APPENDIX. 157 



ship's head to the west is consistent with the preceding il- 

 lustration, except that it appears to he rather too strong. 

 The inactivity of the compasses with the head to the south 

 has heen explained above. The tendency of the north end 

 of the needle to follow the ship's head in consequence of its 

 great attraction, sufficiently explains the cause of the errors 

 noted in the latter part of this remark. 



(9) Here the compasses remained stationary till, by the 

 head of the vessel opening more to the eastward, they 

 yielded suddenly to the power of its attraction. 



(A) The remark here, of the error increasing to the east 

 or west, as the vessel's head was towards either of these 

 quarters, is quite consistent with the preceding illustrations, 

 which show that the north end of the needle had constantly 

 a tendency to follow the head, although the magnetism of 

 the earth had, of course, its effect in keeping the north point 

 of the card between the ship's head and the north. The 

 inactivity in the other semicircle has been already explained. 



(J.) This remark requires no particular explanation, being 

 similar to all the preceding. 



(k) It has been rendered probable by our preceding expla- 

 nations, that the dip of the needle at this time was nearly 

 88°, and consequently the magnetism of the earth reduced 

 from what it was at Sea-horse Point, in the ratio of cos. 

 S6^ : cos. 87^, or cos. SS° ; that is, in the ratio of 2 to 3. 

 At this time, therefore, the magnetic power of the ship pro- 

 bably exceeded that of the earth, and hence the changes re- 

 marked in the deviation of the needle. Moreover the natural 

 westerly errors, arising from the actual variations, had now 

 nearly diminished to nothing, which would cause an appa- 

 rent increase in the easterly errors. 



(/) There can be no doubt of these phenomena being due 

 to the electro-magnetic effect of the Aurora Borealis. In 



