1923. No. 8. THE ATRORAl. SPFCTRUM ANI1 THF ATMOSPHERK. 9 



cathode ravs in mixtures of Nitrogen and Hydrogen, and Nitrogen and 

 Helium. — As a result of these investigations, which will be more fully 

 treeted later on, we can say that in mixtures of N — H the presence of a 

 few per cent (3 — 7^0 say) of H can be deteched in the spectrum, when 

 the N-spectrum appears on the plate with about the same strength as that 

 of the auroral spectrum. In mixtures of N and He the presence of 30 ^ o 

 He can easily be detected. 



It might be suggested that perhaps the energy possessed by the 

 electric rays were sufficient for exiting the N-spectrum, but too low to 

 excite the H and He-spectra — if so these gases might be present in the 

 higher strata of the atmosphere and still give no light. 



But such an explanation is not possible. — The energy of the rays 

 had to be less than that corresponding to a potential fall of 30 volts, but 

 whatever may be the nature of the cosmic rays they must have a much 

 greater energy if they shall get down to a height of 100 km. above the 

 ground. Even if we only take into account the mass of Nitrogen to be 

 traversed, a cathode particle should have an energy corresponding to a 

 fall through several thousands of volts in order to get down to a height 

 of 100 km., and any kind of positive rays would require a much higher 

 voltage. 



The absence of H and He lines in the auroral spectrum then shows 

 that in that region where the auroral light is emitted, the pressure of H 

 and He must be small compared with that of N in the same region. 



Now we know from obser\ations of the height of aurora and the 

 light distribution along the auroral rays.^ that the principal part of the 

 auroral light is emitted in the height interval of 100 — 120 km. Taking 

 the mean value as representing the region of emission we must have that 

 at the height of 110 km. the Nitrogen pressure must be at least 15 times 

 as large as that of H^-drogen and 3 times as large as that of Helium. 



The possible quantities of H and He which may be present, will depend 

 on the assumption we make with regard to the pressure distribution of 

 Nitrogen. — Now the quantity of Nitrogen present at the various heights 

 will depend on the way in which the temperature varies upwards. 



In previous papers dealing with the absorption of electric rays in the 

 atmosphere-, I have, in accordance with Wegener, assumed that the temperature 

 up to a height of 10 km. on an average can be put equal to — 23" C, and 

 above this height put equal to — 53 C. 



In table I\' is given the pressure at various heights (hi for the gases 

 H, He, N and O, corresponding to the assumptions of Wege.ner. 



' L. Vegard and O. Krogness : The Position in Space of the .Aurora Polaris. Geophys. 



publ.. Vol. I, nr. i, p. 140. 

 2 Phil. .MaîT., Vol. XLII, 1 92 1, p. 47. 



