344: Mr. J. H. J. Poole on a Possible Connexion between 



always earthed when the exciting current switch was closed. 

 An ammeter was used to measure the current through the 

 coil, and a reversing switch was also placed in the circuit, 

 so that the direction of magnetization of the iron specimen 

 could be reversed at pleasure. 



The method of procedure in conducting an experiment 

 was, first to test the sensitivity of the electrometer by 

 applying a suitable known voltage to the leaf, and then to 

 adjust the arc so that it ran fairly steadily at the current 

 value required. The time that the leaf took to leak from 

 zero to some fixed scale-division was measured with a stop- 

 watch, (a) with the iron unmagnetized, (b) with the iron 

 surface a north pole, (c) with the iron surface a south pole. 

 Sometimes the order of reading the three values was varied, 

 so as to avoid one value getting unduly affected by some 

 variation in the arc. This process was repeated immediately 

 and as many readings as possible obtained before the arc 

 burnt out its carbons. The arc throughout the course of the 

 experiment was, of course, kept adjusted so as to take as 

 constant a current as possible, but there is no doubt that the 

 quantity of ultra-violet light is not constant but varies in an 

 erratic fashion. However, by taking a sufficiently large 

 number of readings, it is probable that any effect due to 

 chance variations in the light would not very largely vitiate 

 the mean values obtained. 



I give below two tables. Table Number 1. shows the 

 reading obtained in a typical experiment, and Number II. 

 shows the mean results of a series of such experiments. It 

 may be noticed that these tables completely confirm the 

 above view, as, while the individual readings shown in 

 Table I. differ quite appreciably among themselves, the mean 

 results obtained agree with each other surprisingly well. 

 The results for the different experiments in the second table 

 are really not to be strictly compared with each other, owing 

 to variations in the state of the iron surface from day to day 

 and also to the fact that the behaviour of the arc would 

 probably alter when new carbons were inserted. Taking 

 these facts into consideration, the rate of rise of voltage is 

 probably as constant as could be expected. The important 

 point about the table, however, is that it shows practically 

 no evidence for any change in the ionization current when 

 the field is applied in either direction, the rate of rise of 

 voltage being practically constant for each experiment. 



