98 Fisher on the Variation of the Compass. 



The great difference perceptible in these observations is very 

 remarkable, arising not only from the different position of the 

 ship's head, but also from the different parts of the vessel in 

 which the azimuths were taken ; it was, therefore, always ne- 

 cessary to compare the azimuth compasses at the time of obser- 

 vation, with the compass by which the ship was steered, in order 

 to obtain the proper variation to be allowed on the course 

 steered. 



It would likewise be advisable, if it could be done, to have a 

 compass placed in midships, so that the vessel might always be 

 steered by it, which compass should always be considered as 

 the standard compass ; and whenever azimuths or amplitudes 

 are taken in any other part of the ship, a comparison should be 

 made between this and the azimuth compasses, by repeating 

 these observations on opposite tacks and courses, an approximate 

 correction would be obtained, and likewise a knowledge of the 

 proper variation to be allowed on the course steered. 



Indeed, whenever the weather would permit, a knowledge 

 of the true variation might always be obtained by azimuths 

 taken at the mast-head, independent of the ship's attraction ; for 

 the sphere of influence of this attraction appears to me to be ex- 

 ceedingly limited, as plainly appears from the agreement of 

 azimuths taken on the ice close under the ship's sides when they 

 were beset with ice, and those taken on the ice at a considerable 

 distance from the ships. It likewise appeared from the bearing 

 taken with the mast-head compass, agreeing with those taken 

 on the ice. On the 21st July, 1818, at midnight, the sun's 

 azimuth was taken on the ice about eighty yards from the ship, 

 and found to be 27°. W. ; and in order to determine the effect of 

 the ship's attraction on the needle, I immediately placed the 

 azimuth compass on the ice, close under the ship's bow, and 

 allowing for the sun's motion in azimuth, it was still 27°. W.; 

 but on removing it on board, and placing it on the ship's gang- 

 way, it was then 21°.45' W. — the ship's head at the time W.bS. 

 Since, therefore, no sensible effect was produced on the needle 

 by a near approach to the ship's side, much less probable is it 

 that any effect should be produced at the mast-head, where the 



