Pei Fang in 1930 in neighboring Hongya County, to the east in Sichuan, but no further Polystichum yaanense Liang Zhang & Li Bing Zhang, sp. nov. TYPE: China. Sichuan: Ya’an Prefecture, Yucheng Distr., Bifengxia Town, Bifeng Village, Bifengxia, 30°4 '17.51 "N, 102°59 '48.74 "E, 950 m, 15 Mar. 2009, sandstone substrate, in acidic soil, L.-B. Zhang, Y. Wang & L Zhang 4745 (holotype, CDBI; isotypes, CDBI, MO, SAUF). Figures 1, 2. Plants perennial, caespitose, evergreen, 8-15(— 30) cm tall; rhizome short, 0.5-1 cm, ascending; scales ovate to lanceolate, chartaceous, light brown, 0.4-3. 6 mm; roots dark brown when dry, to 10 cm, ca. 0.6 mm diam., sparsely to densely covered with scales. Leaves 5 to 11 per rhizome; petiole 1.5-3(— 7) cm, 0.5-1. 2 mm diam. at mid-portion, adaxially canalic- Novon 22: 244-249. Published on 30 November 2012. doi: 10.3417/2010041 Volume 22, Number 2 Zhang et al. 245 ulate, green; basal petiole scales narrowly ovate to lanceolate (Fig. 1C), 2. 7-3. 4 X 0.8-1. 1 mm, variable in size, thinly chartaceous and brown at mid-portion, membranous and light brown on margin, composed of multiple cell layers, margin subentire, apex acumi- nate or caudate, matte; distal petiole scales similar but narrower and shorter toward rachis apex, lanceolate or narrowly ovate, membranous, brown, 246 Novon (Dryopteric (Dryopteric Polystichum normale (Dryopteridaceae), a New Species in Section Crucifilix from Guizhou, China doi: 10.341' Volume 22, Number 2 2012 Zhang & Wang 251 Polystichum normale (Dryopteridaceae) from Guizhou, China lamina lanceolate, slightly contracted toward base, 1- pinnate, (6-)9-27 X 1. 9-3.3 cm wide at mid-portion, apex acuminate; rachis 0.7-1.3 mm in median diam., without proliferous buds, adaxially sulcate; rachis scales 1.5-3 X 0.3-0. 7 mm, variable in size, membranous, light brown, margin ciliate, apex caudate, matte (Fig. ID); pinnae in 14 to 28 pairs, not overlapping, pointing upward, basalmost pairs 0 g i (Dryopteridaceae) Dryopteris liboensis (Dryopteridaceae), a New Species in Dryopteris sect. Erythrovariae from Guizhou, China Li-Bing Zhang * Missouri Botanical Garden, P.0. Box 299, St. Louis, Missouri 63166-0299, U.S.A., and Chengdu Institute of Biology, Chinese Academy of Sciences, P.0. Box 416, Chengdu, Sichuan 610041, People’s Republic of China Pei-Shan Wang and Xiao-Ying Wang Guizhou Academy of Sciences, Guiyang, Guizhou 550001, People’s Republic of China pswang36@yahoo.com.cn * Author for correspondence: libing.zhang@mobot.org Abstract. Dryopteris liboensis P. S. Wang, X. Y. Wang & Li Bing Zhang, a new species of Dryopteris Adans. (Dryopteridaceae) from southern Guizhou, China, is described and illustrated. Dryopteris liboensis is assigned to Dryopteris subg. Erythrovariae (H. Ito) Fraser-Jenk. on the basis of its stiff and narrow petiole and rachis scales, but the new species is noteworthy within the section because these scales are not bullate. Dryopteris liboensis has stiff, narrowly lanceolate and filiform scales on the petiole and rachis and is therefore considered a member of the D. erythrosora (D. C. Eaton) Kuntze group of Dryopteris sect. Erythrosoriae H. Ito. It is unique among Chinese taxa of Dryopteris sect. Erythrovariae in having leaves 1- or 2-pinnate basally, with zero to three free lobes, and 1-pinnatifid at the blade apex. Key words: China, Dryopteridaceae, Dryopteris, Guizhou, IUCN Red List. Dryopteris Adans. (Dryopteridaceae) is one of the largest fern genera, with about 300 or more species worldwide (Lu, 1993). Dryopteris has been classified into four subgenera as Dryopteris subg. Dryopteris, Dryopteris subg. Erythrovariae (H. Ito) Fraser-Jenk., Dryopteris subg. Nephrocystis (H. Ito) Fraser-Jenk., and Dryopteris subg. Pycnopteris (Moore) Ching (Fraser-Jenkins, 1986), or three subgenera as Dryopteris subg. Dryopteris, Dryopteris subg. Eryth- rovariae, and Dryopteris subg. Pycnopteris (Wu & Lu, 2000). Of these, Dryopteris subg. Erythrovariae is one tions. The circumscription of Dryopteris subg. Erythrovariae by Wu and Lu (2000), however, differs slightly from that of Fraser-Jenkins (1986). Fraser- Jenkins (1986) used overall morphological similari- ties to characterize Dryopteris subg. Erythrovariae and did include a few species without bullate scales. In contrast, Wu and Lu (2000) strictly used the presence of bullate scales to define Dryopteris subg. Erythrovariae. As a result of the differing circum- scriptions, the taxonomic placement of a few species, e.g., D. polita Rosenst., has been controversial. A recent molecular analysis demonstrated that Fraser- Jenkins’s (1986) definition of Dryopteris subg. Erythrovariae is more natural (Zhang et al., 2012). In the summer of 2005, one of the authors (P.-S.W.) collected specimens of an unidentified species of Dryopteris. Additional material was collected in 2009 and 2010 from the same locality. This species is similar to members of Dryopteris subg. Erythrovariae in the shape of the petiole and rachis scales, but the scales are not bullate. The discovery of this species offered an opportunity to test the differing views of Fraser-Jenkins (1986) and Wu and Lu (2000). Dryopteris liboensis P. S. Wang, X. Y. Wang & Li Bing Zhang, sp. nov. TYPE: China. Guizhou: Libo Co., Lihua, Wujia River, 25°21'L71"N, 108°7 '32.26 "E, 520 m, on rock wall by river, 3 Aug. 2005, P.-S. Wang 20052 (holotype, PYU; isotypes, CDBI, MO, Herbarium Pei-Shan Wang [Guizhou Institute of Biology]). Figure 1. & Li Bing Zhang is most similar to species of the D. filiform scales on the petiole base and rachis, but it differs pairs of free lobes only, rather than either 1-j: Terrestrial, evergreen, 13-26 cm tall; rhizomes 5-7 mm, erect or slightly ascending, together with petiole bases densely scaled, the scales brown, narrowly Novon 22: 256-259. Published on 30 November 2012. doi: 10.3417/2012049 Volume 22, Number 2 2012 Zhang et al. 257 Dryopteris liboensis (Dryopteridaceae) from Guizhou, China 259 (Dryopteridaceae) from Selaginella longistrobilina (Selaginellaceae), a New Species from Guizhou, China, and Selaginella prostrata , a New Combination and Its Lectotypification 262 Novon IUCN Red List category. Only two populations of Selaginella longistrobilina were observed at the type locality, with fewer than 20 individuals in total found. The new species might occur elsewhere in a nearby similar habitat. Based on the best evidence available, this taxon can be classified as Critically Endangered (CR) according to IUCN Red List criteria (IUCN, 2008). Etymology. The epithet of the new species is taken from the Latin prefix, “longi-,” meaning “long,” and the Latin “strobilina,” meaning “of the strobilus,” which refers to the long strobili. Discussion. Selaginella longistrobilina has been confused with S. prostrata (H. S. Kung) Li Bing Zhang (comb, nov.; see below) since the publication of the basionym of the latter, Lycopodioides prostrata H. S. Kung (Kung, 1988; Wang & Wang, 2001). When describing a taxon identified as “prostrata” (Kung, 1981; the name not validly published; see later discussion), Kung (1981) cited three gatherings from Guizhou, Sichuan, and Yunnan, which included correspond with Kung’s “prostrata” (Chu, 2006), while the collection from Guizhou is selected herein as the type of the newly described S. longistrobilina. Both taxa are indeed similar to each other in the shape of the microphylls and also share other similarities in the arrangement, shape, and size of the sporophylls. However, the two taxa do differ in the morphology of their strobili, sporophylls, sculpturing of spores, and colpi of microspores as noted by Chu (2006). Selaginella longistrobilina often has strobili branching positions of ultimate branchlets upward. The larger sporophylls are adaxial and clearly navicular. Microspores have verrucate sculpturing with irregularly sized verrucae; the colpi of micro- spores are twisted, usually not forked at the ends, and they reach the equatorial plane of the spore. Irregularly verrucate sculpturing of the microspores is common among species of Selaginella P. Beauv. In contrast, S. prostrata has shorter strobili, to only 3-4 mm, and the sporophylls are located only at the flattened adaxially, with only a few of the upper, adaxial sporophylls slightly navicular. The micro- spores are tuberculate with regularly sized tuber- cules; the microspore colpi are straight, forked at the ends, and do not extend to the equatorial plane. crevices"on hill, 1500 m, 24 June 1978, P.^Wang 75550 (PYU, Herbarium Pei-Shan Wang [Guizhou Institute of Biology]). Selaginella prostrata (H. S. Kung) Li Bing Zhang, comb. nov. Basionym: Lycopodioides prostrata H. S. Kung, LI. Sichuan. 6: 76, pi. 20, parts: 7- 11. 1988. [Latin diagnosis & description in Kung, Acta Bot. Yunnan. 3: 254. 1981.] TYPE. China. Sichuan: Emei Shan, Jiushijiudaoguai, Shouxingqiao, 4 July 1955, X.-J. Zheng 30071 (lectotype, designated here, PE; isolectotype, SZ Discussion. When Kung (1981: 254) described his Selaginella prostrata , he cited three gatherings: X.-J. Zheng 30071 from Sichuan, NE Yunnan Exp. 808 (KUN not seen, PE, PYU not seen) from Yunnan, and P.-S. Wang 75606 (CDBI, MO, PYU, Herbarium Pei-Shan Wang [Guizhou Institute of Biology]) from type, as noted by Chu (2006). Therefore, this name was not validly published by Kung in 1981, following the International Code of Botanical Nomenclature (McNeill et al., 2006, Art. 37.1). Later, Kung (1988: 76) made a new combination, “Lycopodioides prostrata (H. S. Kung) H. S. Kung,” based on his 1981 invalid name in Selaginella. Kung (1988: 76) implicated one of the three gatherings cited in Kung (1981) as the type, X.-J. Zheng 30071 from Emei Shan in Sichuan, by stating that “the type was collected from Emei Shan” in Chinese. This type designation, together with Kung’s (1988: 76) refer- ence to the previously and effectively published Latin description and diagnosis by him (Kung, 1981: 254), validated the publication of L. prostrata H. S. Kung (1988) as a new species (McNeill et al., 2006: Art. 36.1) instead of a new combination or the name S. prostrata H. S. Kung, which Kung (1988) did not intend to publish. Article 37.7 does not apply, since the specific deposit of the type was not required by the Code until 1990. The herbarium of deposit for the type of L. prostrata was not indicated by Kung (1988). There are two specimens known to us for X.-J. Zheng 30071, which are deposited at PE and SZ, and herein we designate the specimen at PE as lectotype. We were able to confirm the taxonomic identity of the PE sheet, but did not see the SZ specimen, and thus the better known PE specimen is chosen. Acknowledgments. We thank Wei-Ming Chu of rizing the differences between Selaginella long- istrobilina and S. prostrata. We thank Faqiang Lii www.mbgpress.info CONTENTS Missouri D /^v4- n i s~\ n 1 Botanical Garden A Journal for Botanical Nomenclature ME 2 2 V 0 L U NUMBER 3 2013 Volume 22, Number 3 May 2013 Novon, A Journal for Botanical Nomenclature from the Missouri Botanical Garden The mission of the Missouri Botanical Garden is to discover and share knowledge about plants and their environment in order to preserve and enrich life. Novon publishes short articles whose primary purpose is the establishment of nomencla- ture in vascular plants and bryophytes. All articles are peer-reviewed by qualified, indepen- dent reviewers. Manuscripts must fully state and justify the reasons for proposing nova. These may include detailed comparisons with similar taxa, short keys to similar taxa, illustrations to similar taxa, and mechanical nomenclature reasons, among others. Manuscripts whose primary purpose is other than establishment of new nomenclature, which usually are longer manuscripts, cannot be accepted for review. These include reviews, revisions, monographs, or other papers that incidentally include nova. Manuscripts must follow the guidelines in the Checklist for Authors. The Checklist may be downloaded from the Garden’s web site, www.mbgpress.info, or authors may contact the editor at novon@mobot.org to request a copy. Novon will not knowingly accept manuscripts that have been simultaneously submitted to other journals for consideration or previously published in some form elsewhere. Victoria C. Hollowell Editor Allison M. Brock Associate Editor Tammy Charron Associate Editor Laura L. Slown Associate Editor Cirri R. Moran Press Coordinator Ihsan A. Al-Shehbaz Consulting Editor Carmen Ulloa Ulloa Consulting Editor George Yatskievych Consulting Editor Kanchi N. Gandhi Nomenclature Consultant Nicholas J. Turland Nomenclature Consultant Roy E. Gereau Latin Editor Novon is included in the subscription price of the Annals of the Missouri Botanical Garden. Price for 2012: $180 per year U.S.A.; $190 Canada and Mexico; $215 all other countries. Four issues per volume. Postal address: Subscriptions: Annals of the Missouri Botanical Garden, P.0. Box 1897, Law- rence, KS 66044-8897; Manuscript submissions: Novon, Missouri Botanical Garden, P.0. Box 299, St. Louis, MO 63166-0299. Editorial queries: novon@mobot.org Orders for back issues: mbgpress@mobot.org Web site: http://www.mbgpress.info Missouri Botanical Garden 2013 © Novon is printed on paper that meets the re- quirements of ANSI/NISO Z39.48-1992. Novon (ISSN 1055-3177) is published quarterly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Periodicals postage paid at St. Louis, MO and additional mail- ing offices (USPS #006-777). POSTMASTER: Please send address changes to Annals of the Missouri Botanical Garden, P.O. Box 1897, Lawrence, KS 66044-8897. Information on the contents of Novon is present- ed in SciSearch®, Research Alert®, Current Con- tents®/Agriculture, Biology, and Environmental Sciences, APT Online, the CAB Abstract/Global Health database, and the ISI® database. The full-text of Novon is available online through BioOne™ (http://www.bioone.org). Volume 22 NO VON Number 3 W 2013 Ocotea grandifructa (Lauraceae), a Novelty from the Brazilian Atlantic Forest Leandro C. S. Assis Universidade Federal de Minas Gerais, Instituto de Ciencias Biologicas, Departamento de Botanica, Belo Horizonte, MG 31270-901, Brazil, leandroassis@ufmg.br, leandassis@gmail.com Matheus F. Santos Universidade de Sao Paulo, Departamento de Botanica, Laboratorio de Sistematica Vegetal, Rua do Matao 277, Sao Paulo, SP 05508-090, Brazil, matheus_fs@yahoo.com.br species of Lauraceae, Ocotea grandifructa L. C. S. Assis & M. F. Santos, from the Brazilian Atlantic forest in Minas Gerais State, is presented. Following the diagnostic criterion for species delimitation, in which a species is delimited by an exclusive feature or an exclusive combination of morphological features, the new species differs from other dioecious species in Ocotea Aubl. by presenting branchlets with rhythmic growth, chartaceous leaf laminas, deciduous perianth at fruit maturity, cupules with a single margin, and large, ovoid fruits. Resumo. Nesta contribuigao taxonomica, uma espe- cie nova de Lauraceae, Ocotea grandifructa L. C. S. Assis & M. F. Santos, e descoberta na floresta atlantica brasileira, no Estado de Minas Gerais. stico para a delimitagao de i especie e delimitada por uma caracterfstica exclusiva ou uma combinagao exclusiva de caracterfsticas morfologicas, a especie nova difere- se das outras dioicas em Ocotea Aubl. por apresentar ramos com crescimento rftmico, laminas cartaceas, um perianto decfduo na maturidade dos frutos, cupulas com uma unica margem e frutos grandes, ovoides. Ocotea Aubl., the largest Neotropical genus of Lauraceae, comprises ca. 300 to 350 species in the Neotropics plus ca. 50 to 60 species in Africa and Madagascar (Rohwer, 1993; Madrinan, 2004a). The genus is highly diversified in the Neotropical lowlands and montane forests (Rohwer, 1986, 1993; Madrinan, 2004a). Its polyphyly has been confirmed by molec- ular phylogenetic analyses (Chanderbali et ah, 2001). However, a new classification as an expression of the phylogenetic relationships has not yet been achieved. Ocotea is not taxonomically well defined, but the species have been characterized by stamens with four pollen sacs arranged in two horizontal pairs and flowers with free perianth, which generally lacks papillae on the adaxial side (Rohwer, 1986; van der Werff, 1991, 2002). In this taxonomic contribution, a new dioecious species of Lauraceae, O. grandifructa L. C. S. Assis & M. F. Santos, discovered from the Brazilian Atlantic forest, is described. As defined here, flower length includes the pedicel. Vegetative and reproductive terminology follow Radford et al. (1974), Halle et al. (1978), Weberling (1992), Rohwer (1993), and Assis and Mello-Silva (2010a, 2010b). Branchlets with short intemodes and cataphylls surrounding buds and inflorescences are associated with rhythmic growth, which can be characterized by its periodicity and by the leaves distinctly clustered at the tip and/or middle of the branchlets (Halle et al., 1978; Assis & Mello-Silva, 2010b). The androecium of Ocotea is composed of four whorls, each one with three elements (Rohwer, 1993). Whorls I and III are doi: 10.3417/2011023 Novon 22: 265-270. Published on 24 May 2013. Volume 22, Number 3 2013 268 Novon 1 0.1 and 5.0 cm. g Erigeron pacalis (Asteraceae), a New Species from Western Canadian Boreal Grasslands on 24 May 2013. 272 Novon Volume 22, Number 3 2013 Bjork 273 Erigeron (Asteraceae) from Western Canada papyrifera Marshall-Pieea glauca (Moench) Voss- Populus tremuloides Michx. forest, and in boreal grasslands with Achillea millefolium L., Anemone patens L., Artemisia frigida Willd., Carex duriuscula C. A. Mey., Comandra umbellata (L.) Nutt., Elymus lanceolatus (Scribn. & J. G. Sm.) Gould, Monarda fistulosa L., Poa cusickii Vasey, and Rosa acicularis Lindl. IUCN Red List category. Despite searches in grasslands along major portions of the British Columbia area of the Peace River and its major tributaries in 2008, no additional populations were found. Specimens of all Erigeron species that could be confused with E. pacalis (i.e., entire leaved and with non-woolly involucres) were examined at ALTA, CAN, PMAE, UBC, and V. No additional specimens of E. pacalis were found in these herbaria. The Peace Grasslands straddle the British Columbia-Alberta border; the Alberta portions of the grasslands have mostly been converted into farmland, and only about 0.5% of the original grassland area retains good- condition native vegetation (Baker, 2005). No approximations are available for the portion of the Peace Grasslands in British Columbia that remain in grassland vegetation has been converted to agricul- ture, intense grazing, or other human uses, or has otherwise seen major degradation of vegetation condition. Major portions of the Peace Grasslands that have not been converted to agriculture have lost native plant cover as the invasive exotic species Bromus inermis Leyss. and Melilotus (L.) Mill, species continue to spread. According to IUCN Red List (2001) criteria B2 (geographical range < 10 km2) and due to ongoing severe threats from human use of the Peace River Valley, Erigeron pacalis falls under the Critically Endangered (CR) i Etymology. The species name comes from the Latin for “peace” or “peaceful,” named after the Peace River Valley, where the type of the species was collected. Discussion. Like members of Erigeron sect. Erigeron (Nesom, 2008), E. pacalis is short- stemmed, monocephalous and possessing of glan- dular hairs. However, the members of that section, unlike E. pacalis, have relatively poorly developed cauline leaves and ray corollas that are more distinctly coiling after anthesis. Additionally, in northwestern North America, the members of Erigeron sect. Erigeron are found only in arctic, alpine, and subalpine habitats. Hence, the affinities of E. pacalis are presumed to be instead with Erigeron sect. Fruticosus G. Don (Nesom, 2008). Like members of that section, E. pacalis has cauline leaves similar in shape and size to the basal leaves and ray corollas that remain straight or coil only slightly after anthesis. Like E. pacalis, several members of Erigeron sect. Fruticosus grow at lower elevations. Several characteristics of Erigeron pacalis sug- gest that it is likely related to E. glabellas (Erigeron sect. Fruticosus ), including its grassland habitat, fibrous roots, narrowly oblanceolate leaf shape, well-developed cauline leaves (i.e., cauline leaves similar in shape and size to the basal leaves), and in the indument of hirsute to hirsuto-villous hairs, similar to those of E. glabellas var. pubescens Hook, (although, in addition, there are shorter glandular hairs on E. pacalis). Erigeron pacalis differs from E. glabellus in its shorter stature (5-7 cm vs. 7-40 cm), in having heads consistently only one per stem (vs. one to 15 heads per stem), glandular hairs on distal parts, coriaceous (vs. softly herbaceous) leaves, and shorter ray and disc florets (5-6 mm vs. 8-15 mm). Compared with Erigeron pacalis, most white- rayed individuals of E. glabellus, as that species is defined in the Flora of North America (Nesom, 2006), are much taller plants, 30-50(— 70) cm, with larger heads (involucres 10-20 mm wide vs. 12- 13.5 mm in E. pacalis ), and with basal leaves acuminate and usually with prominent white midribs (apices rounded or blunt and midribs not so noticeable and white in E. pacalis). These white- rayed plants are common in the Peace River Grasslands and in the Canadian Prairies (east to Manitoba), and have been referred as E. asper Nutt, in earlier Canadian literature. Cronquist (1947) synonymized Erigeron asper under E. glabellus, but the distinctions of that species from E. glabellus appear to be more clearly defined than Cronquist might have been aware of. Erigeron asper blooms mostly earlier (May to June) than the June to August flowering periods of E. pacalis and pink-flowered forms of E. glabellus, as evidenced by the collection dates reported on herbarium specimen labels, my own field observa- tions in British Columbia, Alberta, and Saskatch- ewan, and the phenology observations of Scoggan (1979). Occasional white-rayed individuals are sometimes mixed in populations of pink-rayed E. glabellus s. str. (i.e., minus E. asper), but I know of no populations having only white rays, and those occasional white-rayed plants lack the other distinc- tions of E. asper. 