The Flora of Greenland and its Origin. 5 



in the delimitation of the species. By experimental in- 

 vestigations we may get to the point where we have only 

 to deal with the actually smallest systematic units ; this, how- 

 ever, is out of the question here, where the material con- 

 sists of dried specimens, or specimens preserved in spirits. 

 It is, therefore, inevitable that some difi'erence should be 

 apparent between the views of the various writers as to 

 what should be considered as a species. True, in most cases, 

 the species are taken more or less in a Linnean sense; 

 but in certain genera (e. g. Taraxacum, Hieracium), the minor 

 systematic units are so well distinguished as to be recognis- 

 able even in preserved material. In other genera, especially 

 those where hybridisation is presumably of frequent occur- 

 rence (e. g. Draha, and to some extent also Potentillci), we 

 must needs be content with larger and more capacious 

 species. 



This is, of course, unfortunate, inasmuch as it gives rise 

 to a lack of uniformity in treatment which renders com- 

 parison difficult. Furthermore, it often means the effacement 

 of features that might otherwise have presented valuable 

 indications as to the course of migration of a species, as 

 the more comprehensive species, such as Drapa alpina for 

 instance, have on the whole very wide areas of distribution, 

 and it is likely that, once clear as to the minor systematic 

 units composing such a collective species, we might be able 

 to say whence the form occurring in a given area originated. 

 As it is, this can only be done in rare cases. For the present, 

 we must be content with very wide limits for our species in 

 most of the genera or groups. 



And again, even though we may, in the case of a single 

 large area, have reached a step further in the distinction of 

 species, the greater accuracy thus obtained has to be re- 



I 



HHijgirrtrrrillflSI 



