CAUSES OF MAGMATIC DIFFERENTIATION. 



In petrographical literature in recent years attention has 

 repeatedly been drawn to the fact, that igneous rocks, which are 

 closely connected geographically and in age, are also chemically 

 related to one another, showing a certain " consanguinity " — to 

 use Iddings' 1 very fitting expression — a relationship which makes 

 them form a distinct " petrographical province " (Judd) when 

 compared with igneous rocks of other parts of the world. The 

 cause of this relationship has been sought in the supposition, 

 that all the different rocks of the " petrographical province " 

 come from the differentiation of one common magma, originally 

 homogeneous. 



As to the manner in which the differentiation took place, 

 opinions are divided. We may suppose that it took place during 

 the consolidation of the magma ; in this way, a part of the 

 minerals crystallized out, then were mechanically accumulated 

 and finally reliquified. The differentiation of the original magma 

 into partial magmas could take place in this way, but, as far as I 

 can see, only on a small scale. A silicate magma during its 

 period of crystallization is certainly too viscous to permit of any 

 considerable diffusion. For example, in the reproduction of 

 rocks after the method of Fouque and Levy, in which process a 

 glass is first made having the desired composition, this glass may 

 be completely devitrified (fused), while it remains so viscous 

 that pieces of it neither change form nor adhere to one another. 



Another theory, namely, that the differentiation has taken 



place in the magma while quite fluid, possesses greater probability 



and therefore more adherents. But concerning the details of the 



method opinions differ. While certain petrographers apply the 



1 " Origin of Igneous Rocks." Bull. Phil. Soc. of Washington, 12. 89-214. (1892). 

 This paper contains an extensive bibliography of this subject, to which the reader is 

 referred. 



773 



