Secondary y Radiation. 439* 



radiation as possible is excluded the curves for all substances 

 very nearly correspond. Let us suppose that had it been 

 possible to exclude all the effects of secondary 7 radiation, 

 the curves for all substances would have become identical- 

 This supposition must apparently be very near the truth, for 

 it can be seen from the geometry of the arrangement in 

 fig. 7 that by no means can all the secondary radiation have 

 been excluded in the first set of experiments. We arrive, 

 then, at this result : that, excluding all secondary 7 radiation 

 and its effects, we may consider the Ra to give out two 

 distinct sets of 7 rays, one more penetrating than the other, 

 each practically homogeneous, with values of X/A = '028 and 

 *12 respectively, these values being practically independent 

 of the nature of the absorbing substance. 



When the secondary 7 radiation is allowed to produce its 

 effect in addition to that produced directly by the original 

 7 rays, this law may be modified to a very considerable 

 extent. In the usual arrangement of apparatus used for 

 determining the quality of a radiation the absorbing plates 

 are as a rule placed close to the ionization-chamber, and the 

 values of X deduced from the tangent to the absorption curve 

 — that is to say, it is generally with the dotted-line curves 

 of fig. 8 that we are dealing, and the effects of secondary 

 radiation may, as has already been shown, be in some cases 

 very nearly as important as the effects produced directly 

 from the original 7 radiation. 



This, for example, affords a ready explanation of the effect 

 observed by Eve (Phil. Mag., Aug. 1908), who states :— 

 " It is noteworthy that the lia in 2*2 cm. of nickel-steel 

 gives an effect about 1*5 times as great as when the Ra is 

 in 1 cm. of Pb. From the relative densities we should 

 expect 2*2 cm. of steel to be equivalent to 1'5 cm. of Pb,. 

 and therefore the Ra in the steel cylinder should give, by 

 the density law, two-thirds of the effect of the Ra in Pb. 

 It actually gives one and a half times as much ; thus the 

 primary 7 rays traverse steel much more readily than Pb, 

 but the rays passing through iron arc subsequently absorbed 

 more readily by Pb than if the Ra were in Pb."' 



From the results of the present paper we can say that 

 in going through 2*2 cm. of* steel or 1*5 cm. of Pb, the soft 

 primary y rays will have suffered in each case the same 

 number of collisions, but that the effect of a collision with 

 an atom of Pb, or rather with the constituent of such an 

 atom, is much more definite than in the case of substances 

 of lower atomic weight : that in the case of Pb the collision 



