686 



BELL SYSTEM TECHNICAL JOURNAL 



the later variable into electron-energy by means of the known relation, 

 one gets the distribution -in-energy of the scattered electrons. (Inci- 

 dentally, in Harnwell's apparatus and in that of MacMillen the 

 deflecting field was electric; in Dymond's magnetic.) 



Typical data of Harnwell's are shown in Fig. 9; these are distribution- 

 in-energy curves for electrons scattered by helium, their initial energy 

 having been 75 or 150 equivalent volts (curves on left and right, re- 

 spectively). First, one sees that the great majority of those which go 



200 X 0° SCALE 



e = 16* 



..f^ 



75 X 0° SCALE 

 9 = 8° 



e = o 



ooio-fwa 



120 130 140 150 



ENERGY (volts) 



160 



V, = 150 VOLTS 



40 50 60 70 80 



ENERGY (volts) 

 V, = 75 VOLTS 



Fig. 9 — Distribution-in-energy of electrons scattered from helium atoms, as de- 

 termined by Harnwell. {Physical Review.) 



through nearly undeflected retain their energy. Then, the electrons 

 which are deflected through angles in the neighborhood of 8° are 

 very much fewer (notice the change in the scale of the axis of ordinates) 

 but among them, those which suffer a certain energy-loss are relatively 

 more numerous. The electrons deflected at 16° are fewer yet, but 

 among them those which suffer the same energy-loss are relatively 

 plentiful. With 150- volt corpuscles one sees the same behavior, with 

 differences in detail which I leave to be read from the curves. As for 

 this peculiar value of energy-loss, its "mean value from a large number 

 of observations is 22 volts"; one recalls that the resonance-potential of 

 helium is 19.6, and that there are other critical potentials of inelastic 

 impact ranging from this value upwards. 



Curves obtained by Dymond and Watson, also for helium, appear 

 in Fig. 10. The angle of deflection is about 10°, and the experiments 



