[gray] 



SOFTENING OF SECONDARY X-RAYS 



133 



that the secondary X-rays, at least those that differ in frequency 

 from the primary, arose by the first mechanism, but found that the 

 observed polarization of the secondary X-rays could not possibly be 

 explained if this was the case. Other objections to (1) are that the 

 secondary j3-rays would not give rise to a sufficient intensity of 

 secondary X-radiation, and further, the X-rays so formed would 

 have a much smaller penetrating power than the bulk of the secondary 

 X-rays (see previous paper in these Transactions). 



The possibility of method (2) being the cause has been examined 

 experimentally. As primary rays an unfiltered beam from a Coolidge 

 tube with tungsten target was used, the tube being operated under a 

 potential of about 30,000 volts. The first experiments were similar 

 to those described above. Paraffin wax was used as radiator, the 

 plate P was of aluminium and the angle between primary and second- 

 ary rays 90°. The results obtained are given in Table II. 



TABLE II 



From this table we find that the average mass absorption co- 

 efficient of the primary rays is about 3 . 00, that of the secondary 

 rays 11 per cent, greater. From the observed ratio ^/.B it can also 

 be found that the greater percentage of secondary rays have under- 

 gone softening or lowering of frequency. This, being the case, 

 experiments were then carried out with thin radiators of paper, 

 aluminium, copper and tin, with the intention of finding the relative 

 intensities of the secondary X-rays from these radiators for a purpose 

 discussed below. The angle between the primary rays and the 

 secondary rays examined was again 90°. Table III shows the results. 



