612 Mr. P. T. Beatty on Secondary 



the ratio of secondary to primary continually increases. Hence 

 we may conclude that those radiations which are able to 

 traverse a *265 Al plate are still able to produce air-radiations, 

 for any failure to do so up to that point would be accompanied 

 by a droop in Curve III. 



Hence we need only consider our second hypothesis, that 

 of a transformation of the harder rays. Before examining 

 this supposition, we shall connect the results in Table III. 

 with this curve. 



Draw an ordinate y x to the curve, and at a distance 8 mea- 

 sured along the A'-axis, draw another ordinate y 2 - Then from 

 the definition of P, the penetrative power, 



, or if the curve is plotted on 



semi-logarithmic paper, log Ps=y 2 -— yi, 



.*. log F s = 8x -f- if 8 is small. 



ax 



Hence log V s is proportional to the slope of the curve. 



On the other hand, if we wish to get the penetrative power for 

 different values of 8 when 8 is not small, then, if we always 

 refer to an initial state where no absorbing plates are put in, 

 i. e. the origin of the curve, we get 



log Vs = 2/2-2/1 where 3/1 = 0, 



•'• log P 5 =y 2 . 



Now direct evidence of the transformation of the very 

 penetrating rays has been given above (Table III. 4th reading), 

 and this would indicate that the Curve III., if further worked 

 out, would begin to droop towards the <^-axis. However, 

 when more than 80 per cent, of the secondary radiation is 

 absorbed by the plates, the effect becomes so small that 

 accuracy of measurement is difficult. Experiments are at 

 present being made with a view of increasing the sensitive- 

 ness of the secondary electroscope above described, so that 

 the further course of the curve may be explored. 



In fig. 6 theoretical curves have been constructed embody- 

 ing, in an exaggerated way, the suggestions that the softer 

 rays are only feebly scattered, and that the harder rays are 

 transformed into a softer radiation. For simplicity, the 

 primary beam is supposed to consist of three types of rays 

 only, each type constituting a homogeneous radiation. 



Now if we cut off the main beam by absorbing plates of 

 different thicknesses, and plot, on semi-logarithmic paper, 

 ionization per unit time against thickness of plates, we shall 



