654 SCIENCE PROGRESS 



It became necessary to revise the data relating to con- 

 centrated solutions from this point of view and especially to 

 examine acids and salts under comparable conditions. The 

 work carried out in my laboratory during the past few years 

 has afforded material for what appears to be a rational explana- 

 tion of the behaviour of substances the solutions of which 

 behave as electrolytes. The assumption made is that the 

 formation of composite electrolytes by the dissolution in water 

 of substances such as the acids, the alkalies and salts generally, 

 is a process involving the distribution of both constituents of 

 the composite electrolyte — thus : 



/ H 

 HC1 + OH, «— > HC1< 



\0H 



/CI 

 HX) + C1H <— > H 2 0< 



\H 



and that the occurrence of electrolysis involves the interaction 

 of the two kinds of complex while under the influence of an 

 electromotive force. The extent to which this distribution 

 occurs determines the activity of the composite electrolyte. 



In the case of non-electrolytes, it is to be assumed that only 

 the water is distributed, thus : 



C:H 5 \ C-H 5 \ /H 



>0+0H 2 <— > >0< 



h/ h/ \oh 



The superior activity of salts, etc., in disturbing the proper- 

 ties of water is due to the fact that besides acting mechanically, 

 they also exercise an attractive effect of their own ; in the case 



/IT 



of a solution of hydrogen chloride, for example, the HC1\' ^xj 

 complex in virtue of the presence of the OH radicle, the 



/IT 



H 2 Ov p, complex in virtue of the presence of the CI radicle, 



would both be active in much the same way that the hydrone 

 molecule in water is active. 



A completely distributed binary molecule AB would there- 

 fore exercise the effect of two molecules in disturbing the 

 properties of the water. The further consequences of this 

 hypothesis must be painted on another occasion. 