274 Two New Species of Tropaeolum (Tropaeolaceae) from South America Eva Bulacio Laboratorio de Taxonomfa Fanerogamica, Botanica, Fundacion Miguel Lillo, Miguel Lillo 251, San Miguel de Tucuman, CP 4000, Tucuman, Argentina. evabulacio@yahoo.com. ar Abstract. Two new South American species of Tropaeolum L. (Tropaeolaceae), T. slanisii Bulacio from Salta, Argentina, and T. kieslingii Bulacio from Tarija, Bolivia, found during a revision of the genus for the Flora of Argentina, are described, illustrated, and compared with related species. Both new species belong to Tropaeolum sect. Tropaeolum, which is characterized by the strongly heteromorphic petals, or less deeply incised leaf blades. RESUMEN. Se describen, ilustran y comparan dos nuevas especies de Tropaeolum L. (Tropaeolaceae), T. slanisii Bulacio de Salta, Argentina y T. kieslmgu Bulacio de Tarija, Bolivia, descubiertas durante la revision del genera para la Flora Argentina. Ambas especies pertenecen a Tropaeolum secc. Tropaeolum, diferentes. ciliados en el margen y las laminas de las hojas enteras o mas o menos divididas. Key words: Argentina, Bolivia, IUCN Red List, South America, Tropaeolaceae, Tropaeolum. stamens. The fruit is usually a 3-lobed schizocarp, and each carpel has a single seed. During the study of the Tropaeolaceae for the Flora of Argentina, with plant material reviewed from South and North American herbaria (BAB, CORD, CTES, LIL, LP, MERL, SI, US), several specimens of Tropaeolum could not be identified to species. It should be mentioned that many trips to the countries were made to the localities where the distinctive material was collected and to similar adjacent areas in Argentina and Bolivia, but additional plants were not found. Further examination indicates that the specimens do represent new taxa, and they are described below as new species. By their bioform, T. slanisii Bulacio and T. kieslingii Bulacio both belong characterized by the petals being strongly heteromor- phic, usually ciliate on the margins, often serrate, but not emarginated. The leaf blades are either entire or is incised to a depth less than four fifths of the blade (Andersson & Andersson, 2000). The Tropaeolaceae are widely distributed in the temperate and subtropical lowlands of South America from Mexico to Patagonia, but their highest diversity Andersson, 1991). Tropaeolum L. has been treated with ca. 110 species, 78 of which are principally tropical taxa and 29 of which are primarily temperate (Berry, 1992; Andersson & Andersson, 2000; Watson & Flores, 2010). Recent molecular phylogenies support Tropaeolum as a single genus of two sections, sections Tropaeolum L. and Chilensia Sparre, and smaller segregate genera such as Magallana Cav. and Trophaeastrum Sparre have been submerged in its synonymy (Andersson & Andersson, 2000). Fourteen species of Tropaeolum were previously listed for Argentina (Zuloaga et al., 2008), and the description of one of the new species from Salta brings the total to 15. Most Tropaeolum species are fleshy vines with twining petioles and are easily distinguished by having peltate leaves and flowers with a calycine spur, five clawed petals, and eight . Tropaeolum kieslingii Bulacio, sp. nov. TYPE: Bolivia. Tarija: De Tarija a Narvaez, 2000-2500 m, 19 Mar. 1982, R. Kiesling, O. Ahumada, A. G. Lopez & A. D. Rotman 3741 (holotype, SI). Haec species Tropaeolo meyeri Sparre similis, sed ab eo petalis omnibus integris superioribus biauriculatis 7-10 X Annual climbing herb; stems puberulous, with yellowish white indumentum. Leaves glabrous, mar- ginally peltate, petioles 2-5 cm; blade light green, suborbicular, 1.8-3 X 2.5-4 cm, 3- to 5-lobed, lobes acute, entire or sometimes the upper ones 3-lobate in the middle, mucronate, somewhat cordate or truncate at the base, peltate. Flowers solitary or in pseudo- racemes, pedicels ca. 2 cm. Calyx glabrescent, yellow, bilabiate, lower lobes larger than the upper ones; calyx spur 10-15 mm, funnelform, yellow with purple veins, straight or slightly recurved at the apex; Novon 22: 276-280. Published on 24 May 2013. doi: 10.3417/2005123 Volume 22, Number 3 Bulacio 277 corolla yellow, both lower and upper petals being entire, lower petals lanceolate, acute, briefly mucro- nate, cuneate, upper petals 7-10 X 3-5 mm, with elliptic, unguiculate, biaunculate lobes. Fruit a pubescent schizocarp, with mixed indumentum, long trichomes moderately dense, yellowish white, short glandular trichomes, red-brown, 3 mericarps, 6-10 mm, globose, dark colored at maturity. 278 r t § New Circumscription of the Endemic Brazilian Genus Actinocephalus (Eriocaulaceae) 3 de Sao Paulo, C. Postal 11461, C on 24 May 2013. Costa & Costa & 286 Novon longifolius, based on common features such as the presence of an elongated fertile axis and scapes with trichomes that are T-shaped. 11. Actinocephalus perbracchiatus (Silveira) F. N. Costa, comb. nov. Basionym: Paepalanthus perbracchiatus Silveira, Floral. Mont. 1: 137, tab. 86. 1928. TYPE: Brazil. Minas Gerais: “In campis arenosis inter Serro et Datas, in Serra do Espinhago,” June 1925, A. Silveira 757 (holo- type, R). 12. Actinocephalus phaeocephalus (Ruhland) F. N. Costa, comb. nov. Basionym: Paepalanthus phaeocephalus Ruhland, in Engler, Pflanzenr. IV, 30 (Heft 13): 169. 1903. TYPE: Brazil. Goias: “Cabeceira do Rio Sant’Anna, dans le bois,” 8 Jan. 1895, A. F. M. Glaziou 22314 (holotype, R; isotypes, G [barcode] 00099079, K [barcode] 000640030, P [barcode] 00741971). (Silveira) F. N. Costa, comb. nov. Basionym: Paepalanthus rhizomatosus Silveira, Floral. Mont. 1: 148, tab. 93. 1928. TYPE: Brazil. Minas Gerais: “In campis prope Baraunas, in Serra Geral,” June 1925, A. Silveira 761 (holotype, R). 16. Actinocephalus velutinus (Silveira) F. N. Costa, comb. nov. Basionym: Paepalanthus velutinus Silveira, Floral. Mont. 1: 145, tab. 91. 1928. TYPE: Brazil. Minas Gerais: “In campis arenosis in Serra do Cabral,” May 1910, A. Silveira 577 (holotype, R). Excluded Taxa from Actinocephalus 1. Paepalanthus microphorus Silveira, Floral. Mont. h 149, tab. 94. 1928. TYPE: Brazil. Minas Gerais: “In campis prope Milho Verde, in Serra Geral,” June 1925, A. Silveira 762 (holotype, R). Discussion. Paepalanthus microphorus does not have paraclades, and the flowers are dimerous, in marked contrast to the trimerous flowers and para- clades in Actinocephalus. The species is best placed in Paepalanthus ser. Dimeri (Ruhland) Giul. Paepalanthus macrocephalus (Bong.) Korn., P. praemorsus Ruhland, and P. scholiophyllus Ruhland were included by Ruhland (1903) in Paepalanthus subsect. Aphorocaulon, but these species are mor- phologically distinct from Actinocephalus mainly by the absence of paraclades and a bifid stigma (vs. simple stigma in Actinocephalus ) in P. praemorsus. These three species are best placed in Paepalanthus ser. Variabiles Ruhland. 14. Actinocephalus scytophyllus (Ruhland) F. N. Costa, comb. nov. Basionym: Paepalanthus scytophyllus Ruhland, in Engler, Pflanzenr. IV, 30 (Heft 13): 171. 1903. TYPE: Brazil. Minas Gerais: “Tombador pres Diamantina, dans le campo,” 4 Apr. 1892, A. F. M. Glaziou 19967 (lectotype, designated here, B [barcode] 10 0157349; isolectotypes, K, MO-1664562, P [barcode] 00741986). Discussion. Ruhland (1903: 171) cited two col- lections from Minas Gerais when he described Paepalanthus scytophyllus. The first collection was Glaziou n. 19971 from “Pinheiro, auf trockenen Campos”; the second, uDerselbe [Glaziou] n. 19967,” was noted as being from “Tombador bei Diamantina, auf Campos.” It is necessary to designate the lectotype between these Lwo synlypes. with ihe collection Glaziou n. 19967 chosen because it is more representative of the taxon than Glaziou n. 19971. 15. Actinocephalus trichopeplus (Silveira) F. N. Costa, comb. nov. Basionym: Paepalanthus trichopeplus Silveira, FI. Serr. Min. 46. 1908. TYPE: Brazil. Minas Gerais: Serra do Cipo, Apr. 1905, A. Silveira 353 (holotype, R; isotype, B [barcode] 10 0247641). 2. Paepalanthus macrocephalus (Bong.) Korn., in Martius & Eichler, FI. Bras. 3(1): 379. 1863. Basionym: Eriocaulon macrocephalum Bong., Mem. Acad. Imp. Sci. St.-Petersbourg, Ser. 6, Sci. Math. 1: 630, tab. 33. 1831. Dupatya macrocephala (Bong.) Kuntze, Revis. Gen. PI. 2: 746. 1891. TYPE: Brazil. Minas Gerais: “In Riedel 1036 (holotype, LE [barcode] 00001127, isotypes, B [barcode] 10 0243932, G [barcode] 00099097, K [barcode] 000640027, P). 3. Paepalanthus praemorsus Ruhland, in Engler, Pflanzenr. IV, 30 (Heft 13): 172. 1903. TYPE: Brazil. Minas Gerais: “Tombador, pres Diaman- tina,” 8 Apr. 1892, A. F. M. Glaziou 19960 (lectotype, designated here, B [barcode] 10 0157351; isolectotypes, K, P [barcode] 00741973, 75, 76 [3]). Discussion. Ruhland (1903: 172) cited two collections from Minas Gerais when he described Paepalanthus praemorsus. The first collection was . 19960 from “Tombador, bei Diamantina”; in, “ Derselbe [Glaziou] n. 19966” was noted as being from “Serra dos Cristaes, necessary to designate the Costa & Esenbeckia cowanii (Rutaceae), Epitypification and Emendation Pedro Dias Instituto de Biociencias, USP— Universidade de Sao Paulo, Departamento de Botanica, Rua do Matao 277, 05508-900, Sao Paulo, SP, Brazil. Current address: Instituto de Biodiversidade e Florestas, UFOPA-Universidade Federal do Oeste do Para, Rua Vera Paz, s/n, Sale, 68135-110, Santarem, Para, Brazil, ppdias@gmail.com Renata Giassi Udulutsch Faculdade de Ciencias e Letras, UNESP— Universidade Estadual Paulista, Campus de Assis, Departamento de Ciencias Biologicas, Av. Dom Antonio, 2100, 19806-900, Assis, SP, Brazil. udulutsch@gmail.com Jose Rubens Pirani Instituto de Biociencias, USP— Universidade Sao Paulo, Departamento de Botanica, Rua do Matao 277, 05508-900, Sao Paulo, SP, Brazil, pirani@usp.br Author for correspondence: ppdias@gmail.com Abstract. A complete description of Esenbeckia cowanii Kaastra (Rutaceae) is presented, including new details for floral morphology. The original type of this taxon included only fruiting material, thus we designate an epitype with flowers. In addition, we Esenbeckia Kunth (Rutaceae, Rutoideae, Gali- peeae, Pilocarpinae) comprises 28 species that occur in the Neotropics. A monograph of the entire subtribe was presented by Kaastra (1982) for Flora Neo- tropica, and previous work (Pirani, 1999) contributed two new species to the genus. Esenbeckia cowanii Kaastra was described in 1977 and known from only two collections from the same locality in French Guiana, from the type Cowan 38833 and paratype Cowan 38757 (K not seen, NY, P not seen). Kaastra (1977) described the novel taxon without knowing its flower morphology: both collec- tions, including all duplicates, are non-flowering. Moreover, the chief diagnostic feature used by Kaastra (1977, 1982) to differentiate E. cowanii from Esenbeckia Kunth subg. Lateriflorens Kaastra) was the winged petiole, which is clearly variable. However, the inflorescence typology was not fully described by Kaastra in the protologues (described there as a acul panicle or variation thereof), and this represents a obti Novon 22: 288-291. Published on 24 May 2013. major distinction between these species. Thus, according to Article 9 in the International Code of Botanical Nomenclature (McNeill et al., 2006), an epitype may be designated, which we do here. Additionally, we provide a taxonomic emendation to E. cowanii, including leaf morphology variation, floral morphology, fruit and seed detail, and an updated geographic distribution. Esenbeckia cowanii Kaastra, Acta Bot. Neerl. 26: 481, t. 7, 1977. TYPE: French Guiana. Guyane, Montagne de Kaw, 250-270 m, 14 Dec. 1954 (fr.), R. S. Cowan 38833 (holotype, US- 00101521; isotypes, F0071245, NY 00051857). EPITYPE: Brazil. Mato Grosso: Vila Bela da Santfssima Trindade, est. Ca- choeirinha (cascata), 17 km de Vila Bela da Santfssima Trindade, Parque Estadual Serra de Ricardo Franco, ca. 500 m, na margem direita do riacho, 16 Nov. 2005 (fl., fr.), (epitype, designated here, P. Dias & R. G. Udulutsch 227 [sheets A and B], SPF). Figure 1. Tree 5-15 m tall with brown hyaline pubescence of appressed hairs to 0.4 mm; branchlets 3-6 mm diam., greenish brown and puberulent when young, then glabrescent, with elliptic, pale lenticels. Leaves alternate, unifoliolate, with sessile leaflet; petiole subterete, slightly winged or not, 4-19 mm, the base slightly tumid, puberulent with appressed trichomes; leaflet blades elliptic, (3.1-)6.1-21.7 X 2. 5-9.5 cm, acute to rounded at base, acuminate, with acumen obtuse to retuse at apex, the margin entire and flat doi: 10.3417/2007052 Volume 22, Number 3 2013 Dias et al. 291 Epitypification and Emendation of Esenbeckia cowanii Acknowledgments. of F, IAN, INPA, MG, NY, and US for loans of herbarium specimens and to Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renovaveis (IB AM A) for providing the collecting license N. 080/ 2005-COMON. We are especially indebted to Jacquelyn Ann Kallunki, one anonymous reviewer, and scientific editor Victoria Hollowell for their critical comments and suggestions, which have significantly improved this manuscript. P. D. was supported by Fundagao de Amparo a Pesquisa do Estado da Sao Paulo (FAPESP; Procs. 02/09762-6 and 04/15141-0), R. G. U. by Conselho Nacional de Desenvolvimento Cientifieo e Tecnologico (CNPq; Literature Cited Kaastra, R. C. 1977. New taxa and combinations in R La Acta Bot. Neerl. 26: 471-488. Kaastra, R. C. 1982. Pilocarpinae (Rutaceae). FI. Neotrop. Monogr. 33: 1-198. McNeill, J., F. R. Barrie, H. M. Burdet, V. Demoulin, D. L. Hawksworth, K. Marhold, D. H. Nicolson, J. Prado, P. C. Silva, J. E. Skog, J. H. Wiersema & N. J. Turland clature (Vienna Code). Regnum Veg. 146. (Rutaceae, Pilocarpinae) from Brazil and Bolivia. Bot. J. Linn. Soc. 129: 305-313. Coniogramme bashanensis (Pteridaceae), a New Fern Species from Shaanxi, China Xiao-si Guo and Jian-ping Chen College of Life Sciences, Northwest A&F University, Yangling, Shaanxi 712100, China. gxs6@163.com; ieditor@126.com Abstract. A new species, Coniogramme bashanen- sis X. S. Guo & B. Li (Pteridaceae), is described and illustrated. This new fern species was collected from the northern slope of Mt. Bashan in Langao County of Shaanxi Province, China. It is similar to C. wilsonu Hieron. in its perennial habit and frond veins that are 1- or 2-forked, occasionally connected to form one or two areoles. These fem taxa differ by the shape of the frond base. Pinnules are asymmetrically cuneate or hastate, or divided with a small auriculate segment in C. bashanensis; pinnules are cordate or rounded to Key words: China, Coniogramme, IUCN Red List, Mt. Bashan, Pteridaceae, Shaanxi. The genus Coniogramme Fee (Pteridaceae) con- sists of ca. 50 species, distributed mainly in tropical and subtropical regions of Asia, few of which are distributed in Africa and South America (Wu & Ching, 1991). About 39 species of Coniogramme were documented from China in the Flora Reipu- blicae Popularis Sinicae (Shing, 1990), mainly in the ern China. Recently, in the course of exploration in the Bashan Mountains, several specimens of Con- iogramme were collected. However, these specimens a small auricular segment at the pinnule base, which species. After careful examination of other morpho- logical characters, we determine this to be a new species of Coniogramme, which is described herein. Coniogramme bashanensis X. S. Guo & B. Li, sp. nov. TYPE: China. Shaanxi: Langao, Yihe town, Shiziwan village, 32°16'27"N, 109°4'13"E, 1250 m, 15 Oct. 2004, Y.-s. Chen, Z.-h. Wu, B. Li & X.-s. Guo 1965 (holotype, WUK; isotype, WUK). Figures 1-3. Diagnosis. Coniogramme bashanensis X. S. Guo & B. hastate base or the base divided with a small auricular the latter has lanceolate pinnules somewhat cordate or rounded-cuneate and glabrous on both surfaces. Herbs perennial, 60-90 cm tall; stems basally decumbent; scales dark brown, lanceolate; stipes 40- 60 cm, ca. 5 mm diam., gray-brown or stramineous; base with sparse, brown, linear-lanceolate scales; lamina 30-40 X ca. 25 cm, broadly ovate, bipinnate; lateral pinnae in 4 to 5 pairs, opposite, oblique and slightly incurved, basal pair largest, 15-20 X ca. 10 cm, petioles 3-4 cm, pinnate; lateral pinnules in 1 to 3 pairs, 10-15 X 2.5-4 cm, oblong to lanceolate, apex acuminate and caudate, base rounded to cuneate or hastate, shortly stalked or sessile; medial pinnae simple, oblong or lanceolate, 15-18 X 3-4 cm, petiolules 2-20 mm, otherwise similar to lateral pinnules, hastate or divided with a small auricular segment at base, unequal; terminal pinnules larger than adjacent pinnules, petiolules 1.8 cm, oblong to lanceolate, apex acuminate and caudate, base rounded to cuneate, unequal; fronds herbaceous when dry, adaxially dark green, abaxially gray-green, serrulate with minute, slightly irregular, deltate teeth; 1- or 2-forked, occasionally connected to form 1 or 2 areoles; the veinlets end in hydathodes; hydathodes slightly thickened, linear, reaching to base of teeth. Sori continuous along pinnule veins, extending to 6 mm from the pinna margin. Distribution and habitat. Coniogramme basha- known only from Langao County in Shaanxi Province, China. Langao County is located in the middle of the northern slope of Mt. Bashan. The new species occurs in secondary forest, as well as along locality is deciduous broad-leaved forest or mixed temperate coniferous and broad-leaved forest, and the dominant species are Quercus aliena Blume var. acutiserrata Maxim, ex Wenz., Pinus tabuliformis IUCN Red List category. Coniogramme basha- nensis is known only from one population with mature individuals fewer than 150. Its distributional area is smaller than 100 km2. The only known population of C. bashanensis is severely threatened by developing tourism and overgrazing in the area. Its conservation Novon 22: 292-296. Published on 24 May 2013. doi: 10.3417/2009145 294 Novon r forked once or twice, or rarely , free veinlets. The veinlets end in hydathodes and are that are continuously distributed along the veins. Coniogramme bashanensis is very distinct from other known species of Coniogramme in China and can easily be recognized by having four to five pairs of lateral pinnae. Coniogramme bashanensis is other- wise similar to C. wilsonii in general morphology. Both species have 1- or 2-forked veins, forming a few discontinuous areoles along each side of costae. The new species differs from the latter by its pinnules oblong to lanceolate, the base unequal, rounded to hastate or divided with a small auricular segment, and sparsely short hairs on both lamina surfaces. As for C. wilsonii, its pinnules are lanceolate, also with an unequal base, somewhat cordate or rounded to cuneate, and both lamina surfaces are glabrous. Coniogramme intermedia was also found near the distributional area of C. bashanensis, but the latter species has pinnules with a rounded to cuneate base and veins that do not form Novon United Kingdom. Shing, K.-H. 1990. Coniogramme Fee. Pp. 228-274 in R.- C. Ching & K.-H. Shing (editors), Flora Reipublicae Capituliferae (Labiatae) from on 24 May 2013. Novon Novon i&Ajani 1 2 7 11 the W and IRAN l Novon Basal leaf sizes (mm) Stem leaf sizes (mm), 6-14 X 4-8, mostly in 1 15-20 X 15-20 2-4 X 0.5-1, stems nearly : 1.5-2, glabrate, ciliate 3-4 X 0.5-1, v: 5-10 X 5-10 5-10 X 5-10, leaves equally 8 X 2.5-3^ villous in uppei Corolla shape Nutlet shape (see Table 2). Nutlet shape and size are also significant characters for the taxonomy of the tribe Nepeteae (Budantsev, 1993; Budantsev & Lobova, 1997). Distributional and ecological characters of the three similar species are compared in Table 3. Nepeta natanzensis is another member of Nepeta sect. Capituliferae, but it differs greatly from the discussed species in its clearly lignified caudex and basal parts of the stem and its spine-shaped calyx. The habitat of Nepeta lasiocephala was studied in the Hezar Mountains, in southern Iran, with sampling of two phytosociological releves. Table 1 compares the ecology and species composition of this species with N. sahandica, showing the data of releves. There is no information available for the habitat of N. monocephala because this species is known only from the type collection and has not been recorded since, despite some general, although not detailed, botan- ical surveys by local botanists (H. Maroofi, pers. comm.). The type locale for N. monocephala on the Chang Almas Mountain has not been botanically well explored, and its precise location is not known. It is likely that this species may be found again with botanical investigation. There is no doubt that this species is very rare and locally endemic, and further botanical exploration of its distribution, habitat, and ecology is highly recommended. Table 3. Altitudinal distribution Habitat 3500-3630 scree grounds of subnival zor 2200-2500 2000-4465 ^ no data scree grounds of subalpine Data Deficient (DD) (its Near Threatened (NT) Generic Status and Lectotypifications for Gymnospora (Polygalaceae) Jose Floriano Barea Pastore* and Pedro Luis Rodrigues de Moraes Programa de Pos-Graduagao em Botanica, Departamento de Ciencias Biologicas, Universidade Estadual de Feira de Santana, Avenida Transnordestina, Bairro Novo Horizonte, CEP 44.036—900, Feira de Santana, Bahia, Brazil * Corresponding author: jfpastore@hotmail.com Abstract. Polygala sect. Gymnospora Chodat (Poly- galaceae) is recognized at the generic rank based on J. F. B. Pastore with two new combinations, G. blanchetii (Chodat) J. F. B. Pastore and G. violoides (A. St.-Hil. & Moq.) J. F. B. Pastore. We also present a taxonomic key for species identification and argue that P. blanchetii Chodat should not be treated as a synonym of P. pedicellaris A. St.-Hil. & Moq. and that P. pedicellaris is better treated as a synonym of P. violoides A. St.-Hil. & Moq. (as G. violoides ). Thus, the name P. blanchetii (as G. blanchetii ) is established, in keeping with its traditional treatment. Lectotypifica- tions are provided for the following five names: P. blanchetii, P. laxa Nees & Mart., P. maritima Yell., P. pedicellaris, and P. violoides. Resumo. Polygala segao Gymnospora Chodat (Poly- Gymnospora (Chodat) J. F. B. Pastore, baseado nas evidencias filogeneticas. Nos apresentamos duas novas combinagoes G. blanchetii (Chodat) J. F. B. Pastore e G. violoides (A. St.-Hil. & Moq.) J. F. B. Pastore, uma chave taxonomica para identificagao das duas especies e defendemos que P. blanchetii Chodat nao deveria ser tratada como um sinonimo de P. pedicellaris A. St.