towards a Dynamical Theory of Solutions. 59 



the physical properties at the concentrations we have been 

 studying. Even in fairly dilute solutions, such as 1 per cent., 

 alcohol exists as (0 2 H 5 OH) 2 and acetic acid as (CH 3 COOH) 2 . 

 Yet by the method of the lowering of the freezing-point it 

 has been accepted as proved that in such solutions alcohol 

 and acetic acid exist as C 2 H 5 OH and CH 3 COOH. I think 

 that the method of the lowering of the freezing-point fails 

 here to give the whole molecular weight and yields the 

 weight of the nearly free halves of the double molecules. 

 According to the theory of the present paper it is assumed 

 on the contrary that the trihydrol and dihydrol of water in 

 excess of the alcohols and fatty acids are converted into 

 hydrol consisting of single molecules of H 2 0. But water in 

 all solvents at small concentrations gives only molecules of 

 the form H 2 (the latest reference is that of Walden to 

 Bruni & Amadori, Trans. Faraday Soc. v. 1910). For the 

 reason just given we cannot accept this as proof that water 

 at low concentrations in other solvents exists as hydrol, 

 because the methods hitherto used involve the assumption 

 objected to in that of the lowering of the freezing-point. 

 At all events, I can discover no evidence against the view 

 that in a dilute solution of water in alcohol and in acetic 

 acid the water is changed into hydrol, as in the vapour of 

 water. 



According to the principles of the present paper, the 

 ionization of weak electrolytes takes a new aspect. If the 

 two groups in (CH 3 COOH) 2 are so nearly free tbat each acts 

 as if free in the lowering of the freezing-point of a solvent, 

 then in that solvent the conditions are favourable to the 

 occurrence of dissociation amongst the double molecules 

 according to the laws of probability. It is to be expected 

 then that at infinite dilution in water alcohol (CVH 5 OH) 2 is 

 dissociated into C 2 H 5 OH. To follow the process there would 

 need to be very accurate measurement of the physical pro- 

 perties of dilute solutions of alcohol. In the case of the 

 acids the single molecule like CH 3 COOH is ionized into ions 

 like HjJ and CH 3 COO|?, and electrical methods of easy 

 application become available for studying the changes taking- 

 place. It appears that the double molecule of the fatty and 

 other organic acids is not an electrolyte, but that the single 

 dissociated molecule is an electrolyte. To it applies the 

 principle which I have stated for electrolytes in genera], that 

 their ionization is always complete. Hence in a dilute solution 

 of acetic acid the equilibrium is between (CH 3 COOH) 2 on 

 the one hand and H£ and CH 3 COOt> on the other. The 



