SORBY STRUCTURE OF CRYSTALS. 483 



heat, under a pressure equal to several thousand feet of rock, when 

 water containing a large quantity of alkaline salts in solution was 

 present, along with melted rock and various gases and vapours. 

 Whether or no the presence of this water was instrumental, or even 

 essentially requisite, in producing some of the minerals, still remains 

 to be proved ; but I think no one could compare the drusy cavities 

 in the ejected blocks with the crystalline cavities in the slags of fur- 

 naces, without perceiving that the occurrence of various minerals, 

 placed one over the other in regular order, is a most striking differ- 

 ence, which could be accounted for most completely by the action of 

 water. It would also probably serve to explain why, according to 

 Daubeny (Treatise on Volcanos, 2nd edition, p. 236), the minerals 

 most characteristic of the ejected blocks are never found in the 

 erupted lavas, the crystalline minerals of which were apparently 

 formed when no liquid water was present. I therefore think we must 

 conclude provisionally, that at a great depth from the surface, at the 

 foci of volcanic activity, liquid water is present along with the melted 

 rock, and that it produces results that would not otherwise occur. 



It may perhaps be thought that the spheroidal condition assumed 

 by water in contact with highly heated substances, would explain 

 why it might be present at a less depth, and under less pressures, 

 than those I have described ; but it appears to me that water could 

 not remain in the spheroidal state, unless the vapour could escape, 

 and that the temperature it remains at is essentially connected with 

 the boiling-point at the pressure to which it is exposed, and there- 

 fore the permanent presence of water at such a high temperature 

 necessitates a great pressure, even if it was in the spheroidal state. 

 But I think no one who has made experiments on the subject, would 

 think it possible for water in that state to enter into tubes less than 

 j^i-ijth of an inch in diameter. This, however, has constantly 

 occurred in the minerals of the ejected blocks, and hence it appears 

 to me almost demonstrated that it was not in the spheroidal state, 

 separated by a layer of vapour, but in actual contact with the crystals 

 at a high temperature, and under great pressure. 



The presence of genuine gas- and vapour-cavities side by side with 

 the fluid-cavities, and the existence of so large an amount of salts in 

 solution in the fluid, prove that the water was caught up in a liquid 

 state, and not as vapour so highly compressed as to condense into 

 an equal bulk of water (see Cagniard de La Tour's paper, Annales 

 de Chemie, 1822, t. xxi. p. 127); for in that case, since in the 

 nepheline there is no gradual passage from fluid-cavities to vapour- 

 cavities, we should have to conclude that the two gaseous bodies 

 were not mutually diffusible, and that a very large amount of various 

 alkaline salts was present as vapour along with the vapour of water ; 

 both of which suppositions are I think quite inadmissible. 



Perhaps some may suppose that possibly the water penetrated into 

 the cavities long after the minerals were formed. This, however, 

 would necessitate percolation through the solid substance of the 

 crvstals, a fact differing as much from percolation through a rock, 

 or amongst the minute crystals of which such substances as agate 



