1 86 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



the writer may not agree with either. In adopting the sections pro- 

 posed in that monograph the writer must not be held as approving 

 of that arrangement in all respects, for he feels quite convinced that 

 the grouping might be considerably improved in not a few cases had 

 one the necessary time to devote to it. 



In the section Sinenses there are enumerated six Indian species : P. 

 filipes, P. Listen, P. mollis (fig. 87) , P. geraniifolia (fig. 88) , and P. vaginata 

 the East Himalaya, and P. Clarkei from Kashmir — all of which are from 

 included in Watt's section Petiolaris. The Eastern plants, with the 

 probable exception of P. Listeri, would appear to be rather local and 

 rare. P. filipes is still represented by the unique type-sheet only, and 

 P. mollis, though well known in cultivation, had not been collected 

 since Booth's time until 1905, when J. C. White, then Political 

 Officer in Sikkim, found it in the Bhutan highlands. Comparatively 

 recently, P. obconica — or it may be a species very closely allied to 

 it — has been found in Upper Burma, this being the first record of its 

 occurrence in India. Before leaving this section it may be pointed 

 out that Pax and Knuth denote as having been seen by them a 

 specimen of Cortusa Matthioli from Sikkim. This statement is 

 probably based on a fruiting specimen of P. geraniifolia collected in 

 Sikkim by Hooker and distributed by him as a Cortusa. 



The section Floribundae, Pax, includes the same two Indian 

 species as Watt's section Floribunda : P. floribunda and P. 

 Lacei. 



In the section Petiolares, Pax — which is used in a much more 

 restricted sense than Watt's section of the same name — there are 

 included four Indian species : P. petiolaris, P. Edgeworthii, P. Hookeri, 

 and P. Tanneri. In his treatment of P. petiolaris Watt agrees with 

 that adopted in the " Flora of British India," with the single exception 

 that he retains P. nana as a distinct species. Pax and Knuth, on 

 the other hand, accept all the varieties of the " Flora of British India " 

 except Edgeworthii, which Pax regards as a distinct species. Since these 

 revisions the variety pulverulenta has been brought into cultivation 

 and is known as a distinct species under the name P. Winteri (fig. 89). 

 This plant has been exhibited several times at the Royal Horticultural 

 Society's meetings, and is so well known in horticultural circles that 

 it is necessary here to point out only that P. Winteri is synonymous 

 with P. petiolaris var. pulverulenta, Hook. f. In the writer's opinion 

 P. Edgeworthii and P. Winteri are quite good, valid species, and 

 should not be regarded as varieties of P. petiolaris. But quite as 

 distinct as the two just mentioned are the varieties sulphur ea, Hook. f. 

 and Stracheyi, Hook, f., which the writer would regard as distinct 

 species. The former, which he would name P. sulphurea, is readily 

 distinguished from P. Winteri by the smaller flowers and leaves and 

 by the colour of the farina (from which the name was derived). Un- 

 fortunately the name Stracheyi is already occupied in the genus, so 

 for the latter the writer proposes the name P. Drummondiana, in 

 compliment to Mr. J. R. Drummond,w1io by placing his copious material 



