632 Mr. H. G. J. Moseley and Dr. K. Fajans on 



so that serious error due to faulty insulation would have been 

 introduced had the insulated system not been kept at zero 

 potential. 



The current was measured by connecting the brass plate 

 to the electrometer for exactly one minute, and noticing both 

 the quantity of electricity used in compensation during that 

 time, and the deflexion of the electrometer-needle due to 

 imperfect compensation. 



Measurements were taken in the following order at intervals 

 of two minutes of: — 



I. The current through the first box due to a particles 

 from emanation and active deposit on the disk. 



II. The current through the first box with a field of 

 + 800 volts between the copper gauze and the disk. 



III. and IV. Similar measurements in the second box. 



In a typical experiment the following currents were 

 observed : 1. 298. II. 402. III. 183. IV. 23b\ The 

 differences between II. and I. and IV. and III. were evidently 

 due to a particles from actinium A which had been deposited 

 on the disk. 



The sensitiveness of the first ionization box was known to 

 be 13 per cent, greater than that of the second. 



The ratio of the activities of the actinium A on the disk 

 when opposite the first and second boxes was therefore 

 104:53x1-13. 



If, now, X is the decay constant of actinium A, and t the 

 time taken to pass from the first box to the second 



e-™ = -575. 



The angular separation of the two boxes was 16 0, 4, and 

 the time of one revolution was *037 sec. 



Hence t = • 001 ft 9 sec, giving X = 329 (sec.) -1 , and the 

 half-value period \og e 2/X = "00211 sec. 



These measurements were always repeated about five times 

 in the same order. Separate curves were then drawn showing 

 the variation with time of the currents I., II., III., IV., 

 and values of the decav ratio deduced from simultaneous 

 values of the four currents obtained by interpolation. The 

 growth of active deposit and any steady change in the depo- 

 sition of actinium A on the disk were in this way allowed 

 for. In some sets of experiments irregular changes took 

 place, due no doubt to small changes in position of the ebonite 

 box, whereby both the amount of actinium emanation 

 escaping, the amount remaining as a source of actinium A, 

 and the amount of activity removed from the disk by the 

 contact of the velvet edging were altered. Such experiments 



