242 REPORT — 1901. 



Thus, when an acid and a base are brought together, the neutralisation 

 never takes place quite completely. There always remain as many free 

 hydrogen and hydroxyl ions over as are usually present in pure water. 

 The quantity of ionised water is, of course, so small as to be practically 

 nefflisible in most cases, but its effect becomes very marked when the acid 

 or base of a dissolved salt is very weak. 



If we take, for instance, a salt like potassium cyanide, its acid, hydro- 

 cyanic acid, is very weak, and is still further enormously weakened by the 

 presence of its neutral salt, or, to put it in ionic language, by the presence 

 of excess of cyanogen ions. The water is therefore by virtue of its slight 

 acid properties capable of setting free a considerable quantity of the acid 

 from its salt. 



It might appear at first sight as if the solution should still react 

 neutral, since the acid and base are set free in equivalent quantities. The 

 theory of electrolytic dissociation shows us, however, that this is not the 

 case. If we consider the equilibrium ; 



KCN + HOH^KCn + HCN 



the potassium hydrate exists practically completely in the ionised state, 

 whereas the hydrocyanic acid is almost entirely unionised. Thus we have 

 a large excess of hydroxyl ions in the solution, and it is these that give 

 rise to the alkaline reaction. Expressed ionically the equilibrium will 

 read 



CN' + HOH^^HCN + OH'. 



This theory of Arrhenius has now met with almost universal accept- 

 ance, and has amply justified its adoption as a working basis for all 

 quantitative problems dealing with hydrolysis. 



The conditions for the dissociation of a salt into free acid and base are 

 therefore— 



1. That the acid or base of the salt, or both, be very weak. 



2. That the solvent itself be somewhat ionised. 



Hitherto the phenomenon appears only to have been studied in aqueous 

 solution. If the slight conductivities found for pure alcohol are really 

 due to an ionisation into hydrogen- and ethoxy-ions, then we should 

 expect salts such as sodium phenolate to be also split up to some extent in 

 alcoholic solution. 



For the qualitative detection of hydrolysis, indicators afford the most 

 reliable test. From the results of Ley,^ litmus appears to bo the most 

 sensitive of these. 



Still, the method of simply testing the solution with an indicator 

 might at times give misleading results owing to the presence of traces of 

 acid or alkali in the salt. Ley recommends a more satisfactory method. 

 This is to titrate the solution. If the salt of a weak base, for instance, is 

 really hydrolysed, it will not only react acid in the pure state, but will 

 also continue to react acid even on addition of a considerable quantity of 

 alkali. Thus, whei-eas the least trace of sodium hydrate sufficed to render 

 a solution of magnesium sulphate or barium chloride alkaline, solutions of 

 lead chloride and copper chloride continued to react acid until almost the 

 whole of the hydrochloric acid had been removed by the sodium hydrate. 



As other qualitative methods any proc'e'sses may be used which bring 



> ZnUch: fiir phys. Ctiiim., 30, 203 (1899). 



