Chemical Affinity in terms of Electromotive Force. 279 



slightly lower than the value attained subsequently; which 

 value remained sensibly constant for several hours, and then 

 fell to a greater or lesser extent, as with the stronger solutions. 

 Accordingly, in such cases the lower readings during the first 

 half hour or so were not taken into account in the final ave- 

 rage. For any given combination of plates, the final average 

 thus obtained with weaker solutions was sensibly identical 

 with that obtained with the stronger solutions*. 



(3) The combinations that gave the most constant results 

 on repetition of the experiments were those containing amal- 

 gamated zinc and either electro-copper or amalgamated cop- 

 per; next to which were those with electro-deposited zinc and 

 these same kinds of copper plates. Combinations containing 

 either bright zinc or bright copper (i. e. rods of fused metal 

 or sheets of rolled metal filed, scraped, or sand-papered to 

 perfect brightness) exhibited a considerably wider range of 

 variation. By comparing various bright copper plates with 

 one and the same amalgamated zinc plate, or various bright 

 zinc plates with one and the same electro-copper plate, it was 

 found that differences, amounting nearly to 0*010 volt in the 



* The conclusion that the E.M.F. developed by a given pair of plates 

 immersed, the one in zinc-sulphate, the other in copper-sulphate solution, 

 is sensibly independent of the strength of the solutions when both are of 

 the same density (or, at least, that the variation in E.M.F. caused by 

 variation in strength of the solutions is not outside the limits of experi- 

 mental error), is further corroborated by the results of still more direct 

 experiments on the matter. Three cells were arranged, containing solu- 

 tions respectively of the specific gravities 1-010, 1'090, and 1-175, each 

 containing a recently electro-coppered plate and a freshly amalgamated 

 zinc plate. A number of readings were taken of the potential-differences 

 subsisting between the plates in each case- and then the plates were ex- 

 changed — the pair from the first cell being placed in the second, that origi- 

 nally in the second being transferred to the third, and so on. After a new 

 set of readings had been taken, the plates were again exchanged and a 

 third set of readings taken • so that each pair of plates was read in each 

 cell. The following figures were finally obtained, all readings taken 

 during the first half hours after immersion of the plates being rejected: — 



Specific gravity 1-010. 



Specific gravity 1-090. 



Specific gravity 1-175. 



1st pair of plates 1-1127 

 2nd „ „ 1-1130 

 3rd „ „ 1-1140 



2nd pair of plates 1-1122 

 3rd „ „ 1-1115 

 1st „ „ 1-1135 



3rd pair of plates 1-1125 

 1st „ „ 1-1118 

 2nd „ „ 1-1133 



Mean 1-1132 



Mean 1-1124 



Mean 1-1125 



In each case the value found as the mean for the three pairs of plates 

 differs from the general average 1-1127 by an amount so small as hardly, 

 if at all, to be outside the limits of experimental error. 

 Z2 



