THE EARTH A MAGNET. 21 



greatest easterly excursion at half-past eight in the 

 evening, returns to the west, and attains its greatest 

 westerly excursion at eleven, as at starting. 



Of course, these excursions take place on either side 

 of the mean position of the needle, and as the excursions 

 are small, never exceeding the fifth part of a degree, 

 while the mean position of the needle lies some 20 to 

 the west of north, it is clear that the excursions are 

 only nominally eastern and western, the needle pointing 

 throughout, far to the west. 



Now, if we remember that the north end of the 

 needle is that farthest from the sun, it will be easy to 

 trace in M. Arago's results a sort of effort on the part 

 of the needle to turn towards the sun not merely 

 when that luminary is above the horizon, but during 

 his nocturnal path also. 



We are prepared, therefore, to expect that a varia- 

 tion having an annual period, shall appear, on a close 

 observation of our suspended needle. Such a variation 

 has been long since recognised. It is found that in the 

 summer of both hemispheres, the daily variation is 

 exaggerated, while in winter it is diminished. 



But besides the divergence of a magnetised needle 

 from the north pole, there is a divergence from the 

 horizontal position which must now claim our atten- 

 tion. If a non-magnetic needle be carefully suspended 

 so as to rest horizontally, and be then magnetised, it 

 will be found no longer to preserve that position. The 

 northern end dips very sensibly. This happens in our 

 hemisphere. In the southern, it is the southern end 



