^RT. 2. PETROLOGY AT GOOSE CREEK SHANNON. 39 



The diabases of the Newark series were rather variable in the 

 amount of water which they contained, as indicated by the textures 

 of the consoUdated rocks and the hydrothermal effects produced. 

 Shales are highly hydrated rocks, and the most conspicuous feature 

 of the shales adjacent to the intrusive, both at Goose Creek and else- 

 where, is a loss of the shaly structure and a compacting and hard- 

 ening doubtless due to loss of water. A body of molten magma of 

 diabasic composition, surrounded on all sides by hydrous shales, 

 would certainly tend to increase its content of dissolved water by 

 solution of the highly heated water of the adjacent shales. In sand- 

 stones there would be less necessity for the water to dissolve in the 

 magma since it would be more free to move outward from the heated 

 zone. This may explain the greater frequency of the occurrence of 

 differentiation and other aqueous effects in sills in shales than in 

 those in sandstone or in relatively anhydrous rocks. That water can 

 dissolve in molten silicates under pressure has recently been shown 

 conclusively by Morey.^*' The presence of this water introduces 

 complications into consideration of the problem of crystallization. 

 Whereas in a simple silicate melt all of the constituents can enter 

 into the consolidated product, without regard to the rate at which 

 cooling took place, water or other volatile constituents of the melt 

 will be expelled by crystallization of anhydrous minerals and must 

 of necessity concentrate in the still fluid portions of the magma, 

 resulting in an increase of pressure and a lowering of final consolidat- 

 ing temperature. If the magma crystallizes from the early cooled 

 walls inward there must be a concentric inward expulsion of the 

 water, which in the ideal case would result in a centrally placed peg- 

 matite. Actually this happens in dikes and in some thin or small 

 intrusions, where it is easily demonstable. In larger intrusive masses, 

 however, the volatile constituents seem to concentrate in centers 

 whose location is determined by some unknown factor. Differential 

 movements might result in local areas of lessened pressures and here 

 gases would tend to concentrate and pegmatites might form. These 

 may occur thickly scattered in groups or widely spaced singly in the 

 mass of the rock, and they are well exhibited by coarse and pegma- 

 titic areas, which are common features not only in the gabbroic and 

 diabasic rocks but especially in many areas of granitic rocks. Typi- 

 cal of such "pegmatite chambers" are the pegmatites of the Quincy 

 granites as described by Warren and Palache.-^ 



The mechanism of this concentration of volatile materials in re- 

 sidual magma chambers can not at present be delineated. It is, how- 



M George W. Morey. The development of pressure in magmas as a result of crystallization, Journ . 

 Wash. Acad. Sci., vol. 12, p. 219, 1922. 



" Charles H. Warren and Charles Palache. The pegmatites of the riebeckite-aegirite granite of Quincy. 

 Mass., U. S. A.; their structure, minerals, and origin. Proc. Amer. Acad. Arts and Sciences, vol. 47, pp. 

 146-147, 1911. 



