202 D. Vram— Some new Indian Plants. [No. 5, 



some amendment in the distribution of B. involucrata which, as testified 

 by specimens in the Calcutta Herbarium, is as follows : — 



Typica ; leaves pubescent beneath. 



Upper Gangetio Plain: Moradabad, floZ^eci ! N. Behar : Kurzl 

 Clarke ! Anderson I N. Bengal : Naogaon, Pram I Rangpur, Watt I 

 Assam : Nowgong, 8imo7is I Burma : Minbu Dist., at Sinbok, Gage ! 

 Kyoukmyoung, King^s Collectorl 



VAR. glabra ; leaves glabrous or very slightly pubescent beneath. 

 U. hracteata Roxb. Ic, Ined. 



Upper Gangetic Plain : Gorakhpur, Diithies Collector ! Chota 

 Nag PUR ; Udaipur Tributary State, Prain's Collector ! Vizagapatam : 

 Jaipur State, V. Ball ! East Bengal: Pabna, Clarke ! Jatrapur, Watt ! 

 Assam: Sylhet, Jluls, £foo/<:er^ Thomson! Mann\ Robertsonl Brahma- 

 putra Valley : Gauhati, Jenkins ! Mann ! King^s Collector ! Golaghat, 

 Masters ! Sibsagar, Masters ! Manipur : Imphal, Watt ! Burma : Myit- 

 kyina, Pottinger ! 



To this variety also belongs Wall, Cat. 696 "E montibus Pundooa 

 Sylhet confinibus " which, as Hooker says, is most probably also from 

 Sylhet. 



VAR. parvifolia ; leaflets as in var. glabra^ but less than half the 

 size, (not exceeding '5 in. in length). R. palustrts Ham. MSS. 



N. W. Himalaya : Kamaon, Blinkworth ! Nepal, Wallich ! Behar : 

 Nakeswar, Buchanan I Chota Nagpur : Gamble ! /. /. Wood ! 



The diagnosis between B. involucrata and B. Lyellii is very simple, 

 and may be given as follows: — 



Rambling shrubs, never climbing on banks of slow running rivers 

 below high-water mark or in the beds of jheels and back- 

 waters ; bracts pectinate; flowers sessile ... ... involucrata. 



Climbing over trees on dry ridges j bracts entire ; flowers long- 



pednncled ... ... ... ... ... Lyellii 



In other respects the two roses, as Sir Joseph Hooker truly says, 

 are exceedingly closely allied. Orepin hazards the suggestion that B. 

 Lyellii is a natural hybrid between B. involucrata and B. moschata, a sug- 

 gestion rather difficult to admit even if B. Lyellii had been, as was long 

 supposed to be the case, confined to Nepal. The distribution of B. 

 Lyellii^ which is given below shows, however, in the writer's opinion, that 

 Crepin's hypothesis is untenable. 



Rosa Lyellii Lindl. Monogr. Bos. p. 12, tab. 1 (J820). B. pubes- 

 cms Eoxb. Fl. Ind. i. 514. 



SiwALiKS : " native of the mountains north of Rohilkand" Boxburgh 

 (Ic.) \ Boylel Nepal: in woods, climbing on trees, Wallich I Sikkim : 

 Kurz I Central India : Sagor, Vicary I S. India : Nilgiris, Schmidt ! 



