574 
MESSRS. R. T. GLAZEBROOK AND S. SKINNER ON THE 
in I. and II., keeping their sum constant until there was no deflection of the galvano¬ 
meter, this difference of potential was expressed in terms of the E.M.F. of the 
Leclanclffs, while by connecting A to L and C to M the ratio between the E.M.F. of 
the Clark and that of the Leclanches was found. This method was adopted in some of 
the experiments. In other experiments, C and D were connected together, while A 
was joined to L and B to M. In this arrangement, the electromotive force between 
A and B was the difference between that of the Clark and Pa. The difference was 
expressed in terms of the Leclanches, and then compared as before with that of the 
Clark. By employing a proper current in the main circuit through the voltameters, 
the difference between the E.M.F. of the Clark and Bi could be made small, and thus 
the result was more nearly independent of changes in the Leclanches. 
The sensitiveness of the galvanometer was such that an alteration of the resistance 
in I. by 1 ohm produced a deflection of about 3 cms. on the scale, this corresponds to 
a variation in the E.M.F. between L and M of about one six-thousandth of a Clark 
cell. Thus variations of ’0001 volt could be detected without difficulty. 
§ 9. Details of the Method. 
The theory of the experiment just outlined supposes that the current i in the main 
circuit remains constant. In practice this is not the case; two methods were 
adopted for meeting the difficulty. 
