4/8 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. 



the cooling of a saturated, aqueous solution. There is therefore, in 

 my opinion, no more necessary connexion between the temperature 

 at which the crystals were deposited from this glassy solvent and 

 their own fusing-point, when heated alone, than between the tem- 

 perature at which crystals are deposited from solution in water and 

 their own fusing-point, even if they be fusible. In both cases, the 

 only necessary connexion is, that the crystals could not be deposited 

 in a solid form, except at a lower temperature than that at which 

 they become liquid ; but it might be any heat less than that high 

 enough to cause the glossy solvent to be sufficiently fluid. These 

 facts are of very great importance in the study of igneous rocks, 

 and serve to explain several peculiarities in their structure. Such 

 glass-cavities, however, differ essentially from fluid-cavities, in con- 

 taining bubbles that never move, and do not change their place or 

 disappear when the fragment containing them is heated, unless the 

 heat is strong enough to melt the enclosed glass, which is more 

 fusible than the felspar. 



The best examples of glass-cavities that I have met with in the 

 erupted lavas of Vesuvius occur in the augite. One very excellent 

 case is shown by fig. 65. They contain, at least, two kinds of 

 crystals, which sometimes project beyond the general outline of the 

 cavities, as shown in the figure, as if they were formed at the same 

 time as the augite, and were caught up in it along with the fused 

 material of the glass-cavity, which on cooling deposited other crystals, 

 and by contracting gave rise to a small bubble. In some cases long 

 prismatic crystals have been caught up in the augite, as shown by 

 fig. 66, having two glass-cavities attached to them, one with a bubble 

 and the other without, which is not unfrequently the case in detached 

 cavities, as if, like in some fluid-cavities, the cohesion of the sides 

 had overcome the contraction of the melted glass. 



The leucite in- the lava of Vesuvius often contains many cavities, 

 the material in which has to a great extent become crystalline, and 

 therefore they are very commonly stone-cavities. An example of 

 one, partially stone and partially glass, is shown by fig. 68, which is 

 somewhat analogous to those in the felspar of the pitchstone, repre- 

 sented by fig. 63. Another form is shown by fig. 70, and a very 

 curious, almost circular, flat cavity is seen in fig. 69, containing three 

 different kinds of crystals ; whilst fig. 67 represents a crystal enclosed 

 in the solid leucite, with a small stone-cavity attached to it. In no 

 case have I seen decided bubbles in the cavities in leucite ; but their 

 absence from cavities containing many crystals is easily explained, 

 because many substances expand in crystallizing to such an extent 

 as would compensate for the previous contraction from a high tem- 

 perature. In the felspar of the trachyte of Ponza the cavities are 

 all filled with stony matter, as shown by fig. 71- A very long 

 tubular cavity is represented by fig. 72. 



The general arrangement of these various glass- and stone- cavities 

 is precisely analogous to that of those in crystals formed artificially ; 

 and, independent of the fact that, in all their essential characters, 

 they are identical with the cavities in the crystals in artificial furnace 



