48 CAPTAIN SABINE ON THE DIP OF THE 



of 1821 : the limb is a circle of twelve inches diameter, each quadrant being 

 graduated from 0° at the horizon to 90° at the vertical, and divided into spaces 

 of 20'. The vertical edges of the agate supports, on which the axis of the 

 needle rests, are rendered horizontal by a circular brass plate carrying a spirit- 

 level ; the lower surface of the plate is carefully ground, and being placed on 

 the supports and turned successively in the four principal directions, the ad- 

 justment is made by the foot-screws of the instrument until the bubble of the 

 level is stationary in every direction in which the plate is turned. The divi- 

 sions 90°, and 90° of the circle marking its vertical points, ought then to coin- 

 cide (and should they not do so, there is an adjustment to render them coin- 

 dent,) with the points of conical radii, proceeding from the surfaces of the 

 plate at right angles to it. By means of this plate, the level, by which the 

 horizontal adjustment is effected, is applied directly to the supports, and the 

 graduation of the circle made also to correspond with them. 



The needle which was first used, was a flat needle of the ordinaiy construc- 

 tion 1 1 .5 inches long, .4 broad, and .05 thick, rounded at the extremities. Three 

 distinct observations were taken with it in each of the following positions ; 

 and as the arc was read at both ends of the needle, six readings were obtained 

 in each position. — 1st, with the face of the instrument to the east, and the 

 marked side of the needle also to the east ; 2nd, with the face to the east, and 

 the marked side of the needle to the west ; 3rd, with the face of the instrument 

 to the westj and the marked side of the needle to the west ; 4th, with the face 

 to the west, and the marked side of the needle to the east. The poles were 

 then changed, and the same course of observation gone through as before. 

 The poles were changed by bar magnets in the usual way ; the magnets were 

 passed along the flat surfaces of the needle, ten times on each side ; the needle 

 being laid in a groove, which confined the motion of the magnets to a direction 

 parallel to the needle. The force of magnetism imparted to the needle, on 

 each occasion when its poles were changed, was measured by the time of per- 

 forming 10 vibrations always in similar arcs. The horizontality of the supports 

 was examined afresh every time the instrument was turned in azimuth ; and 

 no pains were spared to obtain results which might be consistent with each 

 other, as I intended afterwards to apply the small screw and weight devised 

 by Professor Meyer to the same needle, and wished to compare its perform- 



