44 



was rather too thin. In fig. 3 are drawn several curves, eacli 

 showing the result of using some particular plug. In order 

 to avoid confusion the representative points are not marked. 

 The eflPect under discussion is clearly soen on comparing, for 

 example, the curve for I'Gl cm. of Pb with that for 3 cm. of 

 Zn. Kleeman was the first to show that the distinction made 

 by the y rays between different atoms in regard to the amount 

 of secondary radiation produced by them could be modified by 

 screening (Phil. Mag., Nov., 1907). Our results agree with 

 his to this extent, but they do not show any true selective 

 absorption, such as he supposes. 



We may add that when using a different specimen of 

 radium, with which we were able to allow the y rays to act 

 with a minimum of previous screening, the emergence radia- 

 tion of Pb was greater than that of C. It would appear pro- 

 bable that with very soft rays other atoms lighter than Pb 

 would surpass C in the same way, and that it is quite con- 

 ceivable that the emergence radiations should increase with 

 the atomic weight throughout the whole range. In a letter to 

 "Nature" (April 2, 1908, p. 509) Cooksey shows that in the 

 case of X-rays the emergence radiations are greater than the 

 incident, thus proving the parallel to our own results in the 

 case of y rays. He also finds that the emergence radiation 

 increases with the atomic weight. This is the opposite to 

 what we have found true of the y rays in most cases ; but it is 

 clear that it can be quite in accord with our theory. 



When the screen through which the y rays pass on their 

 way into the chamber is gradually increased in thickness, the 

 emergence radiation rises rapidly to a maximum, and then 

 slowly decreases. The rise is due to increasing jS radiation 

 from the thickening screen ; the decrease to the absorption of 

 the y rays by the screen. There is a maximum when the two 

 effects balance. This was clearly shown by AVigger floe, cit., 

 p. 429). If we determine the thickness which gives half the 

 maximum value, we obtain an easy and fairly accurate mea- 

 sure of the penetration of the secondary )8 rays. It is easily 

 shown that this particular t*liickness is also that which would 

 absorb half the equally penetrating radiation from a radio- 

 active layer, assuming an exponential law, which it is per- 

 missible to do with sufficient accuracy. 



The results are shown graphically in figs. 4 to 8. In 

 each case the curve A represents the result of an experiment 

 in which the lead plug, 1"61 cm. in thickness, was used to 

 screen the y rays. The curve B shows the res-ult with the plug 

 removed. From each curve can be determined the thickness 

 of screen, which gives half the full value of the emergence 

 radiation ; the points are marked on the diagrams. The re- 



