\ *■ 



228 Efficiency of Recoil of Radium I) from Radium C. 

 Hence the total efficiency E of the recoil is given by 



E f«o-^ =1 la 



Jo c a 2c" 



provided that a is less than c. an assumption which is cer- 

 tainly correct, since the range of the a. particles from 

 radium A is less than that of the a particles from radium C. 



When a = 0, E = l, and when a = c, E = J, below which 

 value the efficiency cannot fall unless the surface is rough 

 or contaminated with grease. 



The determinations of the efficiency of the recoil of 

 radium D were carried out as follows : — A clean platinum 

 plate was exposed for a short time as a cathode in an electric 

 field to a large quantity of radium emanation and then re- 

 moved without passing through mercury or in any way con- 

 taminating the surface. The platinum was mounted with 

 w T ax on a piece of mica with a hole in the middle, and after 

 sufficient time had elapsed for the radium A to decay, the 

 mica was placed on a piece of aluminium foil. In this way 

 the active platinum surface was brought near the aluminium 

 without touching it. A potential difference of 100 volts 

 was maintained between the platinum and the aluminium 

 which was the negative pole. The electric field was main- 

 tained for several hours, during which the radium C decayed. 

 Some of the radium D thus produced remained on the 

 platinum, whereas some reached the aluminium by recoil. 

 By measuring after some months with an a-ray electroscope 

 the amounts of polonium on each plate, the efficiency of recoil 

 could be deduced. 



Two separate experiments were made, in both of which 

 the platinum plates were exposed to the emanation for 

 10 minutes. The recoil was started in one case after another 

 25 minutes, and in the other after 28 minutes. It is easy 

 to calculate the fraction S of the maximum quantity of 

 radium D found before the recoil began under the special 

 conditions of the experiments. If P is the quantity of 

 polonium subsequently detected on the platinum plate, and 

 A the amount on the aluminium plate, the efficiency is 

 given by 



2A 

 E= (P + A)(l-o) (2) 



Two sets of measurements of the activities of the plates 

 were made at an interval of three months and the results 



