680 A TREATISE ON METAMORPHISM. 



It is believed that the process of dehydration is largely caused by 

 pressure. In other words, water is actually separated from its combination 

 and made free water by the pressure, and the freed water is squeezed out of 

 the rocks as water from a sponge. Undoubtedly, also, the increase of tem- 

 perature with depth promotes dehydration. Without exhaustive experi- 

 mental work it is impossible to give any quantitative estimate of the relative 

 importance of pressure and temperature in producing dehydration. While 

 it is certainly true that moderate pressures and temperatures together 

 rapidly produce dehydration of the more hydrous minerals, ordinarily 

 pressure and temperature are not sufficiently great to drive off all water. 

 Apparently liberation becomes more and more difficult as the water increases 

 in amount. Dehydration is always incomplete, and commonly does not 

 reduce the combined water below 1.5 per cent, (See pp. 742-744.) 



It is probable, as explained on pages 665-667, that the larger part of the 

 freed water escapes upward to the belt of cementation, but some part of it 

 is confined in the altered rocks. Thus dehydration explains a large portion 

 of the water inclusions in the metamorphosed rocks. Of course the water 

 inclusions of original igneous rocks are explained differently, although in an 

 analogous manner. The water, or a part of it, occluded in the magmas, 

 separates at the moment of crystallization, and such part as can not escape 

 is included by the crystallizing minerals. 



SOL UTION AND DEPOSITION. 



Solution and deposition are essential concomitants of the chemical 

 reactions of deoxidation, silication, and dehydration, as well as of all other 

 important chemical changes. Cementation, next considered, involves depo- 

 sition. In metasomatism, subsequently treated, recrystallization is the 

 change of greatest consequence, and this is accomplished almost wholly by 

 solution and deposition. 



Whether solution is preponderant in the zone of anamorphism, or the 

 reverse process, deposition, it is difficult to say with certainty. But if intru- 

 sives be ignored, probably solution is slightly preponderant. Elsewhere 

 it is shown that there is no evidence of transfers of material into the zone 

 of anamorphism by the water solutions. (See pp. 665-668, 764-766.) By 

 deoxidation, silication, and consequent decarbonation, and by dehydration, 

 the solids constantly lose material. The water as a whole moves from the 



