ROSA INVOLUCRATA 
is an important one nevertheless, and Col. 1 ). Prain {Join'll. As. Soc. Beiii^. 1904, 
Ixxiii. p. 202) refers Rosa iiivoliicyafa to three varieties (which j^erhaps meets the 
case), but to these should be added Rosa Lyellii Lindl., an even still more moun- 
tainous form, their representative in the lower N.W. Himalaya and from thence to 
Rajputana and South India. Where met with the plants usually regarded as Rosa 
iiivoliicrata are plentiful enough, but between one locality and another a gap of 
many thousand square miles may interpose, over which the plant seems to j)ossess 
no inclination to spread. For e.xample, on passing North-Fast from Silhet it 
disappears, and on the road from thence vid Cachar to iVIanipur, a distance of over 
120 miles in a direct line, although it traverses many situations in which Rosa 
iiivoliicrata might luxuriate, it is nowhere met with until the valley of Manipur 
proper is reached, when, at altitudes of from 2,500 to 4,000 feet, what I have 
suggested as being possibly the Chinese form of the assemblage is found not only 
plentiful but one might almost say characteristic. This sudden appearance and 
disappearance, in Eastern Bengal, is that which the species everywhere manifests 
through its Indian area.” 
The plate shows a plant known in gardens as Rosa clinopJiylla. 
It is a garden form of Rosa involucrata in which the character of toothed 
stipules and bracts has almost disappeared. 
