298 BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, igO2 1903. 



The pressure employed in the discharge-tube was about 0.02 mm., except in the case shown in 

 photograph 7, where it was 0.013 mm. The current-strength was 8 milliamperes with a voltage of 3300 ; 

 and lastly a magnetising current of 33 amperes was employed upon the terrella. 



In taking photographs i, 2, 3 and 4, the axes of the cameras were directed towards the centre 

 of the terrella, and were lying in a plane that passed through the axis of the discharge-tube. This 

 plane formed an angle of 66 l /-2 with the vertical line, and thus formed the same angle with the hori- 

 zontal plane as the axis of the discharge-tube. 



When the angular distances to the axes of the cameras were measured from the axis of the tube 

 in the direction of the cathode, and in the above-mentioned inclined plane with the centre of the terrella 

 as the vertex, measuring contrary to the hands of a clock seen from above, the angles in the four posi- 

 tions were respectively 90, 180, 270 and 315. Photograph 6 was taken with the axis of the camera 

 horizontal in the vertical plane through the axis of the tube, and directed towards the centre of the 

 terrella, and towards its night-side. 



Photographs 5, 7 and 8 were taken in positions that may be described as follows : in three vertical 

 planes through the centre of the terrella at angular distances of 45, 270 and 315 respectively from the 

 vertical plane through the axis of the tube, the axes of the cameras pointing towards the centre of tin; 

 terrella, and forming an angle of 20 with a horizontal plane. 



There are two different phenomena that come out very clearly in these photographs, or rather in 

 the experiments which the photographs reproduce. 



In the first of these, we have the luminous spirals, almost closed rings, that are formed round, and 

 at a certain distance from, the magnetic poles of the terrella. These spirals vary in position with the 

 rotation of the terrella; and I consider them as answering to the auroral zones on the earth. These 

 principal spirals of light form in my opinion the most remarkable phenomenon that I have discovered in 

 my terrella-experiments. The more highly the terrella is magnetised, the narrower does the band of 

 light become, keeping, however, its intensity. The bands of light are here almost coherent; but different 

 degrees of luminosity in the precipitation are easily seen, answering to the various districts of precipita- 

 tion shown by the experiments given in fig. 47 a & b. 



It will be seen from photographs Nos. i and 5, fig. 134, that the spirals begin above as a broad 

 luminous band, indicating a great descent of rays upon the terrella. At the top, to the left of the band 

 in No. i, there is a slight illumination in space outside the terrella, as also in No. 7. These two illu- 

 minations are the beginning and end of the greatest precipitation of rays in the band of light. The 

 principal bands of light can be easily followed in photographs 2 and 6, then in 3 and 7, and 4 and 8, 

 right round the terrella, until they disappear. In No. 8 especially, we see both beginning and end of 

 this long spiral of light round the south pole of the terrella-magnet, which answers to the terrestrial- 

 magnetic north pole. 



These continuous bands of light recall a most remarkable and ingenious hypothesis made by A. E. 

 Nordenskiold ( i ). He assumes that the usual arc of polar aurora seen in Bering Strait was part of a 

 ring of light situated in a plane perpendicular to the radius of the earth, which terminates in a point 

 near the magnetic pole (lat. 81 N., long. 80 W. Gr.). He concludes that the plane which contains the 

 auroral arc, and which is perpendicular to this radius, cuts it at a distance of 125 kilometres below the 

 surface of the earth. In this plane the lower edge of the ring of auroral light would be about 200 

 kilometres above the surface of the earth. 



The second phenomenon, which is clearly visible in the experiments shown in fig. 134, is the 

 presence of portions of luminous rings, also almost circular, which lie considerably nearer to both poles 



(') A. E. Nordenskiold: Vega-Expeditionens Vetenskaplige lakttagelser. Forsta Bandet, p. 417, Stockholm, 1882. 



