FAILURE OF OHM's LAW AT HIGH CURRENT DENSITIES. 



143 



as shown with a high resistance R^ in series and another small resist- 

 ance in shunt to cut down the sensitivity. In the latter part of this 

 work this galvanometer was replaced by another of less sensitiveness. 

 The A.C. detector was a telephone tapped between the same points 

 as the D.C. detector, but with a large condenser in series to prevent 

 D.C. getting into the telephone circuit. The telephone was tapped 

 across a transformer placed in this circuit. In this circuit is also one 

 of the coils of a mutual inductance, M, the other coil of which is in 

 series with the A.C. source, and is not shown. This makes possible 



Figure 3. Details of the bridge connections. 



the elimination of the out-of -phase component by suitable adjustment. 

 The A.C. was prevented from entering the galvanometer circuit by 

 the high resistance in series with it, and by an open key when the 

 galvanometer was not in use. The condenser in the telephone circuit 

 proved an unnecessary precaution, the resistance of the transformer 

 and mutual inductance being sufficient to prevent enough diversion of 

 the D.C. into the telephone line to introduce appreciable error. The 

 condenser was used in most of the work, but in some of the later read- 

 ings it was omitted. The telephone was one of 1100 ohms resistance, 

 made by the Western Electric Co., type 509 W. The transformer was 

 one of the small ones of the General Radio Co. made for this purpose, 

 type 166. 



In Figure 3, the resistances i?5 and R& which are connected to the 

 same points as the sources of the current constitute an auxiliary bridge. 



