306 



Mr. G. C. Simpson on the Electrical 



7 



of j^ and <£, for which the galvanometer deflexion is reduced 

 to zero, be obtained experimentally, and ^ be plotted against 

 tan <£, a straight line such as ABC (fig. 5) will be obtained ; 



Fig. 5. 



COS a 



the values of a. and ^ can at once be deduced for that par- 

 ticular frequency, and for other frequencies in a similar 

 manner. 



In order to carry this theory into practice it was necessary 

 to balance a Wheatstone-bridge containing the bismuth by 

 means of a steady current; then to substitute an alternating 

 current, put a commutator into the galvanometer-arm so as 

 to commutate any current through the galvanometer, and 

 to reduce the consequent deflexion of the galvanometer to 

 zero by means of a change in the resistance of one of the 

 ratio arms. When this was attempted practical difficulties 

 were met with, chiefly due to the fact that the real resistance 

 of the bismuth was continually undergoing a slight change 

 owing to fluctuations in the temperature of the room and in 

 the magnetic field. Although these changes were only small 

 they were quite sufficient to mask the effect looked for if 

 many seconds were required to readjust the bridge. 



This difficulty was overcome by having a switch so arranged 

 that when it was in one position a steady current from a cell 

 was applied to the bridge, and the galvanometer connected 

 straight across AB ; while when it was in another position 