-Hil. & Moq., enquanto que P. pedicellaris e melhor tratada como sinonimo de P. violoides A. St.-Hil. & Moq. (como G. violoides ). Assim, o nome P. blanchetii (como G. blanchetii) e estabelecido, mantendo o seu tratamento mais tradicional. Sao tambem indicados lectotipos para os seguintes nomes: P. blanchetii, P. laxa Nees & Mart., P. maritima Veil., P. pedicellaris e P. violoides. Gymnospora (Chodat) J. F. B. Pastore, here treated as a genus, was traditionally considered by Chodat (1893) and Paiva (1998), respectively, as a section or subgenus of Polygala L. (Polygalaceae). Currently, Gymnospora is comprised of two species that were included in Polygala, G. violoides (A. St.-Hil. & Novon 22: 304-306. Published on 24 May 2013. Moq.) J. F. B. Pastore and G. blanchetii (Chodat) J. F. B. Pastore. A third species originally included in Polygala sect. Gymnospora by Chodat, P. membra- nacea (Miq.) Gorts, is taxonomically excluded from Gymnospora because it lacks the diagnostic morpho- logical features. Gymnospora is endemic to Brazil, apparently restricted to forest margins and savannas. It is characterized by the following set of characters: pubescent pedicels, pubescent and free external sepals, calyx persistent in frutification, noncristate carina, puberulent and chartaceous capsules, and minute seed appendages or caruncles (Chodat, 1893; Marques, 1984). The caruncle, which is well developed in most other genera related to Polygala, can have a strong role in seed dispersal (Forest et ah, 2007). The reduction of the caruncle and the flattened and membranous fruits suggest that Gym- nospora is characterized by wind dispersal rather than by ants. The generic status for Gymnospora is additionally supported by phylogenetic evidence (Forest et al., 2007). Gymnospora, represented by G. violoides, lies as sister to the Badiera DC -Hebecarpa (Chodat) J. R. Abbott -Phlebotaenia Griseb. clade. All traditional sections or subgenera of Polygala are more closely related to other genera than they are to Poly gala, e.g., Acanthocladus Klotzsch ex Hassk. and Bredemeyera Willd. Thus, using the criterion of monophyly, Gymnospora cannot be maintained within the genus Polygala. Analysis of sequence data from G. material of the species deposited in studied has been thus far inadequate for molecular studies (Pastore, pers. obs.). However, morphological characters shared by both species of the genus unmistakably support the similarity between species of Gymnospora. After an analysis of the type collections of all relevant names, we conclude that there are two distinct species: Polygala blanchetii and P. violoides, with P. pedicellaris treated as a synonym of the latter. This is in agreement with the conclusions reached by Marques (1984). doi: 10.3417/2010113 A New Species of Lycoris (Amaryllidaceae) from Hunan, China Miao-Hua Quart* Li-Jun Ou, and Chao-Wen She Department of Life Science, Key Laboratory of Hunan Province for Study and Utilization of Ethnic Medicinal Plant Resources, and Key Laboratory of Hunan Higher Education for Hunan- Western Medicinal Plant and Ethnobotany, Huaihua University, Huaihua, 418008, People’s Republic of China * Author for correspondence: hhqmhl00@163.com Abstract. A new species, Lycoris hunanensis M. H. Quan, L. J. Ou & C. W. She (Amaryllidaceae), is described from Yuanling County in Hunan Province, China. Diagnostic morphological characters that distinguish it from the similar L. straminea Lindl. are presented. The clearest differences are that the perianth of L. hunanensis is pink in bud, lightly yellowish pink in early anthesis, and gradually fading to white later in development, with light red spots apically, while that of L. straminea is always ocher- yellow. Lycoris hunanensis has longer scapes (50-60 cm vs. ca. 35 cm long in L. straminea). The leaf apex different from the obtuse apex of L. straminea, and the leaf blade midveins are noticeably paler in L. straminea. The new species is assessed as Endan- gered (EN), according to IUCN criteria. Key words: Amaryllidaceae, China, Hunan, IUCN Red List, Lycoris. Lycoris Herb, is a small genus with ca. 20 species in the family Amaryllidaceae (Hsu & Siro, 1994) that is mainly distributed in China, Japan, and Korea and seldom occurs in Burma, Nepal, and Indonesia (Hsu & Fan, 1974; Yang, 1976). China is richest in Lycoris species, with 15 species and two varieties, which are mostly distributed in the Zhejiang, Jiangsu, and Anhui provinces in eastern China (Xu et al., 1985; Yuan et al., 2008). During a botanical expedition to Yuanling County in Hunan Province in August 2009, we collected an unusual plant of Lycoris. This 2009 collection provided insufficient material for conclu- sive identification, but in 2010 we collected additional material of this species in Yuanling County, and we now have quite a few living plants at the botanical garden of Huaihua University, where they have come into bloom every year since 2009. Detailed literature studies and comparisons of almost all Lycoris species in China at the Hangzhou Botanical Garden and the Nanjing Botanical Garden Memorial Sun Yat-Sen reveal that this species is indeed similar to L. straminea Lindl. but remarkably doi: 10.3417/2011013 different in its flower characters and leaf shape. In fact, these characters are quite distinctive in the genus and are critical in determining its taxonomy (Fig. 1). The new species is distinguished from L. straminea by characters of the perianth, scape, and leaf. A taxonomic key to the Lycoris species from Hunan Province in China, based on the essential details of the flowers and leaf, is provided below. Lycoris hunanensis M. H. Quan, L. J. Ou & C. W. She, sp. nov. TYPE: China. Hunan: Yuanling Co., Zhuhongxi, moist surroundings, 110°38'E, 28°43'N, 200-350 m, 7 Oct. 2009, M. H. Quan 09007 (holotype, Huaihua University). Figure 1. Herbs perennial; bulbs subglobose, 3-5 c 4-6 cm, with purple-brown epidermis. Leaves 5 to 7, ensiform, 36-39 X 1.3-1. 5 cm, apex acuminate, dark green; leaf midvein not distinctly paler. Scapes 50- 60 cm, 10 mm diam.; involucral bracts 2, each 4-5 X 1.4-1 .6 cm, lanceolate, membranous, light green, apex acuminate, direction changing in bloom. Flowers 6 to 7 per umbel, floral tube 1-1.2 cm; perianth with 6 tepals, 5-6 cm X 6-8 mm, erect in bud, later in development strongly recurved, tepal margin strongly undulate, pink in bud, lightly yellowish pink in early anthesis, later in development gradually fading to white, with scattered pink stripes adaxially and light red spots apically that gradually abaxially; pedicel 0.6-1 cm; stamens 6, filaments longer than the perianth; anther red-brown, 6-7 X 0.8-1 mm; pistil 1, style linear, white but faint fuchsia apically; ovary hypogynous, about 5 mm, green, subspheroidal. Fruit a 3-valved capsule; seeds black, subspheroidal or hemispherical. Novon 22: 307-310. Published on 24 May 2013. Holcus pintodasilvae (Poaceae, Pooideae, Poeae), a New Species from the Island of Madeira (Portugal), and Notes on Macaronesian Holcus Miguel Menezes de Sequeira GBM Centro de Ciencias da Vida, Universidade da Madeira, Alto da Penteada, 9000-390 Funchal, Portugal, sequeira@uma.pt Santiago Castroviejo'f Real Jardrn Botanico de Madrid, Plaza de Murillo, 2, Madrid 28014, Spain Abstract. A new species of Holcus L. (Poaceae), endemic to the island of Madeira (Portugal), is here described as H. pintodasilvae M. Seq. & Castrov. This new species is morphologically closely related to the Macaronesian endemics H. rigidus Hochst. ex Hochst. (Azores Archipelago) and H. mollis L. subsp. hierrensis Stierst. (El Hierro Island, in the Canary Islands) but differs by having the culm nodes light brown, the leaves with a glabrous sheath and subspreading lamina attenuated to an acuminate apex, the glumes blunt, shiny, glabrous (except in the minutely scabrous veins), the upper glume with proximal lateral veins, and the lemma of the upper floret with an awn ca. 4 mm, twice as long as the lemma. Chromosome counts in H. pintodasilvae are tetraploid (2 n = 28). Morphology, ecology, biogeog- raphy, and conservation issues are discussed and related to other Macaronesian Holcus taxa. Holcus (totaling less than 20 individuals), and its IUCN conservation status is Critically Endangered (CR). A new specific status is proposed for H. mollis subsp. hierrensis, as H. hierrensis (Stierst.) Stierst. & M. Seq., based on its chromosome number and morphological Resumo. Descreve-se uma nova especie, Holcus pintodasilvae M. Seq. & Castrov., pertencente ao genero Holcus L. (Poaceae), endemica da Ilha da Madeira (Portugal). Morfologicamente proxima dos endemismos macaronesicos H. rigidus Hochst. ex Hochst. (Arquipelago dos Agores) e H. mollis L. subsp. hierrensis Stierst. (Ilha de Hierro, Arquipelago das Canarias) diferencia-se por possuir: nos de ; folhas com bainhas glabras i apice acumi- nado; glumas brilhantes, muticas e glabras (excepto nas nervuras que se apresentam escabriusculas), gluma superior com as nervuras laterais proximais; lema da flor superior com arista ca. 4 mm (duas vezes da contagem de cromossomas de H. pintosilvae (2 n = 28). Discutem-se, e relacionam-se com outros taxa Macaronesicos do genero Holcus, aspectos relativos a morfologia, ecologia, biogeografia e conservagao. Holcus pintodasilvae foi encontrado em apenas tres populagoes (num total de menos de 20 indivfduos), sendo o seu estado de conservagao IUCN esta criticamente ameagado (CR). Propoe-se ainda a combinagao H. hierrensis (Stierst.) Stierst. & M. Seq. baseada n In a recent revision of the genus Holcus L. (Menezes de Sequeira, 2004), nine species, five subspecies, one \hybrid, and one variety were recognized. More recently, H. azoricus M. Seq. & Castrov. was described as an Azores endemic of possible hybrid origin (Menezes de Sequeira & Castroviejo, 2007). The genus Holcus is restricted to the Old World, but the human-mediated expansion of H. lanatus L. and H. mollis L. subsp. mollis give Holcus a worldwide distribution (Menezes de Se- 2006). Within the genus, five taxa are Iberian Peninsula endemics (H. grandiflorus Boiss. & Reut., H. gayanus Boiss., H. caespitosus Boiss., H. reuteri Boiss., and H. annuus C. A. Mey. subsp. duriensis (P. Silva) Franco & Rocha Afonso), one taxon is a South African Cape Province endemic (H. annuus subsp. setiger (Nees) M. Seq. & Castrov.), and three are Macaronesian endemics (H. rigidus Hochst. ex Hochst., H. mollis subsp. hierrensis Stierst., and H. azoricus ). Recent molecular data seem to group Holcus within the subtribe Holcinae of Poeae, possibly with the genus Vahlodea Fr. (Soreng et al., 2003; Quintanar et al., 2007), rather than sharing a common ancestor within Aveneae (e.g., Deschampsia P. Beauv. and Hierochloe R. Br.), as previously doi: 10.3417/2010109 Novon 22: 311-320. Published on 24 May 2013. 312 Novon suggested by Clayton and Renvoize (1986) and Watson and Dallwitz (1994). The Madeira Archipelago is composed of two main islands, Porto Santo (dating back to ca. 20 Ma) and Madeira (less than 5.6 Ma; Ribeiro et al., 2005). The island of Madeira is situated between 32°38' and 32°52'N and 16°39' and 17°16'W, approximately 600 km northwest of the western African coast, with a land area of 737 km2. The vascular flora of Madeira includes 1137 taxa (species and subspecies), of which 96 taxa are exclusive endemics, with the Poaceae being the most diverse plant family (133 taxa, with six endemics, including Agrostis obtusissi- ma Hack., Anthoxanthum maderense Teppner, De- (Hack. & Bomm.) Buschm., Festuca donax Lowe, and Koeleria loweana Quintanar, Catalan & Castrov.; see Jardim & Menezes de Sequeira, 2008). Hansen and Sunding (1979, 1993) and Cope (1994) listed Holcus lanatus for the islands of Madeira, Porto Santo, and the Selvagens and H. mollis for Madeira. Other authors, such as Menezes (1906, 1914), mention only H. lanatus. In fact, H. lanatus collections date back to the 19th century, and the species’ expansion is clearly human-mediated through habitat change, but its arrival in the flora could have been prior to human settlement (Menezes de Sequeira, 2004; Jardim & Menezes de Sequeira, 2008), at least on marine cliffs and in humid habitats. Although Cope (1994) regarded H. mollis as native, and Robert Brown (1773-1858) listed the plant in his account of Madeiran species (Britten, 1904) (possibly in error for H. lanatus, not included on the list), and recently Vieira (2002), based on Cope (1994), excluded this species in his account of the exotic plants of Madeira, references to H. mollis are very recent. The first known specimen of H. mollis was collected by Costa with the notes “Santa, no Porto do Moniz, vii-1934, Costa, MADM,” possibly corre- sponding to a human-mediated introduction, estab- lished during the massive afforestation that took place in the first half of the 20th century (Menezes de Sequeira et al., 2007; Jardim & Menezes de Sequeira, 2008). The Macaronesian endemics Holcus rigidus and H. mollis subsp. hierrensis (Stierstorfer, 2001; Menezes de Sequeira & Castroviejo, 2007) are small chamae- phytes with aboveground branching, whereas H. hybrid origin (H. rigidus X //. lanatus; Menezes de Sequeira & Castroviejo, 2007). The morphological the genus Holcus, namely the chamaephyte habit, clearly suggests that a specific rank should be applied to these endemics, and accordingly, a new status is proposed for H. mollis subsp. hierrensis. 1. Holcus hierrensis (Stiersl.) Stierst. & M. Seq., comb, et stat. nov. Basionym: Holcus mollis L. subsp. hierrensis Stierst., Feddes Repert. 112: 49. 2001. TYPE: Spain. Canary Islands: El Hierro, SE of Eremita de la Caridad, 1270 m.s.m., UTM 28RBR065742, N-facing slope, Vitrands, blocks of lava, 10 July 1999, Ch. Stierstorfer s.n. (holotype, B-100006407; iso- types, M not seen, TFC not seen). Apart from the new status proposed above as Holcus hierrensis for the El Hierro endemic, recent fieldwork demonstrated the presence of a new species restricted to Madeira. Plants of this new species found in the higher mountains of Madeira show a typical Macaronesian chamaephyte habit, but they also show v new species is described herein. 2. Holcus pintodasilvae M. Seq. & Castrov., sp. nov. TYPE: Portugal. Madeira: Pico do Arieiro, ca. Pico do Gato, 1670 m.s.m., exp. N, 32°44'30.6"N, 16°56'17.4", 16 July 2003, M. Menezes de Sequeira 4368 (holotype, MA- 807339). Figures 1—4. Stierst. & M. Seq. similis, sed ab eis nodis brevissimis natis, glumis nitentibus fere glabris (praeter nervos duplo longiorem) desinente differt. Rhizomatous chamaephyte (Figs. 1A, 3B), with a slender ramified rhizome; culms 55-79 cm tall, branched above ground; nodes light brown, 0.5-1 mm, glabrous (Figs. IB, 3C). Leaves rigid, sub- pungent, subspreading; sheath glabrous (Fig. IB), median leaves with sheath 3.8-6.6 cm, upper leaves with sheath 7-11 cm; lamina minutely scabrous (Fig. 2 A, B), linear-lanceolate (Fig. 3 A), flat to convolute, persistent, attenuate to an acuminate apex, median leaves 57.7-104.2 X 2.6-5.1 mm, upper leaves 11.8-35.7 X 8.4-15.3 mm; ligule 2-2.8 mm (Figs. IB, 3D), asymmetric with long asymmetrical teeth, membranous, glabrous except for the short ciliate margins (hairs 14-26 pm), shortly toothed. Panicle (Figs. 3A, 4) 4.7-7.7 cm, whitish and shiny (Fig. 1C), few-flowered (ca. 100), somewhat compact (spike- like), regular in outline; branches scabrous with Volume 22, Number 3 2013 Sequeira & Castroviejo Holcus (Poaceae) from Madeira 313 acicular hairs. Spikelets (Figs. 1C, 3E, 3F) 4. 5-6.4 mm, falling entire, with a well-marked abscission zone at the base of the glumes, with 2 florets; glumes thinly papery, shiny, slightly unequal, enclosing the flowers, apex acuminate, blunt, keeled, both keel and nerves minutely scabrous; lower glume 4.5-5.8 X 1. 7-2.1 mm, 1-nerved (Fig. 3G), lamina glabrous, shiny, keel scaberulous, prickle hairs 19-109 pm, 314 Novon Volume 22, Number 3 2013 316 Novon Figure 4. Holcus pintodasilvae M. Seq. & Castrov. Photograph of the holotype M. Menezes de Sequeira 4368 (MA-807339). apex acuminate, blunt; upper glume 4.4-6. 1 X 2.3-3 mm, 3-nerved (Fig. 3H), lateral veins proximal, keel scaberulous, prickle hairs 23-82 pm, apex acuminate distal floret, divided in 3 parts: proximal segment short, 0.33-0.53 mm (between the glumes and the first floret), curved, glabrous; the median segment (between the florets) longer, 0.6-0.75 mm, straight, glabrous to puberulous (Figs. 2D, 3F); distal segment (above the upper floret), rudimentary, ca. 0.1 mm, Volume 22, Number 3 2013 Sequeira & Castroviejo Holcus (Poaceae) from Madeira 317 villous and setose. Lower floret hermaphroditic; callus very short, glabrous; lemma cartilaginous (Fig. 3J), 2.2-2.6 mm, shortly puberulous on the keel, deeply emarginate, short ciliate (Figs. 2E, 3K); palea 2. 1-2.5 mm, membranous, trilobed with the lateral lobes as long as or slightly shorter than the median (entire) (Figs. 2E, 3M, 3N), keels (2) pubescent, included within the lemma, adherent to the caryopsis, lodicules ca. 0.7 mm (Fig. 3N), with a turgescent circular-elliptic base, glabrous, acumi- nate, symmetric, with a short lateral tooth, ca. 35 pm, inserted in the distal 2/3; stamens 3 (falling before those of the upper floret), anthers ca. 1.2-1. 3 mm long, blunt. Upper floret staminate; callus 0. 1-0.2 mm long, pear-shaped to rounded, pubescent with hairs 36-915 pm long; lemma 2.2-2.6 mm (Fig. 3L), puberulous, emarginate, with a dorsal awn 3.3-4. 1 mm, geniculate, inserted in the distal 2/3 to 3/4 of the lemma; palea 2-2.4 mm, 2-nerved, shortly trilobed, pubescent; anthers ca. 1.3 mm long, blunt. Leaf anatomy. Although Watson and Dallwitz (1994) indicated that all sclerenchyma cells in Holcus are associated with vascular bundles, the presence of sclerenchyma cells close to the abaxial blade surface not associated with the vascular bundles was for H. rigidus. However, H. pintodasilvae also presents sclerenchyma cells beneath the abaxial surface that are not associated with the vascular bundles (Fig. 2C). Leaf rigidity results from the presence of these sclerenchyma cells not associated of sclerenchyma in all the vascular bundles. Chromosome counts. Due to the scarcity of individuals found in nature, chromosome counts, in accordance with Menezes de Sequeira and Castro- viejo (2007), were performed in only three rhizome portions collected (non-destructive sampling) from three different plants that were then grown at the University of Madeira gardens. Chromosome counts on Holcus pintodasilvae correspond to a tetraploid genome, with 2 n = 28 (Fig. 2F). Due to chromosome pairing, an autopolyploid origin is hypothesized, possibly close to H. mollis. Polyploidy is common among endemic Macarone- sian grasses, for example, the tetraploid Madeiran Dactylis smithu Link subsp. hylodes P. F. Parker (Parker, 1972) and the high 25-ploid Madeiran Koeleria loweana (Quintanar et al., 2006). However, Menezes de Sequeira et al. (2009) have shown that this is not the case for native Festuca L., where all five Macaronesian species were diploid. Holcus pintosasilvae is a tetraploid taxon, as are H. rigidus (Menezes de Sequeira, 2004; Menezes de Sequeira & Castroviejo, 2007) and H. hierrensis (Strierstorfer, 2001). These Macaronesian endemic and tetraploid Holcus species correspond to apoendemics, following the classification proposed by Favarger and Contan- driopoulos (1961) and Contandriopoulos (1988). They are presumed to have evolved from one or several colonization events from H. mollis diploid paleoen- demic-patroendemic (see Favarger & Contandriopou- los, 1961; Contandriopoulos, 1988) Iberian popula- and Menezes de Sequeira and Castroviejo (2007). This was in agreement with Boissier and Reuter (1852), who considered Hispania (the Iberian Peninsula) and part of Mauritania (northwestern Africa) to be the dispersal center for the genus Holcus. References to diploid H. mollis plants, in fact, are restricted to Portugal (Fernandes & Queiros, 1969) and Bulgaria (Walter, 1980), possibly corre- sponding to a formerly continuous Ibero-Turanian distribution affected by Pleistocene glaciations (Me- nezes de Sequeira, 2004). Distribution and habitat. Holcus pintodasilvae is a very rare endemic found so far in only three populations on Madeira. As has been stated by Menezes de Sequeira et al. (2008) for the recently described Teucrium francoi M. Seq., Capelo, J. C. Costa & R. Jardim, there was a positive impact on population numbers and size caused by the removal of goats from the higher mountains of Madeira (Sousa, and this was possibly related to the discovery of H. pintodasilvae. However, the new taxon could have been overlooked by previous collectors, who could have confused it with Anthoxanthum maderense or Agrostis obtusissima, which have a similar growth form. The chamaephytic habit is strongly correlated with susceptibility to damage from grazing, and the rarity of H. pintodasilvae could therefore be related to the long-term effects of grazing by goats introduced immediately after human colonization in the 15th century, which caused massive landscape changes (Menezes de Sequeira et al., 2007). Although Silva et al. (unpublished) noted a quick recovery of the formerly rare plant endemics after goats were removed, they also observed that this recovery was followed by a strong resurgence of invasive plants such as Cytisus scoparius (L.) Link subsp. scoparius, with negative impacts on chamaephytic endemics, such as H. pintodasilvae, by habitat reduction. Holcus pintodasilvae occurs in the Armerio maderensis- Parafestucetum albidae and Teucrio francoi-Origa- netum virentis plant communities (Capelo et al., 1999; Costa et al., 2004). Amomum sahyadricum (Zingiberaceae), a New Species from the Western Ghats, India V. P. Thomas, M. Sabu* and K M. Prabhu Kumar on 24 May 2013. 