SO JOURNAL OF PROCEEDINGS. 
its transit, and for the purpose of a first approximation your 
watch may be used. Bring the telescope upon it at the calcu- 
lated time by moving bodily the frame of the instrument hori- 
zontally. ‘Then level the axis and take the time by the sideral 
clock of the transit of a star very near the zenith. The time 
of the transit of a star near the zenith will not be much affected 
by a deviation of the instrument in azimuth, as all vertical cir- 
cles intersect each other in that point. The difference between 
the star’s R. A. and the clock time will be the approximate 
error of the clock on sideral time. Knowing the clock error 
we can repeat the process upon another slow moving star, em- 
ploying the clock and causing the middle wire to follow the star 
by moving only the azimuth VY. The instrument should now 
be very close to the meridian and can be further tested by other 
stars. All outstanding small deviations represented by the 
small letters, a, b and c, can be found and allowed for by 
methods which do not concern us in this paper. 
