.524 BEPORT— 1886. 



still simpler than the inverse square law — viz., that no element of the surface has 

 any action within the closed conductor. If we suppose that a conductor is a com- 

 plete screen to electrical action, then, whatever the law of the force exerted across 

 an insulator, there will he no action within the conductor. In any null proof it is 

 not sufficient merely to show that there is no action in the null arrangement, but it 

 is also necessary to show that on disturbing- the null arrangement some action is 

 manifested. Now, in the case here considered it is impossible to obtain any action 

 within the conductor in any statical arrangement ; it is only during changes of the 

 system while charging or discharging that we can get a disturbance of the null 

 arrangement. But here new phenomena come in, for we have currents, and there- 

 fore electro-magnetic action. But, even disregarding the different kind of action 

 occurring, the only experiment which I know of on this point was that of Faraday 

 with his electrified cube. While the most ^dolent charges and discharges were 

 taking place on the outside of the cube, so that the null arrangement was probably 

 disturbed, he found no action on his electroscope within. Possibly the actions were 

 alternating, and so rapid that no electroscope of ordinary construction would reveal 

 them. But he himself went into the cube, and he would probably be sensitive to 

 rapidly alternating electromotive forces. It appears to me, then, that we cannot 

 accept this proof, and must fall back upon the more direct proof of Coulomb. I do 

 not know whether Maxwell was aware of this objection ; but it is worthy of note 

 that in the remarkable fragment published since his death, as ' An Elementary 

 Treatise on Electricity,' he returned to Coulomb's proof, and was apparently 

 building up the mathematical theory of electricity in a way quite different from 

 that followed in his larger work. 



10. On the Electrolysis of Silver and Copper, and its Application to the 

 Standardising of Electric Current- and Fotential-Meters. By Thosias 

 Gray, B.Sc, F.R.S.E.^ 



This paper contained an account of a large number of experiments on the electro- 

 lysis of silver and copper, and its application to the standardising of electrical cur- 

 rent- and potential-meters which have been made during the past year in the 

 Physical Laboratory of Glasgow University. It forms a summary of the reports 

 made from time to time to Sir William Thomson on this subject. 



In the earlier experiments pure silver sheet and solutions of pure silver nitrate 

 were used for the electrolytic cells. The results obtained with these cells confirmed 

 those obtained by Kohlrausch, Lord Rayleigh, and other observers as to the great 

 accin-acy obtainable with silver, and also as to the nature of the deposit and the 

 care required in the manipulation of the experiment. The cathode plate consisted 

 in almost all cases of a sheet of pure silver, placed, with its plane vertical, between 

 two parallel sheets of the same metal. In a few experiments a platinum bowl was 

 used very much in the same manner as that described by Lord Rayleigh in his 

 paper to the Royal Society, but generally the plates of sheet silver were preferred. 

 The mode of treating the plates both before and after the deposit was upon them 

 is fully described, and it is pointed out that when the plate is made thoroughly 

 clean, and its size properly proportioned to the strength of the current to be 

 measured, the deposit is finely crystalline, and adheres very firmly to the plate. The 

 best results were obtained when the current density at the cathode is between ~ 

 and 555 of an ampere per square centimetre of the surface on which the deposit is 

 taking place, the solution being supposed to contain five per cent, of silver nitrate. 

 It is also pointed out that the loss of silver from the anode can be used satisfactorily 

 as a check on the amount of sUver deposited on the cathode, but that the current 

 density at the anode must be considerably smaller than that which will give good 

 deposits on the cathode. About ~ of an ampere per square centimetre is stated 

 .as the maximum current density at the anode which can be safely used if the loss 

 of silver is to be weighed. A somewhat greater current density may be used at the 

 cathode if the strength of the solution be increased, but the deposit is more roughly 



' Published in full in the Phil. Mag. for Nov. 1886. 



