191 



value of the effect produced with Zn screens from 56 to 43. 

 In the case of Pb the corresponding change is from 38 to 

 30. In other words, the ratio of the effect produced by a 

 Zn screen to that produced by a Pb screen of the same mass 

 is apparently much the same whether secondary radiation is 

 included or not. It appears almost as if the effect produced 

 in the ionization chamber were due more to the secondaries 

 produced from the hard y rays than to direct effects of these 

 hard y rays. The hard y rays are not apparently broken up 

 as the result of the first collision ; a few may be, but most 

 of the rays seem capable of passing into what we may for 

 convenience speak of as an intermediate stage. A second 

 collision when the ray is in this stage may be more effective 

 in breaking up the y ray and causing the )8 particle to be 

 sent out. In some cases the y ray may still retain the char- 

 acteristics of a y ray even after many collisions. 



It would seem possible for a y ray to produce some 

 ionization in a gas as a result of its passage through that 

 gas, slow-speed delta rays being produced from the atoms of 

 the gas. Where high-speed electrons appear they are to be 

 considered as having been produced from the y ray itself, 

 being originally a part of it, as already suggested by Pro- 

 fessor Bragg. 



It has been found by Eve (Phil. Mag., 1906) that the 

 y rays from Thorium have an absorption curve, measured by 

 Pb, very nearly the same as that for Ra. We should ex- 

 pect, then, that Thorium gave out two homogeneous sets of 

 y rays as does Ra, the values of A. by Pb for these groups 

 being '32 and 1'35. It has also been found by Eve that the 

 value of \ for the y rays given out by Ur is 1*4, and that 

 these rays appear nearly homogeneous. The y rays of Ur 

 appear, then, to correspond very closely with soft bundle of 

 rays emitted by both Th and Ra. 



Again, it seems rather striking that the secondary rays 

 produced by the hard y rays of Ra produce in Pb, and ap- 

 parently in all substances, a secondary, for which the value 

 of \ is of the same order as for the Ur rays and the soft 

 rays, of Ra and Th. 



In the case of the secondary rays produced from the hard 

 y rays of Ra it has been possible to treat them as practically 

 homogeneous over the range of thickness of screens which 

 have been used. It will, however, require much more care- 

 ful research to determine the exact quality of this radiation. 



Again, the value of /\/A for the soft primary bundle 

 of rays has been shown to be nearly four times that of the 

 hard. It seems possible that this may have some connection 

 with the result obtained previously by Professor Bragg and 



