84 Ml", llicharclson on the Positive Ionization produced 



leaks capacities up to 1 uiicrofiirad had to be connected to 

 the quadrants of the electrometer to which the leak was 

 measured. The lengths of time during which the wire had 

 been maintained at the temperature stated are given as well 

 as the actual time at which the observations were iaken ; the 

 currents are in scale-divisions per half minute. During the 

 time which elapsed between any set of readings at a given 

 temperature and the next at a higher temperature^ the 

 temperature of the wdre was intermediate between the two 

 considered. • 



Temporature B. 



! 



Time wire 1 



at e. ! 



Actual Time. 



Current. 



515° C. 



Minute?. 







8 



16 



24 



Hours, j 



\ i 



1 

 2 



Minutes. 1 



37 ! 



45 



53 



1 



15,000 



12,000 



7,000 



4.000 



577° C. 







8 



16 



24 



2 

 o 



2 



20 



28 1 



36 



44 



31,000 

 36,500 

 31,000 

 15,500 



632^ C. 



697° C. 

 751° C. 







8 



16 



3 

 3 

 3 



9 

 10 



18 



280,000 

 58,000 

 28,000 







8 



1 

 8 



i 3 



3 

 3 



24 



32 



36 

 44 



250,000 

 64,500 



. 360,000 

 68,500 



809° C. 



1 







8 

 16 



3 

 3 



; 4 



48 



56 



4 



250,000 

 53,500 



23,000 



! 5,000 

 1 5,500 



1 75P C. 

 697° C. 







8 



; 4 8 



4 16 



i ^ 



1 4 20 



1 4 ' 56 



J 



950 



725 



! 100 

 1 46 



41 

 41 



632° C. 







5 1 



577° C. 







i ^ 



6 



10 

 14 



5] 5° C. 





 4 



5 



; 5 



These results seem to indicate at first sight that at tem- 

 peratures greater than 600° 0. the rate of decay of the 

 variable leak is almost constant and considerably greater 



