Baimashania (Brassicaceae), a New Genus from China Ihsan A. Al-Shehbaz Missouri Botanical Garden, P.0. Box 299, St. Louis, Missouri 63166-0299, U.S.A. ABSTRACT. The new genus Baimashania and the Solmslaubachia (China, Sikkim) and six species of new species B. pulvinata and B. wangii are de¬ Leiospora (C. A. Meyer) Dvofdk (China, Himalayas, scribed. The relationships and characters distin¬ and central Asia) in having median nectar glands guishing Baimashania from Solmslaubachia, Leios- confluent with the laterals, distinct and slender pora, and Pycnoplinthus are discussed. styles, flowers only to 4 mm long, fruit septa with a distinct midvein, obscurely veined, flat-margined During a recent revision of Solmslaubachia Mus- fruit valves that are readily detached from replum chler for the forthcoming account of the Brassica¬ at the fruit apex, replums readily distinguished ceae for the Flora of China, Al-Shehbaz and Yang from the valve margins, and narrow fruits (1-1.3 (2000) excluded three collections previously mis- mm wide) persistent on the pedicel apex. By con¬ identified as S. ciliaris (Bureau & Franchet) Bot- trast, Solmslaubachia and Leiospora lack the me¬ schantsev from the genus for subsequent studies. dian nectar glands and styles, and have much larg¬ These three collections are placed herein in the er flowers 10-26 mm long, veinless fruit septa, new genus Baimashania. Except for the smaller prominently veined, marginally strongly angled flower size, the type specimen of S. ciliaris is in¬ fruit valves apically adnate to the replum, replums distinguishable in all other morphological aspects concealed by the valve margins, and broad fruits, from that of the earlier published S. pulcherrima (3-)5—15(-20) mm wide, readily detached from the Muschler, a species highly variable in petal length pedicel apex. Furthermore, Baimashania also dif¬ and color. In fact, the type of S. ciliaris has petals fers from Solmslaubachia by having forked tri- ca. 1.5 cm long, while S. pulcherrima has petals chomes mixed with simple ones and minutely re¬ (1.5-) 1.7-2(-2.5) cm long. Therefore, the two are ticulate seeds, while Solmslaubachia is glabrous or considered to be conspecific. has only simple trichomes and has papillate, ru¬ The recognition of Solmslaubachia ciliaris in var¬ gose, or coarsely reticulate seeds. It also differs ious works (e.g., Lan & Cheo, 1981; Kuan, 1985; from Leiospora by having entire stigmas, 6- to 12- Lan, 1987; Wang, 1993; Ying et al., 1993; Li et al., ovuled ovaries, nonsaccate lateral sepals, and wing¬ 1995; Huang, 1997; Tan et al., 1999) basically fol¬ less seeds. Leiospora has prominently 2-lobed stig¬ lowed Botschantsev (1955), who did not examine mas, 18- to 50-ovuled ovaries, saccate lateral the type material of the species and erroneously sepals, and often winged seeds. interpreted its limits. In fact, illustrations and de¬ scriptions of S. ciliaris in these works are based on The first collection of Baimashania, Forrest 19605 (see below), in July 1921 was misidentified species of different genera. For example, the ac¬ by Edgar Evans as “Braya sp. aff. forrestii W. W. counts in Kuan (1985), Wang (1993), and Ying et Smith.” Baimashania pulvinata differs from Bray a al. (1993) were based on one