606 DAY AND SHEPHERD WATER AND VOLCANIC ACTIVITY 



other gas by diffusion through pores or by overcoming whatever chemical 

 or mechanical conditions it may encounter. The prospect is not an en- 

 couraging one. The hydrostatic pressure at great depths of the sea would 

 appear to be the only sufficiently powerful agent to drive water against a 

 high adverse temperature gradient, but to invoke this would be to invite 

 nice distinctions of where "magmatic" water begins and "meteoric" water 

 ends. The presence or absence of chlorine is not a conclusive factor one 

 way or the other, because the physical processes of infiltration through 

 porous rock and of distillation are alike of such a kind as gradually to 

 leave the dissolved salts behind; this is illustrated by the fact that the 

 bore-holes yield fresh water except when the infiltration is very rapid. 



To us, therefore, such evidence as there is appears to indicate that the 

 water released from the liquid lava when it reaches the surface is entitled 

 to be considered an original component of the lava with as much right as 

 the sulphur or the carbon. 



