28 



AKSEL S. STEEN. 



M.-N. Kl. 



MN 3 6 9 NOON 3 6 9 MN 



Table 3, which, like Table 2, is drawn up after Göttingen mean time, 

 contains the hourly values of jZ in whole and tenths of / smoothed on 

 the same principle as JX and J Y. The sign -f- denotes, that the ver- 

 tical component of the deflecting force is directed downwards towards the 

 centre of the earth. 



Fig. 6 shows jZ's diurnal variation at the 4 stations, arranged accor- 

 ding to local mean time. 



It will be seen from the 

 figure that the vertical com- 

 ponent of the deflecting force 

 is distinctly influenced by the 

 sun's altitude above the hori- 

 zon of the place, as at all 

 four stations it has its nega- 

 tive maximum at noon, .and 

 this increases with decreasing 

 latitude. There also seems to 

 be a faint indication that the 

 position of the sun with regard 

 to the magnetic north pole is 

 of some significance, as the 

 vertical stroke showing the 

 magnetic-polar noon, marks a 

 slight rise in the positive di- 

 rection, most pronounced at 

 the southernmost station, but 

 almost imperceptible at the 

 most northerly. 

 I have endeavoured to use these complete observations from Paw- 

 lowsk, Vienna, Tiflis and Zi-Ka-Wei, for an approximate calculation of the 

 height of the main part of the current-system above the surface of the 

 earth. 



As already mentioned (p. 25), jX and J Y for Zi-Ka-Wei both have 

 the value o at i r8 a.m. local time, that is to say that the plane in which 

 the curved path of the current may be supposed to lie, must at that mo- 

 ment be tangent to the globe at Zi-Ka-Wei. The course of the current 

 runs through those points in the plane in which the latter is intersected 

 by the vertical of those places on the earth's surface where jZ^o. 



If we now take out from the curves in Fig. 6 the hour for jZ=o, 

 we obtain, 



MW 3 



9 NOON 5 6 9 MN 



Fig. 6. 



