GraminecB.-] CHINA. 241 



S. exaltatum, according to Roxburgh's observations and drawing, has the upper valvelet of the bisexual 

 floret wanting in the pedicelled spikelet, but about equal to the other and acute in the sesile one ; according, 

 however, to N. v. Esenbeck's examination of what he considers the same plant, in Dr. Wight's herbarium, 

 it is the upper valvelet of the pedicelled (not the sessile) spikelet which is equal and similar to the other, 

 while in the sessile one it is much shorter and obtuse. As we have not seen the leaves of the Chinese plant, 

 we are uncertain about its being the same as Roxburgh's; and, besides, that author, in his drawing, repre- 

 sents the upper valvelet much narrower and more acuminated than in our specimens, but probably his figure 

 may be inaccurate in that respect. The neuter valvelet and both those of the fertile florets are ciliated. 



2. S. chinense ,■ paniculis ovatis contracts, ramis verticillatis subcompositis, gluma 

 infera chartacea, flosculi fertilis valvula inferiori nulla, foliis planis margine hispidis. 

 Roxb. Fl. Ind. I. p. 239; (ed. Wall.) \.p. 244. Corom.3. t. 232. Kunth, En. I. p. 474. 



In nostro specimine (in Benghala superiori a Royleo lecto) panicula magis contracta ; sed distinctissimum 

 valvula inferiori flosculi fertilis omnino deficiente ; superiori autem satis conspicua, lineari-lanceolata, ciliata. 

 Valvula neutra ista major, lanceolata, magis ciliata. Lodiculm cuneiformes, truncate, pro magnitudine 

 spiculce satis magnse. Gluma. glabrae. Vagina suprema ventricosa, in folium angustum attenuatum con- 

 tracta. 



10. HETEROPOGON. Pers. 

 1. H. contortus. P. de Beauv.—N. ab E. Agrost. Bras. p. 363 (cum syn.) ; inLinnaa, 

 7. p. 284.— Stipa spicata. Thunb. Fl. Cap. p. 19. 



Hab. Ad Macao"et in insulis adjacentibus ; Millett ; G. H. Vachell, n. 46 (ex parte). 



[After having carefully examined the Chinese plant, and H. polystachyus R. et Sch., of which last N. ab 

 Esenbeck considers the Andropogon contortus of Roxburgh, and perhaps also of Linnaeus, (but not of 

 Brown's Prodr. 1. p. 201,) to be a variety with solitary spikelets, we can see no good grounds for separating 

 them. Both have the peduncles elongated and often fascicled, each with one short-leaved sheath, the male 

 spikelets ciliated, and the leaves linear scabrous and more or less ciliated at the base : in the Chinese speci- 

 mens the hairs on the glumes and leaves are softer, and spring out of inconspicuous tubercles ; in Dr. Wight's 

 specimens of H. polystachyus the hairs are rigid, and arise from pretty large tubercles, but we can see no 

 other difference between the two ; if then our friend be correct in referring our Chinese plant to H. con- 

 tortus, we should be inclined to reunite to it H. polystachyus. 



11. ERIANTHUS. Rich. 



1. E. trislachyus ; spicis terminalibus 2-5 simplicibus griseo-hirtis, glumis ferrugineis 

 basi margineque griseo-hirtis, inferiori plana rigidaque, flosculi hermaphroditi seta spicula 

 triplo longiore, culmo simplici nodis glabris basi bulboso ferrugineo-tomentoso, foliis 

 linearibus basi vaginisque apice hirsutis. — a ; spicis 2-3, vaginis apice tantum hirsutis. 

 N. ab E. in Wight Cat. n. 1691.— Andropogon tristachyus. Roxb. Fl. Ind. I. p. 256 ; (ed. 



Wall.) I. p. 261 A. trispicatus. Sch. Mant. 2. p. 452. Kunth, En. 1. p. 498 — /3 ; spicis 



2-5, spiculis paullo majoribus, glumarum villis longioribus, vaginis basi et apice hirtis. 



Hab. £ ; In vicinia urbis Macao, inque insulis adjacentibus ; G. H. Vachell, n. 35. 



Ab E. aureo, N. ab E. in Wight, Cat. n. 1690, cui proximus, notis memoratis bene distinguitur. Me- 

 morabilia est bulbus culmi caeteroquin gracilis erecti et simplicis, piso major, densissime tomento ferrugineo 

 tectus. Culmi apex tomentoso-hirtus. Lodicula; carneae, crassiusculae, bidentataa. 



2ll 



