CHEMICAL COMPOSITION IN SEDIMENTS 467 
of the slate analyses and 74 per cent. of the pelite schist analyses. In 
the igneous rocks of Washington’s tables this double relationship 
holds in only 44 per cent. of the rocks of classes I and II and in 7 per 
cent. of the rocks of classes III, IV, and V. These relations are 
tabulated in the third column of the table on p. 463. The double 
relationship is therefore of much more diagnostic value than either 
of the single relationships. If we consider the pelites in comparison 
with the igneous rocks of classes I and II, the double relationship has 
about twice the critical value of dominance of only MgO over CaO, 
and over eight times the value of dominance of only K,O over Na,O. 
CRITICAL VALUE OF THE SILICA CONTENT 
Attention has already been called to the fact that certain types of 
metamorphosed sediments, such as quartzites and quartzitic schists, 
are usually readily identified as such from their highly quartzose 
character without the necessity of further chemical study. There 
is however every variation from such extremely siliceous types to 
typical pelites, and it is to be expected that in many of these inter- 
mediate types the high silica content will have diagnostic value. 
It has been shown that the Al,O, content, the ee ratio, and 
a 
Na,O 
metamorphism. The SiO, content, on the contrary, often increases 
very markedly during metamorphism. This is shown by a compari- 
son of the average of the pelite schist analyses with the slate average 
in the table on p. 456. A discussion of the cause of this silication is 
beyond the scope of this paper. It has usually been explained as a 
process complementary to decarbonation, CO, being driven off during 
dynamic metamorphism and carbonates converted into silicates. 
Studies which the writer has in hand have led him to believe, however, 
that while silication as a result of decarbonation undoubtedly takes 
place, it is quantitatively inadequate to explain the large silica increases 
observed. It seems probable that there is actual addition of silica 
from outside, either through the agency of descending groundwaters 
or from magmatic sources. ‘The comparisons of igneous with meta- 
igneous rocks which have been given in this paper afford slight and 
the ratio in a rock exhibit no important change during dynamic 
