Gabbroid Rocks of Minnesota. — WincJicll. 383 
•very recently among basic granular rocks. In fact, no rock 
of this kind is mentioned in the classifications of magmas of 
the authors cited. The only analogous rocks from this point 
of view, but different in certain respects, are, as already noted, 
the cordierite noryte of Pallet, France, described by Prof. A. 
Lacroix, and certain exceptional rocks, which, according to 
the authors who have described them, are all endomorphic 
variations of diverse eruptive rocks of normal composition. 
This fact is entirely in harmony with the explanation given 
above concerning the origin of this rock. If it should be 
demonstrated later that the cordierite noryte was derived 
from an unmetamorphosed magma, it would then be necessary 
to admit the existence of a new magmatic tvpe, whose exist- 
ence from present data does not seem probable. 
The silicoferrolyte (Fig. i) is a type of exceptional char- 
acters. The most marked feature is the enormous develop- 
ment of the iron ordinate, but the excess of alumina is a very 
unusual occurrence, which indicates the existence of an alum- 
inous spinel in the rock. The alkaline-earth triangle is so re- 
duced as to be practically absent, and in fact the rock contains 
no feldspar. The amount of fayalite cannot be determined ac- 
curately from the per cent of silica since free quartz exists in 
the rock. Lang would class this type with the preceding in 
spite of its many peculiarities. It belongs to Brogger's me- 
lanokrate series, and to Michel Levy's albitic magma, which 
is a subdivision of his ferromagnesian. Rosenbusch would 
consider it as derived from the peridotyte magma (;r) since 
the free c[uartz is wholly anomalous. 
It will be noticed that AZ and MAZ of the silicoferrolyte 
are extremely low, and do not fall anywhere nearly within the 
limits considered characteristic of peridotyte magmas by Ro- 
senbusch. This fact is in harmony with the theory that the 
silicoferrolyte is not a rock of purely igneous origin. How- 
ever, very little importance is to be attached to this fact since 
Roth* has shown that these numbers AZ and MAZ possess no 
important significance. 
Passing now to the method recently proposed by Brogger, 
which is a modification of that of Michel Levy, it will be seen 
*J. Roth- Die Eintheilnng nnd die chemischc BeschafFcnlieit dcr 
Eruptivgesteine: Zeitsch. d. d. Geol. Gesellsch. 1891. p. 41. 
