152 H. MØLHOLM HANSEN 
Subularia flag, the water-covered vegetation rich in Th, corresponds 
closely to that of the flöi: concentration of the species in the central 
part, and preponderance of the E species. 
The treatment of the distribution of the species groups and the 
life-forms in the scale of moisture showed that moderately moist 
soil (jadar) caused a relative maximum of E species, hemicryp- 
tophytes and therophytes. This was the case with the vegetation 
on a gently sloping surface with even transitions from one type of 
vegetation to another. However, on moderately moist soil there 
occurs a series of types which, physiognomically, are rather different, 
both mutually and in relation to the jadar, but which, on close in- 
spection, prove to be possessed of the pecularities of the jadar vege- 
tation, though in varying degree, viz. a relatively high H percentage 
and Th percentage. Of these types the valllendi and flag vegetations 
have so far been examined. The first of these types develops on 
the flat cones deposited by the rivulets of melting snow on flat 
ledges. The valllendi soil is thus saturated with water until the last 
snow has melted, i. e. until the geiri is bare of snow. In the flag, 
conditions are otherwise. If it receives any water at all while the 
snow is melting, it is at most as long as there is snow on the mo. 
The bare soil of the flag is thus exposed to the effects of the frost 
(night frost) much longer than the valllendi, which is furnished with 
fresh water daily. Hence it is hardly accidental that the E species 
are more dominant in valllendi than in flag. The E percentages for 
valllendi, jadar, and flag are respectively 75, 53, and 42. 
The Distribution of the Life-Forms in the Scale of 
Snow-Covering. 
Since differences in respect of snow-covering only appear in 
areas not affected by ground-water, where, as previously mentioned, 
Ch, H, and Th are the dominant life-forms, it is principally the re- 
lation of the snow to these which is of interest. Table 30 shows 
the biological spectra of the various classes of snow-covering in the 
localities examined by me. At the bottom of the table are given 
the mean values for all the investigations. 
It appears from the table with all desirable plainness that Ch 
are more abundant on snow-bare soil than on soil with 
a normal snow-covering, and more abundant there than where 
the soil has a constant snow-covering. For H the case is re- 
