34 A LIST OF BRITISH ROSES 



though I do not know how to name it, but I do not think any of 

 the Dovedale forms belong to B. casia. V.-c. Perth, 98. 



E. coRiiFOLiA var. subcollina Christ, Eos. Schw. p. 191. 

 Dingier refers quite a considerable number of examples to this 

 variety, or to forms of it. It should be characterized by refiexed 

 sepals and uniserrate leaflets, narrowed at base, and hairy only 

 on primary nerves beneath. Most of the specimens named by 

 Dingier have spreading-erect sepals, and fit better under B. frute- 

 torum. An Elgin specimen (Armitage), referred here by Dingier, 

 has the sepals refiexed, but has its leaflets densely hairly all over 

 beneath, and not narrowed at base; its peduncles also arelongish. 

 Specimens from Mid Perth {Barclay) may be a little nearer 

 Christ's variety, but are not at all satisfactory. V.-c. 88?, 95'?, 98?. 



E. TOMENTOSA var. OBOVATA Baker, Monogr. p. 218. This, as 

 stated in B. E. p. 98, must fall into the present group, but the 

 type specimens are in too young a state to determine the direction 

 of the sepals. Ley says they are reflexed and caducous, but has 

 almost certainly been misled by a false specimen. A plant from 

 Garve, E. Eoss {Marshall), approaches this variety, but has 

 slender arcuate prickles instead of stout uncinate ones ; its pedun- 

 cles also are much longer .than they should be for this group. 

 Sudre thought the Garve specimen a very curious form of 

 B. tomentosa, to which it perhaps belongs rather than here. 

 V.-c. 66. 



Leaflets Biserrate. 



E. suBCORiiFOLiA Barclay, in Ann. Scot. Nat. Hist. July, 1899, 

 pp. 172-179. This is a somewhat aggregate species, as I believe 

 its author intended it to be, but as such it is fairly well defined. 

 Mr. Barclay has very kindly supplied me with a considerable 

 series of it, all from Perthshire. Tlie specimens all agree in their 

 strongly biserrate leaflets and obovoid or elongate-ellipsoid fruit, 

 but the leaflets vary much in hairiness and in subfohar glands. 

 The glands are mostly few and inconspicuous, and confined to the 

 primary nerves, but they are sometimes as numerous and as 

 scattered and conspicuous as in B. micrantha. _ One of the 

 specimens has quite glandular-hispid peduncles, the setae extend- 

 ing to the base of some of the fruits. The sepals are, as a rule, 

 long and strongly reflexed, but sometimes spread or even rise 

 above the disc. The latter forms run very near var. incana, as 

 indeed many of them do. V.-c. Perth. 



E. c^siA var. incana Borr. in Hook. Brit. Fl. ed. 3, p. 242. 

 Mr. Barclay tells me that all his specimens at S. Kensington, 

 which I mentioned under this variety in E. p. 107, are his 

 B. subcoriifoUa, so that the only authentic specimen I have seen 

 is the very unsatisfactory cultivated one from N. Mimms. I can 

 only say of the variety that I consider that Borrer's description 

 covers B. subcoriifoUa, and that I think that many of, if not all, 

 the examples of the latter may be referred to it. It is impossible 

 in the genus to restrict species and varieties to their desciiptions 

 verbatim, so that the "spreading widely or even recurved" sepals. 



