468 DR. CHARLES G. BARKLA ON POLARISED RONTGEN RADIATION. 



for separate investigation and will not be treated in this paper. It seemed possible 

 that phenomena of exactly the same nature take place in gases, for the fact of 

 ionization implies the emission of negative corpuscles or electrons such as proceed 

 from metals, and, as has been shown, the non-deviable radiation from gases does 

 differ to a small extent from the primary radiation producing it. It had to be proved 

 that the difference in the behaviour of gases and heavy solids is not one of degree 

 merely. 



As explained by Professor J. J. THOMSON (' Conduction of Electricity through 

 Gases,' p. 268), on the hypothesis that Rontgen rays consist of a succession of electro- 

 magnetic pulses in the ether, each ion in the medium has its motion accelerated by 

 the intense electric fields in these pulses, and, consequently, is the origin of a 

 secondary pulse, the intensity of electric force in which is given by the expression 



, where e is the charge on the ion, f its acceleration, r the distance from the 



point considered in the pulse to the origin of the pulse, and 6 the angle which the 

 line joining the point to the origin of the pulse makes with the direction of accelera- 

 tion of the ion. The direction of electric intensity at a point in a secondary pulse is 

 perpendicular to the line joining the point and origin of the pulse and is in the 

 plane passing through the direction of acceleration of the ion. Thus the secondary 

 radiation is most intense in the direction perpendicular to that of acceleration of the 

 ion and vanishes in the direction of that acceleration. 



A secondary beam whose direction of propagation is perpendicular to that of the 

 primary radiation will then on this theory be plane polarised, the direction of electric 

 intensity being parallel to the pulse front in the primary beam. 



If a plane-polarised primary beam be used, then the secondary radiation from the 

 electrons has a maximum intensity in a direction perpendicular to that of electric 

 displacement in the primary beam and zero intensity in the direction of electric 

 displacement. 



Shortly after I arrived at the conclusion as to the origin of secondary radiation 

 from gases, Professor WILBERFORCE suggested to me the idea of producing a plane- 

 polarised beam by means of a secondary radiator and of testing the polarisation by a 

 tertiary radiator. The secondary radiation from gases under ordinary conditions is, 

 however, much too feeble to attempt the measurement of a tertiary. I hoped by 

 means of some metal which was the source of a much more easily absorbed secondary 

 radiation that this experiment might be performed successfully, but further experi- 

 ments, which are described later, show that there is another difference between 

 the radiations from heavy metals and those from gases and light solids in addition 

 to those already mentioned, and that evidence of polarisation cannot be expected from 

 experiments with such metals. 



A consideration of the method of production of X-rays, however, leads one to 

 expect partial polarisation in a beam of radiation proceeding from the antikathode 



