374 Astronomical and ISiauiical Collections. 



effect, will vary directly as the sine of the angular distance of 

 the needle from the ship's head, or from any other given neutral 

 line in which the disturbing force of the ship is found by expe- 

 riment to act, and inversely as the magnitude of the horizontal 



magnetic force ; that is, directly as r y' ( , f- 3) ; r being 



the sine of the bearing of the ship's head, or other " point of 

 change," as ascertained by the actual indication of the compass, 

 and not by the corrected bearing, which has sometimes been 

 employed in a similar calculation, and s the sine of the dip ; the 

 quantity under the radical sign being equal to the square of the 

 secant of the dip increased by 3. 



10. If, for example, the utmost disturbance were found to be 

 5° 40', where the dip is 74° 23', its sine would require to be 

 increased, when the dip became 86°, in the ratio of 1 to 3.523, 

 and the maximum of disturbance would become 20° 21'. It is 

 scarcely possible that the calculation should agree better with 

 the result of the observations made on board of the Isabella : so 

 that we may employ it, with some confidence, for our assist- 

 ance, in correcting the errors arising from the disturbing force 

 of the ship in all ordinary cases. 



11. When the ship's attraction is constant, it is obvious that 

 the two neutral positions, in which it produces no disturbance, 

 will be observed when the ship's head is exactly in opposite 

 directions. But it appears that there is sometimes also an 

 irregular attraction, causing the two neutral points to be within 

 8 or 10 points of each other ; and when this happens, we can 

 only rely on immediate observation, in different parts of the 

 globe, for determining the requisite corrections. This part of 

 the disturbance, however, seems not to increase with the dip, 

 and there is every reason to attribute it to the temporary or 

 induced magnetism of some portions of soft iron ; since it may 

 easily be shown, for example, that a horizontal bar of soft iron 

 will lose its effect on the needle in four positions, at right angles 

 to each other, and a bar, so inclined, as to become perpendi- 

 cular to the dipping needle in the plane of the meridian, will 

 lose its effect in its two opposite positions, in that plane, only, 

 but will act with very different intensities in their neighbour- 



