270 Prof. Barkla and Mr. Ayres on the Distribution of 



reject the electrical theory of the phenomenon. It seems 

 more probable that the cause of the ripples lies in the sound 

 of the spark, and that they resemble the ripples between the 

 nodes in a Kundt tube, rather than the nodes themselves. 

 As evidence in support of this conclusion, it might be mentioned 

 that similar results have been obtained for the ripples in a 

 Kundt tube*, i. e. that the distance apart of the ripples 

 increases as the intensity of the sound increases, and that 

 the diameter of the particles has some influence on the 

 distance apart of the ripples. 



I have pleasure in recording my best thanks to Prof. H. 

 Stroud for his advice in connexion with this experiment, and 

 also for the use of the necessary apparatus. 



The University of Sheffield, 

 December 15, 1910. 



XXXI. Tlie Distribution of Secondary X-rays and the Electro- 

 magnetic Pulse Theory. By C. Gr. Barkla, M.A., D.Sc, 

 Professor of Physics, and T. Ayres, B.Sc, King's College, 

 London] . 



EXPERIMENTS on the phenomena of scattering of 

 X-rays furnish what appears to be the strongest 

 evidence in support of the electromagnetic pulse theory. 

 It has been shown by one of us that, as was to be expected 

 on the pulse theory, the primary beam of X-rays proceeding 

 from an X-ray tube in a direction perpendicular to that of 

 propagation of the cathode stream is partially polarized, that 

 the scattered radiation proceeding in a direction perpendicular 

 to that of propagation of the primary radiation is polarized 

 in the manner predicted on the same theory, and that 

 the ratio of the intensities of radiation scattered in directions 

 approximately opposite and perpendicular to the primary 

 beam is within about 5 per cent, of the value calculated on 

 this theory. Each of the experimental results was predicted 

 before it was demonstrated, and it may be added that each 

 has been verified by various observers. Further evidence 

 for the theory is afforded by the facts that rays varying con- 

 siderably in penetrating power are scattered in the same 



* Robinson, Phys. Zeif. 1908, p. 807. 



t Communicated by the Authors. The expenses of this research hare 

 been partially covered by a Government Grant through the Royal 

 Society. A preliminary announcement of the results of this investi- 

 gation was made at the Congress of Radiology held in Brussels last 

 year ; but a report has not yet been published. 



