THEORY OF SOLUTION AND CRYSTALLISATION. 261 



whilst equivalent quantities of uncombincd water and lower 

 hydrates are pari passu liberated.* 



This continues until a temperature is reached at which some of 

 the anhydrous salt is formed. This compound, however,, although 

 it is much less soluble than the hydrated salt, does not at once 

 make its appearance in crystals, because the solution must first be 

 saturated with it. But no sooner is a portion of this salt deposited, 

 than a new set of conditions is at once established, for the 

 removal of the anhydrous salt from the liquid is, of course, 

 equivalent to the addition of water, and so time must elapse, or the 

 temperature be raised still higher, before any more of the anhydrous 

 salt can be thrown down.f 



There is no reason to suppose that anhydrous sulphate of sodium 

 is insoluble in water as is sometimes maintained, for if it were, it 

 would be possible to throw down the whole of the salt by simply 

 heating the liquid strongly ecough. But this is very far from being 

 the case, for on referring to Lowel's tnble of the solubility of 

 sodium sulphate, we find the rate of decrease of solubility very slow. 



* It is obvious that this dissociation cannot be proportional to the rise of 

 temperature, because the resultants of the dissociation remain in the field and 

 interfere with its progress as already explained. 



t One other circumstance ought also to be taken into account, and that is 

 the tendency of moving (liquid) molecules to attach themselves to one 

 another, and to the fixed (solid) molecules of the crystal. Even when an 

 equilibrium has finally been established, there is still some reason for helieving 

 that an interchange between the faces of the crystal and the dissolved salt is 

 still going on. 



