SORBY — STRUCTURE OF CRYSTALS. 489 



cept by cracking it, is proved by the fact, that when the quartz from 

 Ceylon, in which the fluid expands so as to fill the cavities at about 

 218° C, was heated to at least 350°, some of the fluid-cavities still 

 retained their fluid, though, judging from the force calculated to be 

 necessary to counteract the expansion of the fluid, they must have 

 resisted a pressure equal to about 40,000 feet of rock. Also in my 

 experiments in treating fragments of nepheline to a red heat, a thick- 

 ness of only y^th of an inch must in some cases have resisted a pres- 

 sure of several thousand feet. Since water has not penetrated into the 

 vapour-cavities, or into the vacuities in the glass-cavities, in ancient 

 trappean rocks, it should appear that a considerable pressure for a 

 very long time will not cause it to pass through the solid substance 

 of crystals. In my opinion this could only happen by the formation 

 of actual cracks, which, as sometimes happens with cracked glass, 

 were healed up by the adhesion of the sides, after the fluid had en- 

 tered. This, however, would differ as much from the escape of the 

 fluid when the crystal is strongly heated, as the breaking of an arch 

 by a symmetrical pressure from above would differ from a fracture 

 produced by a pressure from the inside ; and, therefore, if fluid- 

 cavities can resist so great a pressure from within, it appears to me 

 that none at all probable could burst them from without. That the 

 fluid remains permanently in many of the cavities in quartz, is, I 

 think, proved by the fact that, when specimens are obtained below 

 low-water mark on the sea-coast, and afterwards kept in water, the 

 cavities are in no respect different from those in very thin fragments 

 which have been kept dry for years. On the whole, in the absence 

 of any proof of the contrary, I think these reasons are sufficient to 

 warrant the conclusion that the aqueous solutions enclosed in the 

 fluid-cavities in the quartz of granitic rocks, were caught up during 

 the formation of the crystals, and have remained ever since, herme- 

 tically sealed up in their solid substance, without any increase or 

 diminution of the fluid ; and that therefore we may determine from 

 their present condition the circumstances under which the rock was 

 originally formed. 



In my opinion, the water associated with thoroughly melted igne- 

 ous rocks at great depths does not dissolve the rock, but the rock 

 dissolves the water, either chemically as a hydrate, or physically as 

 a gas. In the case of those obsidians and pitchstones which, when 

 heated to redness, give off water having a strong acid reaction, it may 

 probably be in the form of a hydrate, retaining its water when heated 

 under pressure. It is also sufficiently probable that, as suggested 

 by M. A.ngelot (Bulletin de la Societe Geologique de France, 1 ser. 

 t. xiii. p. 178), fused rock, under great pressure, may dissolve a con- 

 siderable amount of the vapour of water, in the same manner as 

 liquids dissolve gases. In either case, if the fused rock passed by 

 gradual cooling into anhydrous crystalline compounds, the water 

 would necessarily be set free ; and, if the pressure was so great that 

 it could not escape as vapour, an intimate mixture of partially melted 

 rock and liquid water would be the result. It is difficult to form 

 any very definite opinion as to the actual amount of this water, and 



