It would seem almost better to replace the name of "ab- 

 sorption coefficient," as it is usually employed, by that of 

 "transmission coefficient," reserving the former as a measure 

 of effects which, as has been explained, can probably be 

 obtained only from a study of very thin screens. 



If the interpretation of the foregoing experiments be 

 correct it seems that the ^ particle in traversing a thick 

 screen may stiff er many collisions and deflections. 



J^ow it has been shown by Allen ("Phys. Review," Aug., 

 1906) that the secondary or reflected ;8 radiation consists of 

 electrons moving on the whole with a somewhat slower speed 

 than the original radiation. 



As the experiments described in the present paper in- 

 dicate that in some cases these reflected electrons have suf- 

 fered many collisions before emerging, it would appear that 

 the loss of energy due to a single collision is as a rule not 

 very great, even though the effect of the collision may have 

 produced a considerable change in the direction of motion 

 of the electron. It is not surprising, then, that some of 

 the returned rays have been found to have practically the 

 same speed as some of the original rays ; they would appear 

 to be electrons which have suffered only one collision of suffi- 

 cient violence to cause them to reverse their original direction 

 of motion, or several minor collisions leading to the same 

 result. 



From the curves shown in figs. iii. and iv. it is seen 

 that for small thicknesses of screen, before much actual ab- 

 sorption has occurred, the number of /8 rays turned back 

 may be large, so that many of the original rays would appear 

 to lose their energy gradually, rather than by a very sud- 

 den stoppage and complete absorption. Since the cathode 

 rays behave in many respects like the )3 rays, it seems diffi- 

 cult to understand how the whole of the energy of the X-rays 

 can be derived from the stoppage of the cathode particles, 

 for, as pointed out by Professor Bragg, (5) the stoppage must 

 be very sudden for this to be the case. 



Summary. 



Experiments with the /S rays of radium support the re- 

 sults previously obtained by Crowther, using uranium, upon 

 the scattering of the rays by thin films of materials. 



The distribution of the scattered j8 rays is unsymmet- 

 rical, about a plane at right angles to the direction of the 

 original stream. 



(5) Trans. Roy. See, S.A., vol. xxxi. fl907). 



