154 H. MØLHOLM HANSEN 
snow-bare soil Ch thrive best, the deeper the snow- 
covering the more does the Ch percentage decrease, while, 
on the other hand, the H percentageincreases, and where 
the snow-covering is deepest, H play the most prominent 
part. This distribution must be regarded as a consequence of the 
geographical distribution of the life forms. Ch, the arctic life-form, 
thrive best where the cold has the strongest effect, while the more 
temperate life-form, H, shows a preference for conditions in which 
there is the greatest protection from the cold. 
From this rule of the distribution of the life-forms in the Ice- 
landic scale of snow-covering there is an interesting, though merely 
apparent, deviation, as will appear from a close investigation of 
conditions in the individual localities. In the highlands the rule 
applies throughout, the slighter the snow-covering, the higher the 
Ch percentage (though in the very lowest class there occurs an in- 
crease), and in the lowlands too it holds good if we consider the 
relation between melar and mo, i. e. between snow-bare and snow- 
covered soil. In other respects conditions in the lowlands seem to 
go against the rule, Ch playing the most prominent part where the 
snow-covering is deepest and of the longest duration, whereas this 
group decreases when the depth of the snow-covering decreases. 
The Ch percentage in the Elyna mo, the comparatively snow-bare 
formation at Bjork, is 28, whereas, in the Arctostaphylos mo, the 
relatively snow-covered formation, it is 39. In the mo at Lyngdalur 
the Ch percentage is 26, in the geiri 35, and in mosathembur only 
10. At Thrasaborg the values for mo and geiri are 25 and 34 re- 
spectively. 
This difference in the distribution of Ch in the highland and 
the lowland scale of snow-covering is due to the fact that the Ch 
vegetation in the lowlands consists especially of E species, whereas, 
in the highland tracts, it consists of A species. This disagreement 
thus handsomely confirms the distribution of the species groups in 
the scale of snow-covering. 
Raunkiær examined the variations in the biological spectra 
in sections from southern towards northern regions (1908, 1911) and 
showed that some life-forms, Ph, K, and Th, decrease in quantity, 
while others, Ch, increase and others again, H, undergo no appre- 