322 Novon Three New Species of Paepalanthus sect. Diphyomene (Eriocaulaceae) from Minas Gerais, Brazil Novon Novon 330 Novon Bauhinia hekouensis (Leguminosae, Caesalpinioideae), a New Species from Yunnan, China Tieyao Tu and Dianxiang Zhang * Key Laboratory of Plant Resources Conservation and Sustainable Utilization, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, Guangdong, People’s Republic of China * Author for correspondence: dx-zhang@scbg.ac.cn Abstract. Bauhinia hekouensis T. Y. Tu & D. X. Zhang (Leguminosae, Caesalpinioideae), a narrow endemic from Hekou County, Yunnan Province, in southwestern China, is described and illustrated. The new species can be diagnosed by a combination of with nine to 11 nerves; terminal, racemose inflores- cences; spindle-shaped, slightly falcate flower buds with acute apices; tubular hypanthium; five freely reflexed sepals during anthesis; and large white flowers with oblanceolate, crisped petal blades. The lorphology suggests an affinity of the new ) series Clavatae (de Wit) Wunderlin of section Phanera (Lour.) Wunderlin in Bauhinia L., especially to the species B. coccinea (Lour.) DC. and B. nervosa (Wall, ex Benth.) Baker. Key words: Bauhinia, Caesalpinioideae, China, IUCN Red List, Leguminosae. The legume genus Bauhinia L. comprises ca. 300 species of trees, shrubs, and lianas. As the largest genus of the tribe Cercideae (Leguminosae, Caesal- pinioideae), it occurs throughout pantropic areas, Asia and South America, extending into southern North America (Wunderlin et ah, 1987). Forty-two species of Bauhinia from China were documented in the Flora Reipublicae Popularis Sinica (Chen, 1988), and several additional species have been subse- quently described or recorded (Zhang, 1993; Zhang & Chen, 1996, 1998). There are 47 species recognized in the recently published Flora of China (Chen et al., 2010). During an exploration for Bauhinia in southwestern China as part of our long- term endeavor on the systematics and evolution of the genus, an apparently new species was discovered. The new species exhibits characteristics of the series Clavatae (de Wit) Wunderlin in Bauhinia, from section Phanera (Lour.) Wunderlin, subgenus Pha- nera (Lour.) Wunderlin, K. Larsen & S. S. Larsen. It shows close affinities to B. coccinea (Lour.) DC. and B. nervosa (Wall. & Benth.) Baker. Bauhinia ser. Clavatae consists of ca. 22 species occurring in tropical southeastern Asia (de Wit, 1956; Larsen et al., 1980; Wunderlin et al., 1987), with one species and one subspecies previously recorded from Yunnan Province of China (Chen et al., 2010). Bauhinia hekouensis T. Y. Tu & D. X. Zhang, sp. nov. TYPE: China. Yunnan: Hekou Co., Nanxi Town, Baishahe, 22°41.237'N, 103°57.597'E, 169 m, near stream, 11 Oct. 2008, T. Tu 511 (holotype, IBSC; isotypes, IBSC, KUN). Figure 1. Haec species a Bauhmia coccinea (Lour.) DC. et B. nervosa (Wall, ex Benth.) Baker petiolo 2.5-6 (nee 1.5-3) Large liana, extending into upper canopies of host pubescent, singly inserted at nodes; branchlets terete, pubescent when young, gradually glabrescent. Leaves ovate to suborbicular, up to 12 X 14 cm; the blades bifid to 1/3 divided with a broad s young to glabrate, blade base cordate, with primary nerves 9 to 11, sLoul. connected by transverse secondary nerves; petioles green, 2.5-6 cm; stipules deciduous, obovate, rotund or orbicular, ca. 3-6 X 2-A mm, glabrous or minutely puberulent. Inflorescences racemose, terminal, unbranched, 5-25 cm; axis greenish pubescent, with up to ca. 30 flowers; pedicels slender, 2. 1-3.4 cm, puberulent; bracts lanceolate, green, 10-15 X 2-3 mm; bracteoles linear, acute, 4-8 mm, inserted above the middle of the pedicels. Flower buds green, spindle-shaped, slightly falcate, apex acute, 15-25 X 3-5 mm; hypanthium tubular, 17-21 X 1. 5-2.5 mm; calyx green, splitting into 5 sepals on the apex of the hypanthium, acicular, subulate, reflexed at anthesis, 1.8-2.5 cm; petals arranged white, crisped, the lower surface and edges of the upper surface sparsely pubescent, the median zone of Novon 22: 332-335. Published on 24 May 2013. doi: 10.3417/2010017 Volume 22, Number 3 2013 Tu & Zhang 333 Bauhinia (Leguminosae) from Yunnan, China the upper surface densely pubescent; the posterior blade of the corolla oblanceolate, the lower part of the blade gradually, the upper part of the blade more abruptly, narrowing to the apices, 14-22 X 5-9 mm; the blades of the lateral petals slightly narrower than the posterior ones, 12-21 X 3-9 mm; the petal claws green, equal to or slightly longer than petal blades, ± pubescent; fertile stamens 3, opposite to the petals; filaments green, glabrous, 32-47 mm, starting outward and then reflexing upward; anthers maroon, elliptic, 3- 4 mm, opening by longitudinal slits; staminodes 2, filiform, 1 mm; ovary densely pubescent, on a 8-15 Distribution and habitat. Bauhinia hekouensis is currently known only from the type locality in a Tu &; Studies in Andean Ocotea (Lauraceae) II. Species with Hermaphrodite Flowers and Densely Pubescent Lower Leaf Surfaces, Occurring Above 1000 Meters in Altitude f base revolute and decurrent on the petiole . . f base plane or sometimes reflexed, but not Volume 22, Number 3 2013 344 346 Novon Figure 3. Holotype of Ocotea gentryi van der Weiff (E. Perez Arbeldez & J. Cuatrecasas 8179, US). covered, the indument floccose, hairs tufted; ten buds densely appressed pubescent. Leaves 7-12 4.5 char cm, alternate, elliptic or narrowly elliptic, f rtaceous, base acute, margin flat, apex acute, i face with some appressed, straight hairs when y< soon becoming glabrous, lower surface densely pubes- cent with appressed hairs, surface completely or largely covered, hairs tufted, floccose; domatia absent; midrib and secondary veins slightly impressed, ultimate venation very finely reticulate on the upper surface, 347 , B, F, NY, P, PORT not s 5 3.5-7 X 2-A cm, a r 6 stamens ca. ] 348 Novon Volume 22, Number 3 2013 van der Werff 349 Studies in Andean Ocotea (Lauraceae) l der Werff (B. Stergios et al. 16107, MO). In silvis subandinis occidentalibus vulcani Tungurahua, Aug. 1901, Sodiro 143/6b (holo- type, B). Otuzco, fiber Succhabamba, 2700 m, 29 June 1914, Weberbauer 6998 (holotype, B). Trees to 30 m, sometimes flowering at 6 m; twigs angular to terete, 3-5 m diam., moderately to densely appressed to ascending pubescent, hairs inconspic- Volume 22, Number 3 2013 ofO. lii Volume 22, Number 3 2013 ||£kst-. Novon 362 7-10 X 2-2.7 s 6, ca. 1.5 X 0.7 mm, (2-2.7 cm vs. 3-6.5 5 (V U), based on IUCN 16°15'S, 67°41'W, 364 Novon Novon Novon 370 Novon Sanchez Vega 4566 (2), 6007 (4), Shonle 165 (21), Silverstone-Sopkin 6539 (10), Smith 2519 (2), 6114 (21), 7191 (2), 13896 (13), Stein 2718 (18), Stergios 16107 (8), 16113 (8), Steyermark 118317 (20). Tipaz 1 (9), 10 (10), 64 (10), 85 (10), 87 (10), Triana 1022 (6). Valenzuela 1764 (1), 8826 (1), 10133 (7), 10161 (7), H. Vargas 3352 (3), 3504 (3), W. Vargas 527 (17), 555 (10), 704 (17), 8095 (22), 9068 (22), Vasquez 25423 (1), 25434 (1), 25464 (21), 25471 (1), 26761 (11), 28546 (14), 28550 (14), 28915 (14), 30186 (2), 30388 (14), 30753 (14), 32728 (14), 34518 (14). Weberbauer 3545 (4), Webster 29471 (3), 30597 (3), van der Werff 7622 (20), 9189 (18), 9192 (18), 9195 (18), 9245 (1), 9283 (18), 9284 (18), 9285 (18), 9321 (18), 9426 (10), 10439 (10), 10594 (10), 10595 ( 9), 10624 (10), 13363 (9), 10468 (10), 14745 (1), 14932 (2), 16852 (1), 16854 (21), 16855 (1), 16884 (2), 17719 (1), 18666 (14), 18670 (14), 18705 (14), 18582 (14), 19474 (15), 19600 (12), 19704 (14), 22680 (1), 22885 (14), 22954 (2), 22957 (2), 22965 (2), 22997(2), 23071 (2), 23077 (14), 23081 (14), 23202 (14), 23203 (14), 23220 (2), Wood 112 (6), Wurdack 1278 (2), 1422 (2), 1462 (2). Zak 2473 (9). A New Woody Bamboo < from Central Brazil, Pedro L. Viana 22: 371-376. on 24 May 2013. cells in rows of two to three, quadrangular to cuplike; the intercostal zone; costal zone composed of silica bicellular microhairs with the apical cell shorter and cells, long cells, bicellular microhairs, short to with rounded apex. Abaxial epidermis consisting of up midsized unicellular macrohairs with sharp apices, to eight rows of cells in the costal zone and up to 10 in apparently with random disposition; intercostal zone 374 Novon Valid Publication of Croton dinghuensis (Eupho •rbiaceae) and Helicia yangchunensis (Proteaceae), Two Species Endemic to Guangdong, Chine Xiang Chun-lei and Peng Hua A New Species of Stephania (Menispermaceae) from South Guangxi, China on 24 May 2013. Novon intact staminate flower. — F. Inflorescence of the staminate plant. G-L. Portion of pistillate vine and flowers. — G. Fertile habit. Lateral view of pistillate flower. — L. Inflorescence of the pistillate plant. A-F drawn from the holotype A. P. Meng 20060002 i P. Meng 20060001 (HIB). Erratum The print date in NOVON 22(2) is 30 November 2012; however, the 28 November 2012 posting of the PDF to BioOne constitutes the effective date of publication. doi: 10.3417/2013020 Novon 22: 383. Published on 24 May 2013. www.mbgpress.info CONTENTS Missouri Botanical Garden A Journal for Botanical Nomenclature 2013 Volume 22, Number 4 Month 2013 Novon, A Journal for Botanical Nomenclature from the Missouri Botanical Garden The mission of the Missouri Botanical Garden is to discover and share knowledge about plants and their environment, in order to preserve and enrich life. Novon publishes short articles whose primary purpose is the establishment of nomencla- ture in vascular plants and bryophytes. All articles are peer-reviewed by qualified, indepen- dent reviewers. Manuscripts must fully state and justify the reasons for proposing nova. These may include detailed comparisons with similar taxa, short keys to similar taxa, illustrations to similar taxa, and mechanical nomenclature reasons, among others. Manuscripts whose primary purpose is other than establishment of new nomenclature, which usually are longer manuscripts, cannot be accepted for review. These include reviews, revisions, monographs, or other papers that incidentally include nova. Manuscripts must follow the guidelines in the Checklist for Authors. The Checklist may be downloaded from the Garden’s web site, www.mbgpress.info, or authors may contact the editor at novon@mobot.org to request a copy. Novon will not knowingly accept manuscripts that have been simultaneously submitted to other journals for consideration or previously published in some form elsewhere. Victoria C. Hollowell Allison M. Brock Lisa J. Pepper Laura L. Slown Cirri R. Moran Ihsan A. Al-Shehbaz Carmen Ulloa Ulloa George Yatskievych Kanchi N. Gandhi Nicholas J. Turland Roy E. Gereau Editor Managing Editor Associate Editor Associate Editor Press Coordinator Consulting Editor Consulting Editor Consulting Editor Nomenclature Consultant Nomenclature Consultant Latin Editor Annals of the Missouri Botanical Garden. Price for 2012: $180 per year U.S.A.; $190 Canada and Mexico; $215 all other countries. Four issues per Postal address: Subscriptions: Annals of the Missouri Botanical Garden, P.0. Box 1897, Law- rence, KS 66044-8897; Manuscript submissions: St. Louis, MO 63166-0299. Editorial queries: novon@mobot.org Orders for back issues: mbgpress@mol ssouri Botanical Garden 2013 © Novon is printed on paper that meets the re- quirements of ANSI/NISO Z39.48-1992. Novon (ISSN 1055-3177) is published quarterly by the Missouri Botanical Garden, 2345 Tower Grove Avenue, St. Louis, MO 63110. Periodicals postage paid at St. Louis, MO and additional mail- ing offices (USPS #006-777). POSTMASTER: Please send address changes to Annals of the Missouri Botanical Garden, P.0. Box 1897, Lawrence, KS 66044-8897. Information on the contents of Novon is present- ed in SciSearch®, Research Alert®, Current Con- Sciences! APT Online, tta CAB Abstract/Global Health database, and the ISI® database. The full-text of Novon is available online through BioOne™ (http://www.bioone.org). ©Mis Volume 22 NO VON Number 4 W 2013 Lectotypifications in Calyptranthes (Myrtaceae) from the Neotropics Novon Information regarding the locations of collections or top sets of particular botanists is taken from Taxonomic Literature 2 (TL2 Online, 2012). When there were no obvious reasons to select one syntype or collection number over another, the better distributed collection was selected. Collections of the following herbaria were visited: BM, E, G, K, M, US, W; the following type collections were available online (separate from JSTOR): C, F, NY, RB, S, U; collections were checked on our behalf by the respective curators of the following herbaria: GOET, HAL, MA, MEL. As suggested by the referees we are using “Sello,” instead of the spelling “Sellow,” within the text when referring to the German botanist and collector. Jahrb. Syst. 19§ 596. 1895. TYPE: E Cuba, May 1889, Eggers 5404 (lectotype, designated here, M; isolectotypes, A, K). The name Calyptranthes glabrescens was based on the single collection Eggers 5404; however, the holotype was destroyed at B. Although Eggers’s top set of the collection is known to be housed at the Copenhagen herbarium, no material from this collection number was found there. Three isotypes were found at other herbaria. Considering that Urban distributed the duplicates himself after he studied them (Stafleu & Cowan, 1981), it is very likely he saw the three isotypes above. The Munich sheet is selected over the others because it bears Urban’s handwriting identifying the collection as C. glabres- cens, whereas the K sheet is identified as C. syzygium, an illegitimate name, and the sheet housed at A has not been identified. The M sheet is a complete specimen; it was closely studied by the authors and corresponds well to the morphological data in the protologue. Myrtaceae (Kiaerskov, 1893: 38). Three years after their publication, Urban corrected this error with his new name C. kiaerskovii Krug & Urb. (Urban, 1898: 43), the holotype of which was housed in Berlin and has since been destroyed. The only known extant material is an isotype at C ( Eggers 3217), which is here designated as the lectotype. 3. Calyptranthes longifolia 0. Berg, FI. Bras. 14(1): 46. 1857. TYPE: Peru, “In silvis primaevis secundus fluvium Amazonum in prov. Solimoes, Brasilia,” s.d., [E. F.J Poeppig 2162 (lectotype, designated here, W; isolectotypes, G, HAL, P, W). Berg (1857) noted only the collection Poeppig 2162 in the protologue without indication of the herbarium. Poeppig’s main type collection set was at W, but he sent duplicates to Berlin where Berg was based (Stafleu & Cowan, 1981). The B sheet is destroyed, and we, therefore, decided to select the collection from W because Poeppig’s top set is housed there and because the sheet label corresponds with the citation for locality by Berg (1857). 4. Calyptranthes multiflora Poeppig ex 0. Berg, FI. Bras. 14(1): 42-43. 1857. TYPE: [Brazil. Prov. Amazonas?:] “ '/lumen Tejfe, Prov. Solimoes," 1831, [E. F.J Poeppig 2684 (lectotype, desig- nated here, W; isolectotypes, G, F [3]. F neg. 68695, HAL, MICH, NY, P). Berg noted two collections syntypic to one another in the 1857 protologue: Poeppig 2684 and Spruce s.n. (Brazil, Para, “barra do Rio Negro,” F, K, M, OXF, W). The Poeppig collection is selected as the lectotype because it is a better specimen, has more complete material, was widely distributed, and presents more information on the label. vii Krug & Urb., Symb. replacement name for Calyptranthes obovata Krug & Urb. Bot. Jahrb. Syst. 19: 600. 1895, nom. illeg. superfl., non Calyptranthes obovata Kiaersk., Enum. Myrt. Bras.: 38. 1893. TYPE. British Virgin Islands. Tortola, 1895, H. Eggers 3217 (lectotype, designated here, C). Krug and Urban (1898) described this species, citing the specimen Eggers 321 7 as the type collection, but under the illegitimate name Calyp- tranthes obovata Krug & Urb.; this epithet was already occupied by Kiaerskov for a different taxon from Brazil, accepted in the Enumeration of Brazilian 5. Calyptranthes obovata Kiaersk., Enum Myrt. Bras. 38-39, tab. IV k-m. 1893. TYPE: Brazil. Rio de Janeiro, Feb. 1889, A. M. Glaziou 17662 (lectotype, designated here, C; isolectotypes, BR, C, F, G, K, NY, P, R-R000008940, US). Kiaerskov cited two collections by Glaziou, numbered as 17662 and 16992 (BR, C, F not seen, G) in his protologue for Calyptranthes obovata, and both were found broadly distributed in several herbaria. Because the author was based at the Copenhagen herbarium, we believe the lectotype should be chosen between one of the two collections housed there. Both collections are equally good Lectotypification of Ranunculus cabrerensis (Ranunculaceae) Eduardo Cires Departamento de Biologia de Organismos y Sistemas, Universidad de Oviedo, Catedratico Rodrigo Urfa s/n, 33071 Oviedo, Spain, and Department of Biology, Research Group Spermatophytes, Ghent University, K. L. Ledeganckstraat 35, 9000 Ghent, Belgium, cireseduardo@gmail.com Candela Cuestra Departamento de Biologia de Organismos y Sistemas, Universidad de Oviedo, Catedratico Rodrigo Una s/n, 33071 Oviedo, Spain, and Department of Plant Systems Biology, VIB, Technologiepark 927, 9052 Ghent, Belgium, candelacuesta@gmail.com Jose Antonio Fernandez Prieto Departamento de Biologia de Organismos y Sistemas, Universidad de Oviedo, Catedratico Rodrigo Una s/n, 33071 Oviedo, Spain, jafp@uniovi.es Abstract. Typification of the endemic Iberian species Ranunculus cabrerensis Rothm. is discussed. A specimen at MA from the exsiccatae for the Plantae Hispanie Boreali-Occidentalis (Rothmaler, 1934) is designated lectotype for this endemic Iberian species. Key words : Ranunculaceae, Ranunculus, Spain. Werner Rothmaler published his “Species novae vel nomina nova florae hispanicae” where he described the taxon “ Ranunculus cabrerensis Roth- maler nov. sp. ... Ex affinitate R. pamassifolii L., a quo differt habitu robustiore, foliis utrinque lanatis, petiolis lanatis, rostro carpelli longiore spiram formante” (1934: 148). In the same work, Rothmaler also noted that the taxon is better considered a geographical variant of the R. pamassiifolius com- plex: “Est forma geographica et melius R. pamassii- folius L. ssp. R. cabrerensis Rothm. nominanda” (1934: 148). The use of a binary combination (“/?. cabrerensis ”) instead of an infraspecific epithet of the International Code of Nomenclature (ICN; McNeill et al., 2012), and although the name is validly published, it needs to be altered to the proper form: R. pamassiifolius subsp. cabrerensis. Similarly, the name was accepted at infraspecific rank by Kiipfer (1974) when he studied the alpine buttercups of R. pamassiifolius, a polyploid complex belonging to Ranunculus sect. Ranuncella (Spach) Freyn (Ranunculaceae), which is widespread and distribut- ed throughout south-central mountains in Europe. A relevant issue is that Rothmaler (1934) simultaneously proposed two different names — Ra- nunculus cabrerensis and R. pamassiifolius subsp. cabrerensis — for the same taxon. According to Article 34.2 of the ICN, these alternative names are validly published because their publication was effective before 1 January 1953. Based on recent investiga- tions on the molecular, cytogenetic, and morpholog- ical characterization of the complex R. pamassiifolius (sensu Kiipfer, 1974), we have found many arguments that support the distinction of R. pamassiifolius subsp. cabrerensis from R. pamassiifolius, with an independent recognition at species level (cf. Cires et al., 2009a, 2009b, 2010; Cires & Fernandez Prieto, 2012). Exsiccatae for Rothmaler’s Plantae Hispanie Bor- eali-Occidentalis are located in the Madrid herbarium (MA), as noted in the protologue for Ranunculus cabrerensis, as “n. 130.” This collection number for Rothmaler appears on two sheets at MA, but there is no in sched. distinction between the two in Roth- maler’s hand. Both sheets were annotated by Gonzalo Nieto Feliner in 1983 as “Material Tipo,” although neither was further distinguished for type. The first sheet (MA-40508, ) includes a fruiting plant with cordate leaves (at upper left of sheet), among seven leafy fragments, and corresponds well to the taxonomic concept because the fruits provide precise diagnostic characters to distinguish R. cabrerensis from the rest of the taxa of R. pamassiifolius. The other sheet (MA- 40509, ) is similar, with three flowering plant fragments among the seven on the sheet. Taking into account the latest evidence toward relationships within Ranunculus pamassiifolius com- plex, and following Article 9.2 of the ICN (McNeill et doi: 10.3417/2012013 Novon 22: 389-390. Published on 18 October 2013. *.f «• Eysenhardtia byei (Leguminosae, Papilionoideae), una Especie Nueva del Noroeste de Mexico doi: 10.3417/2010117 Novon 22: 391-395. Published on 18 October 2013. El doble omas del doble ^imdular, "recto a ligeramente curvo 1-1.5 Four New Species of Anisotes (Acanthaceae) from Madagascar Thomas F. Daniel Department of Botany, California Academy of Sciences, 55 Music Concourse Drive, Golden Gate Park, San Francisco, California 94118, U.S.A. Author for correspondence: tdaniel@calacademy.org Rokiman Letsara Department of Botany, California Academy of Sciences— Madagascar, Enceinte PBZT, Tsimbazaza, Antananarivo, Madagascar, rletsara@calacademy.org Santiago Martm-Bravo Department of Molecular Biology and Biochemical Engineering, Pablo de Olavide University, ctra. de Utrera Km. 1, 41013, Sevilla, Spain, smarbra@upo.es Abstract. Four species of Anisotes Nees (Acantha- ceae) are described from northern and west-central Madagascar: A. hygroscopicus T. F. Daniel, Letsara & Martm-Bravo, A. perplexus T. F. Daniel, Letsara & Martm-Bravo, A. subcoriaceus T. F. Daniel, Letsara & Martin-Bravo, and A. venosus T. F. Daniel, Letsara & Martin-Bravo. A key to the six species of the genus known from Madagascar, all of them endemic to the island, is provided. Morphological features previously unknown in the genus are noted for A. hygroscopicus and A. venosus (hygroscopic trichomes on seeds), A. subcoriaceus (2-colporate, pseudocol- pororate pollen). None of these species can be recognized sections of Anisotes. Data pertinent to the conservation status of each species are provided. Key words: Acanthaceae, Anisotes, IUCN Red List, Madagascar. Anisotes Nees is included in subfamily Acanthoi- deae, tribe Justicieae, where it forms part of a grade of Old World relatives of Justicia L. (McDade et al., 2000; Daniel et ah, 2007). The genus can be characterized by the combination of its strongly bilabiate corollas with ascending cochlear aestiva- tion, relatively short corolla tube (corolla tube:overall corolla length up to 0.56, but usually 0.33 or less), rugulate upper lip, and lower lip that is usually recoiled; androecium of two stamens and no staminodes; and bithecous anthers with thecae mostly subequally to unequally inserted. Anisotes is mor- phologically similar to Justicia, and may not be can be found among the ca. 700 currently recognized species of the latter genus. Novon 22: 396-408. Published on 18 October 2013. Baden (1981a) recognized 19 species of Anisotes tropical Arabia. Baden (1981b, 1984) also treated three species from tropical eastern and southern Africa in the morphologically similar genus Meta- rungia Baden. One of these was subsequently discovered to occur in western Africa (Darbyshire et al., 2008). Yollesen (2010) recognized 24 species in Anisotes, those previously treated in both Anisotes and Metarungia, the sole species previously treated in Chlamydostachya Mildbr., and a newly recognized species that had been treated as a subspecies of A. dumosus Milne-Redh. by Baden (1981a). Baden (1981a) recognized six sections of Anisotes based primarily on differences in inflorescences, bracteolar venation, and pollen sculpturing. Two species have been reported from Madagascar, Anisotes divaricatus T. F. Daniel, Mbola, Almeda & Phillipson and A. madagascariensis Benoist (Daniel et al., 2007), both endemic to coastal or near coastal sandy regions in the dry, spiny forests and thomscrub of the southwestern sector of the island nation. Benoist, an expert on Acanthaceae of Madagascar, annotated several specimens among the Malagasy collections at P as probable undescribed species of Anisotes, but did not publish names for these taxa (Daniel et al., 2007). Recent collections augment those known to Benoist and reveal the existence of other new species of Anisotes from Madagascar. Additional specimens at P were identified by Benoist as Anisotes but these appear to represent species of other genera. Recent collections from Madagascar also appear to represent species of Anisotes (e.g., Letsara et al. 874 at CAS, MO, and TAN), but have insufficient material to characterize fully and thus doi: 10.3417/2012054 Volume 22, Number ■ 2013 Daniel et al. : Anisotes (Acanthaceae) from Madagascar of 3) are well below the threshold to continuing decline (subcriterion b) or extreme 400 Novon hygroscopicus T. F. Daniel, Letsara & Martin-Bravo. - — A .Anisotes subcoriaceus (Daniel et al. 11878, CAS), nodes showing (Daniel et al. 10440), dry seed with appressed trichomes. — D. Anisotes hygroscopicus (Daniel et al. 10440), same seed at same Volume 22, Number ■ 2013 Daniel et al. 401 Anisotes (Acanthaceae) from Madagascar fh l dl n( 1 l on c) in any of the EOO; AOO; Vulnerable (VU) based on criterion D2, because the area, extent, and/or quality of the habitat; number of AOO and the number of locations are smaller than 20 locations; and number of mature individuals (IUCN, km2 and fewer than five, respectively (i.e., guideline 2001, 2011). In addition, the species might qualify as thresholds; IUCN, 2001). However, because the Volume 22, Number ■ 2013 Daniel et al. 403 Anisotes (Acanthaceae) from Madagascar , 0.39-1 times -7 X 0.6-1. 