collection, Qinghai- Xizang Team 73—399 (KUN, PE), which is a mis- forrestii in having flattened, linear fruits, accumbent identified young plant of Leiospora pamirica (Bot¬ cotyledons, forked and simple trichomes, a conflu¬ schantsev & Vvedensky) Botschantsev & ent nectar ring, and much smaller, solitary flowers. Pachomova. By contrast, the descriptions and illus¬ By contrast, Braya forrestii has terete, ovoid fruits, trations in Lan and Cheo (1981), Lan (1987), Li et incumbent cotyledons, exclusively simple tri¬ al. (1995), and Tan et al. (1999) are based on a new chomes, four separate nectar glands, and larger species in a new genus herein described as Bai¬ flowers in distinct racemes. In my opinion, Bai¬ mashania pulvinata, whereas the account in Huang mashania and Braya Sternberg & Hoppe are not (1997) is based on the new species B. wangii. related. Baimashania is named after Baima Shan (some¬ Perhaps Baimashania is most similar to Pycm>- times spelled Beima Shan, Beimashan, or Paimas- plinthus 0. E. Schulz (one species; China, India, han), a mountain in northwestern Yunnan from Kashmir, Pakistan), with which it shares the peren¬ which four of the five known collections of the ge¬ nial, pulvinate, scapose habit, solitary pink flowers, nus are known. It differs from all nine species of persistent entire leaves, thickened petioles, obtuse Novon 10: 320-322. 2000. Volume 10, Number 4 Al-Shehbaz 321 2000 Baimashania from China anthers, 6- to 12-ovuled ovaries, linear flattened lb. Leaves linear, glabrous except for ciliate petiole fruits, entire stigmas, and uniseriate wingless margins and a tuft at leaf apex; flowers mostly in 2- or 3-flowered racemes; ovules 10 to 12 per seeds. However, it differs by having branched tri- ovary; Qinghai.2. B. wangii chomes, free sepals, distinct median glands conflu¬ ent with the laterals, obscurely veined valves, and accumbent cotyledons. Pycnoplinthus consists of 1. Baimashania pulvinata Al-Shehbaz, sp. nov. glabrous plants with united sepals, no median nec¬ TYPE: China. Yunnan: Deqen, Baima Shan, N tar glands, prominently veined valves, and incum¬ side of road, limestone rock crevices, 28°23'N, bent cotyledons. 99°01'E, 4350 m, 30 June 1994, Alpine Gar¬ den Society Expedition to China ACE 829 (ho- Baimashania Al-Shehbaz, gen. nov. TYPE: Bai¬ lotype, K; isotype, MO). mashania pulvinata Al-Shehbaz. Herba perennis scaposa pulvinata, caudice ramoso. Fo¬ Herba pulvinata 0.5-2 cm alta, caudice multiramoso. lia basalia rosulata, integra. persistentia, pilis simplicibus Folia basalia ovata vel oblonga, 2—4 X 1—2 mm, subcar- et furcatis, petiolis crassis complanatis, ciliatis. Racemi nosa, omnino dense pilosa. Flores solitarii. Pedicelli fruc- 2- vel 3-flori vel flores solitarii. Sepala oblonga nonsac- tiferi 3-5 mm longi. Sepala oblonga 1.5-2.5 X ca. 1 mm. eata. Petala rosea, spathulata, obtusa. Clandulae nectari- Petala rosea, 3—4 X 1 — 1.5 mm; ungues 1.5—2 mm longi. ferae confluentes. Ovula 6—12. Fmctus lineares valde Ovula 6-8. Fructus lineares, 4—8 X 1 — 1.3 mm: stylo 0.4— compressi glabri: valvis longitudinaliter striatis nontom- 1 mm longo. Semina 1—1.5 X 0.6-0.9 mm. losis; stylo tenui ad 1 mm longo; stigmata integra. Semina uniseriata. oblonga; cotyledones accumbentes. Herbs pulvinate, 