FiTzGrERALD & CuLLUM — Magnetic Observations at Valentia. 225 
40 cms., the readings being taken on the circle when the magnetic 
axis of the deflecting magnet is at right angles to that of the suspended 
magnet, which is obtained by turning the circle in azimuth until the 
wire of the telescope cuts the centre division of the ivory scale. This 
is done first with the north end of the deflecting magnet to the east, 
and secondly to the west. Half the difference of the mean of the 
readings in the two positions is the angle of deflection required. By 
this process the ratio is obtained of the magnetic moment of the 
deflecting magnet to the Earth's horizontal magnetic force ; the latter 
being to the former as 1 to the sine of the angle of deflection multi- 
plied by half the cube of the distance employed. 
The dip circle has a horizontal circle for setting it in the magnetic 
meridian, and a vertical one for measuring the angles which the 
needles make with the horizon when in that position. There are two 
needles, the means of their readings being taken as the angle of incli- 
nation. They are made of flat, tempered steel, about 3^ ins. long, 
pointed at the ends, having accurately-turned cylindrical axles at 
right angles to the plane of the needles. These axles rest on finely- 
polished agate knife edges. Prom the two elements, inclination and 
horizontal force, the vertical and total forces are calculated by means 
of the formulae 
Vertical force = hor. force x by tan. of the dip. 
Total force = hor. force cos. of the dip. 
Declination. — The same apparatus and magnet is used for this ob- 
servation as for the horizontal force measurement ; the magnet being 
fitted with a double suspension to admit of its being reversed, in case 
the line of collimation of the hollow steel magnet does not coincide 
with its magnetic axis. The position of this is first determined by 
finding the zero of the scale by repeated reversals of the magnet on its 
suspension, and noting the division which the wire of the telescope 
cuts on each side of the centre. The magnet is brought to rest, and 
the instrument turned in azimuth, until the wire of the telescope 
occupies the position so determined, and the reading taken on the ho- 
rizontal circle. Another independent setting is made in this position 
when the magnet is reversed ; but in this case, if the zero is found to 
be on the left of the centre division, it will now be on the same divi- 
sion on the right. A reading of this is taken on the circle and repeated, 
the means giving the readings obtained when the scale is erect and 
inverted. By this plan the difference between the axis of the tube 
and its magnetic axis is eliminated by taking the mean of the readings 
