New Combinations for Chinese (, )

Chris M. A. Stapleton Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3AB, United Kingdom. [email protected]

Li De-Zhu Kunming Institute of Botany, Chinese Academy of Sciences, Heilongtan, Kunming 650204, Yunnan, People’s Republic of . [email protected]

Xia Nian He South China Institute of Botany, Chinese Academy of Sciences, Guangzhou 510650, People’s Republic of China. [email protected]

ABSTRACT. On the basis of morphological char- hirsutissimus (W.D.Li&Y.C. acters inconsistent with their current taxonomic Zhong) Stapleton & D. Z. Li, comb. nov. Bas- placements, new combinations are made for the fol- ionym: hirsutissimum W. D . lowing of woody bamboos within the flora Li & Y. C. Zhong, in J. Res. 16(1): of China: Ampelocalamus breviligulatus, Ampelo- 52. 1997. : China. Guizhou: Guiyang, calamus hirsutissimus, Ampelocalamus melicoides, Bamboo Garden of GF, 5 Mar. 1992, Zhong Drepanostachyum semiorbiculatum, Drepanostach- Yuan-Chun et al. 9203 (holotype, GF not yum ampullarium, Drepanostachyum membrana- seen). ceum, and parvifloscula. To accommodate Ampelocalamus melicoides (Keng f.) D. Z. Li & broader generic concepts, new combinations are Stapleton, comb. nov. Basionym: Drepano- made for qingchengshanensis and stachyum melicoideum Keng f., J. Bamboo Arundinaria spanostachya, and new status is given Res. 5(2): 35. 1986. TYPE: China. Sichuan: to as Arundinaria subgen. Bashania, and Nanchuan Xian, 4 May 1957, G. F. Li 60336 to as Arundinaria subgen. Sarocala- (holotype, NJNU; isotype, SZ). mus. Key words: Ampelocalamus, Arundinaria, Bam- Ampelocalamus and Drepanostachyum are sub- buseae, Bashania, Bonia, China, Drepanostach- tropical clump-forming bamboo genera with type yum, Poaceae, Sarocalamus. species from Hainan Island and the of northwest India, respectively. They differ in several During the preparation of a collaborative account characters. The culm nodes and branching in Am- of Chinese bamboos for the English-language Flora pelocalamus are adapted for a semi-scandent of China, it became apparent that there was a need growth habit, with expanded nodal sheath scars, ge- for several new combinations for species names in niculate lateral branches, a large central branch Ampelocalamus S. L. Chen, T. H. Wen & G. Y. that often remains dormant, and a capacity for ae- Sheng, Drepanostachyum Keng f., Arundinaria Mi- rial root growth on the swollen bases of larger chaux, and Bonia Balansa, and for the provision of branches. Culms of Drepanostachyum are self-sup- names at subgeneric rank within Arundinaria s.l. porting and lack these adaptations. The inflores- for Bashania Keng f. & T. P. Yi and Sarocalamus cence in Ampelocalamus also differs in having larg- Stapleton. er, broader spikelets (to 35 ϫ 5 mm vs. 20 ϫ 3 mm). Ampelocalamus is distributed from Taiwan AMPELOCALAMUS and Hainan Island to the wettest parts of the Him- Ampelocalamus breviligulatus (T. P. Yi) Staple- alayas, especially on calcareous soils, while Dre- ton & D. Z. Li, comb. nov. Basionym: Drepa- panostachyum is found along the Himalayas in dri- nostachyum breviligulatum T. P. Yi, J. Bamboo er subtropical forest types associated with trees in Res. 12(4): 42. 1993. TYPE: China. Sichuan: Quercus L., Schima Reinwardt ex Blume, and Cas- Jiange Xian, 18 Dec. 1983, Yi Tong-Pei 83208 tanopsis (D. Don) Spach. (holotype, SIFS not seen). The two genera were confused by transferral of

