382 



HON. R. J. STRUTT ON THE LEAST POTENTIAL DIFFERENCE 



Leyden jar was also connected across the terminals of the machine. The capacity 

 thus added diminished the effect of any slight irregularities in its action. 



It is not easy to observe the faint sparks which pass between the sparking plates 

 unless the observer gives his undivided attention to watching for them, and if this l>e 

 done it is impossible to read the potential difference at the exact moment when the 

 spark takes place. A telephone was therefore inserted in series with the spark gap, 

 and arranged in a clip so as to be against the observer's ear when his eye was in the 

 proper position for reading the voltmeter. Whenever a spark passed, the telephone 

 gave an easily audible click. By means of these arrangements the measurements 

 could be taken with considerable rapidity and precision. 



A diagram (fig. 3) of the electrical connections is given below. 



Fig. 3. 



It must be remembered that the first spark taken through a gas passes with far 

 greater difficulty than those which succeed it. I have found a gas able to sustain 

 for a short time a potential difference three times as great as that required to produce 

 discharge through it when this initial resistance had been broken down. If the 

 measurements of the spark potential are to be compared with the cathode fall 

 measurements made while a continuous current was flowing through the gas, it 

 is clear that they should be made with the gas in its electrically weakest condition, 

 that is, immediately after it has been vigorously sparked through. This condition 

 was complied with in my experiments. 



In taking a measurement of the spark potential, the pressure of the gas was 

 adjusted to the desired value and read off". The machine was started, and the resist- 

 ance column lengthened until the potential rose sufficiently to cause the first spark to 



