A.— MATHEMATICAL AND PHYSICAL SCIENCES. 33 



electroljrtes. These do not follow the law of mass action. Debye assumes 

 complete dissociation, but with electrical attractions between the ions due 

 to their positive and negative charges. These forces give rise to what 

 may be called potential combination following, however, a different law 

 from that of mass action, and the difficulty in regard to ' strong ' electro- 

 lytes is removed, though only for dilute solutions. The important fact, 

 however, is that some of the zinc goes into the solution, carrying positive 

 charges with it. It goes in not merely by evaporation, as in Nernst's 

 theory, but it is in part pulled in by the SOj-ions carrying negative 

 charges. Again at the copper plate copper is deposited not freely as a 

 vapour might condense, but is retarded by the attractions of the SO^- 

 ions in the solution. Both plates act as semi-permeable membranes, 

 passing selected substances and stopping others. So far as we know, no 

 electricity gets through either of these membranes except as a rider on 

 an ion. At any rate this must be so as long as Faraday's laws of electrolysis 

 hold good. 



On the other hand the membrane (porous pot, &c.) separating the two 

 solutions acts as a membrane more nearly of the metallic kind. Electrons 

 that were riding on SO^-ions get through, leaving their mounts behind. 

 Few membranes will act in precisely this way, and considerable variety 

 may therefore exist in the voltage changes at this membrane. It is not 

 unlikely that the voltage there may be of the same order as that at the 

 outside copper-zinc junction, but of opposite sign ; for in both cases 

 electrons alone are passing. If this is so, then the electromotive force 

 of a circuit may, at least approximately, be the sum of those arising at 

 the metal-liquid junctions. 



From what I have said it will be clear that my opinion is that it is 

 still necessary to be cautious and to avoid dogmatism on this question. 

 Much more detailed experimental knowledge is required before the electric 

 circuit is really understood. The electronic theory in metals still has its 

 difficulties, which it is useless to ignore. It is only by recognising the 

 difficulties that advance is made. On the other hand the experimental 

 difficulties in connexion with the direct measurement of Volta effect are 

 also very great, as all who have made experiments on it must know. 



I had hoped to be able to present to you some new experimental 

 data. I am not satisfied, however, that I understand the meaning of the 

 vagaries that often occur, and I do not mean to publish anything now. 



I wish to say, however, that I am impressed by the excellent and novel 

 work that is being done by Millikan and by 0. W. Eichardson on this 

 question. 



Both the experiments and the theory are associated with great 

 difl&culties. My own opinion is that, though the voltage at the metal- 

 metal junction is likely to be much larger than the chemical school 

 demanded, there is nothing to justify one in going to the opposite extreme 

 and expecting that the whole of the electromotive of a circuit is located 

 at that junction. Opposing schools may both take comfort in the thought 

 that in some respects they are both right. 



Of course no difficulty is introduced if it is concluded that the contribu- 

 tion of an element of the ciicuit to the total e.m.f. is not measured by the 

 heat taken in locally thereat. On both sides of the controversy it is well 



1928 D 



