STUDIES ON THE VEGETATION OF ICELAND 19 
regions. For we cannot take it for granted that a species, because 
it has come to a certain region, will at once assume the most pro- 
nounced life form of the region in question. The dominating life 
form in arctic regions is Ch, yet we find species with another life 
form, and these species must be supposed to be just as well adapted 
to arctic regions as several of the chamaephytes growing there. 
Conversely, in regions having a hemicryptophyte climate, we may 
find Ch. A species like Calluna vulgaris must be assumed to be 
just as well adapted to the Danish climate as many hemicrypto- 
phytes. Even though Raunkiær’s life forms give an ex- 
cellent biological picture of the climate, they do not 
give the most accurate picture of it. The distribution of the 
species must rank first in this respect. However, there must be no 
disagreement between results obtained by means of Raunkizr’s life 
forms and those obtained by investigations based on the geographical 
distribution of the species, though at the outset we must expect 
the latter to give the more pronounced result. Raunkiær’s life 
forms picture the morphological adaptation of the spe- 
cies, the species groups give expression to their physio- 
logical adaptation, both, however, express their adap- 
tation to the same life conditions. 
Various enquirers in various countries have at different times 
worked out a classification of the floras of their respective countries. 
Thus Watson divided the British flora into groups according to the 
geographical distribution of the species. A. Blytt divided the Nor- 
wegian flora on a somewhat similar principle, and various Swedish 
enquirers, i.a. Gunnar Andersson and H. Hesselman, have clas- 
sified the Swedish flora. Ostenfeld has classified the flora of the 
Faeroes, Porsild and Ostenfeld the Greenlandish flora. The 
principle acted upon by all these enquirers was to separate all spe- 
cies haying a pronounced northern or southern distribution from all 
such as showed an equal distribution over the entire area. This 
resulted in the first instance in the setting up of three groups, while 
Watson and Blytt in addition classified the species according as 
they had a pronounced Atlantic or continental distribution. 
An attempt to bring together the species groups of the various 
authors and thus arrive at a classification of the Icelandic flora 
proved impossible, since their species groups overlap to such a de- 
9% 
