The small difference between this method, and the former, is owing to some losses sustained in not finding the exact number answering to a logarithm, and to the advantage of two decimal figures in the difference of latitude and departure, in working the traverse table; therefore the first method is preferable.

OF

ATTRACTION.

IT sometimes happens, that the needle is attracted out of its customary position, by an unknown cause; but most probably owing to some matter near the surface of the earth: as it is frequently observable where there is the appearance of iron-stone; and this attraction will frequently vary in quantity, and in denomination, at every station. Therefore, to take the bearings truly, where the needle is attracted, observe the following directions; as for example,

Suppose a field, bounded as by the Map in the following page, to be so situated; a true survey of which is required.

Note. Particular care must be taken to count the degrees at every station from the same end of the needle; as in this example, I count from the north end.

Seek a place where two stations can be taken, without attraction; this may be tried and proved by back-sights, and the nearer to one corner of the field the better, whether within or without the inclosure, as for instance; suppose I set my instrument at G, and take the bearing to H, and then by a back-sight from H to G, the bearing is the same; therefore no attraction at either of these stations. Then from H, I take the bearing to A, one corner of the field, and it is N. 25° W. then by a back-sight H bears

Note. The dotted lines represent the Magnetic points.

from A, S 22 E. therefore at A, the north point of the needle is attracted 24 degrees westerly; then turning the sights to B, it bears N. 40° W. to this, add the attraction 21 degrees, and it gives the true bearing N. 43° W. distance measured is 7 chains, 20 links. Then, by a back-sight, A bears from B, S. 46° E. therefore the north