762 Prof. F. Horton and Miss A. C. Davies on 



curve that an increasing negative current which begins at 

 20'4 yolts was measured, and that this current began to 

 increase at a more rapid rate after 21*2 volts. This proves 

 that the radiation which results from 20*4 volts electron 

 impacts on helium atoms is a true resonance radiation and 

 that it is absorbed and re-emitted and so passed on from 

 atom to atom, and that the same is also true of the radiation 

 which results from 21*2 volts electron collisions. 



It should be mentioned that a much higher filament tem- 

 perature was needed to measure radiation currents of a 

 given magnitude when the side filament was the source of 

 electrons, than was needed to obtain currents of the same 

 maonitade when the electron stream from the lower filament 

 was used, as in the case of the experiment recorded in the 

 curve of fig. 2. This was to be expected since the radiation 

 cannot directly reach the cylinder E in the former case, but 

 has to be passed on from atom to atom over a much longer 

 distance. Moreover, for a given electron emission from the 

 side filament, the radiation current obtained in a series of 

 observations such as that illustrated m fig. 6 varies greatly 

 with the pressure of helium present in the apparatus. With 

 very low pressures, this current increases at first with 

 increase of pressure, but a stage is soon reached when 

 further increase of pressure results in a reduction in the 

 measured current, showing that less radiation is reaching 

 the platinum cylinder. Such a result presents no difficulty 

 of interpretation if the 21*2 volts radiation ionizes abnormal 

 helium atoms produced by the 20*4 volts electrons. 



In a recent theoretical paper, Kernble*, in discussing the 

 theory of Franck and Eeiche that the normal condition of 

 the helium atom is identical with the unit quantum sharp 

 series state of the crossed orbit system (shown by Lande to 

 correspond to the single-line series), puts forward the sug- 

 gestion that an atom which, as the result of a collision with 

 an electron having 20'4 volts energy, has been changed to 

 the coplanar condition, can only revert to the normal con- 

 dition when by some external means the limitations of the 

 selection principle are modi tied by the temporary introduction 

 of other stationary states, the reversion being accompanied 

 by the emission of radiation in at least two quanta. It is 

 unlikely that radiation emitted in this way could be absorbed 

 and re-emitted by normal helium atoms, and such radiation 

 would therefore be incapable of being passed on from atom 

 to atom. The results of the experiments described in this 

 section are thus not in agreement with Kemble's suggestion. 



* Lot. cit. 



