304 On the Scattering of Rays in X-ray Diffraction. 



reflected beam and shield (3) and (4) from the rays coming 

 from the edges of the primary slit (6). Screen (9) is 

 withdrawn jnst enough to be shielded by (8), and it in 

 turn shields (4) from the rays emerging from (7) and (8). 

 The chamber (4) was filled with ethyl bromide and con- 

 nected with a Wilson tilted electrometer. A rather extreme 

 sensitivity was required, but the readings were easily repro- 

 ducible and the question of proportionality did not have to 

 be considered. 



Readings were made with the scattering substance (3) 

 removed so as to eliminate the effect of any general radia- 

 tion which might be present. This was never more than a 

 few per cent, of the effect to be measured, although the 

 sensitiveness of the apparatus was such that a wire passing- 

 over the slits (7), (8), (9) had to be removed because of the 

 scattering which it produced. The absorbing screen of 

 aluminium was placed successively at (10) and (11). It is 

 essential that (10) be on the opposite side of screens (7), 

 (8), (9) from (3), as otherwise the scattering which it 

 produces when in the reflected beam is comparable with 

 that to be measured. The scatterer (3) was usually placed 

 at the position of second order reflexion from (2), as this 

 gave a larger angle of reflexion and also gave most purely 

 monochromatic rays {i. e. 7 least affected by the overlapping 

 of orders in the general radiation). The distance between 

 the scatterer (3) and the window of the ionization chamber 

 (4) was 10 centimetres. Exposures giving equal ionization 

 in -an auxiliary chamber placed at (5) were regarded as 

 equivalent. 



With this arrangement of apparatus the ionization pro- 

 duced by the scattered rays did not appear to be measurably 

 affected by changing the absorbing screen from position 

 (10) to position (11). The readings were reproducible to 

 within about one per cent. 



An attempt was made to measure the distribution of 

 energy at different angles with respect to the direction 

 of the incident beam with these monochromatic rays, and 

 a few points were obtained on a curve which appeared to be 

 similar to that obtained by Crowther using an aluminium 

 plate as scatterer and heterogeneous rays. A new apparatus 

 suitably designed for this purpose is now nearly completed, 

 and it is hoped that by using monochromatic rays the 

 important problem of " excess scattering " will be much 

 simplified. 



I wish to thank Prof. Sir William Bragg for his helpful 

 interest and kind hospitality. 



