A survey of the Phytogeography of the Arctic American Archipelago 141 
The first of the ten groups includes the ubiquituous species (U), found in all 
the six districts of the Archipelago as well as in each of the three areas of com- 
parison, even if they may not be uniformly distributed over the latter. Closely 
connected with this is the group U 2, species of rather uniform circumpolar range, 
even if they are not yet recorded from some or other of the six districts of the 
Archipelago; that most of them are absent from the list of the CornwalHs Island district 
is rather conspicuous, but may certainly be attributed to the insufficient exploration 
of that region. For the clearing up of phytogeographical questions nearly all the 
62 species on of account their wide distribution inside our area may seem to be of 
no value. Still some of them are to be discussed in the following. Next comes a 
group (S), comprising 18 species more or less generally distributed over the southern 
islands and absent from the northern districts (three of them, however, go a little 
beyond the horder). Of somewhat greater interest is the group of southwestern 
species (SW), 35 in number; in the main they are restricted to Banks and Victoria 
Lands, even if I have thought right to insert some of a wider distribution or such 
as have a western range on the Continent. The corresponding group (SE) includes 
34 species. 
For the study of the immigration of the flora the groups W (16 species), W 
2 (12 sp.), E (11 sp.\ and E 2 (12 sp.) are of the greatest interest. The two first 
mclude plants of a more or less decidedly western distribution, but reaching north- 
ward to Ellesmereland, the two others the corresponding eastern species. They 
will be treated in detail in the following chapter. Fiiially it has been necessary to 
put four species temporarily in a group D, as their distribution is to little known 
at present to allow the placing of them in any of the other subdivisions. 
The borderliues of these geographic groups, ns laid out on map I, of course 
must considerably overlap each other. Map II shows the borderlines of (ireenland 
distribution of the principal species and gr()U{)S <.f american constituents in the 
flora of that country, as contirniing tlie views on the history of the vegetation stated 
in the following. 
