A siirvey of the Phytogeofrrapliy of the Arctic American Archipelago J43 
such possible survivors only among the higharctic and most hardy species, but 
these again are generally circumpolar and thus not apt to give any evidence for or 
against the liypothesis of persistence. 
We now have to see if there exists any possibihty of proving, or at least of 
proferring arguments for, the existence of such preglacical rehcts in the precent 
flora of the Arctic Archipelago. If any plants had survived the iceage in the 
Archipelago they ought, of course, in the tirst line to be sought for amotig the 
most distributed species, as they should, even if they had persisted oiily in a siugle 
place, had the comparatively longest time for spreading. But all species of wide 
range over the islands we find in tlie ubiquituous groups. Most of these are enti- 
rely circumpolar plants, and among tliem we find the most higharctic of all species 
which may, indeed, be thought of as possible survivors. Nothing, however, directly 
speaks in favour of their having lived here during the whole iceage, and as some 
species of wide range in America and showing a more or less northeru distribution 
in Greenland, such as Alopecurus alpinns, Braya purpurascens, Bry as integrifolia 
Saxi fråga Jlayellaris, and S. Hirculus are equally common in the Archipelago, this 
points in another direction. For if we assume the circumpolar species to Uve here 
as relicts of the preglacial vegetation, we can hardly but take the same view of 
the american ones now being equally higharctic and common to the north, but 
then again there is no reason why they should not have lived in the icefree träets 
of northern Greenland also, and thus have had time enough to spread all over the 
latter land. 
A very important point in tlie composition of the flora of the Archipelago 
decidedly speaking against an assumption of elements in it which have there survived 
the iceage, lies in the absence of even a single peculiar type or endemic species. 
Foa evagans cannot come into consideration as such, because it will most probably 
yet be sliowti to have a wider range, and also because it may have develo|)ed in 
postglacial time. The same holds good of Taraxacum phymatocarpum, only known 
from Ehesmereland and Greenland. Of the four endemic species of Greenland 
Carex holostoma, Melandrium triflorwn, and Potentilla Friesiana come very near to 
other species, and Potentilla Ranunculus alone is difficult to account for. The 
absence of endemic types is, as well known, a common feature of fioras of once 
glaciated areas, and in assuming a natural borderhne for the Arctic Regions in- 
stead of the entirely artificial one formed by the Arctic Circle which in some parts 
of each of the three continents runs far down in the wooded country, in other 
parts again leaves large truly arctic träets to the south, we will have very few and 
insignificant endemisras left in the whole arctic flora ^ If we beheve in fioras 
» Engler has in the introduction to his important work Versuch einer Entwicklungsge- 
Hcluclite der Pfiiinzenwelt, anion^r liis »leiten.le Ideen» forniulated bis opinion on tluH point thus 
({). X;: »diisM in Lundern von iiolien» Alter, namentiich in ^ebirgigen Gegenden, deren Vcfretation 
