GEOLOGY OF THE WEST POINT QUADRANGLE, NEW YORK 37 



solid phase. Perhaps both processes are represented. Perhaps the 

 development of a pegmatitic facies is only the rejection of end- 

 product matters from a partly solidified mass. To this degree, 

 therefore, it is a result of fractional crystallization. It is also pos- 

 sible that some of the banding structure and gneissoid habit has a 

 similar history connected with convectional movement or regional 

 deformation during crystallization, and perhaps for some of the 

 differences in the same structural unit one should resort to the 

 principle of liquid separation as an additional factor. Processes 

 of some kind connected with the differentiation have brought about 

 marked differences of composition within short distances and these, 

 in conjunction with movements of all kinds, have caused at least 

 part of the complex structural variety and petrographic confusion 

 of the ancient gneisses of the Highlands. 



It is entirely possible, and indeed as we believe highly probable, 

 that even the basic differentiates, represented by the hornblendic 

 pegmatites and the magnetjte ores, are related to one or more of the 

 granites. If one makes allowance, therefore, for such extremes of 

 composition it adds materially to the difficulty of drawing boundaries 

 between formations in mapping, and it leaves one in much uncer- 

 tainty about the proper correlation of many of the smaller field units. 

 But, although it adds to the difficulty of clearly connecting individual 

 specific occurrences, a recognition of these differentiation possibili- 

 ties aids materially toward an understanding of the great complexity. 

 And since the petrographic variation is so great anyway that it is 

 impossible to make every variety a distinct unit, it leads to more 

 successful understanding of the whole structure to give credit for 

 some of the mineralogic variation and structural peculiarities to the 

 process of magmatic differentiation. 



Magmatic movement. This topic has been discussed to some 

 extent in connection with differentiation phenomena. It is sufficient 

 at this point to review the item for the purpose of emphasizing its 

 comparative importance. Even the simpler granites very rarely have 

 perfectly massive structure. They are almost universally streaked 

 or crudely banded in a way that suggests at once some form of 

 fiowage. In some cases such structure as the rock contains seems 

 to have no necessary relation to the dynamic structure of the region, 

 and it may be that these portions represent simply convectional 

 movements within the magmatic masses. But by far the more com- 

 mon structure is strictly conformable with the regional structure or 

 trend and in so far as it is not derived from outside control through 



