189 



as the secondary rays will be able to emerge from a greater 

 depth of the radiator, when some of them come out more 

 normally, as happens in the case represented by curve 6". 



In the experiment from which the curve 1) is derived 

 the rays emerge very obliquely, and their self-absorption by 

 the radiator is much greater for a given thickness of radi- 

 ator. 



Otherwise the curves are in very good agreement. It 

 is to be noticed that in curve C certain effects may show 

 themselves which are probably not present to such a large 

 extent in the case represented by B. Soft radiation, for in- 

 stance, has a much better opportunity of producing its effect 

 in the former case. 



Returning now to a brief consideration of the second- 

 ary incidence y radiation. Although no attempt has as yet 

 been made to determine whether the radiation from different 

 substances is strictly homogeneous, it was thought advisable 

 to find out how its amount varied for any substance as the 

 Ra was covered with different screen thicknesses of Pb. 



In these experiments C, Zn, and Al showed much the 

 same effect. With no plug, a 5 mm. plug, and a 10 mm. 

 plug over the Ra, the corresponding readings were as 

 100:60:35. Now, since the value of -\ for the soft y radi- 

 ation from Ra is 1*35 for Pb, it should have required about 

 8 mm. of Pb to reduce it to 35 per cent, of its original value, 

 and had the incidence secondary radiation been due entirely 

 to the original soft y rays it should apparently have been 

 reduced to much the same amount. It is shown above that 

 10 mm. of Pb, placed in the path of the primary, reduces 

 the secondary effect to 35 per cent., so that it appears that 

 very nearly all the secondary incidence y radiation from Zn 

 and such substances is derived from the soft bundle of pri- 

 mary y rays. 



In the case of the radiation from Pb the measurements 

 could not be performed with any great accuracy. However, 

 it was found possible to place a 2 cm. Pb plug over the Ra 

 without cutting off all the incidence radiation : the reading 

 was reduced to about 30 per cent. 



The experiment is accurate enough to enable us to say 

 that the secondary incidence radiation from Pb is derived 

 to a large extent from the original hard bundle of y rays : 

 its quality, as tested by Pb domes, shows also that it is of 

 much the same nature as the emergence radiation, which ap- 

 pears to be the product of the hard y rays. 



In all these experiments, and especially in those which 

 deal with the incidence radiation, it must be borne in mind 

 that the radiation has to penetrate a thickness of Pb, cor- 



