THE DIFFRACTION OF ELECTRONS BY A CRYSTAL 97 



this spur at its maximum. The small maxima in the A-azimuths 

 represent the remnants of the "54-volt" spurs. 



We have thus a set of spurs at colatitude 50 degrees in the A- 

 azimuths when the bombarding potential is 54 volts and a set of 

 44 degrees in the B-azimuths when the bombarding potential is 

 65 volts. These spurs are due to beams of full speed scattered electrons 

 which are comparable in sharpness and definition with the beam of 

 incident electrons. This is inferred from the widths of the spurs and 

 the resolving power of the apparatus. 



It is hardly necessary to point out that these sharply defined beams 

 of scattered electrons are similar in their behavior to x-ray diffraction 

 beams. If the incident beam were a beam of monochromatic x-rays 

 of adjustable wave-length instead of a homogeneous beam of electrons 

 of adjustable speed, quite similar effects could be produced. If the 

 wave-length of the x-ray beam were varied, ciitical values would be 

 found at which intense diffraction beams would issue from the crystal 

 in its A-azimuths and others at which such beams would issue in 

 the B-azimuths. The x-ray diffraction beams would indeed be more 

 sharply defined in wave-length than the electron beams defined in 

 voltage. No diffraction beam would be observed until the wave-length 

 of the incident x-rays were very close indeed to its critical value, and 

 the beam would disappear again when the wave-length had passed 

 only very slightly beyond the critical value. This "wave-length 

 sharpness" or "wave-length resolving power" is dependent, however, 

 upon the number and disposition of the atoms involved in the diffrac- 

 tion. If the crystal were only a few atom layers in thickness, or if 

 the x-rays were extinguished on penetrating through only a few atom 

 layers of the crystal, then the x-ray diffraction beams would be much 

 less sharply defined in wave-length ; they would behave more like the 

 electron beams. We may say then that the electron beams exhibit 

 the general behavior of diffraction beams resulting from the scattering 

 of a beam of very soft wave radiation — radiation that is very rapidly 

 extinguished in the crystal. 



Let us try now to forget that what we are measuring in these 

 experiments is a current of discrete electrons arriving one by one at 

 our collector. Let us imagine that what we are dealing with is 

 indeed a monochromatic wave radiation, and that our Faraday box 

 and galvanometer are instruments suitable for measuring the intensity 

 of this radiation. We are to think of the incident electron beam as a 

 beam of monochromatic waves, and of the "54-volt beam" in the 

 A-azimuth and the "65-volt beam" in the B-azimuth as diffraction 

 beams that owe their intensities, in the usual way, to constructive 

 7 



