Thin Metal Partition in a Voltameter, 189 



experiment, which covers the hole in the glass. The end 

 electrodes are of platinized platinum, 4 hy 6 centim. and 

 O'l millim. thick. With CuS0 4 solution, Cu electrodes were 

 used. Very thin plates of metal, even ordinary gold-leaf, 

 were easily sealed over the hole in the small glass plate by 

 laying the foil on smooth paper, carefully melting sealing- 

 wax placed around the hole in the form of powder, and then 

 cautiously picking the thin metal up with the melted wax. 

 This being safely accomplished, sealing the small glass plate 

 over the opening in the partition is easy. Each side of the 

 voltameter could then be filled up to the lower edge of the 

 opening at once : then, by carefully pouring the solution 

 alternately in the two sides a little at a time, using a glass rod, 

 the vessel was filled above the gold-leaf, which showed sur- 

 prising strength in being able to stand a difference of level 

 of several millimetres. 



For the voltameter in fig. 1 we have the following rela- 

 tions : — 



(A) The voltameter included in the circuit and the partition 

 open, 



E-e 



B + r 3 +- j +v 



r x r 2 



We are concerned only with the change in the auxiliary 

 resistance, and not its absolute value, in order to make the 

 current the same when the partition is changed, and hence 

 may simplify the case by representing the sum in the deno- 

 minator by W. Therefore 



E-<? 



■^ 



> 



E— i _E — e-x 



~w~~ w-w > 



(1) 1= -xtt^j with partition open, 



(2) I=— ttt — — , with metal partition, J .*. x = w ( w ) = I 



(A) x=Iiv= polarization on partition. 



E 

 (3) 1= ^f , with voltameter cut out of the circuit. 



VV + u\ — v 



.*. from (1) and (3), 



(B) e=I(io i —v)= polarization on end electrodes. 



In every case the auxiliary resistance was so adjusted as to 



