]W2 THE JOUHXAI. OF BOTAITY 



region, whence it may be assumed to have come, except perhaps 

 that of Klein, requires to be more minutely described. 



The leaves are mostly oblanceolate-rhomboid or narrowly obovate, 

 most of the secondary nerves are sub-furcate ; stipules 3 mm. long, 

 the lowest filiform, but the youngest more linear and therefore 

 more approximating to, but not agreeing with, those shown in 

 Roxburgh's drawing. The specimen does not very closely resemble 

 the drawing, but the main differences between the drawing and the 

 specimen may also, except in the case of the stipules, be found 

 between the drawing and Roxburgh's description ! Although the 

 drawing only shows a single lowest leaf somewhat obovate, the 

 description says " leaves long obovate " and in another place " leaves 

 obovate or wedge-formed " as in the Central Provinces plant ; but all 

 the shapes of the leaves figured could, 1 believe, be found in the Central 

 Provinces plant. There is at first apparently an important differ- 

 ence in the female disc. This is figured as embracing the ovary and 

 reaching three-fourths of the way up and is crenate ; but it is 

 described as exactly as in Cluytia ( = Cleistanthvs) patula. In 

 G. patula, however, the female disc is eupular and deeply lobed. In 

 the Central Provinces plant the disc is eupular and lobed, but in some 

 very young flowers I have found it more closed and crenate as in the 

 iigure. The female tiower of Wallich's B. Hamiltoniana 1 have not 

 dissected, being unwilling to mutilate a type-specimen. The stipules 

 only remain as an important difference, but these are deciduous on 

 some twigs in our specimens, and the youngest, as said above, some- 

 times approach more nearly to Roxburgh's figure ; moreover, Roxburgh 

 employs the term ' withering ' not 'falling' (as in Cluytia colli na) 

 or 'caducous ' (as in B. crenulata) and the possibility of bad drawing 

 has also to be taken into account as to shape. Finally, Roxburgh's 

 plant cannot have been very rare, as he speaks of it as occurring in 

 both the mountains and the lowlands, and it bears a vernacular name; 

 it would be strange if such a species had never been collected since. 

 1 think we should take the name of B. Montana for the Central 

 Provinces plant at least until further material shows us to be wrong, 

 and that B. Hamiltoniana and the Concan plant should be looked 

 upon as separate species or varieties according to the views of 

 individual botanists. 



The following is a summary of the conclusions arrived at: — 



Bridelia montana Hook. f. is not B. montana Willd., but is a 

 mixture of this and another species called by me B. verrucosa. 



Bridelia montana Willd. is a widely distributed species, including 

 the following varities : — 



Var. Hamiltoniana— B. Hamiltoniana Wall, in Fl. Br. Ind. (in 

 part). Habitat — Karakpur Hills with a glabrous form in the Kaimur 

 Hills, both in the province of Bihar and Orissa. 



Var. glabra = B. Hamiltoniana var. glabra Muell. Arg. = 

 B. Hamiltoniana in Fl. Br. Ind. (remaining part). Habitat— 

 Concan, Bombay Presidency. 



Var. Stafifii, var. nov. Habitat — Sambalpur district in province 

 of Bihar and Orissa and Central Provinces. 



Of these three varieties the first two may perhaps form one species 

 as defined in the Prodronius and Fl. Br. Ind. and the third another. 



