66 Mr Palmer, Some Methods of Increasing 



preventing the negative terminal discharging in the form of a 

 brush when it is smaller than the other ball. 



An end of a rod was substituted for the positive terminal and 

 a sphere for the negative as before. As soon as the machine was 

 started the positive brush appeared at the end of the rod and 

 a spark from the negative made the brush increase in size. No 

 effect was observed on taking a spark from the positive side 

 except the momentary annihilation of the positive brush as the 

 potential on that side was lowered. If the terminals were inter- 

 changed so that the positive pole was a sphere and the negative 

 the end of a rod, then a spark from the positive produced a slight 

 increase in the negative brush, and often made one or two other 

 small ones shoot out from other points as well. Beyond the 

 sudden extinguishing of the negative brush no effect was produced 

 by taking a spark from the negative side. 



With two large balls as terminals the effect was the same 

 as with the two smaller, but with the latter the effects of Expts. 1 

 and 2 were easier to obtain with a shorter spark-gap, owing un- 

 doubtedly to the greater want of uniformity in the field. 



There were no indications of any of the above effects if the jars 

 were removed from the machine. 



The Sound produced by the Discharge. 



It is fairly easy to tell when a disruptive discharge is assisted 

 by the brush produced and when it is not, because the sound 

 of discharge is quite different in each case. In the former case 

 the spark may be long but it is not nearly so loud and sharp as it 

 is when no brush is present. In each of the previous cases when 

 the discharge is forced the sound clearly indicates that a brush is 

 present in the gap at the same time. 



When the terminals of the Wimshurst are large balls and the 

 spark-gap great the spark under ordinary conditions is much more 

 abrupt and decisive than when the terminals are small, and this 

 difference is probably due to the fact that there is no simultaneous 

 brush discharge when the poles are large. 



All the preceding effects point to the conclusion that a brush 

 discharge is first formed in each case, and this weakens the 

 resistance to disruptive discharge. Professor J. J. Thomson 

 suggested I should try to obtain the effect in gases in which 

 the brush is more or less difficult to obtain than in air. The 

 terminals were first surrounded with a glass tube 5 cms. in 

 diameter and held in position by rubber stoppers, but the 

 proximity of the tube to the metal of the terminals caused a 

 brush discharge between them, so I used a large spherical bulb. 

 At high potentials a brush always formed on the rubber stoppers 



