280 Mr. A. S. Russell on some Variations observed in 



now continues constant if the successive measurements are 

 properly performed. B was usually from 20 to 30 per cent, 

 greater than A. At this point, instead or" being charged 

 negatively further, the leaf was charged positively, the 



Fig. 1. 



Time in Minutes . 



absolute value of the potential being the same. The leak C 

 obtained is now less than A by about the same amount as 

 B is greater than A. Repeated charging positively gives 

 increasing leaks d, C 2 , C 8 , passing through the original 

 value A and finally attaining a steady maximum D. The 

 value of the leak at D is, within the limit of experimental 

 error, the same as at B. After charging positively at D and 

 getting a leak which neither increased nor decreased, the 

 leaf was charged negatively. The leak immediately dropped 

 to E and is equal to the value of the leak at C. Further 

 charging negatively gave a continuous increase in the leak 

 till a point F is reached at which, as at B and D, the leak is 

 steady. At F, as before, the sign of the charge was reversed 

 and the leak immediately fell off to a value represented by G. 

 At this point the air in the electroscope was thoroughly 

 ionized for 30 seconds by 6*7 mg. of radium bromide and 

 the leak immediately taken in the usual way. A value Hx 

 was obtained equal to the original value A and intermediate 

 between the maximum and minimum values. This was 

 exactly repeated twice, the first time, H 2 , the leaf being- 

 charged negatively, and the second, H 3 , as in H„ positively. 





