APPLICATION OF ELECTRON DIFFRACTION 



593 



pearance, and we therefore concluded that at the time when this 

 pattern was itself greatly weakened the surface of the crystal was 

 covered by many layers of gas atoms. Furthermore, we were able to 

 conclude that the diffraction pattern of the second type had its origin 

 in the film of adsorbed gas, and that many layers of gas were necessary 

 for its formation. 



In Figs. 1-4 are exhibited polar plots of typical electron diffraction 

 beams belonging to the diffraction pattern of the first type. The curves 

 of these figures show the effect of gaseous contamination of the crystal 



(422} REFLECTION 



CONTAMINATED SURFACE 



{422} REFLECTION 



CLEAN SURFACE 



Fig. l^Showing the effect of the removal of surface gas upon the intensity of the 

 electron diffraction beam whose Miller indices are (422) 



surface upon the intensities of diffraction beams arising from the space 

 lattice of the crystal. In these figures the electron diffraction beams 

 have been designated by their Miller indices in accordance with the 

 conventional X-ray nomenclature. Since the refractive index of 

 nickel for electron waves differs from unity by a quite appreciable 

 fraction, in the wave-length range of our experiments, a knowledge of 

 the values of this index was necessary before these Miller indices could 

 be assigned. This knowledge has been supplied by our experiments ^ 

 on electron reflection. (Although Bethe " and others deduced refrac- 

 tive indices correctly from the data which we published in the Physical 

 Review (loc. cit.) and assigned the correct Miller indices to the diffrac- 

 tion beams which we observed, these deductions seem to us to have 

 rested upon a rather inadequate experimental basis.) 



5 C. J. Davisson and L. H. Germer, Proc. Nat. Acad. Sci., 14, 619 (1928). 

 * Bethe, Naturwissenschaften, 16, Hi (1928). 



