DEFORMATION OF CRYSTALLIZING MAGMA 267 



We may pass on now to a consideration of the squeezing out of 

 liquid at a still later stage of crystallization to form a distinct 

 differentiate such as a granophyre mass. The breakdown of the 

 crystal mesh at this stage would still, I believe, be accomplished by 

 fracturing along areas of contact between two adjacent crystals 

 and subsequent revolving of the crystals into a position permitting 

 closer packing. There is later, too, a further growth of crystals 

 dependent upon the amount of liquid ultimately left in the inter- 

 stices, and the final result would be a panidiomorphic granular 

 (not granulated) mass showing normal crystallization textures. 

 The production of broken crystals and of granulation belongs to a 

 later stage, when there is a neghgible amount of interstitial liquid 

 and shearing forces of a much greater order of magnitude must be 

 brought into action. In all probability such forces do come into 

 play in the production of some anorthosites, but far-reaching 

 results can be produced by filter-press action without any necessity 

 for the development of granulation. 



In the case of production of a granophyric body from gabbro 

 magma by expressing of liquid, there* is no necessity that the 

 gabbro should show granophyric interstices, for if the coohng is 

 slow enough the granophyric liquid that remains in the interstices 

 may be used up and normally will be used up by reaction with 

 crystals already separated. No one who has examined a section 

 of a gabbro with granophyric interstices can have failed to see the 

 reaction referred to, interrupted before completion. The reaction 

 between relatively large blebs of granophyre produced by immisci- 

 bihty, while it might be of the same nature, could not possibly be 

 carried to completion. 



In conclusion I wish to confess some surprise at Grout's state- 

 ment that I have made an inaccurate copy of his map of the Duluth 

 gabbro, for the copy was made for me by a competent drafts- 

 man. Furthermore, my surprise has been greatly increased on 

 examining my map, going over it minutely with a pair of dividers 

 and finding that it corresponds absolutely, dimension for dimen- 

 sion, with Figure 6 of his paper, "Internal Structures of Igneous 

 Rocks."^ 



^Jour. Geol, XXVI (1918), 446. 



