672 



BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, 1902 1903. 



circles), when the magnetising was so arranged that the groups formed only a very small angle with 

 one another. It would thus appear that the rays move preferably above and below the plane of tin- 

 equator. 



As the magnetisation is made stronger and stronger, the "node-line" in the form of a luminous 

 circle, will approach the cathode-globe, but suddenly the balance will be disturbed, and the phenomenon 

 will go over into a secondary ring "Saturnian ring" which only developes into full beauty with strong 

 magnetisation and small discharge-current, as represented in fig. 255, 3, where the cathode-globe is 24 

 centimetres in diameter. Applied to the sun, our experiments would imply that we must hen 

 a comparatively low magnetisation, but comparatively high electric radiation. 



One can imagine that among the various kinds of cathode-rays that the sun can emit, then 

 especially a great many that will be brought by solar magnetism to move near the plane of the 

 magnetic equator, possibly bending alternately above and below it. 





Fig. 257. 



Fig. 256 shows phenomena with the large 24 cm. cathode-globe a light that resemble^ the MIM'S 

 corona. ( J ) 



Applied to Saturn (fig. 257), our experiments must lead us to infer that the quantity of rays emitted 

 by the planet was comparatively small, while the magnetisation was comparatively greater than that nl 

 the sun. 



Our experiments with the large cathode-globe (see fig. 255, 3) show that if it is desired to ha\ < 

 ring very thin, it is better to go down to about V 10 milliainpere; but in that case the light will also 

 faint. The ring looks now, however, quite as thick and distinct as with Vj milliampere and with one 

 of the small cathode-globes. 



Let us now simply assume that the current issuing from Saturn is as many times greater t 

 Yio milliampere, as the radius of the planet is greater than that of our globe-cathode. This givi 

 about 50000 amperes from Saturn. Let us assume the tension to be 100 million volts. We then find 



(') As all these figures show, the apparatus has been illuminated beforehand with ordinary light, and the experiments I 

 made and the electric light-phenomena photographed. In this way various reflexions appear in the figures thai have noth 

 to do with the phenomena, but they will not give rise to misunderstanding. 



