ON ELECTROLYSIS IN ITS PHYSICAL AND CHEMICAL BEABINQS. 357 



electricians in the case of copper, and whicli has recently been brought pro- 

 minently into notice in the case of g-old by Prof. Hoberts- Austen, may be cited 

 in illustration of the kind of effect on the solvent which conceivably may arise.' 



The arguments in favour of the dissociation theory appear indeed to be based on 

 a narrow interpretation of the results : the dissolved substance alone is regarded as 

 active, the solvent is neutral ; the indubitable complexity of the phenomena of 

 dissolution and of chemical interchange generally is entirely left out of considera- 

 tion. Many chemists and physicists continue to regard values as simple constants 

 which there is the strongest reason to believe are composite. The law of the 

 constancy of the so-called atomic heats of the elements is an illustration. It is well 

 known that in the case of the majority of the elements the product of specific heat 

 by atomic weight is approximately a constant : we are thereby enabled in many 

 cases to determine atomic weights, but, be it noted, by a measurement effected with 

 the aid of molecules, not of atoms ; these molecules are undoubtedly of very varied 

 atomic complexity, and it would seem that we are bound to assume that the nearly 

 constant value of the so-called atomic heat is the outcome of a kmd of balance — 

 that as, in raising the temperature of a solid, work is done inter- and tn^rrt-molecu- 

 larly, the amount of intermolecular work is greater in some cases and less in others, 

 and the amount of intramolecular work proportionally less or greater, the amount 

 per atom being, however, about the same in most cases. May we not interpret the 

 simple results obtained by Raoult's method in a similar manner? Is it not 

 probable that the effects are in reality often produced by molecular complexes of 

 the fundamental molecules, the latter corresponding to the atoms to which we 

 refer the result in discussing variation of specific heat with atomic weight ? 

 Perhaps it is only in cases in which atomic rearrangement has taken place that 

 Raoult's method affords e\'idence of increased molecular complexity, as in the 

 case of aldehyde and paraldehyde, for example ; in such a case as that of dextrose, 

 ■which, whether freshly dissolved or not, produces an effect such as is required on 

 the assumption that its molecule has a weight corresponding to the formula 

 C^IIijOg, there may be no atomic rearrangement but the mere juxtaposition of 

 molecules involved in the formation of a ' molecular ' compound. 



Arrheuius, Ostwald and others regard both electrical conductivity and chemical 

 activity as similarly conditioned by the degree of dissociation — in their opinion, 

 very active substances, such as sulphuric acid, are to a large extent dissociated in 

 solution ; inert substances, such as acetic acid, are but to a slight extent dissociated 

 in solution. But the adherents of this school all overlook the fact that there are 

 two distinct theories of chemical interchange : the older theory that the interacting 

 molecules initially combine and that the resulting complex then splits up — which 

 may be termed the integration theory ; and the more modern dissociation theory. 

 I am led to regard the former as the more comprehensive and generally applicable, 

 especially as comparatively so few compounds are electrolytes, and I venture to 

 think that physicists also would incline more to my belief if they would assume a 

 somewhat different mental attitude towards the facts, and would seek to fully 

 unravel the entire series of changes involved in chemical interactions. 



The attention of the Committee has in previous years been directed to the 

 question whether electrolytic conduction and metallic conduction are sharply 

 separated from one another, and last year experiments by Prof. Roberts-Austen 

 were described which appeared to show that alloys do not conduct electrolytically. 

 Prof. J. J. Thomson deals with this question in his already-mentioned work in the 

 following passage (p. 296) : — 



' There does not seem any necessity for supposing that the passage of electricity 

 through metals and alloys is accomplished in a fundamentally different way from 

 that through gases and electrolytes. For the chief differences between conduction 

 through metals and through electrolytes are — (1) that in electrolytic conduction the 

 components of the electrolyte appear at the electrodes, and we have polarisation ; 



' Prof. Eoberfs-Austen informs me that the melting-point of gold is lowered to 

 about that of zinc by ^^ of silicon. 



