ON ELECTEOLTSIS IN ITS PHYSICAL AND CHEMICAL BEARINGS. 327 



Table V. 



I have included in the same table the values obtained -when the correction for 

 the current-density {k' = -000012) is introduced, in order that an opinion may be 

 formed as to the magnitude of the difference produced : it -will be seen that the 

 amount of difference corresponds to an error in -weighing of only 0-3 milligramme 

 per gramme. I have given reasons for preferring the value as calculated -without 

 the introduction of this correction, and from the circumstance already pointed out, 

 Tiz., that after the correction k' = -000025 is introduced, the ' 2 wire ' values are 

 minima, it -would seem likely that tlie deposit obtained with the highest current- 

 density may be too great in consequence of the operation of some action -which is 

 not accoimted for. To decide this point requires measurements to a degree of ac- 

 curacy beyond that reached in the present observations, so that -we may set do-wu 

 the ratio of the chemical equivalent of silver to that of copper, as obtained by this 

 method, as being 



3-39983 , 



vfith a mean error of + '00175 in the observations. 



The Atomic Weight of Copper. 



It -will be at once noticed that the value obtained for the ratio of the equivalent 

 diflfers by only 17 parts in 300000 from the number 3-4000, and there is some indi- 

 cation that the observed value is too small ; -we may therefore accept the observa- 

 tions as showing the ratio to approach very nearly indeed to 3-4000. This gives for 

 the ratio of the atomic vs^eight of silver to that of copper the ratio of the -whole 

 numbers 17 : 10. If -we take the atomic -weight of silver as 107*66, the atomic 

 weight of copper will be 63"333. This result gives some striking relations ; it 



gives, in fact, for the atomic weight of silver the number — ; , and for that of 



copper 



19x10 



3 



and therefore gives both as oue-third of whole numbers. 



The 



atomic weight of silver is one of the most accurately determined of all, and the 

 number here assigned to copper is within the limits of error allowed by L. Meyer -, 

 so that these remarkable relations may perhaps offer some reason for considering 

 the value of the atomic weight of copper deduced fi-om Faraday's law as being 

 worthy of consideration. 



