PIERCE. — THE COOPER HEWITT MERCURY INTERRUPTER. 401 



ta^e of the mercury interrupter may lie in the rapidity of recovery of 

 the nonconducting character of such an interrupter after the discharge 

 has passed, so that the condenser which is connected to the transformer 

 may again be charged to a high potential and may thus again quickly 

 accomplish a strong series of oscillations. 



To test this point, a number of photographs (Plates II and III) of the 

 mercury interrupter and the spark in air were taken to ascertain how 

 many times distinct series of oscillations occur during a single cycle of the 

 charging transformer. The pictures were taken upon films or sheets of 

 bromide paper attached to a rapidly rotating disc. The image of the in- 

 terrupter was focussed upon the sensitive paper or film by a lens of short 

 focal length. To avoid overlapping of the pictures on the film when the 

 disc made more than one revolution during the exposure, the lens was 

 mounted in a pendulum, and the exposure was made by swinging the lens 

 behind a diaphragm in front of the revolving film. In this way the image, 

 instead of moving in a circle, was made to trace out a spiral on the film. 

 "With such an apparatus the motion was too slow to resolve the discharge 

 into its separate oscillations. On the other hand, each series of oscilla- 

 tions constituting a complete discharge made an impression upon the plate. 

 The speed of the motor could be varied so as to take from one to four 

 cycles of the transformer per revolution. The period of the transformer, 

 3*0 second, served as a measure of the speed of the motor. 



In order to obtain an understanding of these pictures it must be 

 remembered that the secondary of the transformer was attached perma- 

 nently to the condensers; the primary of the transformer was closed, 

 and the condensers were allowed to charge directly and to discharge 

 through the interrupter. During this action of the interrupter, the 

 pendulum containing the lens was allowed to drop, thus making the 

 exposure. 



Simon and Reich * have already shown by a photographic method 

 that several discharges may occur during a single half-cycle of the trans- 

 former. Evidently, the number of such complete discharges, each with 

 its series of oscillations, will depend on the capacity of the condenser, 

 the inductance of the secondary of the transformer, and the potential at 

 the terminals of the secondary of the transformer. For a given second- 

 ary, as we increase the charging potential by increasing the current in 

 the primary of the transformer, the number of charges and discharges 

 will increase. Their number will also increase with decreasing con- 

 denser capacity. 



* Phys. Zeit, 4, 301 (1903). 



VOL. XXXIX. — 20 



