FORMATION OF CRYSTALS—GAUBERT. et 
it is the mother liquid which is absorbed its purification is impossible. 
The solvent must be changed. 
When the crystals of a determined substance obtained from two 
different solutions do not present the same forms, it is incontrovertible 
that in one of the cases, perhaps in both cases, since we do not always 
know the form of the pure crystal, there has been absorption of the 
molecules of the mother liquor. Sometimes it is the water which is 
absorbed, and this water has been regarded as water of crystallization 
or as water of constitution, according to the temperature at which 
it is driven off. 
When purification is attempted by recrystallization, if the foreign 
substance which passes into the crystal is present in small quantities 
in the mother liquid, the first or the last crystals formed, according to 
the mode of synchronous crystallization, will be the purest. In case 
there is a division of the foreign matter between the crystal and the 
liquid, if the coefficient of its solubility in the crystal and the liquid 
are known, the number of crystallizations demanded for the purifica- 
tion of the crystals may be calculated under proper conditions. 
Vv. 
The natural crystals appear in such varied habits that before Romé 
de l’Isle no one could see the constancy of forms, and the genius of 
Hauy was necessary to establish their derivation. It is known that 
ordinarily the crystals of the same deposit and of the same generation 
are identical, and that those of successive deposits or generations may 
have different dominant forms. All these differences may be ex- 
plained by the rapidity of crystallization, but especially by the con- 
stant presence of foreign substances. Unfortunately it is difficult to 
determine the nature of the latter, since the results of analyses made 
up to the present time have little value in solving this problem. 
Indeed, a very small quantity of matter is required to modify the 
forms of a crystal; sometimes an amount even less than one-one thou- 
sandth of the weight of the latter is sufficient. 
In every case, whether we have to do with natural or artificial 
crystals, we need to determine their form in the pure state, a form 
which is constant and which I have called fundamental. It may be 
distinct from the primitive form chosen by crystallographers. 
In closing, I shall observe that the substances prepared in labora- 
tories seem rarely to show the numerous modifications of form, so 
frequent in the natural crystals. This is due to the fact that the 
artificial crystals are prepared almost always in the same manner, 
with the same reagents and consequently with the same foreign sub- 
stances in the mother liquor. In nature, on the contrary, as the 
