142 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



than are now available ; here I can only tell you that : — P. ambita, 

 Balf. fil., is a glabrous form from a very dry site, with a remarkable 

 involucre. P. barbicalyx, C. H. Wright, has, as its name implies, the 

 calyx bearded, but the plant is very hairy all over, and has somewhat 

 elongated leaves with rounded lobes. These characters are more 

 emphasized in P. oreodoxa, Franch. This plant is the true P. 

 oreodoxa, Franch., and is not in cultivation. The plant often met 

 with in gardens under the name is P. saxatilis, Kom. P. begoniae- 

 formis, Petitm., is a smaller form than the type and less hairy. 

 P. Bonatii, R. Knuth, is not, I believe, a form deserving of a name — 

 it shows no distinctive characters, but I hesitate to suppress the name 

 without more knowledge of the plant. P. parva, Balf. fil., is a re- 

 markable dwarf, small in all its parts, but characteristically xero- 

 morphous. P. Vilmoriniana, Petitm., has leaves twice or thrice the 

 usual size, associated with short scapes and minute flowers, the whole 

 plant very hairy. P. Petitmengini, Bonati, is a grotto plant, also with 

 large leaves, which are membranous and delicate, and the scapes are 

 very short. P. Cavalieri, Petitm., shows an extreme type of the grotto 

 plant, all its parts soft and the lobed membranous leaves, on long 

 petioles, delicately thin. I have left to the last P. sinolisteri, Balf. 

 fil. This is a plant of special horticultural interest. It was introduced 

 in 1908 by Bees, Ltd., grown from Forrest's seeds, and promises 

 to be in our gardens what P. obconica, Hance, ought to have been, but 

 is not. P. sinolisteri, Balf. fil., has not the irritant hairs. It is a 

 free-grower, forming compact masses of dark green, acutely lobed 

 leaves, and the trusses of white (sometimes lilac) flowers are many. 

 In our northern climate it is not quite hardy — like true P. obconica, 

 Hance, in that respect. It was sent out as P. Listeri, King — a venial 

 error of naming — and the name sinolisteri has been given in the hope 

 of making the change of nomenclature less disturbing. P. sinolisteri, 

 Balf. fil., is, I am sure, a plant of horticultural merit. It has, 

 moreover, particular botanical interest because it is the form in the 

 whole series of which I am writing which shows conspicuous acute 

 lobing of the leaf like that in the Himalayan type P. Listeri, King. 

 That Himalayan type is in contrast with the Chinese P. obconica, 

 Hance, a plant of wood and shade, but it too varies. The Manipur 

 and the Kumaon forms of it differ markedly from the Sikkim, and 

 the plant known as P. filipes, Watt, has the form of leaf character- 

 istic of the typical Chinese P. obconica, Hance. Different as the 

 Ichang P. obconica, Hance, and the Sikkim P. Listeri, King, appear, 

 there is a wonderful series of uniting forms. Long ago Sir Joseph 

 Hooker recognized this, as did Franchet also at a later period 

 when he definitely included a number of the Chinese forms in P. 

 Listeri, King. His example was followed by Forrest. But there 

 is one character that may point to a fundamental difference between 

 the Himalayan and Chinese plants — the former gives off an odour 

 like that of our Geranium Robertianum — so strongly indeed that 

 Sir George Watt, the discoverer of the species, proposed to name 



