474 BELL SYSTEM TECHNICAL JOURNAL 



treme right — differ by about 10 volts. We believe that the position of 

 this maximum is given correctly by Curve A, and that in Curve B it is 

 shifted to the right owing to an eccentricity of the tube used in the 

 earlier experiments. It will be noted that the position of the maximum 

 in Curve A is marked in Curve 5 by a shoulder which protrudes from 

 the side of the peak. It is with respect to the positions of the maxima 

 only that the curves of Fig. 4 may be legitimately compared, the ordin- 

 ates in the two cases being proportional to different quantities. Those 

 of Curve A are proportional to the current of full speed electrons 

 entering the collector, while those of Curve B are roughly proportional 

 as has been explained (loc. cit.), to the ratio of full speed electrons, 

 entering the collector to the corresponding current of electrons of all 

 lower speeds. 



There is some doubt in our minds as to whether the maximum in 

 Curve A which occurs at 20 volts truly indicates a maximum in the 

 reflecting power of the crystals for electrons of corresponding speed. 

 The current to the collector is determined primarily by the product of 

 the primary current by the square of the coefficient of reflection, so 

 that a maximum in the collector current must correspond to a maximum 

 in the reflecting power if the current in the primary beam is almost or 

 quite independent of voltage, but not otherwise. This condition is 

 known to be reasonably well satisfied in the range of bombarding 

 potentials above 30 or 40 volts. Below this range, however, the current 

 from the gun is limited partly by space charge, and its variation with 

 voltage is rapid. A maximum in the current to the collector in this 

 region must therefore be regarded with a certain suspicion; it may be 

 due to a maximum in the reflecting power of the crystal with which, 

 however, it will fail to coincide in voltage, or it may signify only that 

 the reflecting power has a trend opposite to that of the primary current. 

 We are not, however, greatly concerned in this investigation with the 

 interpretation of this maximum, nor even of the other maxima of 

 Curve A. 



Measurements have been made of the intensity of the twice reflected 

 beam as a function of the angle 6 for bombarding potentials correspond- 

 ing to the five maxima of Curve A. In some cases intensities have 

 been measured at intervals of 5 or 10 degrees around the entire circle; 

 but for the most part measurements have been made only at the cardi- 

 nal positions ^ = 0, 90, 180 and 270 degrees. The total number of 

 measurements of this kind is about 500. The complete data for bom- 

 barding potential 77 volts, corresponding to the third maximum of 

 Curve A , and for 6 = 270 degrees are given in Table I. 



