1923. No. 10. DISTRIBUTION OF MATTFR IN THE ATMOSPHERE. I3 



§5. 



The views here suggested with regard to the constitution of the higher 

 strata of the atmosphere are in short the following: 



Above 10 km. the temperature decreases, first slowly, then more rapidly, 

 and at a height of say 99 — 100 km. it should have fallen below the freezing 

 point of Nitrogen. The pressure would then be extremely small. But 

 through the photo-electric action of the sun the Nitrogen becomes charged 

 and takes the form of small particles or clusters. These clusters are driven 

 upwards through the eft'ect of the electric field. They may on their wav 

 each an electron and so fall down again. We have an "atmosphere" of 

 electrically charged Nitrogen dust. 



As this dust atmosphere depends on the solar action, we might expect 

 it to be most strongly developed towards the equator. This may, at 

 any rate partly, account for a very marked difference with regard to the 

 auroræ observed near the auroral zone and those appearing in lower lati- 

 tudes. As we pass away from the poles the "ray forms" seem to be 

 more frequent and very much higher. This is also confirmed by the most 

 interesting height measurements carried out by Stormer' at Kristiania 

 during later years. 



He finds much greater upper limits for the auroral rays at Kristiania 

 than at Bossekop and it would seem as if the "height ot the atmosphere" 

 increased towards the equator. 



We should bear in mind, however, that there is another possibility to 

 be taken into consideration. We know that the upper limit of an auroral 

 streamer may vary greatly e\en at the same place. As shown in previous 

 papers- these variations are explained through the effect of the magnetic 

 field. When the rays form a finite angle with the lines of force they will 

 turn about them in screw orbits. When the rays come down nearly 

 parallell to the lines of force there will be an intense light near the bottom 

 edge, but passing upwards the luminosity rapidly diminishes and the visible 

 light may only have en extension upwards of 15 — 20 km. 



If a considerable portion of the cosmic rays enter the atmosphere in 

 directions nearly perpendicular to the lines of force, the number of collisions 

 and the light intensity may be considerable very high up in the atmosphere 

 and we get the long ray streamers. 



Now it was pointed out in my previous papers-^ that the magnetic 

 disturbances would increase the probability for the formation of long ray- 



' C. Stormer, C. R. t. 179 p. 109, 1923. 



2 Forhandlinger ved Geofysikermotet i Göteborg, aug. 191 8, p. 12. The Position in 

 Space etc. Geophys. Pubî. Vol. i, 1920, p. 159. 



3 loc. cit. 



