ON DYNAMIC ISOMERISM. 215 
nitrocamphane is so complete that the pseudo- form cannot be detected in 
the equilibrium, no difficulty is experienced in reconverting it into the 
potassium salt of the pseudo- form. 
Chemical agents, therefore, give very little information as to the 
nature of the equilibrium, and can only be used for separating the 
constituents when isomeric change proceeds slowly under the prevail- 
ing conditions. Thus no difficulty is experienced in separating ammonium 
thiocyanate and thiourea or a mixture of sulphonic acids, because at 
ordinary temperatures and in dilute solution these have no tendency to 
pass into one another, and are, in fact, no longer isodynamic. So 
also Kiister was able to separate the isomeric hexachlorocycloketo- 
pentenes by converting the B- compound into the sparingly soluble anilide, 
C,H;.NH.C,;Cl,0. In the case of substances such as those described in 
Section IV., in which the transference of a mobile radicle takes place 
with the utmost readiness under ordinary conditions and at atmospheric 
temperatures, the majority of chemical agents are useless for this purpose, 
especially as many of them have a very powerful catalytic action, and 
greatly accelerate the isomeric change. The only agent that seems to 
fulfil the necessary conditions is phenyl isocyanate,' which has little or no 
catalytic action, and may even remove any moisture or acid impurity 
already present in the solution ; experimentally it has been found that it 
combines with phloroglucinol but not with ethyl succinosuccinate, indi- 
eating that the former is enolic but the latter wholly ketonic ; the same 
agent indicates that isatin is a ketone, since it gives a substituted urea 
and not a urethane. 
co reg tases aI 
Pu’  CcOH kee H Sco >¢,H,Z co. 
[° NyZF Shee Eich: 2" NOOR)” 
Pseudo-isatin Tsatin 
Benzylidene aniline, C,H,.CH : N.C,H,, although equally ready to com- 
bine with ketones or enols, cannot be regarded as a trustworthy indicator, 
for the addition of a trace of sodium ethoxide appears to determine 
the formation of the enolic, whilst piperidine favours the formation of 
the ketonic addition-product quite independently of the proportions of 
ketone and enol present in the mixture.* 
A different class of agent is typified by ferric chloride, which inter- 
acts only with a minute proportion of the hydroxylic isomeride to form a 
coloured ferric salt. If this were insoluble a complete conversion into 
the hydroxylic form would ensue, but when it remains in solution it does 
not seriously disturb the equilibrium, and so may serve to indicate the 
proportion of hydroxyl present. 
.CH,.CO.2 . CH : COH 2 .CH : C(OFeCl,). 
In the case of the simple nitro-paraffins and isodynamic ketones ferric 
chloride can be used to follow the gradual conversion of one form into 
the other, but in the case of the nitro-ketones it has such a powerful 
catalytic action that equilibrium is established immediately, and no differ- 
ence can be detected either in the intensity of the colour or the rapidity 
! Goldschmidt, Ber. 1890, 23, 257. 
* This conclusion is in agreement with that arrived at by Hartley from a study 
of the absorption spectra. 
* Schiff, Ber. 1898, 31, 601, 
