the Absorption of the Gamma Rays. 143 



therefore, can be explained if we assume that the electrons 

 are packed very close in the atoms of the elements *. The 

 displacement of the electrons influenced by the primary ray 

 should be also dependent on the wave-length of the rays. 

 When the wave-length is exceedingly short, the duration 

 of the influence on the electrons should be very small, and 

 therefore we might expect that the displacement of the 

 electrons should be exceedingly small. The result of the 

 present paper shows that even heavier elements scatter 

 the gamma rays of exceedingly short wave-length, like 

 light elements scatter the X rays of comparatively long 

 wave-length. This ma}' be considered as a strong support 

 of the above explanation. 



(2) Table VII. shows that the values of p,\p and <r\p are, 

 for the elements examined, all about the same order of 

 magnitude for the very penetrating gamma rays from 

 radium. Assuming the relation between absorption and 

 wave-length found by Hull and Rice, the true absorption 

 coefficient of gamma rays in aluminium should be small 

 compared with the scattering coefficient ; but this does 

 not appear to be the case. (Since it is very difficult to 

 determine the values of the corrections to be applied to the 

 observed value of a/p, on account of uncertainty as to 

 the distribution and absorption of the scattered gamma 

 rays, it might be thought possible that, after all, the value 

 of cr/p is large compared with p J \p i or, in other words, that 

 scattering is predominant in ordinary methods of measuring 

 absorption coefficients of gamma rays. To show that this is 

 not the case, the following simple experiments have been 

 made. 



Mercury was contained between two concentric glass 

 spheres A and B (fig. 5), of diameters 6'2 cm. and 2*1 cm. 

 respectively. A small sphere K, of the diameter *77 cm., 

 contained about 80 millicuries of radium emanation, and was 

 kept at the centre of the sphere A, which was filled with 

 mercury. Hence the gamma rays from the source had 

 to penetrate about "67 cm. of mercury, which was nearly 

 equivalent in absorption to the lead plate, 1 cm. thick, 

 in the main experiments described in § 2 ; so that only 

 the penetrating gamma rays from radium C were used. In 

 this arrangement everything is symmetrical with regard 

 to the source, and therefore not only the primary rays 

 but also the scattered rays should be equally intense in 



* Barkla, Phil. Mag. xxxi. p. 222 (1916). 



