OH Till: SILVER VOLTAMETER. 



56, 



abnormal liquid could have diffused through the walls of the pot. An analysis of 

 the strength of the anode and kathode solutions before and after electrolysis was 

 made in other experiments and led to the conclusion that the pots were efficient. 

 The syphon was tested by placing the electrolyte used in observation 30c in the anode 

 lx>wl and a normal solution in the syphon limb and kathode bowl ; the result, 

 I'll 829 (49d), shows that no appreciable quantity of the anode liquid could have 

 diffused into the kathode vessel. 



Deposit on Platinum and on Silver. 



Table VII. gives the results when silver bowls were employed as kathodes. Our 

 successful employment of these bowls is in a large measure due to the use of an 

 electric oven for drying purposes. 



TABLE VII. Comparison of Results with Platinum and Silver Kathodes. The Mean 

 Value of the Results with Platinum Bowls is I'll 827 Milligrammes of Silver per 

 Coulomb. 



Results with Silver Bowls as Kathodes. 



It is apparent that the deposit on a clean platinum surface is the same as that on a 

 silver surface. This result is in agreement with VAN DIJK'S* oljservations. 



GORE, KAHLE,! and RICHARDS and HEIMROD^ found a somewhat larger deposit 

 when the kathode of the Rayleigh form was of platinum with silver deposited on it 

 than when the kathode at the commencement of the observation was platinum only. 

 RICHARDS found that the deposit on a silver kathode was 1 part in 10,000 heavier 



* VAX DIJK, 'Ann. der Phys.,' 19, p. 282. 1906. 



t KAHLK, 'Zeitechr. Inst,' 18, pp. 229-267, 1898. 



} RICHARDS and HEIMROD, 'Proc. Am. Ac.,' 37, p. 418, 1902. 



