222 BELL SYSTEM TECHNICAL JOURNAL 



For values of V below, say, 10,000 volts the X-ray wave-length is 

 much greater than the corresponding de Broglie wave-length. 



The lengths of de Broglie waves of electrons which have been 

 accelerated through potential differences comparable with 100 volts 

 are the same as the lengths of moderately hard X-rays. For this 

 reason crystal diffraction of de Broglie waves is observed with electrons 

 of relatively low speeds — speeds corresponding to 100 volts or less — 

 whereas, to observe the same phenomenon with X-rays, the tube 

 producing the radiation must be operated at potential differences 

 comparable with 10,000 volts. 



The first clear evidence of the diffraction of X-rays was obtained 

 when Laue and his collaborators investigated the scattering of X-rays — 

 of X-ray quanta, shall we say — by a single crystal of zincblende. 

 The analysis of this phenomenon led to the prediction and discovery 

 of the Bragg reflection as a special case of crystal diffraction, and 

 later on to the prediction and discovery of the special case of diftraction 

 by aggregates of small crystals of random orientation. All three of 

 these types of diffraction have now been observed with electrons. 

 The Laue type of diffraction, and also the Bragg type, have been 

 observed and investigated by Dr. Germer and myself. Diffraction 

 by the crystal aggregates has been studied by Thomson and Reid, 

 by Ironside and by Rupp. And observations by the Bragg method 

 have been made also by Szczeniewski and by Rose. 



I must now modify to a certain extent the picture of electron 

 diffraction which I suggested to you a while ago. It is not quite true, 

 as I suggested, that the only difference between the diffraction of 

 light waves and the diffraction of electron waves is that in one case 

 the pattern is formed by light quanta and in the other by electrons. 

 In our investigation of the Laue type of diffraction we find, for example, 

 that the streams of electrons which issue from the crystal do not 

 coincide exactly in direction with the streams of quanta which would 

 issue from the same crystal if the experiment were made with X-rays. 

 In the case of X-ray diffraction the streams of quanta proceed from 

 the crystal in the directions of regular reflection from important sets of 

 atom planes, or nearly so. It is recognized that the Laue beams do 

 not, in general, lie precisely in these directions because of a very slight 

 refraction of the rays by the crystal. The situation in regard to 

 electrons seems to be that electrons also are refracted and much 

 more strongly than X-rays. The refractive indices of a metal such 

 as nickel for electrons of low speed depart from unity much more 

 widely than do the indices for X-rays of equal wave-length. It is a 

 consequence of this difference that the departure from the simple law 



