264 . FRANK F. GROUT 



feldspar and troctolites with no augite. The extremes are 100 per 

 cent different, and are not to be expected from filter-pressing. 



Taking next the case of granite separated from gabbro, we find 

 it a general rule that the separation has proceeded to such a degree 

 that the granophyr diabase is very small in bulk.' Such action 

 develops only after the magma is 80 per cent crystals. The occur- 

 rence of 300 feet of red rock above gabbro with practically no grano- 

 phyr, as described at Duluth, should mean that over 1,200 feet of 

 gabbro was almost completely drained of all 20 per cent residual 

 fluid — so completely that the trifling residue left no determinable 

 mineral and produced no zoning of the feldspars. Such thorough 

 drainage from so large a mass seems very unlikely. And if the 80 

 per cent crystal mesh had any strength at all it must have refused 

 to yield so completely as to leave no interstitial magma at all. 

 Interstitial granophyr would be much more likely to be trapped 

 among crystals than as blebs in an immiscible liquid. 



SUMMARY 



In igneous magmas the movements which are subject to some 

 disagreement are convection, crystal settling, and the expulsion of 

 residual magma from a partly crystalline mass. An attempt is 

 made to remove certain ijiisconceptions of the convection process 

 and its effects. The process of crystal settling seems to be limited 

 to effects which may be estimated to extend not over 100 feet. 

 Attention is called to several difficulties in the application of the 

 filter-pressing idea. The writer favors all three of the ideas as 

 working hypotheses, but certain critical field relations should be 

 sought for before accepting either in specific cases. 



^ Frank F. Grout, "A Type of Igneous Differentiation," Jour. Geol., XXVI (1918), 

 pp. 645-57. 



