44 H. G. SIMMONS. [sec. arct. exp. fram 



American Archipelago, really have reference to broad-leaved forms of 

 D. integrifoUa. Among the many specimens that I have seen in going 

 through the collections of the museums of Copenhagen, Stockholm, Lon- 

 don and Kew, there was not a single D. odopetala from this great 

 area. First further south-vs^estward, in the Rocky Mountains and Alaska, 

 this species again appears. 



The great difficulty of detecting sharply-defined characters for the 

 distingushing of these two specimens, is clearly to be seen from litera- 

 ture. Authors who only have seen D. integrifoUa in western Green- 

 land, or know it from specimens from that country, are not in doubt 

 about its being a good species, but in floristic works about regions 

 where both are found, the author either puts D. integrifoUa as a sub- 

 species under D. odopetala, or he asserts his opinion that the former 

 is a good species without however being able to give a clear exposition 

 of the differences. Hartz, Fan. o. Karkr., p. 320, who has devoted 

 special attention to the Dryas-iorms of north-east Greenland, Dusen, 

 Gefasspfl. Ostgronl., p. 13, Kolderup Rosenvinge, 2 Till., p. 654, for 

 instance, consider it as a sub-species, whereas Hooker, 1. c. I, p. 174, 

 Nathorst, N. W. Gronl., p. 24, Kjellman, As. Beringss. Fan., p. 527, 

 and others, look upon it as a separate species. I must decidedly join 

 with the latter, even if I must admit that the differences between the 

 two species are only relative. 



In its most typical form, D. integrifoUa has leaves with teeth only 

 at the nether part of each edge, and entire for the rest; also leaves, quite 

 without teeth, are found, but besides also such as are dentate along the whole 

 margin. The teeth are however mostly smaller than in D. octopetala and, 

 as a rule, more pointed. The strongly involuted margin of the leaves is 

 indeed characteristic of D. integrifoUa, but in especially shady localities, 

 such as between great boulders, or in clefts of rock, etc., forms with 

 entirely flat leaves are also often found, and as these are besides mostly 

 stronger dentate, the resemblance to D. octopetala becomes rather 

 great, and the specimens are easily taken for that plant, as Hart and 

 Nathorst have done. On the other hand, there are also D. octopetala- 

 forms with leaves as narrowly involuted as in the typical D. integri- 

 foUa; such I have seen, for instance, from Novaja Semlja, collected by 

 Th. Holm. Much more usable is another character, which is also ob- 

 served by Hooker (1. c, I, p. 174), viz., that the veins in D. odopetala 

 are distinctly conspicuous on both sides of the leaves. The upper sur- 

 face then gets a certain corrugated aspect and becomes lustreless, 

 whereas in D. integrifoUa, where they are only a little or not at all 



