ON DYNAMIC ISOMERISM. 113 



all proportion to the decreased concentration of the base. In one 

 extreme case the isomeric change in a solution to \vhich a trace of 

 piperidine had been added practically ceased at the end of a fortnight, 

 although the final condition of equilibrium had not by any means been 

 reached. The more concentrated acid solutions produced an acceleration 

 of isomeric change, but a retardation was noticed when smaller concentra- 

 tions were used : in one case the rotatory power of the solution remained 

 absolutely constant during a period of twenty-four days. 



The earliest observations of arrested isomeric change had led to the 

 conclusion that in non-oxygenated solvents the change was conditioned 

 by the presence of a third substance, probably a basic impurity. The 

 retardation produced by the acid chloroform solutions could not be 

 attributed to mere neutralisation, since an acceleration was observed 

 when benzene was used as the solvent. It was noticed, however, that 

 the chloroform solutions acquired a pungent and unpleasant odour, 

 which was ultimately traced to the formation of carbonyl chloride by 

 oxidation of the chloroform CHCI3 + = C0C1_> + IIC1 ; to this substance 

 the curious behaviour of the chloroform solutions was undoubtedly due. 



The action of the chloride probably depends on its power of converting 

 ammonia and bases such as piperidine into neutral carbamidcs 



COCl, + 2NH3=2HC1 + CO(NH 2) , 

 COC12 + 2NC5Hh = 2HC1 + CO(NC5H,o).. 



It was found that the chloride was capable of retarding or arresting the 

 isomeric change of nitrocamphor dissolved in benzene or in ether, and 

 that acetyl chloride produced a similar effect in benzene and in carbon 

 disulphide. Phenyl-carbimide did not retard the isomeric change of 

 nitrocamphor dissolved in benzene, and acetyl chloride proved inefficient 

 when added to a solution in acetic acid. In the latter case it is probable 

 that the solvent has itself a catalytic action, and this may also be true 

 when water and alcohol are used. The bearing of these observations 

 on the problem of chemical change has been discussed elsewhere.^ 



C. — Relationship between Absorption Spectra and Isomeric 



Change. 



As the lesult of a series of experiments on the absorption spectra of 

 substances of the ethyl aceto-acetate group, Baly and Desch ^ were led 

 to the conclusion that there was a close relationship between the 

 absorption bands which appeared in the spectra of some of these com- 

 pounds and the property which they possess of undergoing reversible 

 isomeric change. In view of the importance of this hypothesis in 

 relation to the subject of dynamic isomerism it was thought to be 

 desirable to test it in the case of a number of optically active 

 compounds. These have the advantage that many of them can be 

 crystallised out in a pure enolic or in a pure ketonic form, that it is 

 possible by means of polarimetric observations to determine the actual 

 velocity with which they undergo isomeric cliange, and that in a number 



' Trans. Chem. Sue, 1908, 93, 110-132. 



' Ibid., If 04, 85, 1029; 190.J, 87, 760, 



1908. I 