NOVON 15: 599–601. PUBLISHED ON 12 DECEMBER 2005. 600 Novon

most species of Ampelocalamus into Drepanostach- fore Drepanostachyum was misunderstood in China, yum (Keng, 1986; Yi, 1993) and eventually by the where it was considered very similar to Ampeloca- synonymizing of Ampelocalamus within Drepano- lamus, which was still recognized (Keng & Wang, stachyum (Li, 1997). However, recent molecular 1996), albeit for only 2 less well-known species. data (Nı´ Chonghaile, 2002) has suggested that they Meanwhile, when species of Drepanostachyum were are not so closely related, with Drepanostachyum in encountered in the Himalayas of southeast , a clade (Bootstrap Support 73%) that did not in- they were placed in an altogether different , clude Ampelocalamus. Franchet. Section Sphaerigemma T. P. Yi The generic characteristics of Ampelocalamus and series Ampullares T. P. Yi were described large- breviligulatus have not been investigated in depth, ly to accommodate these mainly Himalayan species but the large discrepancy in branch sizes with a within Fargesia, because of their semi-orbicular dominant central branch and finer lateral branches, culm buds and swollen culm nodes, not seen in along with the deeply ridged culm internodes sug- other species of the genus. gest that it is more likely to belong in Ampeloca- Fargesia semiorbiculata and F. ampullaris also lamus than in Drepanostachyum. This is reinforced differed substantially from other species placed in by its distribution in Sichuan and Gansu provinces Fargesia in their larger number of more subequal of China, rather than the Himalayas. branches, leaf blades with indistinct transverse Described as related to Ampelocalamus scandens veins, the narrow apex and distally scabrous inte- Hsueh & W. D. Li, with apically arching culms and rior surface of their culm sheaths, and their sub- patelliform nodes, and illustrated as having a tropical rather than temperate habitats. These char- branch complement with strongly geniculate lateral acteristics are shared with the type species of branches, Ampelocalamus hirsutissimus would also Drepanostachyum, D. falcatum, into which genus appear more appropriately placed in Ampelocala- they are now transferred. mus rather than in Drepanostachyum. The ciliolate Fargesia membranacea differs from most other culm sheath margins and prominent leaf sheath oral species in Fargesia in its semi-orbicular branch setae are similar to those of A. patellaris and A. buds producing 13 to 33 subequal branches, its leaf scandens, respectively. blades with indistinct transverse veins, and the nar- The dominant central branch and geniculate lat- row apex to its culm sheaths, characteristics more eral branches, along with broad spikelets suggest suggestive of Drepanostachyum than Fargesia. that Ampelocalamus is also more appropriate than Known only from southern Sichuan province, this Drepanostachyum for A. melicoides. species is well separated from all other species of Drepanostachyum. DREPANOSTACHYUM ARUNDINARIA Drepanostachyum semiorbiculatum (T. P. Yi) Arundinaria subgen. Sarocalamus (Stapleton) D. Stapleton, comb. nov. Basionym: Fargesia se- Z. Li, stat. nov. Basionym: Sarocalamus Sta- miorbiculata T. P. Yi, J. Bamboo Res. 2(2): 40. pleton, Novon 14: 346. 2004. TYPE: Saroca- 1983. TYPE: China. Tibet: Cona Xian, 7 Jan. lamus racemosus (Munro) Stapleton. 1978, Jiang Changgui 1 (holotype, SIFS not Arundinaria subgen. Bashania (Keng f. & T. P. seen). Yi) D. Z. Li, stat. nov. Basionym: Bashania Drepanostachyum ampullarium (T. P. Yi) Sta- Keng f. & T. P. Yi, J. Nanjing Univ., Nat. Sci. pleton, comb. nov. Basionym: Fargesia ampul- Ed. 1982(3): 722. 1982. TYPE: Bashania far- laris T. P. Yi, J. Bamboo Res. 2(2): 18. 1983. gesii (E. G. Camus) Keng f. & T. P. Yi. TYPE: China. Tibet: Zhangmu, 4 Nov. 1979, Arundinaria qingchengshanensis (Keng f. & T. M. L. Zhou 4 (holotype, SIFS not seen). P. Yi) D. Z. Li, comb. nov. Basionym: Bashania Drepanostachyum membranaceum (T. P. Yi) D. qingchengshanensis Keng f. & T. P. Yi, J. Nan- Z. Li, comb. nov. Basionym: Fargesia mem- jing Univ., Nat. Sci. Ed. 1982(3): 725. 1982. branacea T. P. Yi, Acta Bot. Yunnan. 14(2): TYPE: China. Sichuan: Guan Xian, 25 June 135. 1992. TYPE: China. Sichuan: Mianning 1981, Yi Tong-Pei 80037 (holotype, SIFS). Xian, Yi Tong Pei 90172 (holotype, SIFS not Arundinaria spanostachya (T. P. Yi) D. Z. Li, seen). comb. nov. Basionym: Bashania spanostachya The type species of Drepanostachyum, D. falca- T. P. Yi, Acta Bot. Yunnan. 11: 35. 1989. tum (Nees) Keng f., was not known at all in China, TYPE: China. Sichuan: Huili Xian, Beimu while several well-known species of Ampelocalamus Shan, 18 May 1987, Yi Tong-Pei 87249 (ho- had been transferred into Drepanostachyum. There- lotype, SIFS). Volume 15, Number 4 Stapleton et al. 601 2005 Chinese Bamboos

