292 ( 'rookcr — Influence of a iSeries Spark 



tions, (N) when there was no spark, (S) when sparks 

 were passing- slowly and (F) when sparks were passing 

 rapidly. 



With the coil in the position (1) and with no spark, the 

 current X, was ohserved to be a small pulsating one 

 having an irregular and ragged edge. These irregulari- 

 ties are quite noticeable and are probably due to poor 

 commutation at the machines as well as fluctuations in 

 their speed of rotation. ' 



With a few sparks passing, say three per second, the 

 current Sj suddenly jumps to a maximum each time a 

 spark passes and then more gradually falls to zero. The 

 current is always in one direction and its maximum value 

 is larger than N a . 



When the spark gap is adjusted so that sparks pass 

 more rapidly, the current F x has the same shape as S, 

 except that the impulses are crowded closer together. 



Connecting the field coil in the position (2), without 

 spark, the current N 2 is quite constant giving a straight 

 line. 



With only a few sparks per second we have a peaked 

 current form S 2 rising rapidly and decaying more slowly 

 than in Sj. 



For a greater frequency of sparks the ionization comes 

 into play in a more pronounced fashion. The ionization 

 current does not have time to reduce to zero between 

 consecutive impulses from the spark and so the resultant 

 effect is a direct current F 2 with peaks, which correspond 

 to the sparks passing on the other side of the tube, ordi- 

 narily called a pulsating current. 



These current forms are essentially the same when the 

 wire is positive and when it is negative. They show 

 directly that there are no oscillations in either the series 

 spark or the corona tube. If there were still surges or 

 oscillations left, they would have to be exceedingly weak 

 and of very high frequency. Oscillations can no doubt be 

 produced easily by means of transformers, the intermit- 

 tent current passing through the primary coil. It may 

 also be possible to use the corona tube with a third elec- 

 trode of wire net like an audi on. 



While the three methods applied to the corona dis- 

 charge show that the current is only intermittent and not 

 oscillatory, the effect of the spark, destroying the nega- 

 tive beads and giving rise to the positive streamers, is not 



