268 Dr. C. R. A. Wright on the Determination of 



then a cold-saturated solution of copper sulphate (sp. gr. 

 somewhat below 1/2) was carefully floated on the top of the 

 zinc-sulphate solution. In some experiments this disposition 

 of the plates was reversed, the copper plates being lowest and 

 the zinc plates highest ; in these cases saturated copper-sul- 

 phate solution was first poured in, and then a lighter zinc- 

 sulphate solution. To avoid the almost inevitable contamina- 

 tion of the zinc-sulphate solution with faint traces of copper 

 sulphate which occurs when the former is poured on to the 

 latter, no matter how carefully done, it was found more satis- 

 factory to fill the cellhalf full of the zinc-sulphate solution, and 

 then slowly to introduce the copper sulphate at the base of the 

 cell through a U-tube passing through the cork C, the end 

 inside the cell being drawn out to a point and bent downwards, 

 so that the copper-sulphate solution flowed in gradually and 

 lifted up the zinc-sulphate solution without passing into it as 

 a jet and so more or less impregnating it with copper. 



It is evident that, since A and B, E and F are respectively 

 in the same horizontal planes, the resistance of the column of 

 fluid between the plates, R, must be sensibly the same, whether 

 the smaller pair of plates, B and F, or the larger pair, A and E. 

 be employed to generate a current; either of which can at plea- 

 sure be done by simply connecting the mercury-cups G H I K, 

 attached to the appropriate plates, with the extremities of a 

 known external resistance. By measuring the difference of 

 potential E subsisting between the ends of this resistance the 

 current passing, C, is known ; hence if R be known, the 

 E.M.F. of the cell, e, is known, being given by the equation 

 e=E + CE. 



The value of R can be deduced, with a fair amount of pre- 

 cision, from the results of two series of observations with 

 varying currents, made, first, with the smaller, and, secondly, 

 with the larger pair of plates, in the following way: — By divi- 

 ding the actual current-strengths by 2'5 and 5*0 (the super- 

 ficies in square centimetres of the plates respectively), two 

 series of values of E for corresponding " current-densities " 

 (rates of flow per square centimetre of plate surface) are ob- 

 tained, by interpolation from which, for a given current-den- 

 sity D, two values, E l and E 2 , are deduced for the smaller and 

 the larger pair of plates respectively. The E.M.F. with the 

 smaller pair, e^ is manifestly 



^ 1 = E 1 + 2-5DR, 



whilst that with the larger pair is 



* 3 = E 2 +5-0DR. 



