218 REPORT—1904. 
of the stable isomeride, which will again melt when the equilibrium tem- 
perature is reached ; frequently the melting and resolidification proceed 
simultaneously, and only a slight sintering is observed in the neighbour- 
hood of the melting-point. 
Although in the liquid state each isomeride is equally labile when 
alone, and equally stable when equilibrium is attained, in the solid state 
only one isomeride can be stabie at any given temperature, namely that 
which has the least vapour pressure ; this is usually that which has the 
highest melting-point, but the opposite case is not infrequently observed. 
In the absence of a catalyst the labile isomeride may be preserved indefi- 
nitely, and may even be fused without undergoing isomeric change, but 
usually the change commences before the melting-point is reached, and 
at a temperature. depending on the amount of impurity present in the 
crystals. This temperature Knorr has described as the stability limit,! 
but it cannot be regarded as a physical constant of the substance as 
Knorr at first supposed, and his later work has shown that in the purified 
material the stability limit becomes identical with the melting-point.? 
4, Dissolution.—Each isomeride has a true solubility which is inde- 
pendent of isomeric change, and may be determined by measuring the 
concentration of the solution obtained when the solid and solvent have 
been in contact during a period of time sufficient for saturation, but 
not long enough for isomeric change to produce any marked effect. 
Ultimately, however, a condition is reached in which the solution contains 
a stable mixture of the isomerides and is saturated with regard to 
one of them. The apparent or ultimate solubility® is thus dependent on 
the true solubility and the composition of the mixture in equilibrium. 
The measurement of the initial and final solubility affords a method 
of determining the composition of the mixture. Thus in the case of 
{-bromonitrocamphor the initial solubility at 10° C.is 2 grams per 100 c.c. 
of benzene, and the final solubility 8 grams per 100 c.c. ; and it may, 
therefore, be inferred that in the ultimate mixture the isomerides are 
present in the ratio 1 : 3 approximately. The method has the advantage 
of being applicable even when only one of the isomerides can be isolated. 
5. Crystallisation from Solutions.—The form that separates first on 
evaporating or cooling a solution is that which has the smallest apparent 
solubility. Usually it will be the major constituent of the mixture, but 
the minor constituent may separate if its true solubility is small. Slow 
crystallisation will then result in complete isomeric change, whilst rapid 
crystallisation will yield a mixture from which the constituents can 
sometimes be separated by picking out the crystals mechanically. 
The order of solubility can sometimes be reversed by changing the 
temperature or the solvent, or both, and in this way it may be possible 
to isolate and purify more than one isomeride. Thus Dr. Whiteley has 
shown’ that the yellow hydroxylic modification of isonitrosomalonanilide 
separates from chloroform or benzene, but the white pseudoxime from 
ethyl acetate, methyl alcohol, ether, or acetic acid, 
HO.N :C(CO.NHPh), _, so 
(from hydrocarbons) + eal 
(from oxygenated solvents). 
C(CO.NHPh), 
1 Ann. 1899, 306, 70 and 88. 
? Wislicenus, ‘Ueber Tautomerie,’ footnote, p. 225. 3 Lowry, Trans. 1899, 75,231. 
4 Lowry, Pree, 1903, 19, 156, 5 Trans. 1903, 88, 34. 
