92 BELL SYSTEM TECHNICAL JOURNAL 



happens to coincide with a wave-length of one or more of the diffraction 

 beams. In that case the favored beams appear but no others. 



With this picture of x-ray scattering in mind one sees at once the 

 significance of the main results of the present experiments. A homo- 

 geneous beam of electrons is directed against a crystal of nickel, and 

 at certain critical speeds of bombardment full speed scattered electrons 

 issue from the incidence side of the crystal in sharply defined beams 

 — a few beams at each of the critical speeds — the totality of 

 such beams making up a regularly disposed array similar to the 

 array of Laue beams that would issue from the same side of the same 

 crystal if the incident beam were a beam of x-rays. 



The electron beams are not identical in disposition with the Laue 

 beams, and yet it is possible to treat them as diffraction beams, and 

 from their position and from the geometry and scale of the crystal 

 to calculate "wave-lengths" of the incident beam — just as we might 

 do if we were dealing with x-rays or with any other wave radiation. 

 When this is done we arrive at a definite and simple relation between 

 the speed of the electron beam and its apparent wave-length — the 

 wave-length is inversely proportional to the speed. 



Surprising as it is to find a beam of electrons exhibiting thus the 

 properties of a beam of waves, the phenomenon is less surprising 

 today than it would have been a few years ago. We have been 

 prepared, to a certain extent, by recent developments in the theory 

 of mechanics for surprises of just this sort — for the disco\'ery of circum- 

 stances in which particles exhibit the properties of waves. We have 

 witnessed, during the last three years, the inception and development 

 of the idea that all mechanical phenomena are in some sense wave 

 phenomena — that the rigorous solution of every problem in mechanics 

 must concern itself with the propagation and interference of waves. 

 The wave nature of mechanical phenomena is not ordinarily apparent, 

 we are told, because the length of the waves involved is ordinarily 

 small compared to the dimensions of the system. It is only in such 

 small scale phenomena as the intimate reactions between atoms and 

 electrons that the wave-lengths are comparable with the dimensions 

 of the system. Here only are we to expect notable departures from 

 classical mechanics, and here only are we to find evidence of a more 

 comprehensive wave mechanics.* 



The success of this new theory has been confined, up to the present 

 time, to explanations of certain of the data of spectroscopy. In this 

 field the theory has appealed very strongly to all of us because of the 



* It was predicted by VV. Elsasser in 1925 (Naturwiss., 13, 711 (1925)) that evidence 

 for the wave mechanics would be found in the interaction between a beam of electrons 

 and a crystal. 



