RICHARDS, COLLINS, AND HEIMROD. — COPPER AND SILVER. 147 



with the silver; * and the concentration effect prevents the deposition of 

 silver in any place where the silver has become nearly exhausted from 

 the solution. 



It is possible to apply a correction to the various values which have 

 been assigned to the electrochemical equivalents of copj)erand silver with 

 the help of the tables given above. 



This computation is made below ; it does not pretend to be precise, 

 but will show iu a general way the effect of the complications under 

 consideration. 



(Grams per 



Ampere-second.) 



Lord Rayleigh and Mrs. Sidgwick . . . - oom ' 9 _ 0.0011170 t 



J s ° 1.00082 ' 



Fr. and W. Kohlrausch ^J"®? = 0.0011174 



1.00Ub2(Y) 



Kahle (fresh solutions) -———^ = 0.0011173 



v } 1.00082 



Patterson and Guthe ' ' - = 0.0011174 



1 .UU— Uo 



The average of these results is about 0.0011173, but perhaps 0.001 1172 

 (the mean between the two extremes) is a safer value to choose. 



From this value of the electrochemical number for silver, that for 

 copper may be calculated simply by multiplication with the ratio of the 

 chemical equivalents ; for we have shown that its true value corresponds 

 to Faraday's law within one part in five thousand. Hence it may be 



taken as 0.0011172 X ,.-^-57; = 0.0003292 grams per ampere-second. + 



2 1 0.0 o 



* Lord Rayleigh and Mrs. Sidgwick, Phil. Trans., 175, 470 (1884). 



t The probable reason for the lowness of this result is the fact that Lord 

 Rayleigh washed and ignited the silver at high temperatures. All the other 

 values, including our own, may he a trifle high because of traces of occluded 

 mother liquor. No attempt is made at present to correct the results for this pos- 

 sible error. The results of Masquart (J. de Phys., [2], 3, 283 [1884]), Pellat and 

 Potier (Ibid, [2], 9, 381 [1890]), Kopsel, Heydweiller, and others are omitted from 

 this table, since they throw no further light chemically. Some are too low and 

 others too high, hence they would not have much effect on the average. Compare 

 Patterson and Guthe's paper. 



% The value 0.0003294 found by F. E. Beach from a solution containing chlorine 

 is probably affected by the presence of cuprous salts. (Am. J. Sc. [3], 46, 81 

 [1893].) All the other directly determined values for this quantity are lower than 

 our value. 



