48 



viour of soft y rays alone. The form of the curve for both 

 Pb and Sn seems to suggest that the operation has been a 

 success. The half-values are now '024 and '029 respectively. 

 Thus the )8 rays due to these soft y rays have only one-quarter 

 the penetration of those derived from the harder y rays. Pos- 

 sibly we have here an association with the facts stated by 

 H. W. Schmidt (Ann. der Phys., Bd. 21, p. 654), viz., that the 

 )8 rays emitt-ed by RaC consist of two groups, of which one 

 has four times the penetration of the other. The thickness 

 of Al screen necessary to reduce the harder set to half-value 

 was "53, which agrees with the '50 given in the table above ; 

 for the softer rays the value was '131, nearly a quarter of the 

 other. If there are two main groups of )8 rays we should cer- 

 tainly expect to find two corresponding groups of y rays. 

 Kleeman has also argued in the same direction, though for 

 different reasons. 



As will be seen from figs. 6 to 8 we have made the same 

 attempt to sort the soft from the hard rays in respect to Cu, 

 Al, and celluloid as in the case of Pb and Sn. The propor- 

 tion of soft y rays converted into i3 rays is now much less 

 relatively. The results for soft rays are therefore of less 

 accuracy, and depend appreciably on the allowance made for 

 the hard rays. Nevertheless, they are of considerable 

 interest. They show clearly the difference between the 

 effects of hard and soft rays, and the varying proportion of 

 soft to hard from substance to substance. 



It may be gathered from consideration of all these curves 

 that secondary y rays play a very small part in ionizing the 

 air within the chamber. If there were any appreciable 

 amount of it we should expect a gradual rise long after the 

 effect of the secondary )8 rays had reached its maximum 

 value. 



§ II. 



We may now give a short summary of the results of this 

 and our previous paper : — 



(1) When y radiation is diminished in quantity in con- 



sequence of its passage through matter, )8 radiation 

 appears in its place, moving at the outset in the ori- 

 ginal direction of the y radiation, and subsequently 

 undergoing scattering in the ordinary manner of J8 

 rays. 



(2) The speed and penetration of the j8 radiation thus 

 produced increase with the penetration of the y 

 radiation to which it is due. 



(3) The speed of the jS radiation does not depend on the 



nature or condition of the atom in which it arises. 



(4) In the case of radium the speed of the jS radiation 