0.5-2 cm tall; caudex many Herbs perennial, scapose, pulvinate, with a well- branched, to 3 mm diam. Trichomes simple, to 1 developed, few- to many-branched caudex covered mm long, mixed with fewer forked stalked ones, by leaves or leaf remains of previous years. Tri- restricted to leaves. Basal leaves rosulate; petiole chomes simple mixed with stalked, 1-forked ones, 2— 5 mm long, expanded at base and 0.5—1 mm restricted to leaves. Stems absent. Basal leaves pet- wide, ciliate, persistent; leal blade ovate to oblong, iolate, rosulate, simple, entire, persistent, with stra¬ 2^4 X 1—2 mm, somewhat fleshy, densely pilose mineous, thick, basally flattened, ciliate petioles. throughout, base subattenuate, margin entire, apex Cauline leaves absent. Racemes 2- or 3-flowered obtuse. Flowers solitary. Fruiting pedicel slender, and ebracteate, or flowers solitary on short pedicels 3— 5 mm long, ascending, glabrous. Sepals oblong, originating from axils of rosette leaves. Fruiting 1.5-2.5 X ca. 1 mm. Petals pink, spatulate, 3^4 X pedicels slender, erect to ascending, often hidden 1-1.5 mm; claw 1.5—2 mm long. Filaments 2-2.5 among basal leaves. Sepals oblong, deciduous, mm long; anthers 0.3—0.5 mm long. Ovules 6 to 8 erect, equal, base of inner pair not saccate, margin per ovary. Fruit linear, 4—8 X 1—1.3 mm; valves membranous. Petals pink; blade spatulate, apex ob¬ longitudinally striate, without a distinct midvein; tuse; claw slightly differentiated from blade, sube¬ style 0.4-1 mm long. Seeds 1—1.5 X 0.6-0.9 mm. qualing sepals. Stamens 6, slightly tetradynamous; Flowering June—July, fruiting July—August. filaments not dilated at base; anthers oblong, not apiculate at apex. Nectar glands confluent and sub¬ Habitat. Moist gravelly meadows, limestone tending bases of all stamens; median nectaries pre¬ rock crevices; 4200-4600 m. Yunnan. sent. Ovules 6 to 12 per ovary. Fruit dehiscent, Baimashania pulvinata was illustrated in Lan capsular siliques, linear, strongly latiseptate, ses¬ (1987) as Solmslaubachia ciliaris. The illustration sile; valves with an obscure midvein, glabrous, not accurately shows the habit of the plant, leaf shape torulose, longitudinally striate; replum rounded, and trichomes, and mature fruit, and, therefore, visible; septum complete, membranous, with a dis¬ should be consulted for reference. The holotype tinct midvein; style slender, to 1 mm long; stigma was misidentified as Braya cf. forrestii, though capitate, entire. Seeds uniseriate, wingless, oblong; nothing was mentioned about it in the detailed ac¬ seed coat minutely reticulate, not mucilaginous count of the expedition that led to its discovery when wetted; cotyledons accumbent. (Grey-Wilson, 1996). Two species: endemic to China (Yunnan and Qinghai provinces). Paratypes. CHINA. Yunnan: Baima Shan, 28°18'N, 99°10'K, G. Forrest 19605 (E, K, US); Atuntze, Baima Key to the Species he Baimashania Shan, Shulung, T. T. Yii 9324 (KUN, PE); Deqen, Baima la. Leaves ovate to oblong, densely pilose through¬ Shan, Qing-Zang Team 2779 (KUN); Deqin Xian, Baima out; flowers solitary; ovules 6 to 8 per ovary; Shan, mts. N of W pass, 28°23’N. W O I 'E. Alden. Alex¬ Yunnan. 1. B. pulvinata ander, Long, McBeath, Sol tie & Watson 765 (E). 