5 404 Novon mm, lobes lmea/ 0.5-2 X 0.3-0.5Pmm; stamens 25- 406 Novon secondary veins 5 to 7 per side. Spikes axillary, mostly opposite at leaf nodes, (1 or)2 to 4 per axil (sometimes with a vegetative branch in axil as well), densely bracteate, pedunculate, peduncles 4-17 mm, glabrous or sparsely pubescent with antrorsely appressed eglandular trichomes to 0.1 mm long, fertile portion of spike subcylindric, 10-26 mm (excluding corollas), rachis not visible, pubescent with antrorsely appressed eglandular trichomes 0.5-1 mm long; bracts green or sometimes tinged with maroon, subcoriaceous, imbricate, 4-ranked (2 adja- cent rows fertile, thus spikes ± secund), oblate to subcircular to broadly elliptic to obovate, 4.5-10.5 X 4.5-8 mm, rounded at apex and either entire, 2-fid, or with a V-shaped split in center, abaxial surface inconspicuously puberulent with erect subglandular trichomes to 0.05 mm long (glandular puberulent) or appearing glabrous (and often pubescent near base with trichomes like those of rachis as well), ± conspicuously veined, major veins 7 to 9, subparallel to midvein, margin densely ciliate with erect to flexuose eglandular trichomes 0.1-0.6(— 1) mm long; proximal pair (or pairs) of bracts sterile and smaller than fertile pairs, 2.5-6 X 2.5-6 mm; bracteoles linear to linear-oblanceolate to lance-linear, 4.5-6.5 X 0.8-1. 2 mm, abaxial surface subglandular puber- 0.8 mm long proximally, only midvein evident, margin ciliate like bracts. Calyx 7.5-8 mm, tube 1- 1.5 mm, 0.14-0.23 times as long as lobes, lobes lanceolate. 6.5 -7 X 1-1.4 mm, abaxial surface and margin pubescent like bracteoles or with midvein pubescent with antrorse eglandular trichomes its entire length, margin hyaline, ciliate; corolla pale cream-yellow to yellow-green, with maroon markings on lower lip and maroon tinged on internal surface of upper lip, 31-41 mm, externally pubescent with an overstory of retrorse (proximally) to antrorse (distally) eglandular trichomes 0.2-0.5 mm long and an understory of ± erect glandular trichomes to 0.1 mm long, tube subcylindric proximally and expanded distally, 13-16 mm, 0.36-0.43 times as long as corolla, 1.5-1. 7 mm diam. near midpoint, upper lip 17-25 mm, internally rugulate, apically entire, lower lip recoiled, 15-21 mm, lobes 1.3-1.8 X 0.8-1.2 mm; stamens 17-23 mm, inserted near apex of corolla tube, filaments pubescent with eglandular trichomes proximally, glabrous distally, thecae par- allel, 2-3.5 mm (distal theca longer), unequally inserted (overlapping by 1.2-1. 5 mm), glabrous, both thecae with a conspicuous basal appendage 0.2-0.6 mm; pollen 2-colporate, 4-pseudocolpate, with re- gions (bands) of exine between colpi and pseudocolpi sometimes ± breaking up into insulae toward poles; style 27-30 mm, proximally sparsely pubescent, distally glabrous, stigma inconspicuous, lobes (if present) not evident. Capsule and seeds not seen. Distribution and habitat. Anisotes subcoriaceus is endemic to northern Madagascar (Antsiranana; Fig. 3) where plants occur in subdeciduous dry (to somewhat mesic) forest (western dry forest, fide Moat & Smith, 2007) and windswept scrub on sandstone at elevations between 100 and 570 m. IUCN Red List category. This species is known from three collections (two of them from the same 4 km2 grid) that correspond to two subpopulations (IUCN, 2001). The E00 (ca. 30 km2), A00 (8 km2), and number of locations (maximum of 2) would place this species in the Endangered (EN) category under criteria B1 and B2, but lack of information for subcriteria b or c preclude designating this conser- vation category at the present time. According to criterion D2, Anisotes subcoriaceus might also be considered as Vulnerable (VU) based on the A00 and number of locations, if a plausible threat to the species could be identified. No such threat is currently known. One of the subpopulations occurs in a reserve (community-based protected land under auspices of the Malagasy NGO, FANAMBY) in the Andrafiamena transitional forest. Between 75 and 100 plants of various sizes and ages were observed in 2011 along a 1-km segment of trail in this forest. Given the lack of a knowledge of threats to A. subcoriaceus, it must be considered Data Deficient (DD) at this time. Phenology. Anisotes subcoriaceus was collected in flower in June and July; fruiting collections are Etymology. The epithet derives from the subco- riaceous texture of the leaves and bracts of this Discussion. Pollen of Anisotes subcoriaceus (Dan- iel et al. 11878 and Phillipson 2010 ; Fig. 4C, D), with two colpori and four pseudocolpi (i.e., no insulae in trema region), resembles pollen of several species of Justicia (Graham, 1988; Daniel, 1990, 1998), but was Metarungia by Baden (1981a, 1981b). Based on other characters used by Baden (1981a) to delimit sections of Anisotes (e.g., inflorescence type, bract venation, and the ratio of the corolla tube length to the overall corolla length), A. subcoriaceus would 407 Validation of the Name Ulmus kunmingensis (Ulmaceae) Deng Yunfei Key Laboratory of Plant Resources Conservation and Sustainable Utilization, South China Botanical Garden, the Chinese Academy of Sciences, Guangzhou, 510650, People’s Republic of China. yfdeng@scbg.ac.cn Chen Xin College of Forest Resources and Environment, Nanjing Forestry University, Nanjing 210037, People’s Republic of China, chenxinzhou@hotmail.com Abstract. Ulmus kunmingensis W. C. Cheng (Ul- maceae), originally described in 1963, was an invalidly published name, because two collections, types. The name is validated here with the designation of the fmiting collection W. C. Cheng 11002 (NF) as the holotype. Key words: China, Ulmaceae, Ulmus. Ulmus kunmingensis W. C. Cheng (Cheng et al., 1963) was described based on material from Kun- ming, Yunnan, China. Unfortunately, it is invalidly published under Articles 8.1, 40.1, and 40.2 of the Melbourne Code (McNeil et al., 2012) because two collections, both kept at NF, one flowering ( W . C. Cheng 11002 ) and the other fruiting (T. C. Chao 1516), were cited concurrently as types in the protologue. In the original description, Cheng et al. (1963) compared the species with U. glaucescens Franch. However, U. kunmingensis is easily distin- guished from the latter by its seed apex not reaching the notch, and these two species were placed by Fu (1980) in different series, Ulmus ser. Glabrae Moss, and Ulmus ser. Nitentes Moss., respectively. Fu et al. (1979) reduced the name to the rank of variety as U. changii W. C. Cheng var. kunmingensis (W. C. Cheng) W. C. Cheng & L. K. Fu, but no validation of the basionym was undertaken at that time. Since then, some authors have treated U. kunmingensis as an independent taxon at the rank of species (Fu, 1980; Fu et al., 1982; Liu, 1985; Bao, 1990; Ma & Zhou, 1997; Mao, 2005; Li, 2006), while others have considered it a variety of U. changii (Li, 1982; Wu, 1984; Fu & Xin, 2000; Fu et al., 1998, 2003). In fact, U. kunmingensis is easily distinguished from '|X' \ changii by its flowers from mixed buds, scattered in basal or sub-basal bract axils on young branches and its leaf blades that abaxially bear tufted hairs in axils of veins. In contrast, U. changii is distinguished by its flowers in fascicled cymes from floral buds, not mixed buds, and its leaf blades being either glabrous or pubescent along the veins. Given this morpholog- ical distinction, we therefore prefer to treat kunmingensis as a distinct taxon at the rank of However, until now, the invalidity of Cheng’s name was not recognized in recent floristic works. The present paper validates the name Ulmus kunmingen- sis under Article 33.1 (McNeill et al., 2012) by designating T. C. Chao 1516 as type. The selected holotype is a specimen with fruits in the herbarium NF and was cited in the original publication. The fruiting collection is preferred over the flowering one (Cheng 11002), both mentioned by Cheng (Cheng et al., 1963: 12), because fruit is one of the significant diagnostic characters to identify species in Ulmus. Validation is also supported by direct reference to Cheng’s Latin description (Cheng et al., 1963: 12). The flowering collection, W. C. Cheng 11002, is designated as the paratype. We ascribe the name, U. Cheng, which is supported by Article 46.2 (McNeill et al., 2012). Cheng, sp. nov. Ulmus kunmingensis W. C. Cheng, Sci. Silvae Sin. 8(1): 12. 1963, nom. inval. TYPE: China. Yunnan: Kunming, Xishan, 20 Mar. 1934 (fr.), T. C. Chao 1516 (holotype, NF; isotype, PE). Distribution and habitat. Ulmus kunmingensis is endemic to China, occurring in Guangxi, Guizhou, Sichuan, and Yunnan provinces. It is known from forests at altitudes from 600 to 1800 m. Acknowledgments. We are grateful to the curators of the herbaria at Nanjing Forestry University (NF) and the Institute of Botany, Chinese Academy of doi: 10.3417/2010098 Novon 22: 409-410. Published on 18 October 2013. Primorsky Territory, Russia from the Roman V. Doudkin 412 Novon '2* / ' t 1 mm % Volume 22, Number 4 2013 Doudkin & Volkova Boechera (Brassicaceae) from Russia 413 Berkutenko, A. N. 1988. Brassicaceae. Pp. 38-115 in S. S. Kharkevich (series editor), Vascular Plants of Soviet Far East, Vol. 3. Leningrad. [In Russian.] Berkutenko, A. N. & N. N. Gurzenkov. 1976. Chromosome numbers and distribution of Cruciferae in the south of (Moscow & Leningrad) 61: Magadan Region. Bot. Zhi 1595-1603. Doudkin, R. V. Lozovy (Chandalaz) range, Prymorye region. 83(3): 107-111. Aristea (Iridaceae, Aristeoideae), a Subgeneric Classification r«isr Volume 22, Number 4 2013 Goldblatt 41 7 Subgeneric Classification of Aristea (Iridaceae) Goldblatt, P., A. Le Thomas & M. Suarez-Cervera. 2004. Phylogeny of the Afro-Madagascan Aristea (Iridaceae) Bot. J. Linn. Soc.S144: 41-68. P ^ & McNeiH, J., F. R. Barrie, H. M. Burdet, V. Demoulin, D. L. Hawksworth, K. Marhold, D. H. Nicolson, J. Prado, P. C. Silva, J. E. Skog, J. H. Wiersema & N. J. Turland (editors). 2006. International Code of Botanical Nomen- clature (Vienna Code). Regnum Veg. 146. Pax, F. 1888. Iridaceae. Pp. 137-157 in A. Engler & K. 2(5). Wilhelm Engelmann, Leipzig. Weimarck, H. 1940. F ' Univ. Lund. 2, 36(1): ! Typification of the Name Kochia saxicola (Chenopodiaceae) Duilio Iamonico and Gudrun Kadereit Laboratory of Phytogeography and Applied Geobotany, Department DATA, Section Environment and Landscape, University of Rome, Sapienza Via Flaminia 72, 1-00196 Rome, Italy. Corresponding author: d.iamonico@yahoo.it Institut fur Allgemeine Botanik, Johannes Gutenberg— Universitat Mainz, 55099 Mainz, Germany. clausing@uni-mainz.de Abstract: The typification of the name Kochia saxicola Guss. [= Eokochia saxicola (Guss.) Freitag & G. Kadereit] (Chenopodiaceae) is here discussed, and a specimen from the Gussone collection (NAP) is designated as the lectotype. The distribution of the species is given and remarks on its IUCN conserva- tion status are provided. Keywords: Chenopodiaceae, Eokochia, Italy, IUCN Red List, Kochia. Recent molecular, morphological, and biogeo- graphical studies of the tribe Camphorosmeae in the Chenopodiaceae (Kadereit & Freitag, 2011) revealed three major lineages, informally named as the Chenolea Thunb., Sclerolaena R. Br., and Bassia- Camphorosma clades. The phylogenetic analyses showed that all nonmonotypic genera, except Cam- phorosma L. and Neokochia (Ulbr.) G. L. Chu & S. C. Sand., were polyphyletic. Among these, Bassia All. [= Kochia Roth] was split into five clades and four new genera: Eokochia Freitag & G. Kadereit, Grubovia Freitag & G. Kadereit, Sedobassia Freitag & G. Kadereit, and Spirobassia Freitag & G. Kadereit. Eokochia belongs to the Chenolea clade, which consists of four disjunctly distributed and morpho- logically divergent genera with this group altogether comprising only five species: the monotypic Chenolea from South Africa; the monotypic Spirobassia from Europe to South Siberia; Neokochia, with two species in North America; and Eokochia, whose single species is a rare endemic of the area surrounding the Tyrrhenian Sea west of Italy. Kadereit and Freitag (2011) concluded that these four genera are the extinction, and that their present-day ecology might represent the ancestral ecology of the entire tribe Camphorosmeae. Eokochia saxicola (Guss.) Freitag & G. Kadereit is a subshrub, growing on coastal cliffs exposed to sea spray. The leaves are strongly succulent and almost terete. The species has single flowers in the axils of regular leaves and the persistent perianth shows five distinct wings in the fruiting stage (Fig. 1). The new monotypic genus Eokochia Freitag & G. Kadereit is based on Kochia saxicola Guss. (Gussone, 1855) which is, as far as we are aware, yet untypified. This study is part of the research work carried out by one of the authors (D. Iamonico) for the project “Taxa of vascular plants described from Italy: Inventory, typification, and biosystematic investiga- tions” under the auspices of the Italian Botanic Society (e.g., Iamonico, 2010, 2012a, 2012b, 2012c, 2013; Iamonico et al., 2011; Iamonico & Jarvis, 2012; Iamonico & Peruzzi, 2012; Iamonico & Reveal, 2012). Eokochia saxicola (Guss.) Freitag & G. Kadereit, Taxon 60(1): 72. 2011. Basionym: Kochia saxicola Guss., Enum. PI. Inarim.: 275. 1855. Bassia saxicola (Guss.) A. J. Scott, Feddes. Repert. 89: 108. 1978. TYPE: [Italy. Campa- 10 Aug. 1850, G. Gussone s.n. (lectotype, designated here, NAP). Figure 1. In order to select an adequate lectotype for Eokochia saxicola, the specimens collected by G. Gussone in BI, CAT, FI, NAP, PAL, and RO were examined (acronyms according to Thiers, 2011). Gussone’s collections came from two islands in the Tyrrhenian Sea: Ischia (locus classicus) and Capri (Campania region). Most were collected after the year 1855, so they cannot be chosen as the lectotype (see Art. 9.3 of the ICN; McNeill et al., 2012). Two specimens (kept in NAP) are from 1850 and each includes two fragments and one label. Both labels report the locality “Ischia ne scogli di S. Anna,” which matches the locus classicus by Gussone (1855). However, one of the specimens (collected on 28 May 1850) bears a plant without an inflorescence, the structure of which is important to the identification of the species. The other, collected on 10 Aug. 1850, is a good specimen, bearing two parts of one plant whose features (both sexual and vegetative characters) are in accordance with the Novon 22: 418-421. Published on 18 October 2013. doi: 10.3417/2011058 Volume 22, Number 4 2013 419 lamonico & Kadereit Typification of Kochia saxicola (Chenopodiaceae) protologue. After the diagnosis Gussone (1855) cited a reference to an “Ic. [Iconography] nostra T. XIII.” This plate shows a detailed drawing that includes a portion of the plant with both sterile and fertile branches and details of flowers and seeds. The illustration can be considered original ] perfectly matches the protologue. Among these selected elements (specimens from Gussone’s collection in NAP and the illustration by Gussone), we prefer to choose one of the specimens lIMIf Casearia maynacarpa (Salicaceae), a New Name for Carpotroche parvifolia Peter M. j0rgensen and Ron L. Liesner Missouri Botanical Garden, P.0. Box 299, St. Louis, Missouri 63166-0299, U.S.A. peter.jorgensen@mobot.org; ron.liesner@mobot.org Abstract. A new name is proposed for Mayna parvifolia (J. F. Macbr.) Sleumer [= Carpotroche parvifolia J. F. Macbr.]; the species belongs in Casearia Jacq. (Salicaceae). However, the basionym name Casearia parvifolia Tub has already been used twice by both Willdenow and Tulasne, and we therefore establish Casearia maynacarpa Liesner & P. Jprgensen as a replacement name. Key words: Bolivia, Carpotroche, Casearia, Mad- agascar, Mayna, Peru. During the preparation of the Salicaceae for the catalogue of Bolivian vascular plants, we discovered that a name was missing for numerous specimens from Peru and Bolivia of what so far have been called Mayna parvifolia (J. F. Macbr.) Sleumer. The species was originally based on fruiting material by Macbride (1932) in the genus Carpotroche Endl., but it belongs in Casearia Jacq. The three genera in question are easily distinguished by the following morphological Casearia, to two to five in Mayna Aubl., to four to eight in Carpotroche. The number of petals and sepals is identical (five) in Casearia, whereas Mayna and Carpotroche have more petals than sepals. The fruit in Casearia is a capsule that splits into three(four) valves; in Carpotroche, the fruit is also a capsule, but one that splits late, and maybe only if pressed, into three to five(to seven) valves. In Mayna, the fruit is an indehiscent berry that occasionally splits distally, perhaps also as an effect of pressing. Casearia has no pulvinus associated with the leaves, while both Mayna and Carpotroche do. The leaves in Casearia obovate in Mayna and Carpotroche. Casearia has leaf blades that are pellucid linear punctate, which does not occur in Mayna or in Carpotroche (Sleumer, 1980). The placement of the species in the correct genus would not be problematic if flowering material were available, but even sterile and fruiting speci- mens have enough characters to safely and correctly place them in Casearia. Macbride (1934: 390) stated: “I venture to describe it even in the absence of flowers,” but he was somewhat uncertain of its placement, noting that “the leaves are exceptionally small for the genus, suggesting more the related group Mayna.” Sleumer (1938) was confident in moving the species to Mayna, but provided no argument or morphological character to support the move. Subse- quently, he indicated that the flowers are unknown (Sleumer, 1980), but had evidently seen only four collections. We have now seen 33 collections and still have not encountered flowering specimens; all material are either sterile or fruiting. However, the species undoubtedly belongs to the genus Casearia despite the fruit having long projections, as many species of Mayna and Carpotroche do. This is the sole character that the species has in common with these two genera, and a number of exceptions to the circumscription of Mayna will disappear with the suggested realignment, such as the appearance of linear glands on leaves, the thick dehiscent capsular fruits with three valves, and the absence of a pulvinus. Casearia maynacarpa Liesner & P. Jprgensen, nom. nov. Replaced name: Carpotroche parvifolia J. F. Macbr., Candollea 5: 390. 1934, non Casearia parvifolia Willd., Sp. PL, Ed. 4: 628. 1799, nom. illeg., nee Casearia parvifolia Tul., Ann. Sci. Nat., Bot., ser. 5, 9: 328. 1868, nom. illeg. Mayna parvifolia (J. F. Macbr.) Sleumer, Repert. Spec. Nov. Regni Veg. 45: 12. 