MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 245 



crystallization might take place. (See the chap. pp. 93-100 in my 

 treatise on anchi-monomineralie and anchi-eutectic rocks, 1908.) 



In highly viscous melts, inoculation causes no crystallization. 

 Thus rhyolites contain glass mainly of the composition of quartz 

 and feldspar, even though crystals both of quartz and feldspar 

 have been segregated. 



The entire lack of glass in deep-seated rocks is an evidence of 

 quite extraordinarily slow cooling. The crucial point for the 

 distinction between the deep-seated rocks, on the one hand, and the 

 effusives and dikes on the other, is not pressure, but time. Many 

 dikes have been solidified at very great depths, thus under the same 

 pressure as the adjacent deep-seated rocks. 



Granites of practically the same chemical composition show 

 at times considerable variation in the size of grains. Thus the 

 feldspar individuals in some granites have an average length of 

 0.3-0.5 cm., in others 1-1.5 cm. The relatively coarse grained 

 granites form at more places much smaller areas (of, e.g., 10-50 

 km.^) than those of liner grain (with areas of, e.g., 100-500 km.^). 

 The decisive point for the size of the grain thus can neither be 

 duration of crystallization nor solidification in more or less great 

 depth. The cause may, I suppose, be due to the fact that the 

 magma of the relatively coarse grained granites had less viscosity on 

 account of a higher content of H2O, etc. 



In this connection, it may be noted that the hypersthene- 

 granite from Birkrem, etc. in the Ekersund District is medium fine 

 grained, which may be due to the fact that the magma of this 

 granite, as stated above,^ contained relatively little H2O. It might 

 be interesting to examine whether all hypersthene-granites are 

 characterized by only medium size of grains. 



The extreme size of the grains of the pegmatite dikes of granite, 

 augite- and nephelite-syenite, etc., proves, as maintained by many 

 investigators, an extreme fluidity of the magma, depending on 

 relatively large contents of H2O, etc. At one atmosphere pres- 

 sure the rate of crystallization of many minerals, by the segrega- 

 tion from very thin melts, is on an average i mm. per minute or 

 3 cm. per half hour, corresponding to about 1.4 m. per day (twenty- 



' The Journal of Geology, XXX, 670.. 



