38 THE BRITISH ROSES 
me all over an inch long, with a dilated leafy point, the largest 
entirely ee of scars and acicles. ies comes nearest the 
Belgian [R. coronata] var. subnuda Crép., but that has lades 
race as densely glandular + taiiaatth as in the Rubiginose.” 
Webb’s specimen from Hoylake, which is Baker's type, has 
its main ee short, declining, and subulate rather than acicu- 
late, with very few small scattered acicles. Leaflets of moderate 
size, oval, subobtuse, glabrous above or very nearly so, thinly 
scattered hairy beneat , and scabrous or somewhat glandular on 
midrib, fully but not het? strongly ecard fy teeth quite 
men in Déséglise’s herbarium from the same station differs in the 
sepals, only some of which have an occasional very short pinna, 
which makes me think that the occasional strong pinnation 
noticed by Mr. Baker is abnormal. 
his variety is very near var. levigata, and if it be allowed that 
the sepals on the type specimen are abnormally developed, the 
I of 
present variety. think both seem better placed under the 
fi. eu-canine x pimpinelitfolie than in the present group, d 
not feel sure enough to displace them. The biserrate hairy leaflets 
could of course come Rip. some species allied to R. canescens 
oval or suborbicular, mibvinesst or ve ly 
leaflets having two or three teeth with denticles. They are quite 
glabrous above, hairy on midrib, an 
Rosa Witsont 
Borrer in Hooker, British Flora, ed. 3, p- 231 (1835). 
‘Prickles crowded, roe a straight, intermixed with set, 
ser. dise diem ndular, calyx simple, 
of slender habit, well 