322 Novon 2. Buimashaiiia wangii Al-Shehbaz, sp. nov. Acknowledgments. I am indebted to Henk van TYPE: China. Qinghai: Nangqen Xian, Xia- der Werff for correcting the Latin. I thank Sue oqu, S bend, under rocks, 13 June 1982, 4100 Zmarzty for her permission to deposit at MO an rn, Wang Weiyi 0011 (holotype, HNWP; frag¬ isotype of Baimashania pulvinata. I also thank the ment, MO). directors and curators of the herbaria cited, as well as Zhu Guanghua, He Si, and Song Hong for their Hed>a pulvinata 1—1.5 cm aha, caudice pauciramoso. help in the translation of labels from Chinese. Folia linearia, 4—15 X 0.5—1.5 mm, subcamosa, praeter apicem pilosam glabra, petiolibus ciliatis. Racemi fructi- feri 2- vel 3-fructi vel fructus solitarii. Pedicelli fructiferi Literature Cited 3—5 mm longi. Flores ignoti. Ovula 10-12. Fructus line- ares, 0.7—1.3 cm X ca. 1 mm; stylo 0.4—0.7 mm longo. Al-Shehbaz, I. A. & G. Yang. 2000. A revision of Solms¬ Semina ignota. laubachia (Brassicaceae). Harvard Pap. Bot. (in press). Botschantsev, V. 1955. De Cruciferis notae criticae. Bot. Herbs pulvinate, 1—1.5 cm tall; caudex few Mater. Gerb. Bot. Inst. Komarova Akad. Nauk S.S.S.R. 17: 160-178. branched, to 6 mm diam. Trichomes simple, to 1 Grey-Wilson, C. 1996. The Alpine Garden Society Expe¬ mm long, restricted to petiole margins and leal api¬ dition to China. Quart. Bull. Alpine Card. Soc. 64: 122- ces, these mixed at leaf apex with fewer forked 164. stalked ones. Basal leaves rosulate; petiole 2—5 mm Huang, R. F. 1997. Cruciferae. In: S. W. Liu (editor), FI. long, expanded at base and 0.5—1.5 mm wide, cil- Qinghai. 1: 410-505. Qinghai People’s Publishing House, Xining. iate; leaf blade linear, 4—15 X 0.5-1.5 mm, some¬ Kuan, K. C. 1985. Solmslaubachia. In: C. Y. Wu (editor), what fleshy, base attenuate, margin entire, apex FI. Xizang. 2: 381-384. Science Press, Beijing. with a tuft of hairs, obtuse. Fruiting racemes with Lan, \. Z. 1987. Solmslaubachia. In: T. Y. Cheo (editor), 2 or 3 fruits, sometimes fruits solitary. Fruiting ped¬ FI. Reipubl. Popularis Sin. 33: 326—339. Science Press, Beijing. icel slender, 3—5 mm long, ascending, glabrous. -& T. Y. Cheo. 1981. On the Chinese genus Solms¬ Flowers not seen. Ovules 10 to 12 per ovary. Fruit laubachia Muschler (Cruciferae). Acta Phytotax. Sin. linear, 0.7—1.3 cm X ca. 1 mm; valves not torulose; 19: 472-480. style 0.4—0.7 mm long. Mature seeds not seen. Li, X. W. et al. 1995. Cruciferae. In: C. Y. Wu, C. Chen Fruiting June. & S. K. Chen (editors), FI. Yunnan. 6: 1 — 121. Science Press, Beijing. Tan, Z. M., Z. Q. Zhang, Y. Zhao & S. I). Zhou. 1999. Baimashania wangii, which is named after the Cruciferae. In: Z. M. Tan (editor), FI. Sichuan. 14: I - collector of the holotype, is known thus far only 181. Sichuan Nationality Press, Chengdu. from the type collection. The type was annotated in Wang, W. T. 1993. Cruciferae. In: W. T. Wang et al. (ed¬ 1995 by Huang Rong-fu as Solmslaubachia ciliaris, itors), Vascular Plants of the Hengduan Mountains. 1: 618—652. Science Press, Beijing. and it is very likely that the record of S. ciliaris Ying, T. S„ Y. L. Zhang & D. E. Boufford. 1993. The from Qinghai (Huang, 1997) was erroneously based Endemic Genera of Seed Plants of China. Science on the holotype of B. wangii. Press, Beijing.