5. Melica onoei Franchet & Savatier, Enum. Pl. Jap. 2: 603. 1879.
广序臭草 guang xu chou cao
Melica matsumurae Hackel; M. kumana Honda; M. scabe-rrima (Nees ex Steudel) J. D. Hooker var. micrantha J. D. Hooker.
Perennial, tufted. Culms few, 75–150 cm tall, 2–2.5 mm in diam., many-noded. Leaves all cauline; leaf sheaths much longer than internodes, scaberulous or puberulous, lower sheaths with reduced blades, often retrorsely setose; leaf blades broadly linear, 10–25 cm × 6–14 mm, abaxial surface smooth, adaxial surface hispid or pilose, often sparsely; ligule 0.2–0.5 mm. Panicle lax, 15–35 cm; branches 2 or 3 per node, ascending or spreading, branched, up to 15 cm, spikelets diffuse. Spikelets linear-lanceolate, 5–9.5 mm, green, fertile florets 1 or 2, 1 smaller sterile floret raised on ca. 2 mm internode; glumes unequal, clearly shorter than adjacent florets, lower glume narrowly ovate, 2.5–3.5 mm, upper glume lanceolate-oblong, 4–5 mm, both obtuse to acute; lemmas lanceolate-oblong, lowest 4.8–5.5 mm, herbaceous, granular-scaberulous, 7-veined, additional weaker veins sometimes present, upper margins and apex broadly membranous, apex obtuse; palea keels scaberulous near apex, otherwise smooth. Anthers 0.8–1.5 mm. Fl. and fr. May–Oct. 2n = 18.
Woodlands, damp shady places on hillsides, gullies, roadsides; 400–2500 m. Anhui, Gansu, Guizhou, Hebei, Henan, Hubei, Hunan, Jiangsu, Jiangxi, Ningxia, Shaanxi, Shandong, Shanxi, Sichuan, Taiwan, Xizang, Yunnan, Zhejiang [Japan, Kashmir, Korea, N Pakistan].
This species is mainly distributed from Japan to Yunnan, but a few gatherings are known from the W Himalayas. The combination of broad leaf blades, short ligule, narrow, green spikelets and smooth palea keels distinguishes it from other Chinese Melica species with a large, open panicle.
Melica onoei var. pilosula Papp (Acad. Romana, Mem. Sect. Sti., ser. 3, 12: 242. 1937) was described as having a simple, contracted panicle, smooth leaf sheaths, a long pilose adaxial leaf surface, and a pointed, ca. 5 mm ligule. The panicle and ligule definitely exclude it from M. onoei. It is based on a specimen from Beijing that has not been seen.