V. THE ICELANDIC HIGHLAND FORMATIONS. 
N° thorough investigations of the highland vegetation which might 
serve as a basis for comparison with the lowland vegetation 
and the vegetation in other localities in the highlands having so far 
been carried out, I have endeavoured to furnish such a basis by 
my investigations. By its geographical position, its height above 
sea level, and the grandeur of its scenery, Tvidägra, the highland 
plateau west of Langjökull, seemed to me best fitted for such a pur- 
pose. Hence the investigations described below have been carried 
out there, that is, more precisely indicated, in the region round Ulfs- 
vatn on Arnarvatnsheidi. 
Before I proceed to a more detailed description of my own in- 
vestigations I shall give an account of what is known about the 
highland vegetation. The first more comprehensive description is 
that of St. Stefansson (1894). On p. 199 Stefansson writes: “At 
the Vatnsdalshals and on the high plateau we especially meet with 
three formations, viz. the myri, melar and heather mo vegeta- 
tions.” The latter I found best developed in the so-called “Helga- 
vatnsnupar”, gravelly heights on the eastern margin of the Vatns- 
dalshals. The depressions between these are more or less, and some- 
times entirely, covered with a greenish-brown carpet formed of 
Empetrum nigrum, Vaccinium uliginosum, and Betula nana completely 
mixed up with each other. Salix herbacea and glauca also occur 
almost everywhere throughout this carpet, whereas Salix lanata 
appears only here and there. Loiseleuria procumbens is found in no 
small quantity in several places, and Cassiope hypnoides is met with 
here and there. In several places the dwarf birch predominates 
over the crowberry (Empetrum nigrum), but as a rule the latter is 
dominant. Of other plants I shall mention first Dryas octopetala, 
which is hardly ever absent from any Icelandic heather mo, next 
Thymus serpyllum, Polygonum viviparum, Galium silvestre, Silene 
