On the North-ioest Magnclic Pole 91 



itous, as we generally observe the aurora horealis to act in this 

 direction. 



It is utterly impossible to attempt to account for the constant 

 increase of the variation, without supposing that the north-west 

 magnetic pole has a constant motion round the north pole ol tiie 

 earth in longitude, on a parallel of latitude ; or in an elliptic 

 curve. Though the variation, when first discovered, was only 

 11° 15' east, there can be little doubt but that those who are 

 destined to exist in the year 20^0, or about that period, will find 

 the variation as much east, as it is now west. The north pole 

 appears to attract more powerfully than the north-west magnetic 

 pole, as must be the case on the supposition of a revolutionary 

 movement indispensable for the formation of any tolerable theory 

 of the variation. The great difficulty in the way of a theory of 

 magnetic revolution, arises from that of accounting for the fact 

 of no variation found in some places. The solution of this diffi- 

 culty may be found in a fair supposition, that the north pole, 

 the moving magnetic pole, and the place of no variation on the 

 smface of the earth, may be at the time nearly in one line. I 

 found, by continued observations during two years, that the va- 

 riation at Bencoolen was 1" 7' to \^ 1 1' east ; the vibrating va- 

 riation giving a returning swing of about four minutes of a de- 

 gree. Capt. Cook found the variation in the Straits of Sunda 

 to be r west. — At Condore in 8" 6' north, and 106° IS' east 

 longitude, the variation was 0° 14' west. Now these, and many 

 other places of nearly no variation, are nearly in the line, verti- 

 cal plane, or section of the two poles; and, consequently, the 

 variation must, necessarily, be little or nothing. The well-known 

 fact that the variation is constantlv changing in one and the 

 same place, furnishes no small proof in favour of the theory of 

 movement of the secondary magnetic cause, or north-west mag- 

 netic pole. — The variation in London was nothing in lGti2, or 

 158 years ago. Supposing the new pole to i)e situated in 100" 

 of west longitude, it would require 5GS years, nine months, and 

 eighteen days, to effect its revolution under the parallel of itsi 

 supposed movement. — In 2\\i years the dip of the needle ap- 

 pears to have diminished in London only 59 miinites of a de- 

 gree. This would seem to indicate that the movement of the 

 magnetic pole is more in a straight line, nearly in an east and 

 west direction, than in a circular or elliptical curve, round the 

 north pole of the earth. Bond makes the variation nothing in 

 London in 1057. The observations regularly taken by our li- 

 brarian, at the rooms of the Koval Society, may be relied on. 

 It would appear fr(>in thcni, that the west variation has ceased, 

 or turned. The variation, therefore, has taken (allowing the 

 change to have been iu 1818) about IGl years tu attain its nt- 

 M 2 nntht 



