126 Dr. 8. W. Richardson on the Magnetic 



The brushes a a of the secohm meter D were joined to the 

 mercury cup 1 of the rocker K x and to the point e respectively. 

 This rocket serves the double purpose of throwing the current 

 in and out of the primaries and of obtaining a zero. This 

 second purpose is considered under a separate heading. The 

 three arms of the rocker on either side are soldered together. 

 In the initial position the points 2 and 3 are connected up, 

 and the current flows through the branch circuit Z. On 

 reversal, the points 1 and 2 are joined, and the current then 

 flows through the primaries of X and Y. 



The Galvanometer. — The resistance of the galvanometer 

 finally used in the experiments was about 7 ohms. It was 

 made as sensitive as possible. A weighted needle was used 

 in the first experiments, but was afterwards discarded as it 

 was found that the gain in steadiness was more than out- 

 balanced by the loss of sensitiveness. The needle was never 

 absolutely at rest, owing to the difference of the time constants 

 in the two secondaries, and the variability of that of X. 

 This effect was not sufficient to prevent the detection of 

 a deflexion of ^ millim. on the scale. The scale was read by 

 means of a telescope placed at some distance from the galva- 

 nometer. The mutual inductance Y was placed with its axis 

 vertical, and at such a distance from the galvanometer as not 

 to disturb the needle. 



To obtain a Zero. — The mercury cups 4 and 6 in the key K x 

 are connected. On reversing the rocker, the galvanometer is 

 first cut out and then reconnected. The lengths of the legs of 

 the rocker are such that the current is reversed before the 

 galvanometer is reconnected. If the rocker be turned over 

 slowly, the spot of light moves off continuously from the 

 zero. 



The Mutual Inductance. — The coil Y was wound with one 

 primary and three secondaries. The values of the mutual 

 inductances for the separate secondaries were 9*04 x 10 6 , 

 7*014 x 10 5 , and 4'536 x 10 5 respectively. It was hence pos- 

 sible to vary considerably the value of the mutual inductance 

 used in the experiments. The values given above were de- 

 termined by comparison with a standard coil whose mutual 

 inductance had been carefully calculated. 



Comparison with the Ballistic Method. — It was thought 

 desirable to compare the results obtained by this method with 

 those obtained from one of the methods in common use. For 

 this purpose a laminated iron ring was wound with primary 

 and secondary coils and was compared with the standard in- 

 ductance ( = 9'040xl0 6 ), first by the Ballistic Method and 

 secondly by the Balance Method. 



