ON ELECTROLYSIS. 



353 



able to feeble E.M.F.'s ; and that most perfectly pure bodies, undisturbed by heat 

 or by admixture with foreign matter, have their molecules in this condition ._ When, 

 however, two or more substances (like salt and water for instance) are mixed to- 

 gether, their simple molecules combine into somewhat indefinite molecular aggre- 

 gates or hydrates of complex structure, in which it may happen, either that some 

 of the outlying atoms are not held with full vigour and so become partially free 

 and able to interchange with other similarly placed atoms, or else that by the 

 coUision of the miwieldy aggregates with each other the atoms of one molecule are 

 brought so close to the respectively opposite atoms of another molecule that a mutual 

 interchange occurs. 



For it is well known that it is quite unnecessary to postulate atoms as hovering 

 around in an entirely free or disemmoleculed condition. All that is wanted to 

 explain the facts of electrolysis is a certain number of atomic interchanges 

 occurring at random ; for the slightest E.M.F. will then suffice to exert a directive 

 influence one way or the other on the atoms during their infinitesimal moment of 

 freedom. 



The operation is perhaps more simple to contemplate if the atoms are imagined 

 to be actually, instead of only (.so to speak) potentially, free, but the result is the 

 same ; and of course the conductivity will depend upon the number of such ' free ' 

 atoms existing in the solution, the molecular conductivity k\in representing the pro- 

 portion of such free atoms to the whole. It is upon this very same proportion, 

 according to Arrhenius, that the chemical activity of the substance depends. 



Whatever be the most satisfactory form of hypothesis to hold at present, I am 

 bound to say that the hypothesis of Dr. Armstrong does not commend itself to me. 

 And that for several reasons. 



I make no objection to his notion of residual affinity, nor to the formation of 

 molecular aggregates by means of it. Those are points for the consideration of 

 chemists. The objectionable, and I venture to think: fatal, part of his hypothesis, 

 is where he supposes dissociation to be produced by means of the applied E.M.F., 

 instead of independently of it. Few things are more certain than this, that an 

 electrolyte is incompetent to resist the smallest E.M.F. really applied to it (i.e., not 

 applied only to electrodes). If the molecules have to be torn asunder by any 

 action depending on the magnitude of the applied E.M.F., it must be possible to 

 choose an E.M.F. too weak to effect decomposition. A substance which needs an 

 E.M.F. to tear its molecules asunder in the interior of its mass is ipso facto a di- 

 electric — it may be a very weak one — but it is not an electrolyte. All that a slope 

 of potential can possibly achieve in the interior of an electrolyte is to direct a pro- 

 cession of otherwise randomly mo^'ing atoms. This is, of course, the foundation of 

 the theory of Williamson and Clausius ; and all experimental knowledge acquired 

 since the time it was first promulgated with regard to the obedience of electrolytes 

 to Ohm's law, down to the researches of Professor Fitzgerald and Mr. Trouton, 

 communicated in part to the last British Association and still going on, tends but 

 to confirm and strengthen that position. 



Of course, at surfaces of discontinuity (i.e., at electrodes) electrolytes need 

 a finite E.M.F. to liberate their ions, but the range over which the stress here 

 concerned acts is of atomic dimensions, say 10—'* centimetre from each electrode. 

 There is no such finite stress needed, or possible, in the interior of a homogeneous 

 liquid. 



A minor objection to Dr. Armstrong's hypothesis may be made at the point 

 where he supposes endosmose to precede conduction, and to be a phenomenon inde- 

 pendent of surface contact. It is not easy, again, to imagine why his molecules 

 should be travelling past each other in the fluid, nor why, even if they did, this fact 

 should assist their previously incompetent forces to disrupt each other. 



Unless every molecule is supposed to be in a condition extremely like every 

 other molecule, which is quite contrary to the usual doctrine of averages as applied 

 to molecules, it is difficult to beheve that a number of molecules are in such a state 

 of strain as to be made to break up with mere gentle locomotion, and yet that none 

 of them shall be able to break up without such assistance. 



1887. A A 



