TART I. ON MAGNETIC STOKMS. CHAP. III. 30! 



It is with a view to a careful study of the conditions connected with the positive polar storms that 

 I have endeavoured to bring out in my terrella-experiments the directions in which the rays descend 

 tangentially to the terrella's surface at various times of day in the polar regions, by the aid of narrow 

 phosphorescent screens. 



Owing to an accident to my discharge-tube, the final results of these investigations will not appear 

 until the next section of this work; but I nevertheless have so many photographs of experiments that I 

 have made, that I seem already to have a tolerably clear idea of the phenomena. We will first look 

 again at some of the experiments already described, namely those shown in figures 38, 46 and 47 



These experiments show indeed perfectly clearly that there are bundles of rays that graze the 

 terrella from east to west along the auroral zone, corresponding, in my opinion, to the conditions on the 

 earth during positive polar storms, and also bundles of rays that graze the terrella from west to east, 

 corresponding to negative polar storms. 



Fig. 38 b shows a tongue of light on the screen, down towards the "auroral zone" of the terrella, 

 which is not found on the other side of the screen in the position observed. We will call the first 

 side of the screen the a-side, and the other the 6-side. The tongue of light does not appear upon the 

 screen in the position shown in fig. 38 a, but it is found on the a-side of the screen in fig. 38 c, where, 

 however, it does not extend so far in towards the terrella; and on the other hand we also see already 

 on the 6-side, on the opposite part of the screen, a considerable amount of precipitation. In the position 

 shown in fig. 46 a, which forms a direct continuation of the experiment in 38 c, the precipitation does 

 not even extend so far on the a-side, while on the 6-side it has become very marked, and goes right 

 down to the terrella, indicating rays that glance past the terrella from west to east, though without 

 doubt single rays curve in towards the terrella, and form narrow loops before they go out again, very 

 much as shown in the diagram, fig. 50 a. 



In fig. 47 b, we see a powerful precipitation on the 6-side of the screen, produced by the same 

 kind of rays. 



The precipitation on the a-side of the screen in fig. 38 distinctly shows that a wedge-shaped tongue 

 of rays is thrust in towards the terrella, reaching farthest on the afternoon and evening side; the rays 

 turn back as shown in fig. 50 b, and in my opinion correspond to the rays that occasion positive polar 

 storms on the earth. 



These conditions are confirmed and rendered still clearer by the experiments represented in the 8 

 photographs in fig. 136. 



The first five of these refer to an experiment in which the position answers to that of the earth 

 in the winter solstice, and to about noon at the earth's magnetic north pole, and the last three to another 

 experiment in which the position represents an equinox, and midnight at the same magnetic pole. From 

 the north pole of the terrella issue three narrow, phosphorescent screens, 3 millimetres in height and 

 about 3 centimetres long, by the aid of which it was intended to determine the direction of the rays in 

 the various instances of precipitation in the polar regions. 



The five positions of the camera, from which photographs i to 5 of the first experiment were 

 taken, may be determined as follows: 



The axes of the cameras pointed towards the centre of the terrella, and were situated in vertical 

 planes, at angular distances of 45, 90, 180, 270 and 315 from the vertical plane through the axis 

 of the discharge-tube. In each case the axes of the cameras were at an angle of 20 with the horizon. 

 In the three positions from which photographs 6, 7 and 8 were taken, the axes of the cameras were 

 situated in three vertical planes, at angular distances of 45, 90 and 135 from the above-mentioned 

 vertical plane, the axes being pointed towards the centre of the terrella, and forming the same angles with 

 the horizon as before. It will easily be understood from these last three photographs, that the object 



