5 66 



lilRKEI.AND. THK NORWEGIAN AURORA POLARIS EXPEDITION, 1902 1903. 



An endeavour was made to determine the positions of the screen in which the precipitation in 

 these three cases was nearest to the terrella at the equator. 



The first precipitation was nearest at an hour-angle of about 30". In the second precipitation it 

 was not easy to determine the position, as the precipitation became much fainter as the screen was 

 turned; but it appeared to be at from ioo"to 120". The position in which the third precipitation was nearest 

 to the equator was hardly capable ot determination, but it must have been somewhere between 250 

 and 310". 



Finally, it was observed that when the screen had a position answering to an hour-angle of 153, 

 there was very marked precipitation upon its east side, nearest the corner where the screen is bent. 

 This is also visible in No. 11 of lig. 203. It consists of returning rays. It is possible, as we have 

 said, that they turn right round and give rise to the secondary precipitation upon the terrella. There 

 is yet another circumstance which we will mention here. When there was comparatively much gas in 

 the discharge-tube, there appeared, as already mentioned, continuous, luminous polar bands. These were 

 not closed circles, but were somewhat spiral in form, as they la}' at a higher latitude on the day-side 

 than on the night and morning-side. This circumstance we have previously shown in photographs, but 

 it is also applicable here where the magnetic axis is the axis of rotation. 



110. Experiments in which the Terrella is Surrounded by a Horizontal Screen. The terrella 

 was surrounded by a horizontal screen of aluminium after the vertical screen had been removed. The 

 new screen, which is shown in fig. 205, had three holes or slits cut in it, so situated in relation to one 

 another that the angle between the median lines of the first and second slits was no", of the second and 



third no", and of the third and first consequently 140. 

 To the terrella itself were attached two almost 

 radially projecting wires, as fig. 206 shows. They, 

 were placed there in order that conclusions might be 

 drawn, from their shadows upon the terrella and screen, 

 respecting the course of the rays. 



Nos. i , 2 and 3 of fig. 206, are from an experiment 

 in which the pressure was 0.0012 mm., the discharge 

 current 20 milliamperes, the tension 3600 volts, and the 

 magnetising current 8 amperes. All the photographs of 

 experiments with; this screen were taken from positions 

 in which the screen was viewed from above at an angle 

 of 20". The first slit is so placed that its median line 

 forms an angle of 147 with the central line between the 

 centres of the terrella and the cathode. For the sake of 

 brevity, we will say that the hour-angle of the median 

 line was 147". The photographs were taken from posi- 

 tions with eastern hour-angles of 90", 180" and 270. 



It will be seen that the first line of precipitation falls quite to the east of the slit, with the result 

 that no second or third precipitation appears on the terrella. Nos. 4, 5 and 6 are from an experiment 

 where the pressure was 0.0018 mm., the discharge-current 22 milliamperes, the tension 2800 volts, and 

 the magnetising current 8 amperes. The first slit is placed so that the hour-angle of the median line is 

 155, and the photographs were taken from positions of which the hour-angles were 60, i8o u and3io. 

 The first line of precipitation falls more or less over the first slit, so that the rays pass through it 



205. 



