40 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. G6. 



ever, a variant of crystallization-differentiation and can not be con- 

 sidered to be liquid immiscibility. Such volatile constituents would 

 doubtless carry with them constituents of the magma from which 

 they were expelled, principally silica and alkalies, especially soda, and 

 the residual mineralizer-rich magma is probably always enriched in 

 material thus derived. The diabase pegmatites of the Goose Creek 

 locality are of this nature. Whereas the main body of the dia- 

 base cooled too rapidly to be greatly differentiated, these small cham- 

 bers, having high molecular mobility due to dissolved gases with low- 

 ered consolidating temperature and under greatly increased pressure, 

 had an opportunity to crystallize more slowly with the production 

 of very large crystals. The minerals which crystallized out of these 

 melts were at first and in greatest amount the same as those which 

 were formed in the normal diabase, namely iron and titanium rich 

 augite and moderately calcic plagioclase. These grew outward from 

 the walls of the relatively small chambers and apparently left a central 

 residuum of liquid out of which subsequently crystallized albite and 

 diopside, leaving a final interstitial liquid which formed quartz-albite 

 micropegmatite. The process was interrupted at various stages by 

 strains which ruptured the consolidated surrounding diabase and 

 carried the liquid, at whatever stage, out as a dikelike mass into the 

 crack. Zoned crystals of the augite surrounded by diopside and of 

 calcic plagioclase surrounded by albite occur but are not common. 

 The most difficult feature to explain is the hiatus between the crystal- 

 lization of the augite and labradorite of what is called normal diabase 

 pegmatite and the relatively pure albite with diopside and abundant 

 quartz which form what has been described as albitic pegmatite and 

 aplite. Apparently the liquid attained its final very salic composition 

 solely by the crystallizing of relatively basic minerals. It behaved 

 as though the quartz-albite mineralizer magma were a solute and the 

 basic plagioclase, magnetite, and augite were dissolved materials which 

 were influenced in their crystallization from solution by the molecular 

 attraction of the same minerals in the adjacent diabase. The final 

 acid fluid was not stable in contact with these minerals at the close 

 of the magmatic phase as shown by the evident reaction between the 

 crystals and the residuum, with replacement of plagioclase by albite, 

 of augite by diopside, and of magnetite by titanite. This rarely if 

 ever reached equilibrium, however, probably because fissuring of the 

 mass of the diabase was going forward and the release of the pressure 

 of the dissolved gases, which must have been enormous in the later 

 stages, and upon which the fluid depended for its lowered crystal- 

 lization temperature, caused complete consolidation and release of 

 the volatile constituents. These constituents carried with them a 

 load of dissolved solids and continued to react in the same manner 

 upon whatever material they were in contact with, replacing plagio- 



