Spectrum of the Rout yen Rays from a Focus Tube. 577 



at different points along this band would show different pro- 

 perties, since they were produced by cathode particles which 

 differed from one another, presumably in velocity. 



That a more or less complete separation o£ the Rontgen 

 rays into a spectrum was actually effected was made plain by 

 dividing the band into halves lengthwise, and photographing 

 it while one-half was covered by a sheet of copper 0004:4: cm. 

 thick, the other half being uncovered. Fig. 2 shows a photo- 

 graph thus obtained. The width of the band was limited by 

 brass strips with bevelled edges about 0*3 cm. apart. The 

 Rontgen rays produced by the least deviated cathode particles 

 are at A, those produced by the most deviated cathode 

 particles at B. The difference in absorbing power of the 

 copper at the two ends of the band is plain. The variations 

 in intensity along the uncovered half of the spectrum give a 

 rough idea of the distribution of the different kinds of 

 rays. 



The existence of relatively selective absorption in the 

 case of one pair of metals, aluminium and silver, was 

 shown by photographing one half of the spectrum through 

 a sheet of one metal, the other half through a sheet of 

 the other. The aluminium was 0*16 cm. thick, the silver 

 0*002 cm. Fig. 3 shows a set of these photographs. In that 

 figure, CD is a comparison spectrum, photographed without 

 the interposition of any metal. In the photograph EF, the 

 silver was on the left and the aluminium on the right. The 

 thicknesses of the two sheets were so chosen that the rays 

 at E, corresponding to the least deviated cathode particles, 

 were equally absorbed in them. Under these circumstances, 

 the rays at F were transmitted by the silver in much greater 

 quantity than by the aluminium. To show that this effect 

 was not to be explained by a difference in thickness of 

 either sheet at different points along the spectrum, the 

 sheets were turned in their own plane through two right 

 angles, and the photograph G H was then made, showing 

 the effect unchanged. Secondary radiation from the sheets 

 was not present in appreciable quantity at the photo- 

 graphic plate; for if it had been, it would have resulted 

 in a blurring of the boundaries of the shadows on the plate, 

 the distance between the latter and the sheets being about 

 2 cms. 



Another pair of metals, aluminium and tin, has been found 

 to show relatively selective absorption to some extent, but the 

 effect has not yet been obtained sufficiently well marked for 

 reproduction. 



Phil. Mag. S. 6. Vol. 13. No. 77. May 1907. 2 R 



