the High" Tension Magneto. 287 



Experimental Methods. 



3. For the determination of the periods of the oscillations 

 a Helmholtz pendulum was employed. The pendulum broke 

 in succession two electrical contacts, the distance between 

 which could be varied by moving one of them relatively to 

 the other by means of a micrometer-screw. The first contact 

 carried the primary current, the second made connexion 

 between the circuit in which the oscillation was to be 

 measured and an instrument for measuring the potential in 

 that circuit. Consequently, when both contacts had been 

 broken that instrument registered the potential in the circuit 

 at a time after the break of the primary current determined 

 by the distance between the two contacts. 



The speed of the pendulum over the distance between the 

 two contacts was usually 250 cm./sec. One division of the 

 scale of the micrometer-screw corresponded to a movement 

 of 0*001 cm. Consequently a time interval of 4xl0~ 6 sec. 

 could be measured and reproduced. 



The contacts consisted of pivoted steel rods resting against 

 the ends of permanent magnets. The attraction of the 

 magnet ensured that the rod always took up a definite 

 position, and that the electrical resistance of the contact was 

 approximately constant. For the successful working of the 

 method it is necessary that at successive operations of the 

 break of the primary current the potentials developed in 

 the circuits should be the same ; this condition will be 

 fulfilled only if there is no sparking at the break. In a 

 subsequent note the circumstances which are necessary and 

 sufficient for the obtaining of a sparkless break will be 

 described ; for the present it need only be paid that so 

 long as the current broken is below a certain limit (deter- 

 mined by the nature. of the contact and of the circuits), it is 

 quite easy to obtain a sparkless break. In most of the 

 measurements to be described the current broken was 

 0*5 amp. Preliminary observations had shown that the 

 maximum potential developed in the primary or secondary 

 circuit was very nearly proportional (as it should be 

 according to the theory) to the current broken in the 

 primary. Accordingly it was assumed that the periods of 

 the oscillations did not depend on the current broken, and 

 that a change in that current changed nothing but the 

 amplitude of the oscillations. 



4. The instrument for registering the potential attained 

 needs some attention. Since the potential to be registered 

 often amounted to several thousand volts, it was clearly 

 impossible to obtain the desired result by simply breaking 



X2 



