— 239 — 



for X W and 24"+ 180° 



for y 14°— 90° and 24° + 90° 



differ nearly 180°. 



All tliis points to a circular patli, in which the two foei 

 have a diametrical position. 



If we adopt the path as being' the same for both 



l \ 3 4 



foei, having lts eentrum at Xo = ^—z '—= — 2.8, and 



— 10.7 -10.2 ^^^ 1 .1 . • . ^• 



yo = — = — 10. .5 and the foei oeeupying dia- 



metrical positions, the values: 



(X— Xo)n'S (y— yo)(ii+6)S — (X— Xo)(n4-18)i', — (y— yo)(n+18)h 



must represent the same quantity, and we may derive a 

 new set of 12 values by taking the mean of each group of 

 tour. The harmonie formula eomputed for this set is: 

 X+ =- — 2..3 + 14.5sin(t + 22°) + 1.3sin2 (t + 28°). 



The term of the seeond order has stlll more diminished, 

 besides the amplitudes, viz. the angles 2 X 19°, — 2 X 2°, 

 — 2 X 15° and —2 X 50°, showed no regularity at all, 

 accordingly we are justifled in negleeting the term of the 

 seeond order and in putting forth the conelusion: 



The positive and negatioe foei of arctir veetors of daily 

 ineqiialitji of disturhing foree move daily in a rlochvise diredion, 

 with eonstant velociti/ and diametriecdly opposite to each other in 

 a circidar path of 14*^.5 radius, having its centrum in 79" N 

 and 780 j^r 



Combining the two results, viz.: 



1°. The aretie veetors point to a couple of foei, which 

 rotate daily clockwise around the pole of disturbanee; 



2°. The eharaeter of the vector-diagrams is ehiefly 

 dependent on the distanee from this same pole; it is quite 

 natural to draw the conelusion: 



The daily inequality of disturhing foree is caused by a constant 

 field of foree rotat ing from Eto W around the axis of disturbanee. 



This ('onstancy must be understood in such a manner, that 

 the rotating field suffers only minor alterations, when the 



