1898 1902. No. 2.] VASCULAR PLANTS OF ELLESMERELAND. 53 



description in the same work, but from the description in the Fl. Alt., 

 it is evident, that here there is no question of any form of P. nivea. 

 As I have had the opportunity of seeing original specimens of Bunge's 

 plant in the Copenhagen herbarium, I can affirm at least, that it does 

 not belong to P. nivea. In the Index Kewensis, it is referred to P. 

 multifida, L., where its right place seems rather to be. As the name 

 P. nivea var. pinnatiflda seems only to be used by Lehmann for the 

 plant of BuNGE (he only has one locality for it in Revis. Potent., p. 169) 

 it is out of the question, and there is the choice between var. penta- 

 phylla, Lehm., and var. subquinata, Lange. As the former is the elder, 

 it must be upheld, there being not the slightest cause to look'upon it 

 as„ a species as Rydberg has done, probably because he has had no 

 opportunity of studying the plant from nature ^ 



Rather might it be justifiable to distinguish all the arctic forms 

 with deeply incised leaflets, from the typical P. nivea, which has them 

 more rounded and feebler dentate. I have, however, had too little 

 opportunity of studying them from nature, to be able to give any de- 

 finite opinion about it. In a collection of herbarium specimens, it 

 appears as if a continual series of intermediate forms existed. Ryd- 

 berg has, however, not only (1. c.) established a new species P. quinqiie- 

 folia for the plant here in question, but he has also put it in a separate 

 group, Suhjugae, together with P. subjuga, Rydb. Between the two 

 species, there is not a single character in common, unless the "pinnate 

 tendency" (Rydberg, Mon. Amer. Potent., p. 37) is regarded as such, 

 which is said to appear therein that the terminal leaflet is petiolate. 

 This characteristic, however, appears very often in the principal form 

 of the P. nivea, and there can be no doubt but that the group Sub- 

 jugae is merely artificial, and has nothing to do with affinity. Perhaps 

 Rydberg has seen arctic specimens of P. rubricaulis, and has referred 

 them to his P. subquinata, (Lange) Rydb., as later he has called it 

 (Furth. Stud. Potent.). That would perhaps account for his arrange- 

 ment. 



P. rubricaulis var. arctica in Ellesmereland is found generally in 

 rookeries and vegetation-covered slopes, where it is loosely tufted and 

 thriving, whereas when it grows in open gravel- or clay-plains, it be- 

 comes stunted and very densely tufted, so as to be rather like P. 

 Vahliana in habit. It begins to flower at the end of July, and is 

 flowering all through the summer, besides bearing fruit abundantly. 



Occurrence. Grinnell Land (?). Hayes Sound and Fram Harbour 

 district: Beitstad Fjord, Skrahng Island in Alexandra Fjord (1373), Fram 



