DEMONSTRATION OF ELECTRON WAVES 561 



non-uniformity or asymmetry in the number of electrons scattered 

 into the various segments of a ring, and is due to an asymmetry in the 

 source of the electrons constituting the beam incident upon the gold 

 foil. These, instead of coming directly from a filament or from a 

 high voltage discharge, as is ordinarily the case, are electrons which 

 have already been scattered through 90 degrees by a metal target. 

 The primary beam of electrons moved in a line parallel to the diffrac- 

 ting film and at some distance away; it fell upon a metal target; some 

 of its electrons were scattered through 90° by this target, forming a 

 secondary beam which was the one diffracted by the film of gold. 

 Azimuths about a secondary beam formed in this way are not indistin- 

 guishable : one of them is unique in containing the primary beam which 

 fell upon the target. It is in this azimuth that the electrons are most 

 copiously scattered by the diffracting film, and the rings are most 

 dense. 



The sense of the effect may be stated in a different way. If the 

 density of the rings were independent of azimuth, the mean total de- 

 flection from the primary beam (the deflection at the target plus the 

 deflection at the film) of the electrons forming the pattern would be 

 90°: but actually, the mean deflection is greater than 90°. The elec- 

 trons in the secondary beam have a bias toward a further deflection 

 in the same sense as that which they received at the target. 



Rupp has further shown that the eft'ect of passing the secondary 

 beam (the beam from the target to the film) through a longitudinal 

 magnetic field is, so to speak, to rotate the azimuth of polarization; 

 the azimuth of maximum density is displaced from its original position 

 (without magnetic field) in a sense which depends upon the direction 

 of the field and by an amount proportional to the length and intensity 

 of the field. 



The effect of a transverse magnetic field is difi'erent for different azi- 

 muths; if the direction of the field is normal, to the plane of the primary 

 beam incident upon the target and the secondary beam the eft'ect is 

 nil. If it is parallel to this plane the polarization of the secondary 

 beam decreases as the field strength is increased ; at a critical intensity 

 depolarization is complete, and at intensities which are higher still, 

 the polarization is reversed — the azimuths of maximum and minimum 

 density are interchanged. In these tests with transverse magnetic 

 fields the force on the electron due to its motion through the magnetic 

 field is balanced by the force due to a suitably adjusted electrostatic 

 field. 



The results obtained by Rupp in these experiments may be ex- 

 plained by means of the concepts of electron spin and corresponding 



