OF A NORTH-WEST PASSAGE. 53 



the needle in this neighbourhood, a considerable error may have been occa- 1821. 

 sioned by the effects of local attraction, produced by the iron in the ship, a \^-k_> 

 phenomenon of which navigators were not then aware. If the magnetic 

 pole were at that time situated near its present position, a difference of no 

 less than four or fire points of the compass may have arisen in consequence 

 of a change in the direction of the ship's head from east to west, as was now 

 the case with us. No accurate deduction therefore can possibly be made, 

 respecting the change which the variation has undergone, from obser\ations 

 made on board a ship at an early period, especially in the neighbourhood of 

 either of the magnetic poles of the earth *. 



The phenomena we had, for some time past, observed in the traversing of 

 the compasses on board the ships, were similar to those noticed on the pre- 

 ceding voyage, though they had not as yet occurred to so great an extent. 

 In proceeding to the westward, up Hudson's Strait, where, by a gradual 

 approach to the magnetic pole, the dip of the needle regularly increases, a 

 proportional increment in the effects of local attraction was also found to 

 take place, displaying itself as well in the amount of what has been termed 

 the deviation, as by the sluggishness with which the compasses tra- 

 versed. About the time of our making Southampton Island, the card of 

 Walker's azimuth compass which, on account of its graduated metal rim, is 

 more heavy than the others, became too sluggish to depend upon. Those of 

 Alexander, Avhich were the lightest and best of our steering compasses on the 

 common construction, began also to require constant tapping or shaking. Cap- 

 tain Rater's excellent azimuth compasses, which unite lightness, sensibility, 

 and accuracy, required, though in an infinitely smaller degree, the same 

 precautions to assist them in traversing. These phenomena, the observa- 

 tions on which are given in detail in the Appendix, had for the present sea- 

 son attained their maximum. Repulse Bay being the nearest approach 

 that can be made to the magnetic pole, by sea, in this direction. Accord- 



* Middleton has, in his published log, set down the variation in Repulse Bay as 50° west, 

 and at Cape Frigid 45°, making a difference of five degrees in a distance of eleven or twelve 

 leagues. Rapid as the changes in the variation are here, this difference appears to me too 

 great to attribute to any thing but a change in the Furnace's course ; and I cannot but con- 

 sider it as extremely creditable to Middleton to have faithfullj' recorded a fact, of whicli. at 

 that time, no probable explanation could be given, and which might, therefore, have sub- 

 jected him to a charge of inaccuracy or misrepresentation. 



