[gray] 



SOFTENING OF SECONDARY X-RAYS 



131 



the intensity of the secondary X-rays should be the same in either 

 case. 



Let us suppose that the two radiations are identical in quality. 

 We will also assume that the plate P lets through a fraction p of the 

 rays and that the radiator R sends towards the chamber E a fraction 

 k of the rays falling on it. If the primary beam has an initial intensity 

 I and the plate P is in a position A, the intensity of the primary rays 

 reaching R will be pi and that of the secondary rays entering the 

 chamber E will be kpl. In a similar manner it can be shown that the 

 intensity of the secondary rays entering E when the plate P is placed 

 in position B will be equal to pkl. Now let us suppose that the prim- 

 ary beam consists not of one but of n types, of intensities /i, . . . 

 /;., . . . /„, respectively. In both positions of the plate P the 

 intensity of the secondary X-rays entering the chamber E will be 



r = n 



V 



2 p^ k, /,. 

 r = l 



On the other hand, if the secondary rays have undergone a 

 softening or decrease in frequency, the intensity of the rays entering 

 the chamber E will be smaller when the plate P is in position B than 

 when it is in position A. Table I shows the results obtained when 

 the angle between the primary and secondary rays was 110°. By 

 ratio ^/-B is meant the ratio between the intensity of the secondary 

 rays entering E when the plate P was in position A to that when it 

 was placed in position B. 



TABLE I 



The above table shows that the secondary X-rays are softer 

 than the primary rays, i.e., that there has been a transformation to 

 rays of lower frequency. We find, for example, that a plate of 

 aluminium of 1.768 grammes per cm.^ lets through 7.1 per cent, of 

 the primary rays but only 7 . 1/1 .6 or 4.44 per cent, of the secondary. 

 From these results it can be shown that the primary rays have a mass 

 absorption coefficient in aluminium of 1 .49, while the secondary rays 

 have one of 1.76, an increase of 18 per cent. This increase in ab- 

 sorption coefficient was, within experimental errof, independent of 



