26o FRANK F. GROUT 



involves differential pressures, and no such pressures are evidenced 

 by the great majority of granites having pegmatitic facies. Barrow 

 also refers to a case of straining off of magma into crevices too small 

 for crystals to get in. Such an action is easily understood but would 

 not be expected to produce large bodies of rock. Barrow goes on 

 to say, however, that "the continuance of the pressure will still 

 further force the liquid from the sohd crystals, leaving at last just 

 sufficient of the magma to fill the interstices between them." This 

 can hardly be considered a clear outline of what happens " further" 

 than the oozing of magma into cracks too small for the crystals 

 to enter, Judd^ has described andesite and pitchstone intrusions 

 which he thinks separated by a process of "liquation" — -apparently 

 a growth of crystals along the walls leaving a residue of different 

 composition. Harker, referring to the suggestion, says "it is easy 

 to conceive of various modes of liquation and decantation, straining 

 and filtration by which a partial separation may be brought 

 about." Bowen has shown the improbability of diffusion sufficient 

 to yield much differentiation by any liquation in this sense. "De- 

 cantation" is not a process of filter-pressing. To the present 

 writer "straining and filtration" of a magma on any large scale 

 are not easily conceived. 



Harker reviews these descriptions, emphasizing the case of 

 pegmatites and that of pitchstones related to andesites. He does 

 not refer to any indication of a process of compression except in the 

 case of numerous small pegmatite intrusions related to large 

 granite gneisses. Even here the larger masses of pegmatite are 

 better explained by a separation of some other sort. 



Bowen presents the latest development of ideas on filter-pressing.^ 

 His first analogy is to the squeezing out of water from sand on a 

 wet beach. There is little question that such a mesh of crystals 

 as he suggests, up to 80 per cent of the total mass, might, when 

 subject to differential pressure, give way and result in closer packing. 



The supposed filter effects are described for three different cases, 

 but no mention is made of the case of pegmatitic separation sug- 



' J. W. Judd, Quart. Jour. Geol. Soc, XL VI (1890), p. 379. 

 ^ N. L. Bowen, "Crystallization-Differentiation in Igneous Magmas," Jour. Geol., 

 XXVII (1919), p. 393- 



