PART II. POLAR MAGNETIC PHENOMENA AND TKRRELLA EXPERIMENTS. CHAP. IV. 575 



of the horizontal screen. Precipitation A is in direct connection with the first precipitation upon the 

 terrella (see fig. 66, p. 151, Section I). 



Photographs 7, 8 and 9, from three experiments, were taken from one position with an hour-angle 

 of 320, and from above at an angle of 15 with the horizon. The experimental conditions were the same 

 in all three experiments, except that the hour -angle of the screen was respectively 70, 50 and 40. 

 The pressure in all three was 0.0105 mm > the magnetising current 6.5 amperes, the discharge-current 

 22 milliamperes and the tension 2600 volts. The photographs were taken for the purpose of studying 

 the shadows of the two vertical wires upon the vertical screen. 



No. 7, with the hour-angle of the screen 70, shows two distinct shadows, comparatively far down 

 in the precipitation. If the angle were made greater than 70, the shadows sank still lower, and suddenly 

 also made a partial appearance in precipitation B on the horizontal screen. It was, as we have said, 

 quite clear that the lowest rays on the right of the vertical screen were rays that would have fallen 

 upon the horizontal screen -- precipitation B -- if they had not been intercepted by the vertical screen. 



No. 8 shows two coincident shadows of the two wires. A plane through these wires, in this posi- 

 tion, passed approximately through the centre of the cathode. The impression given was that the rays 

 which threw the shadow upon the vertical screen in this position, fell normally upon the screen. For 

 the next experiments, therefore, a slit was cut in the screen in very much the same direction as that 

 in which the shadow fell. 



No. 9, which is taken with the hour-angle of the vertical screen 40, shows that the shadows have 

 now gone towards the left margin (looking from the cathode) of the precipitation. If, during the ex- 

 periment, the angle were made less than 40, the shadows drew up towards the edge, and became 

 very long. The rays here evidently soon bend straight up, and they are seen to strike against the 

 roof and floor of the discharge-box (see photograph of this during discharge, fig. 200). 



In order to investigate more closely the rays that went in at right angles to the vertical screen, a 

 slit was cut, as we have said, at the place in question. A new wing was moreover added to the 

 screen at an angle of about no with the original screen, and in the manner shown in the photograph, 

 where it appears with sufficient distinctness. The purpose of this enlargement of the screen was to 

 catch the returning rays that had passed through the slit that had just been cut. The terrella was 

 moreover furnished with a small movable screen, also to be seen in the photographs. This screen could 

 be turned from outside by magnetic means, and also served in the investigation of the course of 

 those rays which passed through the slit. The way that the rays went, however, made it difficult to 

 observe them upon this movable screen; at any rate no photograph was obtained that could be of any 

 use, so this small, movable screen on the whole did little service. 



Nos. i, 2 and 3 in fig. 210 were taken during experiments with a pressure of 0.0095 mm -> a 

 discharge-current of 20 milliamperes, and 6.5 amperes to the terrella. The photographs were taken 

 from places with hour-angles of 130, 180 and 320. 



Precipitation of returning rays that have come through the slit, is distinctly visible in No. i. A faint 

 continuation of the luminosity upon the screen nearest the terrella is observable; a clear wedge of light 

 could be seen running right in towards the surface of the terrella. The position of this precipitation 

 answered to about j p. in., and the precipitation was of such a kind that these returning rays of group B 

 could very well have given an explanation of the positive polar storms. (Compare also the previously- 

 described beautiful experiments shown in Nos. 1315, fig. 204, in which 25 amperes were employed 

 for the terrella.) 



At Kaafjord, however, positive storms, with sharply-defined maximum occurred at 6 p. m , during 

 the six winter months for which we have the material for judging of the conditions there, (See Chap. 

 Ill, Table XCVI, p. 539). 



