102 BELL SYSTEM TECHNICAL JOURNAL 



The radiation to which our electron beam is equi\alent is extremely 

 soft as already noted. Its intensity suffers a considerable decrement 

 when the beam passes normally through only a single layer of atoms. 

 This characteristic is inferred from the low resolving power of the 

 crystal, and is consistent with what we know of the penetrating 

 power of low speed electrons. When the beam passes through a 

 layer of atoms at other than normal incidence the decrement in its 

 intensity is greater still — and in the limit as the angle of incidence 

 approaches grazing to the atom layer the intensity of the transmitted 

 beam will approach zero. Thus we may expect that when a diffraction 

 beam leaves the crystal at near grazing emergence the contributions 

 to the resultant beam which come from the second and lower layers 

 of atoms will be much less important than when the beam emerges 

 from the crystal at a higher angle. Near grazing the radiation 

 proceeding from the second and lower layers will be heavily absorbed 

 in its passage through the overlying layers. Within a limited angular 

 range near grazing the diffraction beam will be made up almost 

 entirely of radiation scattered by the uppermost layer of atoms. 

 The diffracting system becomes essentially a single plane grating 

 and what we should observe is ordinary plane grating diffraction. 



The first order diffraction beam from a line grating appears at 

 grazing emergence when the wave-length of the incident radiation 

 is equal to the grating constant. The grating constant for diffraction 

 into the A- and B-azimuths is 2.15 A. and grazing beams should 

 appear in both azimuths when the wave-length of the incident electron 

 beam has this value. The bombarding potential corresponding to 

 wave-length 2.15 A. is 32.5 volts, and at just 32.5 volts diffraction 

 beams appear at grazing in both these azimuths. As the bombarding 

 potential is increased the beams move up from the surface to satisfy 

 the relation X = (/ sin 6. Ten or fifteen degrees above the surface 

 radiation from the second and lower layers escapes in sufficient amounts 

 to reduce the intensity of the resultant beam through interference, 

 and at a somewhat higher angle the beam disappears. 



An exactly similar beam is found at grazing in the C-azimuth. 

 The grating constant here is 1.24 A. and the bombarding potential 

 corresponding to wave-length 1.24 A. is 97.5 volts. The beam appears 

 at grazing at just this voltage. These three beams occurring and 

 behaving exactly as required by the theory constitute the strongest 

 evidence we have in favor of the wave interpretation of electron 

 scattering. 



We have been less successful in trying to eiccount for the occurrences 

 of the remaining 21 sets of beams. We do not know why they occur 



