THE HAMILTON ASSOCIATION. «19 
viated to the left. The object, S, will now be seen at a distance 
to the left of the intersection of the wires which measures the 
angle D1, C D2, which is twice the angle D C D1, or the devia- 
tion of the line of collimation from the perpendicular D C. In 
whatever position the instrument may be placed the rotation 
axis is an imaginary line passing through the central points of 
the pivots, and the axis of collimation an imaginary line drawn 
from the optical centre of the object-glass perpendicular to the 
rotation axis, and describes a great circle in the heavens when 
the telescope revolves. Now the position of this great circle 
is fullv determined when we know the position of the rotation 
axis, or in other words, when we know the altitude and azimuth 
of the points where the rotation axis produced meets the celes- 
tial sphere. The instrument is said to be in the meridian when 
the great circle described by the axis of collimation is in the 
meridian. The axis of rotation is then perpendicular to the 
plane of the meridian and lies in the intersection of the prime 
vertical and the horizon. Now, if the centre thread on which 
a star is observed is in the axis of collimation, the time of ob- 
servation is that of the star’s transit over the meridian. The 
sidereal time at that instant is equal to the star’s R. A. The 
error of the clock on sidereal time is obtained at once by tak- 
ing the difference between that R. A. and the clock time of 
transit. These conditions are very rarely exactly fulfilled, 
but small corrections must be made to the time of observation 
for small deviations in the three principal adjustments. These 
corrections are denoted by the small italic letters, a, b and c—a 
is the excess of the azimuth of the axis above go degrees (reck- 
oned from the north), and is called the azimuth constant; b is 
the elevation of the west end of the axis, and is called the level 
constant; c is the inclination of the sight line to the collimation 
axis, and is called the collimation constant. I have already 
shown how DB and ¢ are reduced to zero, or a very small quan- 
tity. To reduce a toa small quantity, or to make the approxi- 
mate adjustment in the meridian, we must have recourse to ob- 
servation of the stars. The star most suitable for this purpose 
is the Pole star (Ursae Minoris). Compute the mean time of 
