﻿GEOLOGY OF THOUSAND ISLANDS REGION 1 87 



ages of this amphibolite, and column 4 a calculated mixture of 

 granite and amphibolite in the ratio of 24$ of the former and j6f 

 of the latter. The agreement in this case is not so close as in the 

 former one, but in large part the differences can be ascribed to the 

 fact that the granite here is not normal, but of the pegmatitic type, 

 higher in mineralizers and in iron, and poorer in alumina, lime and 

 magnesia than the normal rock. Comparison of analyses 3 and 4 

 seems definitely to suggest this, the iron being higher in the grani- 

 tized rock than in the amphibolite, and the lime and magnesia much 

 lower. It is thought that if an analysis of the neighboring granite 

 was available for use in the computation, the agreement would be 

 much closer. But the result is somewhat disappointing, and the 

 importance to be assigned to the discrepancies a debatable matter, 

 likely to vary with the personal equation of the reader. Consider- 

 ing all the circumstances the agreement seems to us as close as 

 could be hoped for, and indicative of the correctness of interpreta- 

 tion of the field evidence, namely, that these are true mixed rocks. 



