536 Mr. N. R. Campbell on Radiation 



distance between the sides will cause no increase in the 

 resulting ionization. The part of the curve due to this cause 

 will increase rapidly to a constant value. Combining the 

 two, we should expect a rapid rise of the curve near the 

 origin followed by a straight line cutting the axis of ionization 

 on the positive side. This is precisely the form of the curves 

 in the figure. 



Reasoning from the assumption of the existence of these 

 two radiations, we can obtain some information as to their 

 intensities. 



Let each sq. cm. of the w r alls give out (a) an amount of 

 absorbable radiation which, when totally absorbed by the air, 

 causes ionization s, and (b) an amount of penetrating radia- 

 tion which causes ionization v^ per c.c. ; (c) let the external 

 penetrating radiation discovered by Cooke * cause ionization 

 v 2 per c.c. 



Then, if x is the distance apart of the movable sides, c and 

 d the lengths of their edges, the surface of the box exposed 

 to the air inside is 



2cd + 2(c + d)z 



and the volume of the air contained is cdx. 



When x is so great that all the radiation (a) is absorbed, 

 the ionization in the vessel will be 



{2cd + 2(c + d)x\s due to (a) 



{2cd + 2(c + d)x}v 1 cdx due to (b) 

 cdxv 2 due to (c). 



If y is the total ionization the equation of the curve remote 

 from the origin is 



y = 2cds + { 2 (c + d)s + 2c 2 d\ + cdv 2 \x + 2cd(c 4- d) v x x 2 . 



Now we see that the part of our curves remote from the 

 origin is indistinguishable from a straight line ; hence we 

 may put 1^ = and obtain 



y = 2cds 4- {2{c + d)s + cdv. 2 }x. 



Produce the straight portion of the curve to cut the axis 

 of?/ ; let the intercept on this axis be p. Then 



p = zeds or s = . 



V 

 The tangent of the angle which the straight line makes 

 with the axis of ,v is tan = 2(c + d)s + cdv 2 . From the 



* H. L. Cooke, Phil. Mag. Oct. 1903. 