1938. TYPE: (fr.) L. Williams 5260 (holotype, F, F photo 50666; isotypes, G, G fragm. ex F, USj* A combination based on Carpotroche parvifolia in Casearia is blocked because the epithet has already been used in the genus twice. The oldest name by Willdenow from 1799 is illegitimate because it places C. decandra in synonymy and is thereby in conflict with Principle III of the Melbourne Code (McNeill et al., 2012). The second name by Tulasne from 1868 is a later homonym, for which a replacement name was proposed by Baillon in 1886 in Guidonia Mill., which was subsequently transferred into Casearia as C. tulasneana (Baill.) Warb. Novon 22: 422-423. Published on 18 October 2013. doi: 10.3417/2013004 Tropidia namasiae , a New Species of Orchidaceae (Epidendroideae, Tropidieae) from Southwestern Taiwan doi: 10.341' Volume 22, Number ■ 2013 Synthetodontium kunlunense (Mielichhoferiaceae, Musci), a New Moss Species from the Kunlun Mountain Range, China 430 Novon peristome. — B. Inner surface of peristome Mamtimin 09336 (HBNU). Etymology. The species epithet refers to the type locality on the Kunlun Mountains in the Xinjiang Uyghur Autonomous Region in China. Discussion. Superficially the endostome seg- ments of Synthetodontium kunlunense appear to be positioned opposite the exostome teeth, as in the Funariales and Encalyptales. Further observations using scanning electron microscopy reveal two or three vertical lines on the ventral endostomial surface. Shaw (1984) explains the endostomial structure of S. pringlei, noting that there are two or three columns of cells on the inner surface of the endostome. These give rise to the cilia that are fused to each other as well as to the exostome teeth. Thus, the endostome is not clearly organized into segments and cilia. This appearance can lead to a misunder- standing of the peristome pattern, but this is in fact a moss with segments that alternate with the exostome 432 Novon n and protonema development of Synthetodontium kulunense J. C. Zhao & Y. Y. Liu. — A. Spore. — chloronema. — F. Bud (arrow) on the caulonema. — G. Young gametophyte. Micrographs of material taken from the holotype S. Mamtimin 09336 (HBNU). teeth. This structure is also one of the most important features used in identifying Synthetodontium (Shaw & Rooks, 1994; Shaw, 2000). Based on the morpholog- ical features of S. kunlunense and current conceptions of the family (Shaw, 2009), the new species surely belongs to the Mielichhoferiaceae. Synthetodontium kunlunense differs from S. prin- glei, the only other species in the genus, in several morphological characters. The new species has longer leaves (0.5-1. 3 mm vs. ca. 0.5 mm in S. pringlei ), shorter median leaf cells (8.6-21.7 pm vs. 40.6-72.6 pm), decurrent leaf bases (vs. not decurrent), a cells (vs. green). Furthermore, geographic nental masses. Synthetodontium pringlei is restricted to North America (Mexico) (GBIF Data Portal, 2010); S. kunlunense is thus far found only in the Kunlun Mountains of China. The gametophyte of Synthetodonuum kunlunense is similar to that of Mielichhoferia, especially M. sinensis Dixon. However, there are morphological differences between the two species. Synthetodontium kunlunense is defined by its double peristome, in which the endostomial membrane and segments tightly adhere to the exostome, whereas M. sinensis has a single exostomial peristome (Li, 2006; Zhang et al., 2007). Distal and medial leaf cells of S. kunlunense are shorter than those of M. sinensis (77-125 X 10-20 pm), and the proximal leaf cells are Bryum algovicum is similar to Synthetodontium kunlunense in that the endostome is fused to the exostome teeth (Shaw, 1984; Li, 2006; Zhang et ah, 2007). The cilia stand alone, and segments are perforate in B. algovicum, while the cilia of S. kunlunense are fused to each other and adhere to the in S. kunlunense are narrow and nonperforate, appearing to lie opposite the exostome teeth. In addition, B. algovicum can generally be distinguished from S. kunlunense by its taller plants, 5 mm (Crum & Anderson, 1981) to 10 mm (Noguchi, 1988; Li, 2006; Zhang et ah, 2007), the larger leaves 2-3 mm, the long-excurrent costa, and the leaf border composed of two to four rows of linear, thick-walled cells. Synthetodontium kunlunense are observed as shorter plants, 2-3 mm, with smaller leaves, 0.5-1. 3 X 0.4- 0.6 mm, the costa usually ending at or near the apex, and the marginal cells somewhat narrower, with the two or three rows not forming a distinct border. Synthetodontium kunlunense may be closely relat- ed to Haplodontium Hampe or Plagiobryoides J. R. Spence [Bryum Hedw. subsect. Areodictyon (Mull. Hal.) Broth.], as indicated by its similar leaves and laminal areolation. However, Haplodontium consists of species with an extremely reduced peristome, typically of one layer (Spence, 2005). Those of Plagiobryoides have leaf margins showing a strongly differentiated border, the capsules zygomorphic with an oblique mouth, and a double-layered peristome. These features distinguish Haplodontium and Pla- aim ei« Two New Species of Passiflora supersect. Decaloba (Passifloraceae) from Eastern Mexico John M. MacDougal Department of Mathematics and Natural Sciences, Harris-Stowe State University, 3026 Laclede Avenue, St. Louis, Missouri 63103, U.S.A. threebrane@sigmaxi.net Abstract. Two new species of Passiflora L. (Passi- floraceae) endemic to eastern Mexico are described and illustrated and placed in subgenus Decaloba (DC.) Rchb., supersection Decaloba in a clade distinguished by no laminar nectaries. Passiflora dictamo DC. is a synonym of P. biflora Lam., and the rare diminutive species previously assigned to P. dictamo in Killip’s monograph is described as P. lauana J. M. MacDougal. A second species, P. complanata J. M. MacDougal, is described and placed in the P. sexflora Juss. species group, where it is notable for its clusters of large flowers to 4 cm diam., flattened stems, and unusual 1/2 phyllotaxy. From Mexico we now recognize 75 native species of Keywords: IUCN Red List, Mexico, Passiflora subg. Decaloba , Passifloraceae. Presented here are two new species of passion- flowers from the states of Oaxaca and Puebla, Mexico, bringing the number of native species of Passiflora L. known in that country to 75, of which 59 are in subgenus Decaloba (DC.) Rchb. This subgenus is comprised mostly of smaller-flowered species with small purple berries, and it contains a number of clades and species groups that have radiated and diversified in the seasonal temperate or wet moun- tains of Mexico. Many of the species in subgenus Decaloba have limited distributional ranges, some- times confined to one part of a mountain range, and the two new species described herein are local endemics. The first species to be described below was originally treated by Killip in his monograph (1938) under the misapplied name Passiflora dictamo DC. Killip (1938: 144) assigned that name to two Mexican specimens collected by C. A. Purpus “with much hesitation, for that [name] may be only a synonym of P. biflora [Lam.].” De Candolle had coined the name P. dictamo in 1828 (p. 324) for one of the “fl. mex. ic. ined.,” or one of the unpublished plates from the Royal Botanical Expedition to New Spain, 1787- 1803, sometimes referred to as the Sesse and Mocino Expedition. Reported as DC. Plate no. 28, a colored copy (McYaugh, 1980), the plate conserved at G-DC is actually an original painting and bears the Fl. Mex. leones number 362 in Mocino’ s hand, according to McYaugh (1980: 131). A virtually identical, but slightly more detailed, original painting is in the Tomer collection at the Hunt Institute for Botanical Documentation, Tomer coll. no. 0363, labeled as “Passiflora Normalis” and “362.” The DC. Plate no. 28 at G-DC was designated as lectotype of the name P. dictamo by McVaugh (2000: 427). Although only one painting is the lectotype, it is useful to consider evidence from both in this analysis because they are simultaneously executed (McYaugh, 1980), and the one at the Hunt Institute is slightly more detailed. The two paintings show a fruiting passiflora with bilobed leaves with obtuse to rounded, divergent tapering lobes, and no laminar nectaries. The fmits depicted show extended androgynophores, approach- ing the length of the peduncles. The lack of leaf glands is unusual in bilobed passionflowers and was noted by both de Candolle and Killip. At MA and F there are specimens from the expedition labeled “Passiflora normalis [L.],” Sesse & Mocino 3302 (F, fragm. seen; MA, photo seen), with a range of leaf shapes strikingly similar to the range of leaf shapes shown in the paintings described above. This specimen is identified and corresponds to Passiflora biflora Lam. The specimen shows small inconspicuous leaf glands typical of that species, but it is remarkable that the range of leaf shapes, especially the angle of the lateral veins with most lobes widely divergent but with some more narrow lobes, matches closely the same range seen in the paintings, and the leaves closely resemble those in the paintings. The specimen has no fmits, but the relatively long androgynophores seen in the paintings are typical of P. biflora. It appears, then, that the inconspicuous leaf glands were overlooked by the original artist, and that the name P. dictamo and its derivative 6 leca dictamo (DC.) M. Roem. (1846: 2: 146) are later synonyms of P. biflora. Planch., Novon 22: 434-441. Published on 18 October 2013. doi: 10.3417/2002149 436 Volume 22, Number 4 MacDougal 439 2013 Passiflora (Passifloraceae) from Eastern Mexico 4 mm, narrowly triangular to narrowly lanceolate. Peduncles 2 per node (rarely 1 per node on lowermost nodes or on young plants), usually richly branched to 3 or 4 orders (rarely unbranched or branched only 1 order at lowermost nodes or on young plants), (1- to) 7- to 20-flowered, 0.5-1.2(-1.4) cm to the first branch, other branches typically 0.3-1 cm, pedicels 0.6-2 cm; bracts (6-)8-17 X (2.5-)5-12 mm, widely elliptic or ovate to obovate, concave, the apex usually acute and abruptly long-cauduate with a 3-5 mm apiculus, margins entire or more usually irregularly shallowly few-cleft or long-toothed, especially near apex, light green, bract position anomalous by exaggerated displacement of each bract distally. Flowers 3-4 cm diam. (to 4.5 cm pressed), green- white or pale green-yellow, often with dull purple centrally, the corona nearly white to cream, the odor mild, fresh, greenish, slightly sweet, with a hint of 10 mm diam.; sepals 14-23 X 6.5-10 mm, narrowly ovate-triangular with no projection, light green-yellow abaxially, nearly white to pale green-yellow adaxially; petals 11-18 X 4-6 mm, narrowly ovate or narrowly ovate-oblong, nearly white to green-white to very pale green-yellow; coronal filaments in 2 series, the outer 8- 12 mm, ca. 0.6-0.7 mm diam. (fresh), filiform, slightly curved-falcate in distal third, short-attenuate at very apex, cream and unmarked or cream with dull violet or purple on the basal 2 mm, the inner 3-4.5 mm, capillary, apically subclavate, erect; operculum 2-3.5 mm, membranous, plicate, unmarked or with a hint of purple-red; limen edge long-papillate, the floor usually dull or pale purple-red; androgynophore 8-9.5 mm, nearly white except pale purple-red ca. 2 mm at very base, the free staminal filaments 4.5-5 mm, light green, the anthers 3-3.5 mm, the pollen yellow; ovary 2-3 X 1. 5-2.9 mm, widely ellipsoid to subglobose, densely white-hirsute, ca. 102 to 138 ovules per flower (N= 3), styles 6-7.5 mm, including stigmas. Fruit 13-20 mm diam., subglobose (oblate to widely ellipsoid), estipitate, blue-black, glaucous, lightly pubescent; arils clear-gelatinous, sweet, very 1.4-1. 6 mm, black, transversely sulcate with 6 or 7 sulci, the intervening ridges slightly verrucose, the chalazal beak inclined toward the raphe. Germination epigeal. Passiflora complanata typically has very short petioles and leaves that are broad, thick, and shallowly 2-lobed. The leaves resemble those of the only distantly related P. costaricensis Killip. The whole plant is densely white, short-pubescent, the stem is flattened, and the flowers are clustered at the nodes into sometimes massive balls of light green bracts, buds, and pale green-yellow or nearly white flowers. There can be 20 to 40 flowers clustered at a node, though only one to five of the ca. 4 cm flowers will open each day at any one node (greenhouse observations). Passiflora complanata is assigned to subgenus Decaloba supersect. Decaloba (DC.) Rchb. on the basis of its cemuous shoot tip, eglandular petioles, bilobed leaf, plicate operculum, purple berry, transversely sulcate seeds, and lack of chloroplast intron rpoCl (Hansen et al., 2006: 149). It has no laminar extrafloral nectaries, which is a synapomorphy of a recently recognized unnamed large clade in supersection Decaloba (Krosnick et al., 2013), and the new species is assigned there. Morphological characters of the leaf shape, lack of variegation, broad floral bracts, branched peduncles, and very small seeds place the new species in the P. sexflora group (MacDougal, 1989). This placement was confirmed on the basis of four genes by Krosnick et al. (2013) who used this new species in their analysis and placed it in the P. sexflora alliance as sister to P. rugosissima Killip. In the Passiflora sexflora alliance, P. complanata most resembles P. manantlanensis J. M. MacDougal in its leaf, the Chiapas form of P. rugosissima in ffe floral bracts and stem, and P. porphyretica Mast, in the size of its flower. Passiflora manantlanensis, from the mountains of southwestern Mexico, is very similar the leaves is the same, but its leaves are usually more deeply bilobed with acute apices, the bracts are very deeply 5- to 9-cleft, and the smaller flowers have an androgynophore of only 4.5-7 mm. The closely related P. rugosissima of Guatemala to Hidalgo, Mexico, is also similar to P. complanata in its thick leaves, several-branched inflorescences, wide bracts, and somewhat flattened stems. Passiflora rugosissima grows at higher elevations (1700-2600 m), however, and has much smaller flowers with an androgyno- phore of only 3.8-5 mm; the lobe angle of the leaves is wider; the length of the peduncle to the first branch is (1.0-)1.4-4.0 cm; and the floral bracts are usually smaller and often deeply lacerate. Passiflora rugo- sissima sometimes has a similar flattened stem, as discussed below. an androgynophore of 8-9.5 mm, is approached in this species group only by Passiflora porphyretica, which can have flowers almost as large, with an androgynophore length of 4.8-8 mm. That species also has large concave bracts, but they are uncleft and sometimes colored brown or olive and are usually glossy-glandular. It can immediately be distinguished from the new species by its unbranched peduncles, Two New Species of Dracaena (Asparagaceae, Nolinoideae) from Central Africa, with a Note on the Identity of D. braunii Geoffrey Mwachala East African Herbarium, National Museums of Kenya, P.0. Box 45166, 00100 Nairobi, Kenya. gmwachala@museums.or.ke Eberhard Fischer Institut fur Integrierte Naturwissenschaften, Biologie, Universitat Koblenz-Landau, Universitatsstr. 1, D-56070 Koblenz, Germany, efischer@uni-koblenz.de Abstract. Two new species of Dracaena Vand. ex L. (Asparagaceae, Nolinoideae) from Central Africa are here described and illustrated. Dracaena longipetio- lata Mwachala & Eb. Fisch. from Gabon is distin- guished from D. phrynioides Hook, by its long petioles and elliptic leaves. Dracaena litoralis Mwachala & Eb. Fisch. is described from Cameroon and also occurs in Equatorial Guinea, Gabon, and Congo-Brazzaville. This new taxon differs from D. braunii Engl, by its smaller flowers whose perianth lobes are shorter than the tube. The identity of D. braunii is discussed and its affinity with D. litoralis analyzed. Key words: Asparagaceae, Central Africa, Came- roon, Congo-Brazzaville, Dracaena, Equatorial Guin- ea, Gabon, IUCN Red List. Herb with subterranean rhizome; plants to 35 cm tall, petioles 37-85 cm, 2-3 mm wide, wingless. Leaf lamina elliptic, widest at middle, rounded at the base, acuminate at the apex, 25-30 X 8.5-10 cm, main parallel veins visible on adaxial surface, reticulate venation with secondary venation anasto- mosing between main parallel veins prominent on abaxial surface, midrib conspicuous abaxially, ending before leaf apex, discolorous, dark green on adaxial surface and rusty brown on abaxial surface. Rachis of inflorescence to 35 cm, 1.5-2 mm diam.; inflorescence paniculate, appearing spikelike, fruits on lateral glomerules (condensed lateral partial inflorescences), each glomerule with ca. 5 pedicels; pedicels not exceeding 1 mm. Flowers not known. Fruit an ellipsoid, 3-lobed berry, 15 X 8 mm. Dracaena Vand. ex L. belongs to the family Asparagaceae, subfamily Nolinoideae, according to the APG III classification (Stevens, 2008; APG III, 2009; Chase et al., 2009). This subfamily is represented in Central and East Africa by three genera, namely Dracaena, Eriospermum Jacq. ex Willd., and Sansevieria Thunb. (APG III, 2009). While preparing a revision of Dracaena for Flora of Tropical East Africa, several unidentified specimens came to light that represent new species, two of which have already been described (Mwachala et al., 2007; Mwachala & Cheek, 2012). Two further new taxa are described in the present paper. Dracaena longipetiolata Mwachala & Eb. Fisch. sp. nov. TYPE: Gabon. Crystal Mtns., 12 km SW of chutes de Kinguele [Kinguele Falls], 18 Feb. 1966, N. Halle & J. F. Villiers 5362 (holotype, P). Figure 1. Distribution and habitat. Dracaena longipetiolata is known only from the type collection from the Crystal Mountains in Gabon. The new species was found in riverine vegetation, in submontane rainforest. IUCN Red List category. Dracaena longipetiolata is assessed as DD or Data Deficient, according to IUCN (2001) criteria. More exploration in the Crystal Mountains is required to establish the range and habitat requirements for this species. This new taxo Dracaena aubryana Brongn. Mwachala (2005), based on usually not exceeding 1 m in height and its possession of a distinct petiole that is at least half as long as the lamina. It differs from D. phrynioides Hook, by its greatly reduced stems and the wingless petioles. These are between 37 and 85 cm in length, and over twice as long as the leaf lamina, which is elliptic. While also having very short stems, D. phrynioides has a leaf lamina much shorter and not exceeding 25 cm, about equaling the petiole, and is ovate-lanceolate in shape. Novon 22: 442-446. Published on 18 October 2013. doi: 10.3417/2009138 Volume 22, Number 4 2013 Mwachala & Fischer Dracaena from Central Africa 443 Despite the fact that flowers are presently unknown and the specimens bear only mature fruits, we describe this new species as it shows very distinctive features and can be easily recognized. Dracaena braunii Engl, in Baker and Engl., Bot. Jahrb. Syst. 15: 479, taf. XX (20). 1893 [1892]. TYPE: Cameroon: “Kamerun, Malimba,” 1889, /. Braun 329 (holotype, B). Figure 2. Volume 22, Number 4 2013 Mwachala & Fischer Dracaena from Central Africa 445 Dracaena Braunzi Eng l . 446 Novon New Species, Combinations, and Lectotypifications in Neotropical and Northern Mexican Frangula (Rhamnaceae) Amy Pool Missouri Botanical Garden, P.0. Box 299, St. Louis, Missouri 63139-0299, U.S.A. amy.pool@mobot.org Abstract. In preparing the family Rhamnaceae for an upcoming volume of Flora Mesoamericana, it became apparent, based on recent molecular work (Bolmgren & Oxelman, 2004; Bolmgren, pers. comm., 2013), as well as historical (Giubov, 1949; Kartesz & Gandhi, 1994), and personal morpholog- ical studies, that recognization of the genus Frangula study has been on Mesoamerican species and four species from Mesoamerica, F. circumscissa A. Pool from Costa Rica, F. darienensis A. Pool from Panama, F. grandiflora A. Pool described from Honduras and also found in Chiapas, Mexico, F. pendula A. Pool described from Costa Rica and also found in Panama, and one species from Oaxaca, Mexico, F. mconspicua A. Pool, are published here as new. Four Mesoamer- ican species and two varieties published in Rhamnus L. are transferred to Frangula as: F. breedlovei (L. A Johnst. & M. C. Johnst.) A. Pool [= R. breedlovei A. Johnst. & M. C. Johnst.], F. mcvaughii Johnst. & M. C. Johnst.) A. Pool [= R. mcvau A. Johnst. & M. C. Johnst.], F. oreodendron Williams) A. Pool [= R. oreodendron L. 0. Williams F. wendtii (Ishiki) A. Pool [= R. wendtii Ishiki], capreifolia var. grandifolia (M. C. Johnst. & L. A Johnst.) A. Pool [= R. capreifolia var. grandifolia IV C. Johnst. & L. A. Johnst.], and F. discolor var mesoamericana (L. A. Johnst. & M. C. Johnst.) A Pool [= R. sphaerosperma var. mesoamericana L. A Johnst. & M. C. Johnst.]. The following new combinations are made for five Mexican species north of the Flora Mesoamericana area: F. chimala pensis (R. Fernandez) A. Pool [= R. chimalapensis R Fernandez], F. hintonii (M. C. Johnst. & L. A Johnst.) A. Pool [= R. hintonii M. C. Johnst. & L. A Johnst.], F. longistyla (C. B. Wolf) A. Pool [= R longistyla C. B. Wolf], F. scopulorum (M. E. Jones) A. Pool [= R. serrata var. scopulorum M. E. Jones], and F. surotatensis (Gentry) A. Pool LE Gentry]. Four South American species are trans- ferred: F. acuminata (Maguire & Steyerm.) A. Pool [= R. acuminata Maguire & Steyerm.], F. chiman- tensis (Steyerm. & Maguire) A. Pool [= R. chiman- tensis Steyerm. & Maguire], F. marahuacensis (Steyerm. & Maguire) A. Pool [= R. marahuacensis Steyerm. & Maguire], F. neblinensis (Maguire & Steyerm.) A. Pool [= R. neblinensis Maguire & Steyerm.]. Rhamnus sphaerosperma var. longipes M. C. Johnst. & L. A. Johnst. (from Hispaniola and Puerto Rico) is transferred to Frangula and raised to the rank of species, necessitating the publication of the new name F. longipedicellata A. Pool. Lectotypes are designated for F. polymorpha Reissek, R. citrifolia Rusby, and R. riojae Perkins. Key words: Frangula, IUCN Red List, Mesoamer- ica, Mexico, Rhamnaceae, Rhamnus, South America, West Indies. The genus Frangula Mill. (Rham of approximately 50 species and is native to North America, the Neotropics, Europe, and Asia. Grubov (1949) published a worldwide revision of Rhamnus L. s.l. and provided names in Frangula for those species known to him. His generic recognition of Frangula was accepted in Europe for the Flora Europaea (Tutin, 1968) and for taxonomic treatments in North America (Kartesz & Gandhi, 1994; G. Nesom, Flora of North America, in prep.). However, most modem Neotropical works (e.g., Standley & Steyermark, 1949; Nowicke, 1971; Johnston & Johnston, 1978; Fernandez N., 1986; Johnston, 2001; Steyermark & Berry, 2004) have continued to treat the species of Frangula in Rhamnus. Johnston and Johnston (1978), who recognized Frangula at the subgenus characters used by Grubov (1949 [Russian]) to distinguish Frangula from Rhamnus, and Kartesz and Gandhi (1994) provided a similar table of distinguishing characteristics, which is slightly revised here (Table 1). The most convincing characteristics for recognizing Frangula include the bud scales absent, the pyrenes (Fig. 1A) indehiscent, i a thickened cartilaginous basal rostrum that is exserted through the base of the pyrene. In contrast, in Rhamnus, the bud scales are present, the pyrenes (Fig. IB) ventrally dehiscent, and the seeds (Fig. 1C) grooved, without a basal thickening and completely enclosed within the doi: 10.3417/2013009 Novon 22: 447-467. Published on 18 October 2013. Volume 22, Number ■ 2013 Pool Frangula (Rhamnaceae) 449 452 Novon cruce de Rio Humo al Tunel, 9°42'10"N 83°47'20"W, 1700 m, 8 Mar. 2000, L. Acosta 454 Novon trichomes. Leaves with blades 5-10 X 2-3.5 cm, Volume 22, Number ■ 2013 (Rhamnaceae) 457 Pool Frangula (Rhamnaceae) 459 (2.8-B.5 1.75-2.2 mm), and sepals spreading to recurved (vs. 