﻿om 



Ionization in Gaseous Mixtures by Rontgen Radiation. 317 



exactly with that of the light which is passing off tl 

 pTutino- on the side towards which light is reflected £r 

 the steep sides of the grooves. It is worthy of remark that 

 the tops of the ridges will be illuminated by diffracted light 

 of the right wave-length, though this circumstance alone is 

 not sufficient to account for the narrowness of the bright 

 line. The shift of the dark band towards the region of 

 shorter wave-lengths produced by lightly rubbing the sur- 

 face of the grating should prove an important clue, as well 

 as the unequal rate of advance of the bright line and dark 

 band along the spectrum as the grating is rotated. 



If any further experiments are required to test any 

 theory which may be developed, they can be made without 

 difficulty. 



XXVIII. Ionization in Gaseous Mixtures by Rontgen Ra- 

 diation. By Prof. C. G. Barkla, ALA., iJ.Sc, and 

 L. Simons, B.Sc, University of London, King's College*. 



rglHE intimate connexion between the ionization in a gas 

 JL by X-rays and the secondary radiation from the 

 elements constituting the gas was first shown by one of the 

 writers in a letter to ' Nature ' on. " Ionization by Rontgen 

 Rsys " f. It was shown that the emission of a fluorescent 

 X-radiation from a constituent element of a gas traversed 

 by X-rays was accompanied by a considerable ionization in 

 the gas itself. Later papers J established the generality of 

 the connexion, and showed that a considerable portion of the 

 increased ionization was due to the secondary corpuscular 

 radiation. It was estimated that something of the order of 

 one third of the total ionization was due to the corpuscular 

 radiation ; but it was pointed out that as the assumptions on 

 which the calculation was based were only approximately 

 true, the precise fraction could not be determined. Experi- 

 ments by Bragg J showed the same thing, for though the 

 results published indicated that about 104 per cent, of the 

 ionization produced by X-rays was due to secondary corpus- 

 cular radiation, it is evident that the agreement between the 

 whole ionization and that calculated as due to corpuscular 

 radiation is quite accidental, several of the assumptions 

 giving no more than an approximation to the truth. Even 

 the last assumption that the absorption of air is entirely due 

 to nitrogen and oxygen is wrong by an amount which would 



* Communicated by the Authors. The expenses of this research were 

 partially covered by a Government Grant through the Royal Society, 

 t * Nature/ April 15, 1909. + Phil. Mag. Aug. 1910. * 



§ Phil. Mag. Sept. 1910. 



