Mr. Christie on the Aurora Borealis. 273 



about 1 in extent, gradually increased towards the west, and 

 decreased towards the north, until the needle reached 

 IS. 55 30' W., at which time there was a strong stream of light 

 from the magnetic north. When this had disappeared, the 

 needle returned gradually, and very steadily, towards north ; 

 at length, reaching N. 34 W. : at 10 h. 15 m , the direction of 

 the needle was N. 34 40' W. Soon after this, I again ex- 

 amined the needle, and then made the following observations : 



At 10 h. 30 m. the needle vibrated between N. 40 W. and N. 39 W. 



10 33 the direction of the needle was . N. 36 30' W. 



10 35 . . . . . N.35W. 



10 37 . . . . . . N.34 W. 



10 37 . . . . N. 3330'W. 



10 39 the needle vibrated between N. 34 20' W. and N. 33 40' W. 



10 42 . . . . N. 36 W. and N. 35 W. 



10 44 . . . . N.37 W. andN. 36 W. 



At this time there were no streamers, and the light was very 

 faint in the north : barometer 29.94, thermometer 42. I 

 regret that circumstances would not allow of my continuing 

 my observations throughout the night, which I was very 

 desirous of doing. The next morning at 7 h. 20 m. the needle 

 pointed N.40W. 



The mean direction of the needle, when uninfluenced by the 

 aurora, I consider to have been N. 37 W. As the needle 

 assumed this position in consequence of the attractive force of 

 the earth, acting on its south pole towards north, and of the 

 repulsive force of the magnet, acting upon the same pole in 

 the opposite direction, a deviation towards west would indicate 

 a diminution in the terrestrial horizontal intensity, and a devi- 

 ation towards north an increase in that intensity, the intensity 

 of the magnets remaining the same. 



In a paper published in the Cambridge Philosophical Trans- 

 actions for 1820, I first pointed out that the change in the 

 direction of the horizontal needle, arising from extraneous 

 action, would be best determined by referring the action to a 

 needle freely suspended by its centre of gravity, and then refer- 

 ring the direction of this to the horizontal plane; and stated, 

 that in this manner we should be able to account for the 

 changes which have taken place in the variation and dip of the 



