436 ‘TRANSACTIONS OF SECTION B. 
are produced which interact under the influence of the electric stress." In 
the case of alcohol no such interchange takes place. It may be that the 
alcohol is hydrolated to some slight extent hut the hydrol must be less 
basic than hydronol; probably, like ammonia, alcohol exists in solution for 
the most part in the hydronated state :— 
ae H 
H,0¢ ELHOC EtHO—OH, 
Nor OH 
Hydronol, Ethanol-hydrol. Ethanol-hydrone, 
Much more must be learnt of the properties of solutions before definite 
decisions can be arrived at with regard to such delicate and refined issues. 
I would apply the interpretation here given of the nature of conducting 
solutions generally to the explanation of all cases of chemical change: in 
other words, I assume that in all cases correlative systems are present 
which are formed by the reciprocal distribution of the interacting sub- 
stances. Irom this point of view the solvent is no mere medium but an 
active participant in the series of interchanges of which, as a rule, only 
the final product is noticeable. 
The solution thus offered of the complex problem discussed very fully in 
my Address in 1885, which has ever since occupied my thoughts, will, I 
trust, be found to be helpful, although by no means complete in all its 
details. 
In effect, the doctrine makes no demand which the chemist should not 
be able to grant forthwith, as it is generally supposed that hydrols are 
easily formed—to give an example, in the case of the conversion of chloral, 
CCl,.COH, into chloraldehydrol (chloral-hydrate), CCl,,.CH(OH),. The 
novelty of the conception lies in supposing that the occurrence of electro- 
lysis involves the interaction of the hydrol and its correlative and the 
explanation which it affords of the difference between electrolytes and non- 
electrolytes. 
It is essentially an association theory, although it involves the dissocia- 
tion of the interacting substances but never the production of separated ions. 
In the case of aqueous solutions the amount of the distributed substances 
may be taken as the measure of the activity—of the degree of ionisation, 
so-called. A wrong view prevails that the so-called molecular con- 
ductivities are measures of activity ; they are in reality only measures of the 
relative activities under corresponding conditions of the substances to which 
they refer. The molecular conductivity of an acid is at a maximum in 
its weakest solutions: presumably it is then present to the maximum extent 
in its simplest state and in the active hydrolated state; but as a hydrolytic 
agent its activity is at a maximum near to the opposite end of the scale. In 
other words, the hydrolytic activities of a series of acids are in the order 
of their molecular conductivities in solutions of comparable strength, but 
molecular-hydrolytic and molecular-electrolytic activity run in opposite 
directions; the maximum electrolytic conductivity of an acid solution, 
which is manifest at a particular degree of concentration—presumably at 
the point at which the two forms of the distributed materials most nearly 
balance—is also in no way identical with maximum molecular hydrolytic 
activity. On these assumptions not a few of the deductions based on the 
ionic dissociation hypothesis are clearly fallacious. 
It has been asserted that the association hypothesis does not admit of 
quantitative treatment, and that therefore it is at a disadvantage; but if 
the quantitative meaning given to various results in accordance with the, 
* I would repeat the plea I put forward in 1885 that the use of the term hydro- 
chloric acid as applied to hydogen chloride is undesirable if not unjustifiable; the 
solution of the gas may be said to contain chlorhydric acid, HCl (OH,)z, From 
my point of view, oxygen is a constituent of all acids ; 
