302 



is conceivable tliat the effect of varying the atomic weight 

 of the radiator might depend on the velocity of the ;8 particle, 

 and that the smaller atomic weights might return a relatively 

 small amount of fast primary rays. Assuming the velocity 

 of the secondary rays to be of the order of that of the pri- 

 mary, it would then follow that the constitution of a hete- 

 rogeneous primary beam would be altered by scattering, the 

 alteration depending on the atomic weight of the scattering 

 substance or radiator, and being of the nature indicated by 

 experiment. Crowther (Phil. Mag., Oct., 1906) has described 

 an experiment which appears to show a considerable effect 

 of this kind. He found that the secondary ionisation due 

 to the /3 rays of UrX was much more nearly proportional 

 to atomic weight than in the case of radium, as studied by 

 McClelland, and he has suggested that the difference ''may 

 be due to the presence in the radium radiation of compara- 

 tively slowly-moving /3 rays." On the other hand, H. W. 

 Schmidt has recently found (Ann. der Phys. Bd., 23, 1907) 

 that the behaviour of substances under radiation from the 

 /3 rays of UrX is not very different to that found by McClel- 

 land in the case of radium. As it was important to settle 

 this point, if we were to find the true interpretation of our 

 own experiments, we put together the apparatus of the form 

 used by Crowther, and repeated his experiments with UrX,. 

 using also radium and actinium. The apparatus is not very 

 well suited, as Schmidt points out, for obtaining accurately 

 the proportion between secondary and primary ; but it gives 

 a ready answer to the question as to whether or no the speed 

 of the )8 particle has any effect. 



Subtracting the effects of the carbon plate from alV the 

 others, and setting lead equal to 200, we obtained the follow- 

 ing results 



about 10 cm. of air, and enough to cut off all a rays. 



UrX — No screen. 



Ra 1— Thin mica + tinfoil, the latter '00366 cm. thick.. 



Ra 2 — Thin mica -1- tinfoil, the latter '0293 cm. thick. 



Ra 3— Thin mica -f tinfoil, the latter '0586 cm. thick. 



Ra 4 — Thin mica + lead, the latter equivalent to '11 cm, 

 of tinfoil. 