Although it is strongly suspected that Bashania ansa also had these characteristics (Xia, 1996). The Keng f. & T. P. Yi and Sarocalamus Stapleton are characteristics of Monocladus parviflosculus are not probably not closely related to the North American well known as the only collections made so far are genus Arundinaria (Nı´ Chonghaile, 2002; Stapleton poor, but it would appear that it also represents a et al., 2004), the morphological distinctions are less species of Bonia rather than any other genus. pronounced than those between Arundinaria and all other Chinese genera. As the Acknowledgments. The corresponding author follows a primarily morphological classification sys- thanks Missouri Botanical Garden for funding this tem, these three genera will be merged into a rather research and the Keeper of the Herbarium of the broadly interpreted Arundinaria. Hopefully further Royal Botanical Gardens, Kew, for providing work- molecular data will provide clarification of the true ing facilities. relationships between these groups. New combi- nations in Arundinaria are required for two species Literature Cited of Bashania, while B. fargesii and the remaining Keng, P. C. 1986. A new discovery of bamboo genus Dre- two species of Sarocalamus already have combi- panostachyum Keng f. in China. J. Bamboo Res. 5(2): nations in that genus, and the affinities of three 28–40. further, more recently described species of Bashan- & Z. P. Wang (editors). 1996. Gramineae (Po- ia remain unclear. aceae), Bambusoideae. Flora Reipublicae Popularis Sinicae 9(1). Science Press, Beijing. [In Chinese.] Li, D. Z. 1997. The Flora of China Bambusoideae Project: BONIA Problems and current understanding of bamboo taxon- Bonia parvifloscula (W. T. Lin) N. H. Xia, comb. omy in China. Pp. 61–81 in G. P. Chapman (editor), The Bamboos. Academic Press, London. nov. Basionym: Monocladus parviflosculus W. Nı´ Chonghaile, G. 2002. Molecular Systematics of the T. Lin, J. Bamboo Res. 12(3): 3. 1993. TYPE: Woody Bamboos (Tribe Bambuseae). Trinity College, China. Guangdong: Zhaoqing, Qixingyan, Z. Dublin [unpublished Ph.D. thesis]. K. Li 84126 (holotype, CANT). Stapleton, C. M. A., G. Nı´ Chonghaile & T. R. Hodkinson. 2004. Sarocalamus, a new Sino-Himalayan bamboo ge- The genus Monocladus Chia, H. L. Fung & Y. L. nus (Poaceae–Bambusoideae). Novon 14: 345–349. Yang et al. was established for tropical bamboos Xia, N. H. 1996. A study of the genus Bonia (Gramineae: Bambusoideae). Kew Bull. 51: 565–569. from south China with solitary branches and spike- Yi, T. P. 1993. New taxa of Drepanostachyum and other lets with a well-separated basal floret. It was not new combinations of bamboos in China. J. Bamboo Res. realized that species of the earlier genus Bonia Bal- 12(4): 42–47.