Dr. J. A. Orowther on "J" Radiation. 723 



screens. The third column gives the ratio of the current in 

 the chamber 0$ to that in the chamber Cp, and thus measures 

 the relative intensity of the primary and secondary rays after 

 passing through the thicknesses of aluminium given in the 

 first column. The ratios are in arbitrary units. The fourth 

 -find fifth columns give the corresponding logarithms. 



It will be seen that the ratio of the secondary to the 

 primary radiation (column III.) steadily diminishes as the 

 thickness of the absorbing screens is increased. The secon- 

 dary radiation is therefore distinctly more absorbable than 

 the primary. 



Calculation of Results. 



The observed difference in quality between the primary 

 and secondary radiations might be due to two causes. In 

 the first place it is conceivable, though not probable, that 

 the primary radiation might suffer a small change in wave- 

 length by the process of scattering, and that the true 

 scattered radiation might thus be more absorbable than the 

 primary from which it was produced. The coefficient of 

 absorption of the primary rays can easily be determined 

 from the figures in column IV. of the table. If these values 

 are plotted against the corresponding thicknesses of the 

 absorbing screens the resulting graph (fig. 2) is approxi- 

 mitely a straight line, showing that the absorption of the 

 primary rays is practically exponential, the coefficient of 

 absorption (X) in aluminium being 2*81 per cm. The 

 radiation P entering the chamber C p is thus represented by 

 the formula 



where d is the thickness of the absorbing screen. If, in the 

 process of scattering, the radiation suffers a slight increase 

 in wave-length and a consequent increase in coefficient of 

 absorption to some higher value \ l5 the radiations entering 

 C s should be given by $ = 8 Q e~ Xld . We should have, 

 therefore, 



s/p=s /p e- (Xi - X) ^ 



On plotting the logarithms of S/P given in column V. 

 against the corresponding thickness the graph should be a 

 straight line. The experimental curve is shown in fig. 2 and 

 is distinctly not straight. It would appear then that this 

 suggestion is not tenable. 



A second explanation is that the radiation from the 



