222 Description of the Instruments 
The sources of error are numerous, and consist, first, 
in an error in the graduation of the divided circle; 
secondly, in the axis of the needle not being truly in the 
centre of the circle in which it revolves; thirdly, in the 
centre of gravity not being precisely in the axis of sus¬ 
pension ; fourthly, in the magnetic axis, or poles of the 
needle, not coinciding with its points; fifthly, in the 
graduated circle not being constructed of pure unmag- 
netic material; and, sixthly, from the axis of the needle 
not being truly cylindrical. 
To counteract these numerous sources of error, ob¬ 
servations are multiplied in every possible manner, with 
the needle and circle in different positions; and it will be 
advantageous, in order to give distinctness to our ideas, 
to recapitulate the sources of error, and the means em¬ 
ployed to obviate or overcome them, in the numerical 
order in which they are before mentioned. 
1st. The error in the graduation of the divided circle 
is got rid of by reversing the instrument in azimuth, i. e. 
observing the positions of the needle with the face of the 
instrument first to the east, and then to the west. 
2nd. By observing both ends of the needle at the 
same time ; consequently, any error in excess in one end 
w r ill be in defect in the other. 
3rd. By reversing its poles, making the one south that 
was north, and vice versa* 
4th. By inverting the needle on its axis; that is to 
say, observing with its face to , and from, the instrument. 
5th. The error arising from this source ought to be 
guarded against in the construction of the instrument: 
* This is effected by rubbing the north pole of a mugnet on t’ne 
north pole of the dipping needle, which deprives it of its northern 
polarity, and imparts southern. The south pole of another magnet 
is similarly applied to the south pole of the dipping nee«Vle, which is 
then changed into a north pole. ly COh 
\ 
