413 Jour., Bom. Nat. Hist. Soc, Vol. XXXII, No. 3. {Jan. 15, 1928. 



based hairs, spinulosely serrulate ; sheaths glabrous, lower ones compressed ; 

 ligule very short, villous. Panicle 7-13 cm. long, contracted, lower branches 

 long, few in a whorl, rhachis and branches minutely hairy ; spikes solitary, green 

 or brownish. Sessile spikelets subcylindric, 4 mm. long, callus long, densely 

 bearded with rusty hairs all round. Lower involucral glume laterally compres- 

 sed above, minutely truncate, glabrous below, hispid above, obscurely 4-nerved, 

 tip 2- dentate, upper chartaceous, hispid above on the keel and sides, tip 2- 

 lobed, awn as long as the glume or shorter. Lower floral glume linear-oblong, 

 2-nerved, ciliate, upper consisting of an awn with a narrowly dilated 2-lobed 

 base, awn 50-65 mm. long, column hispid. Pedicelled spikelets nearly 12 mm. 

 long, lanceolate, pubescent ; pedicel truncate, margins shortly villous. Lower 

 itrvolucral glume glabrous or pubescent, 7-nerved, awn longer than the glume, 

 keels ciliate, upper lanceolate, 3-nerved, awa as long as the glume or shorter. 

 Lower floral glume oblong, 2-nerved, ciliate, upper very narrow, ciliate* 

 nerveless. 



Locality : S. M. Country : Castle Rock (Bhidel). — N. Kanara : Jog to Sid- 

 derpur, open grass land (McCann A273!). 



Distribution : Madras, Nilgiris, Burma, Assam. (Hackel mentions a speci- 

 men gathered in Ceylon but, according to Hooker f ., it seems to be a starved 

 specimen of Chrysopogon zeylanicus, Thw.). 



6. Chrysopogon montanus, Trin. in Spreng. Neue Entdeck. ii, 93; Haines Bot, 

 Bihar and Orissa 1037— Andropogon monticola, Schult. Mant- (1824), 665 - 

 Kunth Enum. PL i, 506 ; Steud. Syn. Gram. 395 ; Plack. Monogr. Androp. 557 

 (excl. var. velutinus) ; Hook f. in F. B. I. vii, 192, cum omnibus var-: Cke- ii, 

 985 ; Hole in Ind- Forest Mem. i (1911), 108—^. Sprengelii, Kunth Rev- Gram. 

 166.— Pollinia fulva, Spreng. PugilL ii, 93— Andropogon Trinii, Steud- Syn, 

 Gram 395 ; Hack. Monogr. Androp- 558. — A ciliolatus cceruleus et 

 increscens, Steud. 1. c— Chrysopogon ciliolatus, Boiss, FL Or. v. 458 (excl. van 

 Aucheri, Boiss); Duthie Grass. N. W. Ind- 22-— C u?ruleus y Duthie 1. c. 23, 

 Fold. Grass. N. Ind. 39, t. 60— C. increscens, Nees ex Steud. 1. c 396— C. 

 Wightianus var. leucanthus, Thw- Enum. PL Zeyl. 366. — C. serrulatus, Trin. 

 in Mem. Ac. Petersb. ser. 6, ii (1833), 318, Spec Gram. t. 331— C. Esenbeckii, 

 Arn- in Steud. Syn. 395— Raphis ciliolata et cosrulea, Nees ex Steud- Syn .396, 395. 



This synonymy requires an explanation. Chrysopogon montanus, Trin. as 

 understood in this place comprises Hackel's two species Andropogon monticola 

 Schult. and A. Trinii, Steud., and is identical with Hook, i-'s- A. monticola, 

 Schult. with all its varieties. 



Hackel has two species and he distinguishes them by the following characters : 



A. monticola : Upper involucral glume of sessile sp'kelet keeled, the keel 

 from the base up to f-f of its length densely pectinate ciliate with long, rigid, 

 rufous hairs, shortly white hispid in the upper \. 



A. Trinii: Upper involucral glume of sessile spikelet keeled below the apex 

 only, keel white-ciliate, the lower f-f not keeled and glabrous. 



Hook. f. in F- B. I. makes of these species two varieties : var monticola proper 

 and var. Trinii, and includes them under A. monticola, Schult-, adding a third 

 ariety : var. robustus- 



At the same time Hooker confesses: ' I am unable to classify the varieties of 

 this common and variable plant in accordance with geographical areas or 

 ether considerations. This, if possible, must be effected by field-botanists in 

 India. There is every gradation from the coarsely hirsute keel of monticola* 

 to the perfectly smooth of some states of Trinii; from the awnless to long 

 awned gl. I of the pedicelled spikelets, and from the glabrous to the pubescent 

 of the same organ ; the colour of which affords no character ; nor does its 

 length, or that of the cilia on its keels. ' 



Cooke (ii, 985) has adopted the name/i. monticola, Schult- with Hooker's 

 description and evidently also the latter's varieties. But his opinion does not 

 count in this case as he has not seen any specimens from the Presidency and 

 ' was therefore ' as he says himself, ' unable to fix definitely the variety to 

 which the Bombay species belong. They will probably belong to var. Trinii 

 H- f .' What induced Cooke to say that they probably belong to var. Trinii we 

 cannot understand, especially as Hooker came to the conclusion that he was 

 not able to classify the varieties according to geographical areas. 



We have examined a great number of specimens from all parts of the Presi- 

 dency, except Sind, Cutch and Kathiawar and we have been able to separate 



[6] 



