372 ft. A. DALY THE OltANAGAN COMPOSITE BATHOLITH 



gressive fusion of main contact walls by original hot magma aided by its 

 water and other solvents; secondly, of the progressive stoping of blocks 

 (xenoliths) from the roof and walls of the batholith or stock, followed by 

 the abyssal solution of the xenoliths in the hot interior of the magmatic 

 body. Systematic differentiation of this new, widely extended, sub- 

 crustal, compound magma accompanies and follows the assimilation. 

 The principal cause of differentiation has been sought in gravitative ad- 

 justment stratifying the magmatic couche according to the law of up- 

 wardly decreasing density (meaning, in general, increasing content of 

 silica from below upward -in the magmatic strata). A subordinate cause 

 of differentiation commonly develops basified contact zones by the diffu- 

 sion of basic materials to the surfaces of cooling. In the nature of the 

 case, this latter kind of differentiation belongs to the late magmatic period 

 immediately preceding crystallization; and, as illustrated in the Castle 

 Peak stock and the Similkameen batholith, appears to form a basified 

 zone along the roof as well as along the walls. 



Partly for that reason it has proved impossible to discover a law of in- 

 creasing density with depth in the Similkameen granite. A series of 

 fifteen fresh specimens of the rock were collected at altitudes varying from 

 1,200 to 8,050 feet above sea, and their specific gravities were carefully 

 determined. The difference between the densities of specimens taken near 

 or at the two extremes of vertical distance was found too small to allow 

 of a definite conclusion, though the difference, small as it is, favors the 

 law of density stratification. It must be remembered, however, that the 

 concentration of volatile matter, such as water vapor, dissolved in the 

 magma but largely expelled during crystallization, would possibly be 

 greatest at the roof. The specific gravities of the crystallized rocks may 

 therefore not afford direct values for the total density stratification during 

 the fluid state of the magma. Then, too, the observed relative uniformity 

 of the Similkameen granite is a function of the scale of the subcrustal 

 magma couche. It was unquestionably very thick; strong density differ- 

 ences are probably not on any hypothesis to be expected in a vertical sec- 

 tion less than several miles in depth. 



The objection has very often been made to the idea of extensive assim- 

 ilation that signs of actual digestion about blocks broken off from roof or 

 wall commonly fail. This is, however, just what would be expected on 

 the stoping hypothesis. The very position of such a block shows that at 

 the time when it was broken off it was floated in a magma too toughly 

 viscous to allow the block either to rise or sink. Under those conditions 

 the solvent or assimilative power of the magma is at or near its minimum. 

 In other words, contact decrepitation persists some time after contact 