'fili 1 462 Novon s 6-15 X 2.4^6.3 cm, e Nomenclatural Transfers in the Pantropical Genus Myrsine (Myrsinaceae) Jon M. Ricketson Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166-0299, U.S.A. jon.ricketson@mobot.org John J. Pipoly III University of Florida-IF AS/Broward County Extension, Parks & Recreation Division, 3245 College Avenue, Davie, Florida 33314-7719, U.S.A. jpipoly@broward.org Abstract. The names of 14 taxa of Rapanea Aubl. are transferred to Myrsine L. (Myrsinaceae). The 14 new combinations include: M. acutiloba (Mez) Ricketson & Pipoly [= R. acutiloba Mez]; M. amischocarpa (A. C. Sm.) Ricketson & Pipoly [= R. amischocarpa A. C. Sm.]; M. boivinii (Mez) Ricketson & Pipoly [= R. boivinii Mez]; M. comorensis (Mez) Ricketson & Pipoly [= R. comor- ensis Mez]; M. courboniana (Mez) Ricketson & Pipoly [= R. courboniana Mez]; M. crassiramea (A. C. Sm.) Ricketson & Pipoly [= R. crassiramea A. C. Sm.]; M. daphnoides (Mez) Ricketson & Pipoly [= R. daphnoides Mez]; M. forbesii (Mez) Ricketson & Pipoly [= R. forbesii Mez]; M. grijfithiana (Mez) Ricketson & Pipoly [= R. grijfithiana Mez]; M. hadrocarpa (A. C. Sm.) Ricketson & Pipoly [= R. hadrocarpa A. C. Sm.]; M. longipes (A. C. Sm.) Ricketson & Pipoly [= R. longipes A. C. Sm.]; M. polyantha (A. C. Sm.) Ricketson & Pipoly [= R polyantha A. C. Sm.; M. seychellarum (Mez) Rick- etson & Pipoly [= R. seychellarum Mez]; and M. striata (Mez) Ricketson & Pipoly [= R. striata Mez]. Six binomials are lectotypified: M. boivinii (Mez) Ricketson & Pipoly; M. comorensis (Mez) Ricketson & Pipoly; M. forbesii (Mez) Ricketson & Pipoly; M. grijfithiana (Mez) Ricketson & Pipoly; M. seychella- rum (Mez) Ricketson & Pipoly; M. striata (Mez) Ricketson & Pipoly. Resumen. Se transfiere 14 especies pertenecientes al genero Rapanea al genero Myrsine. Las trans- ferencias estan enumeradas en la lista a continu- acion: M. acutiloba (Mez) Ricketson & Pipoly [= R. acutiloba Mez]; M. amischocarpa (A. C. Sm.) Ricketson & Pipoly [= R. amischocarpa A. C. Sm.]; M. boivinii (Mez) Ricketson & Pipoly [= R. boivinii Mez]; M. comorensis (Mez) Ricketson & Pipoly [= R. comorensis Mez]; M. courboniana (Mez) Ricketson & Pipoly [= R. courboniana Mez]; M. crassiramea (A. C. Sm.) Ricketson & Pipoly [= R. crassiramea A. C. Sm.]; M. daphnoides (Mez) Ricketson & Pipoly |— R, daphnoides Mez]; M. forbesii (Mez) Ricketson & Pipoly (= R- forbesii Mez]; M. grijfithiana (Mez) Ricketson & Pipoly [= R. grijfithiana Mez]; M. hadrocarpa (A.C. Sm.) Ricketson & Pipoly [= R. hadrocarpa A. C. Sm.]; M. longipes (A. C. Sm.) Ricketson & Pipoly [= R. longipes A. C. Sm.]; M. polyantha (A. C. Sm.) Ricketson & Pipoly [= R. polyantha A. C. Sm.; M. seychellarum (Mez) Rick- etson & Pipoly [= R. seychellarum Mez]; y M. striata (Mez) Ricketson & Pipoly [= R. striata Mez]. Se lectotipifican seis binomios: M. boivinii (Mez) Rick- etson & Pipoly; M. comorensis (Mez) Ricketson & Pipoly; M. forbesii (Mez) Ricketson & Pipoly; M. grijfithiana (Mez) Ricketson & Pipoly; M. seychella- rum (Mez) Ricketson & Pipoly; M. striata (Mez) Ricketson & Pipoly. Key words: Myrsinaceae, Myrsine, Philippines, Rapanea. The systematic position and circumscription of the genus Myrsine L. (Myrsinaceae) have been discussed in a number of taxonomic works over the last two decades. As summarized regarding transfers into Myrsine, Ricketson and Pipoly (2010: 628) stated: [t]he relationship among Myrsine L. and its satellites (inter alia Rapanea Aubl. and Suttonia A. Rich.) has been a historical focal point for treatment of many species as Rapanea by regional floristicians, while monographers have determined that the variation in androecial structure among staminate flowers, and the gynoecial for the pistillate support the concept in the more broadly defined, monophyletic concept of Myrsine. Regional monographers recognizing the more inclusive concept include Hosaka (1940), Backer & Bakhuizen van den Brink (1965), Smith (1973), Fosberg & Sachet (1975, 1980), Sachet (1975), Li (1978), Wagner et al. (1990), Pipoly & Chen (1995), Pipoly (1996, 2007), Jackes (2005), Takeuchi and Novon 22: 468-472. Published on 18 October 2013. doi: 10.3417/2012082 Volume 22, Number 4 2013 Ricketson & Pipoly Nomenclatural Transfers in Myrsine (Myrsinaceae) 469 Pipoly (2009), and Ricketson & Pipoly (1998, 2010). Among all of these papers, the most detailed discussion is provided by Pipoly (1996) in treating Myrsine of the Philippines. After an extensive survey, including detailed examination for each of the remaining entities associated with the genus Rapanea Aubl., we have determined that the following names are still orphaned and in need of being transferred into Myrsine. Any remaining names in Rapanea without a corresponding name in Myrsine are considered taxonomic synonyms and are not in need of transfer. 1. Myrsine acutiloba (Mez) Ricketson & Pipoly, comb. nov. Basionym: Rapanea acutiloba Mez, Pflanzenr. (Engler) IV. 236(Heft 9): 390. 1902. TYPE: Colombia. Magdalena: Santa Marta Mpio., Sierra Nevada de Santa Marta, Tomar- azon bei Rio Hadea, s.d. (fl.), H. Karsten s.n. (holotype, W-0030344, F neg. 31984 [W image]; isotype, F [W fragm.]). Currently, Myrsine acutiloba is known only from the type collection from the Sierra Nevada de Santa Marta in Colombia. Images of the W holotype and the F isotype are available from the respective institu- 2. Myrsine amischocarpa (A. C. Sm.) Ricketson & Pipoly, comb. nov. Basionym: Rapanea ami- schocarpa A. C. Sm., J. Arnold Arbor. 54(2): 280-281. 1973. TYPE: Vanuatu. Tafea: Anei- tyum Island, Anelgauhat Bay, 60 m, 12 Feb. 1929 (fr.), 5. Kajewski 751 (holotype, A; isotypes, BISH, K, NY, P, US). A tree 12-25 m tall in the rainforests and mountain slopes of the Vanikoro Islands (Solomon Islands) and from the islands of Erromango, Ambrym, and Aneityum (Vanuatu). Images of all the listed isotypes are available at JSTOR Plant Science. 3. Myrsine boivinii (Mez) Ricketson & Pipoly, comb, nov. Basionym: Rapanea boivinii Mez, Pflan- zenr. (Engler) IV. 236(Heft 9): 376. 1902. TYPE: Comoros. Mahore: lisiere des bois au pied de Chongui au dessus de Dappani, July 1848 (fr.), L Boivin 3268 (lectotype, designated here, G-DEL [barcode] G00359617; isolecto- types, P [2, barcodes P00184211, P00184212]). Mez’s (1902: 376) original description of Rapanea boivinii, “Insel Mayotte ( Boivin n. 3268. — Herb. Deless., Wien)” indicates the Boivin 3268 collection as the type, but the collection was listed from two institutions, thus necessitating the need to designate a lectotype. Unfortunately, Dr. Bruno Wallnofer, Curator of the Vascular Plant Collections at the Naturhistorisches Museum Wien (W), was unable to locate this material in the collection (Wallnofer, 2012, pers. comm.). However, the type collection is still represented at G [G00359617] and we elect to designate this collection as the lectotype. An image of the lectotype is available from the G online database, and in addition, images of the two P isolectotypes (P00184211, P00184212) are available from the P web site. Finally, there is an L. Boivin sheet available at W identified as R. boivinii; unfortunately, it is without a number and is therefore eliminated from the selection of the lectotype. 4. Myrsine comorensis (Mez) Ricketson & Pipoly, comb. nov. Basionym: Rapanea comorensis Mez, Pflanzenr. (Engler) IV. 236(Heft 9): 375-376. 1902. TYPE: Comoros. Ngazidja: [Grande Comore], sur le volcan (Vouni), May 1850 (fr.), L. Boivin s.n. (lectotype, designated here, W; isolectotypes, BM [BM000925363], P [2, bar- codes: P00184189, P00184190]). Mez (1902: 376) designated a number of collec- tions from Berlin and Wien as syntypes for his new species Rapanea comorensis: “Comoren: Insel Johan- na auf Bergen 1500 m. ii. (Hildebrandt n. 1692 [Bf]), GrofS Comoro, auf dem Vulkan 1000-2000 m. ii. (Kersten [s.n.. Exped. v. d. Decken, Bf]), ohne Standortsangabe (Boivin, Humblot n. 326, 1534). - Herb. Berlin, Wien.” Extant syntypes from the last two collections cited by Mez were found: Comoros, Mahore [Mayotte], Foret de Combani, 18 Oct. 1884 (fl.), L. Humblot 326 [in sched., 1326\ (BM [barcode] BM000925362, K [barcode] K000226049, P [3, barcodes P00184192, P00184193, P00184194]); and Comoros, Ngazidja [Grande Comore], sur le volcan Anjouan, 1886 (fl., fr.), L. Humblot 534 [in sched., 1534 ] (LE, P [2, barcodes P00184195, P00184196], W). Thus, a lectotype needs to be selected from this material. Any material located at the Berlin Herbarium was lost in 1943 (Hiepko, 1987); it is also impossible at this time to determine which of these collections were lost at B. According to Dr. Bruno Wallnofer, the only collections remaining in the herbarium are the L. Humblot 534 (1534) and L. Boivin s.n. collections (Wallnofer, 2012, pers. comm.). The Boivin collection is annotated by Mez, but the Humblot collections do not possess such annotation; thus, we elect to designate the L. Boivin s.n. at W as the lectotype. Images of the two P [P00184189, P00184190] isolectotypes are available at the P web site. 470 Novon 5. Myrsine courboniana (Mez) Ricketson & Pipoly, comb. nov. Basionym: Rapanea courboniana Mez, Pflanzenr. (Engler) IY. 236(Heft 9): 395. 1902. TYPE: Uruguay. Montevideo: St. Gabriel, 10 May 1854 (fl.), A. Courbon s.n. (holotype, P [barcode] P00649879). Myrsine courboniana is known only from the Montevideo area of Uruguay. It is interesting to note that there are two A. Courbon collections of M. courboniana at Paris (P), both marked as types. However, Mez (1902: 395) clearly designates the material collected in May as the type when he states “Montevideo: bei St. Gabriel (Courbon). - Bliiht im Mai (Herb. Paris).” We can only assume that A. Courbon must have discovered the species in fruit in November 1853 and returned to the same area the next May and recollected the type material, unfortu- nately just after anthesis. 6. Myrsine crassiramea (A. C. Sm.) Ricketson & Pipoly, comb. nov. Basionym: Rapanea crassir- amea A. C. Sm., Bull. Torrey Bot. Club 70(5): 547. 1943. TYPE: Fiji Islands. Mba: Viti Levu, Tholo North, Nandarivatu, 900 m, 16 Dec. 1927 (fr.), J. Gillespie 4374 (holotype, BISH [barcode] BISH1003567). Myrsine crassiramea is currently known only from the type collection in the Fiji Islands. An image of the type is available at JSTOR Plant Science. Myrsine daphnoides (Mez) Ricketson & Pipoly, comb. nov. Basionym: Rapanea daphnoides Mez, Pflanzenr. (Engler) IV. 236(Heft 9): 357. 1902. TYPE: India. Near Courtallum, Feb. (fl.), R. Wight 1 779 (types, G-BOISS not located, W not located). According to Mez’s (1902: 357) original descrip- tion of Rapanea daphnoides, the Wight 1779 collection was mixed, “Vorderindien: bei Courtallum (Wight n. 1779 e. p.). — Bliiht im Februar (Herb. Boiss.-Barbey, Wien).” Unfortunately, the staff at the Conservatoire et Jardin Botaniques de la Ville de Geneve (G) were unable to locate the Boiss.-Barbey type (L. Loze, 2012, pers. comm.). In addition, Dr. Bruno Wallnofer was also unable to locate this material (Wallnofer, 2012, pers. comm.). At this time we elect to defer our selection of the lectotype until we are able to locate additional duplicates. 8. Myrsine forbesii (Mez) Ricketson & Pipoly, comb, nov. Basionym: Rapanea forbesii Mez, Pflanzenr. (Engler) IY. 236(Heft 9): 362. 1902. TYPE: Indonesia-Sumatra. Mt. Tengamoes, Lampongs, 6900-7100 ft. [2103-2164 m], 1881-1882 (fr.), H. Forbes 1870 (lectotype, designated here, L [barcode] L0006815; isolectotypes, A [fragm.], GH, LE). In Mez’s (1902: 362) original description of Rapanea forbesii, he cited two Forbes collections, both from the Berlin Herbarium “Sumatra: ohne Standorlsangabe ( Forbes n. 1870, 1882. - Herb. Berlin).” Unfortunately both of the Berlin collections were lost in 1943 (Hiepko, 1987). Several duplicates of the Forbes 1870 and 1882 collections are known from the same locality as cited, at A [fragm.], GH, L, and LE, and are available to choose from for the lectotype. Unfortunately, none of the material from the Harvard University Herbaria (A and GH) has been imaged. It is interesting to note that the fragment packets from Arnold Arboretum Herbarium (A [barcode] 00025638) are both mounted together, likely fragments from Gray Herbarium sheets (GH [barcodes] 00025639; 00025640). Although we have not seen the LE material, Imkhanitskaya (1996: 40) indicates that duplicates of both syntypes are available at LE. Fortunately, both the types at L (Forbes 1882 L [barcode] L0006814, and Forbes 1870 L [barcode] L0006815) have been imaged and are available online. Thus, we designate one of these specimens, Forbes 1870 (L [barcode] L0006815), as the lectotype. 9. Myrsine griffithiana (Mez) Ricketson & Pipoly, comb. nov. Basionym: Rapanea griffithiana Mez, Pflanzenr. (Engler) IV. 236(Heft 9): 360- 361. 1902. TYPE: Myanmar. Malacca, s.d. (fl., fr.), W. Griffith 3543 [A] (lectotype, designated here, P [barcode] P00466721; isolectotypes, BR, GH, LE). In Mez’s (1902: 360-361) original description of Rapanea griffithiana, he cited a mixed Griffith collection represented at several institutions: “Birina: ohne Standorlsangabe (Griffith n. 3543 (e. p.). - Herb. Berlin, Brussel, A. Gray, Paris).” The Griffith 3543 collection is mixed from what we here designate the above type material pro parte, as “[A]” and the remaining material, which we have identified as Myrsine porteriana A. DC as Griffith 3543 [ B ]. From the syntypes identified by Mez, the Berlin (B) material was lost in 1943 (Hiepko, 1987). Images are available of the Paris (P [barcode] P00466721) sheet and the Brussels (BR [barcode] BR0000005232284) sheet. Unfortunately, only data, and no image, are available from the Harvard University Herbaria website. Thus, we here elect to designate the P material as lectotype as clearly representative of this species. Volume 22, Number 4 Ricketson & Pipoly 471 2013 Nomenclatural Transfers in Myrsine (Myrsinaceae) 10. Myrsine hadrocarpa (A. C. Sm.) Ricketson & Pipoly, comb. nov. Basionym: Rapanea hadro- carpa A. C. Sm., J. Arnold Arbor. 54(2): 290-291. 1973. TYPE: Fiji Islands. Yiti Levu: Kambara, 2- 7 Mar. 1934 (fr.), A. Smith 1245 (holotype, BISH; isotypes, GH, K, NY, S, UC, US). Myrsine hadrocarpa is a small tree to 8 m tall. It is island forests of the Fiji Islands. 11. Myrsine longipes (A. C. Sm.) Ricketson & Pipoly, comb. nov. Basionym: Rapanea longipes A. C. Sm., J. Arnold Arbor. 54(2): 287-288. 1973. TYPE: Samoa. Savai‘i: forest above Letui, Tuisivi Range, 1600-1700 m, 26 Sep. 1929 (fr), E. Christophersen 779 (holotype, BISH; isotypes, K, NY [2], P, UC, US). 12. Myrsine polyantha (A. C. Sm.) Ricketson & Pipoly, comb. nov. Basionym: Rapanea polyan- tha A. C. Sm., J. Arnold Arbor. 54(2): 289-290, figs. 84-85. 1973. TYPE: Fiji Islands. Viti Levu: Mba prov., summit ridge of Mt. Tomanivi, Victoria, 4 May 1962, R. Melville, E. Melville & J. Parham 7059 (holotype, K; isotypes, A, BISH, SUVA). 13. Myrsine seychellarum (Mez) Ricketson & Pipoly, comb. nov. Basionym: Rapanea seychel- larum Mez, Pflanzenr. (Engler) IV. 236(Heft 9): 375. 1902. TYPE: Seychelles. Common on mtn. top in Mahe & Silhouette, 1874 (fr.), J. Home 442 (lectotype, designated here, K [barcode] K000226047). In Mez’s (1902: 375) original description of Rapanea seychellarum, he cited two J. Horne collections “Seychellen-Inseln Maho und Pelonette: haufig auf den Gipfeln der Berge (/. Hooker \Home\ n. 442, 588. - Herb. Kew).” Both collections are still present at K, in fact both are mounted on the same sheet. Home 588 (K [barcode] K000226048) is sterile material and Horne 442 (K [barcode] K000226047) is fertile. We therefore designate the Home 442 specimen at K as the lectotype. Images of both the K lectotype and syntype are available online. 14. Myrsine striata (Mez) Ricketson & Pipoly, comb. nov. Basionym: Rapanea striata Mez, Pflanzenr. (Engler) IV. 236(Heft 9): 357. 1902. TYPE: India. Malabar, Concan, s.d. (fl.), J. Stocks & J. Law s.n. (lectotype, designated here, P [barcode] P00467028; isolectotype, GOET [barcode] GOET008184). In Mez’s (1902: 357) original description of Rapanea striata, he cited a single collection by J. Stocks and J. Law “Vorderindien: Malabarkiiste ( Stocks u. Law)" but failed to designate where the type was located. Of the two syntype duplicates known from P and GOET, the P sheet is by far the better material, and we here designate the material at P as the lectotype (image available online). An image is available of the GOET material at JSTOR Plant Science. It should be mentioned that a third sheet, “India. Karnataka: Mysore District, Bababoodim hills, s.d. (fl), J. Law s.n. (K [barcode] K0007 56358 — the image is mistakenly labeled as uLaw 191" and the “191” is not a collection number but an incomplete year from an annotation label), was rejected as a possible syntype. The sheet does not pose an annotation by Mez and this material is from a different region than the type. Acknowledgments. We gratefully wish to thank Dr. Bruno Wallnofer, Curator of the Vascular Plant Collections at the Naturhistorisches Museum Wien (W), Yvette “Tivvy” Harvey of the Royal Botanical Gardens, Kew (K), and Dr. Laurent Gautier, Head Curator - Phanerogams and the staff at the Conservatoire et Jardin Botaniques de la Ville de Geneve (G) for their assistance in locating much of this type material. This project was aided greatly by the availability of herbarium images and data online. We deeply appreciate the ability to have public A, BISH, BM, BR, F, G, GH, GOET, K, L, NY, P, S, UC, US, W. The Virtual Herbaria Austria project is available at . We also gratefully acknowledge the uses of the JSTOR Plant Science web page at . Finally, we appreciate the availability of the Biodiversity Heritage Library, an outstanding project which is “a consortium of natural history and botanical libraries that cooperate to digitize and make accessible the legacy literature of biodiversity held in their collections and to make that literature available for open access and responsible use as a part of a global biodiversity commons” at < http ://www. biodiversitylibrary.org > . Backer, C. A. & R. C. Bakhuizen van den Brink Jr. 1965. Myrsinaceae. Pp. 194-203 in Flora of Java (Spermato- phytes only), Vol. 2. P. Noordhoff, Groningen. 472 A New Species of Pavetta (Rubiaceae) from South Africa Stefan J. Siebert A. P. Goossens Herbarium, School of Environmental Sciences and Development, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa. Corresponding author: stefan.siebert@nwu.ac.za Elizabeth Retief National Herbarium, South African National Biodiversity Institute, Private Bag X101, Pretoria 0001, South Africa. E.Retief@sanbi.org.za Abraham E. van Wyk H. G. W. J. Schweickerdt Herbarium, Department of Plant Science, University of Pretoria, Pretoria 0002, South Africa. braam.vanwyk@up.ac.za Abstract. Pavetta glaucophylla Retief, S. J. & A. E. van Wyk, a new species of the Rubiaceae (Ixoroideae) from South Africa, is described and illustrated. A member of Pavetta L. subg. Pavetta, it has a restricted range and is near endemic to the Sekhukhuneland Centre of Endemism. Hitherto P. glaucophylla has mainly been confused with P. zeyheri Sond., from which it differs in being a taller plant with larger, blue-green leaves in living condition instead of dark green to gray-green leaves and flowers in which the calyx lobes are 0.5-0.75 mm long and triangular-ovate, and 1.25-2 mm long and Key words: IUCN Red List, Pavetta, Rubiaceae, Sekhukhuneland, South Africa. The rubiaceous genus Pavetta L. belongs to subfamily Ixoroideae, tribe Pavetteae, and comprises ca. 400 species from the Old World tropics, excluding Madagascar (Retief & Leistner, 2000; Bridson, 2003). Sub-Saharan Africa is a center of diversity for the genus, with 21 described species known from South Africa, all restricted to the summer rainfall area (Germishuizen et al., 2006). Although not as diverse as it is further north in Africa, Pavetta is the largest of the 53 genera of Rubiaceae native to South Africa (Germishuizen et al., 2006). Members of Pavetta are trees, shrubs, or dwarf shrubs (geoxylic suffrutices), and the majority are, among others, readily distinguished from most other woody Rubia- ceae by the presence of symbiotic bacterial nodules in the leaf blades (Bridson, 2003). The genus is further characterized by stipules partly to completely connate, mostly long-acuminate to awned; terminal, umbellate inflorescences with numerous fragrant, tetramerous flowers with secondary pollen presenta- ially white, contorted corolla lobes, s the corolla (Retief , 2000). The calyx limbs of Pavetta have four distinct lobes, and the fruit are fleshy, 2-seeded, and indehiscent (Retief & Leistner, 2000). Even though general floral morphology is very similar in Pavetta, the calyx is quite variable and provides some of the best criteria for the infrageneric classification of the group (Bremekamp, 1929; Bridson, 2003). In southern Africa a number of Pavetta species with restricted range are known from recognized centers of floristic endemism (van Wyk & Smith, 2001), for example, P. tshikondeni N. Hahn from the Soutpansberg Centre of Endemism (Hahn, 1999) and P. barbertonensis Bremek. from the Barberton Centre of Endemism (Bremekamp, 1929). In this paper we describe a new species of Pavetta near endemic to the Sekhukhuneland Centre of Endemism in northeastern South Africa. In the past, this species often has been included under P. zeyheri Sond., but Siebert et al. (2001) and von Breitenbach et al. (2001) first recognized it as an undescribed species in the literature. Subsequently, Schmidt et al. (2002: 642) referred to it as a “large-leaved form” of P. zeyheri that “may prove after further research to be a distinct species.” Coates Palgrave (2002) also mentions the Sekhukhuneland taxon under P. zeyheri but specu- lates that it is probably conspecific with P. micro- specimen collected at Komatipoort in easternmost Mpumalanga (Bridson, 2003), outside the range of the new species. A possible link between the new species and P. microlancea is refuted by Herman (2005), based on the generally smaller habit, 0.3-1 m, of the latter species, and on P. microlancea having fewer- flowered (four t doi: 10.3417/2011025 Novon 22: 473-477. Published on 18 October 2013. 