19U-15.] BOTANICAL SOCIETY OF EDINBURGH 323 



poeuliformis, Hook, f., from specimens grown by Messrs. 

 A eitch at Chelsea from Maries' seeds. Hooker says of it : 

 " Probably a very variable plant. The earliest flowering 

 specimens sent by Mr. Veitch were less hairy and had 

 rounded and nearly entire leaves." It would appear then 

 to be likely that Maries collected both the nearly entire 

 leaved form which is now, I think, the dominant one in 

 cultivation, and also the form with more oblong and lobed 

 leaf which is. so far as I know, not so common. 



I add also in Plate XXXVI a figure of specimens in the 

 Paris Herbarium collected, in February 1869, by David, on 

 rocks in the Province of Moupin. a region far north of 

 Hupeh. The plant has some resemblance to that shown in 

 Plate XXXV. but the leaves are larger, more oblong, and 

 lobed. and then the plant is densely villous, with long 

 reddish hairs. Why I include specially this figure is that 

 the specimens shown are amongst the type-specimens 

 upon which Franchet founded his variety (a.) hispida of 

 P. obconica. Hance. On p. 313 I quoted from Franchet 

 the types of his variety (a) hispida. Through the 

 courtesy of the Director of the Botanical Department 

 of the Jardin des Plantes, Paris, I had the privilege of 

 examining three of the types named — that collected in 

 Moupin by David, Delavay's Xo. 317 bis from Szechwan. 

 and Delavay's Xo. 317 from Ichang — and I have figured 

 them in Plates XXXVI, XXIII. and XXXV. The fourth, 

 which I have not seen, from the Paris Herbarium, is a 

 plant collected by Consul TVatters about Ichang (see what 

 I say on p. 313). and is therefore probably a portion of the 

 collection which furnished Hance. in the first instance, 

 with material for the description of P. obconica, Hance 

 (Plate XXI). They show that Franchet allowed wide 

 latitude to variation in his variety (a) hispida within 

 P. obconica, Hance. I make no apology, therefore, for 

 segregating the type-specimens he included in his P. 

 obconica, Hance, var. (a) hispida. Delavays Xo. 317 l>>$ 

 I place, as I have already said, in the typical P. obconica, 

 Hance. and Watters' specimen I place here also. Delavays 

 Xo. 317 and David's Moupin plant alone could merit the 

 designation hispida. but the character is hardly diagnostic 

 of a microform. 



