104 Fisher on the Variation of the Compuss. 



determining the magnetic dip with two dipping needles, made by 

 Troughton and Jones. 



If the mean variation near the latitude of 80° N. alters at all 

 in the cours6 of time, it does not alter in any degree as it does 

 in more southern latitudies ; for the variation by Fotherby in 

 Magdalena Bay in 1614, in latitude 79°.34', was found to be 

 25°. W., and by Poole 18°.16' W. in Cross Road in latitude 

 79°.15N., in the year 1610. 



In proceeding northwards above the latitude of 79°§, the va- 

 riation was found by observation on shore and on the ice sen- 

 sibly to decrease ; and it further appears to be a maximum near 

 the meridian of Greenwich, in about 66° N. latitudes, as far as 

 can be deduced from ship azimuths. 



Since, therefore, the westerly variation has been determined 

 by observation sensibly to diminisli above the latitude 80°, and 

 that the magnetic dip also increases, which is very nearly equal 

 to the latitude ; it is plain that one of the magnetic poles must 

 coincide nearly with the pole of the earth. 



The state of atmospheric electricity in these regions, not only 

 from the analogy of the aurora borealis and australis to elec- 

 trical light, but also from its probable influence and connexion 

 with the phenomena of the magnetic needle, was on this occasion 

 an inquiry exceedingly interesting. The opportunities, however, 

 which occurred, were from unavoidable circumstances, far from 

 being as numerous as could be wished. The apparatus used for 



