208 SCIENCE PROGRESS 



The variations in the products ev\ in the above table are prob- 

 ably within the limits of experimental error, and it is clear 

 that, at least in the case of this particular solute, the Nernst- 

 Thomson rule may have not only a qualitative but actually a 

 quantitative significance. It is necessary, however, to call 

 attention to the fact that the solute selected was not a typical 

 salt, but a somewhat complex organic compound. Whilst, 

 therefore, the simple relationship ev\ =f(a) may hold good in 

 certain limiting cases, it is more than doubtful whether it would 

 survive if applied to compounds exhibiting such strong in- 

 dividuality as is displayed by the metallic salts. 1 



Inadequacy of the Physical Theory of Ionisation. The 

 Solvent as a Source of Energy 



Having shown in the above discussion the important re- 

 lationship which exists between specific inductive capacity and 

 ionisation, it is necessary to add some criticisms of the physical 

 explanation of the phenomena of electrolytic dissociation. In 

 the first place it should be noted that the explanation is in- 

 complete, in that no hint is given as to the origin of the different 

 coefficients of ionisation for different solutes ; the physical 

 theory of the action of the solvent in electrolysis must therefore 

 in any case be supplemented by a further theory — presumably 

 chemical in character — in reference to the function of the solute. 

 In the second place it should be noted that the mere weakening 

 of the electrical forces between the ions does not afford an 

 adequate explanation of the dissociation of the salt. 



The latter point bears directly on the fundamental difficulty 

 of the theory of electrolytic dissociation — the source of the 

 energy required for ionisation — and is of such importance as 

 to demand separate discussion. ' To take a concrete case : the 

 formation of solid potassium chloride from its elements is 

 accompanied by the liberation of 105,600 calories, and a corre- 

 sponding amount of energy must be supplied if the salt is to be 

 decomposed again into metallic potassium and gaseous chlorine. 

 The dissolution of the salt in a large quantity of water is 

 accompanied by an absorption of 4,400 calories, but the products, 

 according to the theory of electrolytic dissociation, are simply 

 potassium ions and chlorine ions. We are therefore presented 



1 Contrast, for example, the different behaviour of silver chloride and potassium 

 chloride towards water and ammonia. 



