562 ISIRKKI.AND. TllK NORWKGIAN AURORA POLARIS KXPKDITION, 19021903. 



Photographs 7, 8 and 9 were taken under the same physical conditions as the previous ones, only 

 lhat now the hour-angle of the terrella screen is 175", so that it nearly coincides with the first line of 

 intersection of the rays. 



We see at once that now the majority of the rays pass through the slit in such a manner that the 

 second and third light-figures upon the terrella become very much as they were in Nos. i, 2 and 3, where 

 the screen did not act perceptibly. It will be seen, indeed, that something is wanting in some of the 

 uppermost polar percipitation in the second and third light-figures here, and this probably arises from the 

 slight precipitation of rays seen on the screen in No. 7 ; for it is certain that the rays that keep nearest to 

 the magnetic equator in their journey round the terrella, have not exactly the same lines of intersection 

 as those rays which intersect the equator at large angles. The angular distance between the consecu- 

 tive lines of intersection of the former rays is greater than that of the latter. The rays that form pre- 

 cipitation in the auroral zone, however, are just such rays as, in their discursion above and below the 

 magnetic equator, intersect that plane at great angles. 



In these light-figures there are nearly always 110 between corresponding points in the precipita- 

 tion when 8 amperes are employed as the magnetising current. The rays in the third precipitation, 

 which are farthest up in the polar regions, intersect one another at the equator at an hour-angle that 

 is smaller by from 15 to 20 than that of the rays belonging to the more equatorial parts of the precipitation. 



When a stronger magnetising current is applied to the terrella, several instances of secondary precipita- 

 tion appear on it, as we shall see; but there will always be three principal districts of precipitation in the 

 polar regions, lying about no 100 from one another. The fact that the position of these districts is 

 so independent of the magnetising conditions, is an exceedingly important one, as we may thus venture 

 to transfer the results to the earth, where the magnetic moment is so enormously great. There is, in 

 fact, on the earth, with regard to aurora, something which distinctly points to these fixed districts of 

 precipitation in the polar regions. In the north of Norway, for instance, from about 9 to 10 p. m., and 

 sometimes also between 4 and 5 a. m., there is a distinct culmination in the aurorae. Whether there is 

 any aurora at about 2 in the afternoon it is impossible to say, on account of the light-conditions; but at 

 any rate, during the darkest time of the year I have observed aurora several times at 4 in the afternoon 

 rom the top of Haldde in 1899 1900, aurora which grew fainter and disappeared, only to return again 

 later in the evening with increased strength. I think we are justified in concluding, from analogy with the 

 experiments, that the rays that descend in the auroral zone are just those that come most perpendicularly 

 down to the earth, and therefore those that make their way farthest down into the atmosphere. 



Photographs 10, n and 12 in fig. 203 were taken during experiments in which the terrella was 

 magnetised with 28 amperes. The pressure, indeed, according to measurement, was somewhat lower than 

 before, namely, 0.0005 mm - 1 Dut '* was subsequently proved that the statements of pressure here below 

 0.002 mm. are very unreliable, as there was vapour in the discharge-tube, which we had not troubled 

 to condense, as it was of little consequence, in these experiments, whether it were there or not. A cur- 

 rent of 23 milliamperes was sent through the discharge-tube, and the hour-angle of the screen was 155", 

 The photographs were taken as before. 



Photograph 10 shows distinctly how the first line of intersection of the rays falls just over the 

 slit. The continuation of the line of light is seen upon the screen in a lengthening of the slit. 



Photograph n shows the second principal precipitation and the beginning of the third; but be- 

 tween them are two instances of secondary precipitation, which are especially distinct in the polar 

 regions. 



Photograph 12 shows distinctly the third principal precipitation, and in addition a number of others, 

 fainter, and following one upon another, closer and closer, with increasing hour-angle. We have occa- 

 sionally been able to count nearly 20 of them, fairly distinct. 



