• • 



es 3. Bisei^ual spikelets cylindric, obtuse, with long brovra 

 illy rounded, ripe polished dark brown, hispidulous tON^ 



212 CLTXiii. GUAMiXEJS. (J. D. Hookcr.) [AntJiistma. 



sc^brirl towards the tip only, callus long pungent. Kunth. -Enum. Pi.i. 

 481; Steiid, Syn. aram. 401. A. Forskahlii, Kunth Bevis. Gram. i. i^j 

 Enum I. c. 483 ; Stend, L c. 402. A. vulgaris, Hack, in Engl. Sf PranU. 

 Naturl Vlanzenf. ii. II. 29. Themeda triandra, Forsk. FL ^gyptArab. 

 178. T. poly^ama, Gmel. Syst. 149. T. Forskahlii, HacTc, Monogr. 

 Androp. 659 ; Duthie Fodd. Grass. N. Ind. 43.— Anthistiria, Wall. Lox. 

 n. 8764, A.C. E. F. 



Hotter drier parts of India, from Upper Bengal to Travancore, aseending the 

 Himalaya to 3-4000 ft. (Var. Roylei to 10,000 ft.). BuRHA. Ceylon, ascending w 

 4000 ft.— DisTRiB. Warm regions of the Old World. , 



Sletn 1-6 ft., stout or slender, subsimple or branched. Leares 3-10 m., linear 

 rather rigid, glabrous or sparsely hairy, green or subglaucous beneath ; Ug 

 short, ciliolate. Head of fascicles of spikes few or many, long- or s^^^^'P .^f 

 eled ; spathes glabrous or ciliate with usually very small tubercle-based na«^» 

 proper spathes longer than the spikes, cymbiform, tips acute or subacute, /ny^' ' 

 *p/*<r/e«* ^ in., oblong-lanceolate, acute or acuminate, red brown, ^^^^f^J^y^^^^^J, 

 ciliate or glabrous, glumes 

 hairs ; gl. I terete, dorsally 



the tip only ; awn J-2 in, long, base capillary. Pedicelled spikelets very ^ 

 glabrous or ciliate with a few tubercled- based hairs, keels ciliolate. — A very var 

 grass, some states of which are with difficulty disting:uishable from A. ^^^*^, ' ij, 

 earliest names of this plant are The^neda triandra, For^k., and Anthistiria *»*^^ ' 

 Rc'tz. Hrtckel has abandoned both, substituting first Anthistiria vulgans,^^^ 

 Themeda Forskahlii, on the ground that A. imberhiB was perhaps not ^?^^^* ^ 

 T. triandra (why then call it Forskahlii 1) of which no type specimen exists, a ^ 

 because triandra indicates a character of no individual value in grasses. ^^ 



doing he overlooked Gmelin's* name of T. polygama (Syst. 149). Having rega^^ 

 to the wide range of A. imherbis, from Australia to Africa, its presence m A" 

 might well be anticipated; and that it is a native of that country is »<>* P^ ^jj^ 

 by Schweinfurth's finding Hackel's var. glauca in that country. This ^^^^'^|j|.g 

 var. (which is local, and not Indian) the type of the species, and, ^^ ^^^^ties. 

 name of Themeda is to be retained, necessitates a rearrangement of the ^^* 1^^^ 

 To me it appears most expedient to retain Retz's name which applies to the pr® 

 form over the area of distribution as the specific one. - ^e 



The following is Hackel's arrangement of the Indian forms which maj^^^ 

 variously reduced or added to. It has the great merit of systematizing^, 

 authorities fairly well. Of these I have kept up var. laxa as n species, lol o ^^ 

 Auderason, though not without doubts. Of others, as A. polystachyaf .^^ 

 doubtful, judging from Roxburgh's description and its being an annual, ^.^^^^joly 

 may be a state ofciliata. Andersson included it under A, arguenSj which itcer 



• sum 1-3 ft. Leaves 2-8 by ^i^-| in. Panicle 6-12 in., nodes 2-3.^^ ^^^ 



branches simple below. Gl. I of invol. spikelets not winged or narrowly 



side only. . 4, 



Var. vulgaris. Hack. I.e. 660; spikelets with many tubercle-based hairs. ^ 



oiliata, Setz. l.c.{non Linn.f.); Lamk. III. t. 841, f. 2 ; Car. Jc. *• ^r'/igSd), 

 Fl. Ind. i. 217; Wight. Cat. n. 1708; Anderss.in Nov. Act. Sc. Upsal.^^- K p 

 in Walp.Ann. vi. 1057; Grah. Cat, Bomb. FL 239; Dalz. Sf Gths. -^^'"jQg; 

 304 ; Thw. Enum. PI. Zeyl, 366 (eoJcZ. var. major) ; Trim. Cat. CeyL f ^- j>|, 

 Benth. FL Austral, vii, 542. ? A. polystachya, Roxb. I. c. 248; Kunth is-^^ * 

 i. 481 ; Steud. I. c. nariog^I 



Var. iinherbis, Hack. 1. c. 661 ; invol. spikelets glabrous or very sp* ^^^ 

 hairy. A. ciliata var. imberbis, Nees in Linneea vii. (1832) 284. A. ciiia*^^^^ 



• Gmelin has, by oversight probably, assumed the first word of Forskabl « 

 diag-oosis to be hi» specific name. 



