﻿390 Mr. J. A. Crowther on the Coefficient of 



of substances whose absorption was measured no deviation 

 from the simple exponential law could be detected. This 

 seemed to show that the secondary radiation produced by the 

 /3 rays from uranium was much less in amount than that set 

 up by the radium rays. It was considered desirable to have 

 some direct evidence upon this point. 



Owing to the much smaller intensity of the radiation from 

 uranium as compared with that from radium, it was impossible 

 to use any method similar to that of McClelland. The 

 method employed consisted in measuring the ionization 

 between a charged plate and a parallel wire-gauze electrode, 

 due to the uranium radiation ; and finding the increase in 

 the ionization when the gauze was backed by a thick plate of 

 the metal under observation. The first reading was a measure 

 of the intensity of the incident radiation I ; the second was 

 proportional to the sum of the incident and the return radia- 

 tion, that is to I + R ? if R is the amount of radiation sent 

 back from the plate. From these two readings therefore we 

 can calculate the value of R/I . 



The uranium oxide was contained in a circular depression 

 2 mms. deep and 4 cms. in diameter, in the centre of a plate A, 

 8 cms. in diameter, which was kept at a high potential by 

 means of a battery of small storage-cells. The uranium was 

 covered in as before with thin aluminium to cut off all a. 

 radiation. The gauze electrode, 8 cms. in diameter, was placed 

 parallel to the plate A, and connected to the electroscope 

 through a mercury key, in the same manner as in the 

 absorption experiments. 



A small amount of primary radiation may escape sideways 

 without striking the upper plate. This, though reckoned in 

 the incident radiation, does not in reality add its quota to the 

 secondary radiation. However, by placing the electrodes 

 sufficiently close together, we can make this error as small as 

 we please. The measurements were actually made with the 

 plates only 5 mms. apart, and it was found that there was 

 very little alteration in the value obtained for the ratio R/I 

 even if the distance between the plates was increased to twice 

 that amount. 



The results obtained showed that, as expected from the 

 absorption-curves, the return radiation was much smaller than 

 that obtained by McClelland with radium rays, and only those 

 substances whose absorption-curves departed from the ex- 

 ponential law gave very appreciable amounts. The following- 

 brief table gives a comparison of the results obtained by this 

 method for uranium rays, with McClelland's results for 

 radium. 



