CIRCULATION OF WATER. 663 



the centvospliere upward into the zone of anamorphism. But it is probable, 

 as will be seen below, that these are not important sources of water. I 

 suspect that, on the average, the mechanical water and the water of dehy- 

 dration are the main sources of water supply of the zone of anamorphism. 

 My reason for this belief is that rocks which, containing large quantities of 

 combined water, like the sediments or the porous lavas, are readily recrys- 

 tallized in the zone of anamorphism, as fully explained on pages 741-748; 

 whereas great masses of igneous rocks in which there is a deficiency of 

 combined water are very frequently not recrystallized, but granulated. 

 The fact that massive igneous rocks containing little combined water do not 

 readily recrystallize is rather clear evidence that sufficient water to do this 

 ' work is not driven downward from the belt of cementation, or upward from 

 the centrosphere into the zone of anamorphism. 



We are now prepared to consider the circulation of the water in the 

 zone of anamorphism. It has been seen that the available openings are 

 mainly those of subcapillary size. It is through these openings that the 

 water must circulate, unless it be assumed that under the high pressures 

 and temperatures obtaining in parts of the zone of anamorphism the water 

 can make its way through the intermolecular spaces of the solid crystallized 

 minerals. This has been held by some geologists. It has been pointed 

 out that water is certainly occluded within magmas; but this is a very 

 different thing from the existence of free water between the molecules of a 

 crystallized substance. While the possibility of the transmission of water 

 in this manner is not denied, it is a mere hypothesis ; and therefore the only 

 openings which we know to be available for water circulation are those of 

 subcapillary size, with possibly a relatively few of capillary size between 

 the mineral particles. 



The movement of the water through these openings must be slow 

 because of their subcapillary size. It has been explained on page 143 

 that in openings of this kind the attraction of the mineral particles extends 

 from wall to wall, and the films are strictly adherent and therefore not 

 free water at all. But as the temperature becomes high the viscosity of 

 the water becomes much less, the molecular attraction between it and the 

 rock is decreased, the adherent film is less fixed, and its mobility is there- 

 fore very greatly increased. Also, in so far as the temperature of the water 

 exceeds its critical temperature, it is in the form of a gas; between a gas 



