s 6 ROYAL SOCIETY OF CANADA 



ioni'^ing power of fi niys of high velocity is not .so great as that of 

 those moving mure slowly, it follows that with increasing magnetic 

 fields the ionising power of the rays introduced would be less than that 

 of the rays cut out and hence a drop in the ionisation values woidd 

 occur. This drop in the conductivity would continue until ultimately 

 all the deflectable f:i rays were swept past the chamber. As the layers 

 of tinfcjil were gradually increased in thickness the more slowly moving 

 ft rays would be absorbed and the first effective sheaves transmitted 

 would consist of rays possessing higher and higher penetrability and 

 consequently of rays with less and less ionising power. It follows 

 then that while a maximum conductivity would be obtained with each 

 thickness of tin foil the value of the maximum would decrease with the 

 thickness of the absorbing layer. It is evident, too, since with in- 

 creasing thicknesses the first effective sheaves of transmitted rays 

 would possess higher and higher velocities, that the field required to 

 deflect the axis of these difïerent sheaves into coincidence with the 

 axis of the chamber would increase. Hence the maximum conducti- 

 vities, when absorbing layers of increasing thicknesses were used would 

 be obtained by fields excited by currents of greater intensit}', and this, 

 as the curves A B, C, and D shew, is actually what happened. 



The numbers corresponding to the saturation currents obtained 

 with the different absorbing layers when the rays were deflected up- 

 wards and away from the chamber by the magnetic fields are given 

 in columns II to V of Table II, and curves representing them are 

 shewn in Fig. 4. From these it will be seen that with each 

 absorbing layer the ionisation fell away as the raj's were deflected 

 upwards and soon reached a value which was constant, and which 

 represented the natural conductivity of the air in the chamber 

 together with that impressed upon it by the undeviable ra^'s from 

 the radium and by the secondary rays which they excited. 



These limiting curves it will be seen exhibit an effect already 

 pointed out and emphasized by MacKenzie' and others, that when 

 the thickness of a plate or wall upon which y rays are allowed to 

 fall is gradually increased the gain in ionisation at the back of the 

 plate from the secondary radiation is at first greater than the loss 

 produced by the absorption of the primary rays. This result is well 

 exemplified by the curves A', B', C, and D', which correspond to 

 absorbing layers of increasing thickness and which shew that the 

 limiting value of the ordinate of B' is greater than A', that of C is 

 equal to that of B', and that of D' is again less than that of C. 



I MacKenzie, Phil. Mag.. July, Ut07. 



