RICHARDS, COLLINS, AND HEIMROD. — COPPER AND SILVER. 127 



the deposit must be too small, while from a cuprous solution the deposit 

 must be too large. Both errors decrease as the temperature decreases ; 

 but it is evidently impossible to obtain results directly with the copper 

 voltameter which will exactly correspond with the atomic weight of 

 copper, except by a chance combination of circumstances. This com- 

 bination would be attained when the amount dissolved from the cathode 

 jy the side reaction Cu+ + + Cu ±=; 2 Cu+ just equalled the extra amount 

 of copper deposited from the trace of cuprous salt in the solution. Such 

 a fortuitous condition could not be maintained for any length of time, 

 because the amount of cuprous salt present must steadily increase by the 

 action at the cathode as well as at the anode. At first sight, it might be 

 supposed possible to increase the current density to such an extent that 

 all the copper in the neighborhood of the cathode should be completely 

 deionized ; but it will be shown that hydrogen begins to be deposited 

 instead of copper before this point can be reached. 



Besides these sources of error another may exist, to which Foerster 

 and Seidel have called especial attention. This fourth eri'or consists in 

 the possible semi-deionizing (or partial reduction) of the cupric salt by 

 the electric current without the deposition of metal. Thus positive 

 electricity may make its exit from the solution without a corresponding 

 increase in the deposit, and the observed atomic weight of copper will be 

 too low. Evidently, however, this action cannot result in the formation 

 of more cuprous salt than the small proportion corresponding to its 

 equilibrium with the cupric remaining in the neighborhood. Hence this 

 fourth cause of error is essentially identical with the first, the simple 

 dissolving of copper in cupric sulphate. The question then arises as to 

 whether or not the equilibrium is established more quickly under the 

 influence of the current than without the assistance of outside electro- 

 motive force. Both Gore and Gray, in the papers already cited, give 

 evidence showing that reasonably small current-densities tend to increase 

 the apparent corrosion of the copper rather than to diminish it. This 

 might have been predicted, for it is easy to imagine that the equilibrium 

 is established more quickly with the help of nascent cuprous ions than 

 with the resting metallic surface of the copper plate. The authors just 

 cited give no evidence to prove, however, that a strong current causes a 

 less proportionate gain in weight than a weak one ; indeed. Gray contends 

 that a strong current may even j^rotect the electrode from corrosion.* 

 The phenomena which led him to this conclusion will be shown later to 

 be due at least in part to another cause. 



* Phil. Mag., (5), 25, 182 (1888). 