474 Novon also indicated by Bridson (2003). Herman (2005) suggests that our new species, already collected as early as 1939 ( Mogg 725, PRE), has been dumped into a taxonomic “waste bin,” namely P. zeyheri. Pavetta glaueophylla Retief, S. J. Siebert & A. E. van Wyk, sp. nov. TYPE: South Africa. Mpumalanga: Sekhukhuneland, Thorn cl iffe Mine, entrance rd., 1 Dec. 1997 (fl., fr.), J. E. Burrows 6198 (holotype, PRE; isotype, BNRH). Figure 1. altioreque, foliis majoribus conspicue glaucis (46-) (non triangulari-lanceolatos) diviso differt. Erect, deciduous shrub or small tree, (1.5-) 2.5(— 3.5) m tall; young twigs glabrous; branches fascicled, sessile, glabrous; lamina narrowly elliptic to oblanceolate, (49-)72(— 93) X (8-)13(— 21) mm, coriaceous, blue-green in living condition, apex acute, base decurrent, margin entire; midrib on lower surface pale and prominent; domatia absent; bacterial nodules dotlike or elliptical, unevenly scattered in lamina; stipules interpetiolar, limb triangular. Inflo- rescences sub-umbellate, branched, (32-)37(— 44) mm wide; primary inflorescence branches 4-6 mm; flowers (14-)17(— 21) clustered on short side branch- es 10-30 mm long; peduncles 6-8 mm; pedicels 3.5- 4.5 mm; bracts up to 2 mm, densely white silky-hairy to almost glabrous inside; bracteoles present but inconspicuous. Flowers (14-)17(— 21) per inflores- cence; calyx 4-lobed, lobes short, triangular-ovate, 0.5-0.75 mm, glabrous, persistent in fruit, never reflexing; corolla 4-lobed, white, tube cylindrical, (10-)13(— 15) mm, lobes (15-)19(-23) mm; style slender, distally slightly thickened into an elongated club-shaped pollen presenter, (21-)26(— 32) mm; stigmatic surface confined to bidentate apex; stamens 4, arising in corolla mouth; filaments very short; anthers exserted, dorsifixed near the base, linear to oblong, 5-6 mm. Fruit a drupe, spherical, slightly 2- lobed, (5-)6(— 8) X (5-)7(— 8) mm, black when ripe, with 1 or 2 pyrenes. Distribution and habitat. Pavetta glaueophylla has been collected in the provinces of Mpumalanga and Limpopo in South Africa (Fig. 2) on calcrete, dolerite, dolomite, gabbro norite, magnetite, and, rarely, shale. It is common on stony, rocky soil in open mountain savanna (bushveld), rarely as part of copses of woody vegetation in grassland, and has been recorded from 500 to 1200 m. The species is confined to that part of the Central Bushveld Bioregion associated with the northeastern Great Escarpment of South Africa, specifically the following vegetation units: Sekhukhune Mountain Bushveld, Sekhukhune Plains Bushveld, and Ohrigstad Moun- tain Bushveld (Mucina & Rutherford, 2006). Biogeo- graphically, it is near endemic to the Sekhukhune- land Centre of Endemism (mainly on ultramafic substrates), where it ranges, usually on dolomite, into the bordering Wolkberg Centre (van Wyk & Smith, 2001). IUCN Red List category. Pavetta glaueophylla is assessed as of Least Concern (LC) according to IUCN Red List criteria (2001) due Lo the extent of occurrence, area of occupancy, and the existence of more than 30 known populations (Siebert et al., 2002). Pavetta glaueophylla is, however, increasingly subjected to habitat degradation as a result of extensive mining activities and urban expansion in the Sekhukhuneland region. Phenology. Flowers of Pavetta glaueophylla were collected in early summer (November and December) and fruits in summer and autumn (January to May). Etymology. The epithet glaueophylla derives from the Greek and refers to the conspicuous blue- green foliage leaves of the species. Vernacular names. The local Pedi use the leaves of Pavetta glaueophylla as a vegetable, and in Northern Sotho (Sepedi) the plant is called mabudel- alelane, with budela meaning “sour,” a reference to the watercress-like taste of the leaves (from Barnard 1446, PRE). English and Afrikaans vernacular names for the species are Sekhukhune bride’s bush and sekhukhunebruidsbos, respectively (van Wyk et al., 2011). Discussion. Older herbarium collections of Pa- vetta glaueophylla are mainly identified as P. zeyheri, and, more particularly, the infraspecific entity is sometimes recognized as P. zeyheri subsp. zeyheri (Herman, 2005). The two species are superficially rather similar owing to resemblances in gross morphology and geographical distribution. Differenc- es between P. glaueophylla and P. zeyheri include the distinctly blue-green and larger, (49-)72(— 93) X (8-)13(— 21) mm, leaves of the former, compared to the dull dark-green to gray-green and smaller, (24-)38(— 50) X (7-)8(— 11) mm, leaves of the latter. In P. glaueophylla the calyx limb-tube is divided at midpoint into relatively short (0.5-0.75 mm), trian- Volume 22, Number 4 2013 Siebert et al. Pavetta (Rubiaceae) from South Africa 475 gular-ovate lobes, but in P. zeyheri the calyx limb- Bridson, 2003). Pavetta glaucophylla is also a more tube is shorter than the lobes, with the lobes robust and larger plant, (1.5-)2.5(— 3.5) m high, with relatively long (1.25-2 mm) and triangular-lanceolate P. zeyheri generally slender and smaller, (l-)2(— 3) m (for calyx limb types in southern African Pavetta, see high. 476 Novon 477 A New Species of Muellera (Millettieae, Leguminosae) from Brazil Volume 22, Number 4 2013 da Silva 479 Muellera (Millettieae, Leguminosae) from Brazil obtuse to rounded, or, less commonly, retuse to emarginate, membranaceous to papyraceous, discol- orous, adaxial surface bright green, glabrous, abaxial surface pale green, strigulose; venation brochido- dromous, midvein and secondary veins impressed on both surfaces, the secondary veins forming angles of 35° to 45° with the midvein. Pseudoracemes axillary, solitary with the flowers in pairs along the rachis, congested; peduncle 2-8 mm, rachis 2-4.4 cm, both angulate, glabrous or shortly tomentose, trichomes Euphorbia rayturneri (Euphorbiaceae), a New Species from Southwestern New Mexico, United States Victor W. Steinmann Institute de Ecologfa, A.C., Centro Regional del Bajfo, Apartado postal 386, 61600 Patzcuaro, Michoacan, Mexico. victor.steinmann@inecol.edu.mx Eugene Jercinovic 6285 Algodon Road SW, Deming, New Mexico 88030, U,S.A, gjercinovic@earthlink.net Abstract. Euphorbia rayturneri V. W. Steinm. & Jercinovic (Euphorbiaceae), a new species endemic to southwestern New Mexico, United States, is described and illustrated. This summer annual occurs in desert grassland and is known from three collections made in Grant, Hidalgo, and Luna counties at elevations of 1400-1700 m. Given its limited distribution and small population size, we Critically Endangered, according to IUCN Red List criteria. The new species belongs to Euphorbia L. sect. Anisophyllum Roep. on the basis of possessing interpetiolar stipules and opposite leaves that are a pubescent, prostrate herb with serrulate leaves that often have a red blotch in the center and an ovary with undivided styles. Key words : Euphorbia, Euphorbiaceae, IUCN Red List, New Mexico, United States. Despite the flora of the southwest United States having been relatively well studied, our basic knowledge of the plants in the region is still far from complete, and new species continue 1 tered and described at a constant rate (Ertter, 2000). New Mexico ranks fourth among the 48 continental United States in terms of plant diversity (Legler, 2010). In the past decade at least 10 new species of vascular plants have been described from the state, and 350 new records have been reported (Kelly Allred, pers. comm., 2010). With at least 43 native species, the genus Euphorbia L. (Euphorbiaceae) is one of the most speciose genera in New Mexico (Allred, 2012). It is therefore not surprising that recent field exploration and the study of existing collections have resulted in the discovery of a distinctive annual species of Euphorbia that does This new species is described here, and information is provided about its habitat, distribution, conserva- tion status, and taxonomic affinities. Euphorbia rayturneri V. W. Steinm. & Jercinovic, sp. nov. TYPE: U.S.A. New Mexico: Grant Co., Pitchfork Ranch, 32°25.072'N, 108°21.169'W, 1605 m, 23 Sep. 2008, E. Jercinovic 827 (holotype, NMC; isotypes, IEB, KSC, MICH, MO, UNM). Figures 1, 2. Herb annual, prostrate, mat-forming with various primary stems extending radially from a slender taproot. Stems to 8 cm, terete, strigulose with white, recurved hairs 0. 1-0.2 mm, pubescence distributed intemodes 0.6-2. 2 cm. Stipules interpetiolar, sepa- rate, filiform, 0.6-0.9 mm, pilose or strigulose; leaves opposite; blades ovate to elliptic, often slightly falcate, 0.5-1. 1 X 0.2-0.5 cm, apex acute, margin sharply serrulate, base asymmetrical, one side cordate, the other round to attenuate, upper surface toward the center, lower surface strigulose, only the midvein evident; petiole 0.5-0. 9 mm. Cyathia solitary; peduncles 0.5-1. 6 mm. Involucre obconical, 0.9-1. 2 X 0.8-1. 2 mm, outer surface strigulose, inner surface sparsely pilose, lobes subulate, 0.2-0.4 mm, pilose; glands 4, circular to transversely oblong, 0.2 X 0.2— 0.3 mm, green, yellow, or light pink, convex; appendages absent or if present reduced to a narrow margin flanking the outer portion of the gland, less than 0.1 X 0.2-0.3 mm, entire, white to pink. Staminate flowers 5 to 8. Gynophore glabrous, exserted from the involucre 1.4-1 .8 mm; ovary 3- lobed, strigulose to canescent; styles undivided, 0.3- 0.4 mm, tapering to the stigmatic tips, pink to red, glabrous. Capsules oblate, 3-lobed, 1.7-2 X 2.2-2. 7 mm, strigulose; columella 1.5-1.9 mm. Seeds not carunculate, broadly ovoid, nearly triangular in cross Novon 22: 482-485. Published on 18 October 2013. doi: 10.3417/2011014 Volume 22, Number 4 Steinmann & Jercinovic 483 .. Fertile habit. — B. Flowering branch. - section, 1.2-1. 4 X 1. 0-1.1 mm, tan to light gray when immature, becoming brown to blackish brown, apex a low, blunt point, base truncate, dorsal keel promi- nent, with 2 well-defined transverse ridges that do not pass through the dorsal keel. Distribution and habitat. Euphorbia raytumeri is known only from southwestern New Mexico, United States, where it occurs in Grant, Hidalgo, and Luna counties at elevations of 1400-1700 m. Given the close proximity of the Gray Ranch to the Mexican border (less than 10 km at its southernmost limit), E. raytumeri may also occur in the adjacent state of Chihuahua. The type locality, the Pitchfork Ranch, is located about 17 miles north of Interstate 10 along Separ Road. The ranch property consists of roughly 12,000 acres just west of the Continental Divide. It includes several miles of the Burro Cienaga, which Figure 2. Euphorbia raytumeri V. W. Steinm. & Jercinovic. Seeds. Taken from the isotype, E. Jercinovic 827 (MICH). Petrocosmea melanophthalma , a New Species in Section Deianthera (Gesneriaceae) from Yunnan, China Wang Huan-Chong Institute of Plant Science, School of Life Science, Yunnan University, Kunming 650091, China; and Key Laboratory of Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650204, People’s Republic of China, hchwang@ynu.edu.cn Zhang Li- Bing Missouri Botanical Garden, St. Louis, Missouri 63166-0299 U.S.A.; and Chengdu Institute of Biology, Chinese Academy of Sciences, P. 0. Box 416, Chengdu, Sichuan 610041, People’s Republic of China. Libing.Zhang@mobot.org He Zhao-Rong Institute of Plant Science, School of Life Science, Yunnan University, Kunming 650091, People’s Republic of China. Author for correspondence: zhrhe@ynu.edu.cn Abstract. Petrocosmea melanophthalma Huan C. Wang, Z. R. He & Li Bing Zhang, an attractive new species (Gesneriaceae, Petrocosmea sect. Deianthera W. T. Wang) from Yunnan, China, is described and illustrated. It is similar to P. kerrii Craib in the zygomorphic calyx and brevirostrate anther. The new species differs from the latter by its peltate leaves and the blue corolla with two striking black spots on the tube, as well as the geniculate filaments of the stamens that thicken above the middle. Petrocosmea melanopthalma was found on basalt cliffs in evergreen broad-leaved forest on Mopan Shan. Key words: China, Gesneriaceae, IUCN Red List, Petrocosmea, Yunnan. Petrocosmea Oliv. established by Oliver with the single species P. sinensis Oliv. (Oliver, 1887). Plants in this genus are perennial, acaulescent herbs with entirely basal leaves. The inflorescences are cymose, axillary, 1- to white and is zygomorphic and bilabiate. There are two stamens with basifixed anthers and parallel thecae; there are three staminodes, but these can be only one or absent. There is no disc; the ovary is 1-loculed with two placentas in parietal orientation; the single stigma is terminal, capitate, and undivided (Li & Wang, 2004). Taxonomic revision of the genus Petrocosmea was first provided by Craib (1919) and later by Wang (1985). Twenty-seven species and four varieties were recognized by W. T. Wang, classified into three sections: Petrocosmea sect. Petrocosmea, Petrocosmea sect. Anisochilus Hemsl., and Petrocosmea sect. Deinanthera W. T. Wang (Wang, 1985). New reports and taxonomic novelties of this genus have been consistently reported (e.g., Wang et al., 1990, 1998; Li & Wang, 2004; Wei & Wen, 2009; Guo et al., 2010; Middleton & Triboun, 2010; Zhao & Shui, 2010) such that Petrocosmea currently comprises 32 species and four varieties. The genus is native to Asia, occurring in China, northeastern India, Myanmar, Thailand, and southern Vietnam. Approximately 24 species are concentrated on the Yunnan-Guizhou Plateau and most species restrict to relatively small geographical areas in distributions. Herein we describe a new species of Petrocosmea, in section Deinanthera sensu W. T. Wang (1985), from basalt cliffs ii lima Huan C. Wang, Z R. He & Li Bing Zhang, sp. nov. TYPE: China Yunnan: Xinping Co., Mopan Shan 23°56.60'N, 101°58.49'E, on basalt cliffs in evergreen broad-leaved forest, 2200-2300 m 26 May 2005, H. C. Wang 20050321 (holotype YUKU; isotypes, CDBI, KUN, MO, YUKU) Figures 1, 2. ellipticis duabus ornata atque filamentis supra medium incrassatis geniculatisque quam antheris longioribus differt. Perennial, small herbs; rhizomes short, ca. 5 mm. Leaves 10 to 25, basal, crowded, the inner leaves Novon 22: 486-490. Published on 18 October 2013. doi: 10.3417/2011035 Volume 22, Number 4 2013 Wang et al. 487 Petrocosmea (Gesneriaceae) from China Figure 1. Petrocosmea melanophthalma Huan C. Wang, Z. R. He & Li Bing Zhang. — A. Fertile plant habit. — B. Flower. — C. 20050321 (YUKU). 7 Y with petioles short or absent, the outer leaves with long petioles to 7 cm; leaf blades flaccid, slightly papery when dry, the outer blades narrowly elliptic, oblong or ovate-oblong, rarely lanceolate, 4-7(— 10) X 1-2. 5(— 4) cm, the inner blades ovate or suborbicular, 1-3.5 X 0.6-2 cm, the blade apex acute, the base peltate, margin repand, adaxial blade surface densely pubescent, abaxial surface purple, short glandular, densely adpressed villous, lateral veins 4 to 8 on either side of midrib; petioles densely adpressed, 488 Novon l C. Wang, Z. R. He & Li Bing Zhang and similar 9.3 X 3.8-9 cm; with P. Author Index: Novon Yol. 22} 492 Novon H He, H. & L.-B. Zhang. Polystichum oblanceolatum, a New Species in Section Haplopolystichum (Dryopteridaceae) from Guangxi, China, 160- 165 He, H., Y.-Q. Yang & L.-B. Zhang. Polystichum loratum (Dryopteridaceae), a New Fern Species from Guizhou, China, 166-169 He, S.-y., W. Wu, P.-t. Li & M.-l. Zeng. Hoya daimenglongensis (Apocynaceae, Asclepiadoi- deae), a New Species from Yunnan, China, 170-173 He, Z.-R. see Wang et al. 486-490 Heo, K.-L .see Lee et al. 25-31 Holmes, W. C. see Mink et al. 56-59 Hovenkamp, P. H. see Rakotondrainibe & Hoven- kamp. 75-77 Husby, C. E. see Laubenfels et al. 8-15 Iamonico, D. & G. Kadereit. Typification of the Name Kochia saxicola (Chenopodiaceae), 418- 421 Jercinovic, E. see Steinmann & Jercinovic. 482^485 J0rgensen, P. M., N. Muchhala & J. M. MacDougal. Passiflora unipetala, a New Bat-pollinated Species of Passiflora supersect. Tacsonia (Pas- sifloraceae), 174-179 J0rgensen, P. M. & R. L. Liesner. Casearia maynacarpa (Salicaceae), a New Name for Carpotroche parvifolia, 422-423 K Kadereit, G. see Iamonico & Kadereit. 418-421 Karamian, R. see Ranjbar et al. 78-81 Kim, S.-C. see Lee et al. 25-31 Laubenfels, D. J. de, C. E. Husby & M. P. Griffith. Further Nomenclatural Action for the Cypresses (Cupressaceae), 8-15 Pseudostellaria (Caryophyllaceae) from Korea, 25-31 Letsara, R. see Daniel et al. 396-408 Li, J.-D. see Tang et al. 227-234 Li, J.-Q. see Zhang, Z.-G. et al. 379-382 Li, P.-t. see He, S.-y. et al. 170-173 Li, R., J. Yang & Z. Dao. Meconopsis xiangchen- gensis (Papaveraceae), a New Species from Sichuan, China, 180-182 Liao, C.-K., T.-P. Lin & M.-S. Tang. Tropidia namasiae, a New Species of Orchidaceae (Epidendroideae, Tropidieae) from Southwest- ern Taiwan, 424^427 Liao, J.-P. see Zou et al. 128-130 Liesner, R. L. see j0rgensen & Liesner. 422^423 Lima, M. I. S. see Ferreira et al. 155-159 Lin, T.-P. see Liao et al. 424-427 Liu, J. see Zhang, L. et al. 244-249 Liu, K.-W. see Liu et al. 43-47 Liu, Y. see Tang et al. 227-234 Liu, Y.-Y., J.-C. Zhao & S. Mamtimin. Syntheto- dontium kunlunense (Mielichhoferiaceae, Mus- ci), a New Moss Species from the Kunlun Mountain Range, China, 428-433 Liu, Z.-J., L.-J. Chen & K.-W. Liu. Neuwiedia malipoensis, a New Species (Orchidaceae, Apostasioideae) from Yunnan, China, 43-47 Loiola, M. I. Bezerra & M. F. de Sales. Erythroxylum ayrtonianum (Erythroxylaceae): A New Species from Brazil, 48-50 Lorence, D. H. & K. R. Wood. Psychotria kosraensis (Rubiaceae), a New Species from Kosrae, Caroline Islands, Micronesia, 51-55 Lucas, E. see Araujo & Lucas. 385-388 Luo, Q. & L.-B. Zhang. Dryopteris jiucaipingensis (Dryopteridaceae), a New Species in Dryopteris sect. Hirtipedes from Guizhou, China, 183-185 Luo, Q. & L.-B. Zhang. Polystichum tiankengicola (Dryopteridaceae), a New Species from a Karst Sinkhole from Guizhou, China, 186-190 Luo, Q. & L.-B. Zhang. Woodsia guizhouensis (Woodsiaceae), a New Species from a Lime- stone Area in Guizhou, China, 191-195 M Ma, X.-J. see Wang et al. 109-113 MacDougal, J. M. see j0rgensen et al. 174-179 MacDougal, J. M. Two New Species of Passiflora supersect. Decaloba (Passifloraceae) from East- ern Mexico, 434^441 Mamtimin, S. see Liu et al. 428-433 Mansano, V. de F. see Marquete & Mansano. 196- 206 Marquete, R. & V. de F. Mansano. Taxonomic Revision of the Casearia ulmifolia Complex (Salicaceae), 196-206 Martin-Bravo, S. see Daniel et al. 396^408 Meng, A.-P. see Zhang, Z.-G. et al. 379-382 Mink, J. N., J. R. Singhurst & W. C. Holmes. A New Species of Hymenoxys (Asteraceae, Helenieae, Tetraneuridinae) from Texas, 56-59 Monfils, A. K. see Shiels & Monfils. 87-90 Subject Index: Novon Yol. 221,2 Novon 22: 496-502. Published on 18 October 2013. Volume 22, Number ■ 2013 Subject Index 497 lombacoideae, 152-154 'ombax crenulatum, 152-154 Irassicaceae, 109-113, 411-414 irazil, 48-50, 137-140, 152-154, 155-159, 196- 206, 265-270, 281-287, 288-291, 304-306, 325-331, 371-376, 478-481 Bahia, 1-7, 82-86, 137-140 Minas Gerais, 265-270, 325-331 Caesalpinioideae, 91-95, 332-335 Callitropsis, 8-15 Callitropsis junebris, 8-15 Calyptranthes, 385-388 Calyptranthes brasiliensis, Calyptranthes fastigiata, Calyptranthes glabrescens Calyptranthes kiaerskovii, Calyptranthes longifolia, Calyptranthes multiflora, Calyptranthes obovata, Calyptranthes pteropoda, 385-388 Calyptranthes ruiziana, 385-388 Calyptranthes schiedeana, 385-388 Calyptranthes tetraptera, 385-388 Cameroon, 442-146 Canada, 60-61, 271-275 British Columbia, 271-275 Carex, 118-124 sect. Racemosae, 118-124 Carex albonigra, 118-124 Carex orestera, 118-124 Caroline Islands, 51-55 Carpotroche, 422-423 Carpotroche parvifolia, 422—423 Caryophyllaceae, 25-31 Casearia, 196-206, 422-423 Casearia cotticensis, 196-206 Casearia maynacarpa, 422-423 Casearia murceana, 196-206 Casearia parvifolia, 422-423 Casearia ulmifolia, 196-206 Cercideae, 148-151 Chamaefistula, 91-95 Chamaefistula tocotana, 91-95 Cheilanthes, 62-66 Cheilanthes dichotoma, 62-66 Cheilanthes orbignyana, 62-66 Chenopodiaceae, 418^21 China, 109-113, 114-117, 128-130, 160-165, 166-169, 170-173, 180-182, 183-185, 186- 190, 191-195, 227-234, 244-249, 250-255, 256-259, 260-263, 292-296, 307-310, 332- 335, 377-378, 379-382, 409^10, 428-133 Guangdong, 114-117, 377-378 Guangxi, 160-165, 379-382 Guizhou, 166-169, 183-185, 186-190, 191- 195, 250-255, 256-259, 260-263 Hainan, 128-130 Heilongjiang, 109-113 Hunan, 307-310 Inner Mongolia, 227-234 Shaanxi, 292-296 Sichuan, 180-182, 244-249 Xinjiang, 428-433 Yunnan, 43-17, 170-173, 332-335, 486-190 Chondrites verticillatus, 240-243 Cladoxylopsida, 240-243 Clethra, 212-216 Clethra concordia, 212-216 Clethraceae, 212-216 Colombia, 91-95, 96-108, 336-370 Congo-Brazzaville, 442-446 Coniogramme, 292-296 Coniogramme bashanensis, 292-296 Coniogramme wilsonii, 292-296 Convolvulaceae, 137-140 Croton, 377-378 Cruciferaefl09-1 13 Cupressaceae, 8-15 Cupressus, 8-15 Cupressus abramsiana, 8-15 Cupressus benthamii, 8-15 Cupressus duclouxiana, 8-15 Cupressus junebris, 8-15 Cupressus goveniana var. pygmaea, 8-15 Cupressus lindleyi, 8-15 Cupressus montana, 8-15 Cupressus nevadensis, 8-15 Cupressus torulosa, 8-15 Cymodoceaceae, 16-24 Cyperaceae, 87-90, 118-124 D Decandrae, 148-151 Dictyosporae, 223-226 Diporisporites, 223-226 Diporisporites giganticus, 223-226 Ditepalae, 235-239 Dracaena, 442-146 Dracaena braunii, 442-446 Dracaena litoralis, 442—146 Dracaena longipetiolata, 442-146 Dracaena phrynioides, 442-146 Illli llii .11 .1111 499 , 75-77, 141-147, ■ 501 Volume 22, Number 1, pp. 1-132 of NOVON was published on 10 July 2012. Volume 22, Number 2, pp. 133-264 of NOVON was published on 30 November 2012. Volume 22, Number 3, pp. 265-384 of NOVON was published on 24 May 2013. Volume 22, Number 4, pp. 385-504 of NOVON was published on 18 October 2013. www.mbgpress.info CONTENTS