SORBY STRUCTURE OF CRYSTALS. 465 



may, I think, as fully rely on the ratios deduced from the minute 

 tubes enclosed in solid crystal, as from the larger tubes employed in 

 our experiments. Sometimes, indeed, it is difficult to determine that 

 ratio accurately ; but very often a close approximation can be made 

 by means of the microscope-micrometer, as already described. 



§ 2. Crystals formed by sublimation. 



In this case, if no fluid be present, of course no fluid-cavities can 

 be formed; but irregularities in the growth of the crystals cause 

 them to catch up gas or vapour, so as to form what may be con- 

 veniently distinguished by'the names gas- or vapour-cavities. Such 

 cavities are well seen in hydrochlorate of ammonia, as shown by 

 fig. 38, or in corrosive sublimate, fig. 39. They are, in fact, like 

 bubbles of more or less regular form, enclosed in the solid crystal, 

 and differ from full fluid-cavities in having a broader and darker 

 margin. If formed at a high temperature and under great pres- 

 sure, since the enclosed highly compressed vapour might be condensed 

 on cooling, the cavities might be covered with small crystals, or 

 contain some liquid, according to the nature of the enclosed vapour. 



§ 3. Crystals formed by fusion. 



The formation of crystals from a state of igneous fusion is in every 

 respect analogous to what takes place when crystals are formed in 

 water. It is simply the deposition of crystals from solution in a 

 liquid that becomes solid at a high temperature, or the crystallization 

 of that liquid itself, in the same manner as when crystals are deposited 

 from solution in water, or the water itself freezes. Nevertheless the 

 temperature at which water and melted rocks become solid are so very 

 different, that the two processes may be conveniently classed under 

 different heads. It is quite inaccurate to suppose that the presence 

 of water is essential in the formation of crystals ; for, as every chemist 

 well knows, many can be formed without a trace of water being present. 



A glass is a liquid which, on cooling, becomes more and more 

 viscous, and at length solidifies without undergoing any sudden and 

 definite change in physical structure. If, however, the liquid, after 

 cooling to a certain temperature, crystallize, it undergoes a sudden 

 and entire physical change, and the structure becomes stony. In 

 most cases the crystals thus formed possess double refraction, and 

 therefore depolarize polarized light ; whereas the uncrystalline glass 

 has no more influence on it than such a liquid as stiff Canada 

 balsam. 



The best illustration I have met with of the characteristic structure 

 of crystals formed artificially by the cooling of a heterogeneous mass 

 in the state of igneous fusion, are the thin flat crystals of basic 

 silicate of protoxide of iron, so common in the slags of copper- and 

 nickel-ores. Their surfaces are covered with deep striae ; and when 

 crystals sufficiently thin to be partially transparent are mounted in 

 Canada balsam and examined with a microscope, it is seen that many 

 such strise have been covered up, so as to form tubular cavities 



