FIELD OBSERVATIONS IN NORTHERN NORWAY 713 
in the field, as stated above, a series of olivine-rich rocks all around 
the marginal zone of the eruptive body and of nearly the same age 
as the environing norite. These olivine rocks can only have been 
individualized during that period of the crystallization process 
when free olivine crystals were really suspended in the crystallizing 
marginal zone. It is obvious that even the slightest accumulation 
of these crystals at certain places would at once bring them in 
excess, and prevent them from being wholly resorbed. _ 
Consequently, olivine would become a part of the norm compo- 
sition at those places. We may conclude, therefore, that such an 
accumulation of olivine crystals has really taken place. That 
gravity separation has not played a prominent part in it is obvious 
from the distribution of the olivine rocks. We are forced to con- 
clude that convection currents and other movements in the magma 
have been able to bring about such an accumulation, possibly also 
a conglutination of already formed crystals. 
The explanation is confirmed by two facts: First, in all the 
olivine rocks the olivine crystals are surrounded by resorption 
rims, proving that resorption has been going on, but stopped at a 
certain point. Secondly, the olivine crystals in all these fields 
have nearly the same proportion, Mg,SiO,:Fe,Si0,, quite independ- 
ent of the quantity of olivine, and consequently have all separated 
out from a uniform magma. 
We get a natural explanation of the varying quantities of 
plagioclase and pyroxenes in the olivine rocks and of their “swim- 
ming” character in the norite magma. 
After the termination of the resorption period and the con- 
solidation of the olivine rocks, the rest of the magma in the marginal 
zone should obviously be expected to have become poorer in mag- 
nesia and richer in silica than the magma of original composition 
in the still fluid central part. Field observations teach us, however, 
that the olivine-free norite in a very thick marginal zone has a 
more basic composition than the central part. 
To explain this, we might think of compensating currents in the 
magma prior to the resorption period, which in combination with 
the resorption of some of the olivine might produce this basic 
