146 H. MØLHOLM HANSEN 
There are, however, a few interesting deviations from the rule. 
Thus Calluna vulgaris is of common occurrence in the mo at 
Bjørk. At Lyngdalur it is absent, or practically absent from the 
mo, whereas it occurs very abundantly in the snow patches. At 
Thrasaborgir it occurs neither in the mo nor the geiri. 
Deschampsia flexuosa occurs both in the mo and the geiri in 
the south country; in the south-west, the north, and the highland 
tracts it is either entirely absent or occurs only in scattered growth 
in the mo, whereas it is very abundant in the geiri (and forest). 
Thalictrum alpinum is a mo plant in the south country but 
shuns geiri; in the south-west and north country it is still a mo 
plant, but here it is also met with in the more snow-covered types 
of vegetation as forest and geiri. 
It seems natural to suppose that these deviations are due to 
differences in temperature. 
Vaccinium uliginosum is another interesting example. Its F.- 
percentage varies as follows in the scale of snow-covering. In I a 
(mosathembur) it has an average F.- percentage of 0.5, in I b (melar) 
4. Class II, i.e. the mo, has an average F.-percentage of 54 and 
class III 75. The species is thus a pronounced geiri plant though 
with strong tendencies towards the mo. At Bjork it plays a promi- 
nent part in the mo, especially the Arctostaphylos mo, while the 
Elyna mo is less favourable. In the mo at Lyngdalur it is but 
sparsely represented, while it is dominant in geiri. The same 
applies at Thrasaborgir. In the highland tracts, at Lækjamôt and 
Norötunga, it is not only peculiar to the areas with a deep snow- 
covering, it also occurs with a high F.-percentage in the mo. On 
melar it does not occur, however. 
These peculiarities must no doubt be put down to differences 
in the scale of snow-covering between the south country on the 
one hand and the rest of the country on the other. In the high- 
land tracts and the north country the snow-covering is more constant 
than in the south country, the mo of which is sometimes covered 
with snow, sometimes bare. 
Table 30 shows the distribution of the species groups 
in the scale of snow-covering. The signatures are the same 
as in table 23. The average values are given at the end of the 
table. It appears from the table with all desirable plainness that 
the A percentage is highest in the class most bare of snow, i. e. in 
