of the Equatorial Instrument. 115 
time be variable, from a variety of causes, such as the pos- 
sible settling of the walls of the building; the partial or irre- 
gular expansion of the instrument, from the sun-beams acci- 
dentally falling upon it ; from the effect of a fire in the room, 
or the heat of a person's body in cold weather ; from the 
sun's heat upon the meridian mark, eastward in the morning, 
and westward in the afternoon ; from the same effect upon 
the observatory ; and lastly, possibly from a lateral refrac- 
tion of the ray coming from the meridian mark, from irregu- 
lar vapours floating near the surface of the earth. From 
some, or all of these causes, this quantity of error will be 
found to be exactly the same hardly for two days together ; 
but I have never yet known it to exceed 13" of a degree, = 
o",7 in time, during a period of more than a year, and very 
rarely above 6" or 7", sometimes on one side, and sometimes 
on the other. When this error is known, and set down, 
move the instrument about the polar axis by its handle t, till 
the divisions 360 and 180° are bisected by the equatorial 
wire ; this done, move the declination circle by its handle u, 
till the hanging level stands true ; and note the division on 
the declination circle cut by the micrometer wire, for 
that is the horizontal point, from whence the altitudes are 
to be reckoned. If the order of the divisions be such as to 
shew declinations, this division will be the angle of the co- 
altitude of the polar axis ; but if the divisions be such as in 
my instrument, it will be equal to the altitude of the polar 
axis, which should be equal to the latitude of the place. But 
as it is suited only to one latitude it is not given here. Tables, something similar to 
this, may be seen in Mr. Lud lam’s Astronomical Observations, Cambridge, 1769 ; 
and also in the Connoissance des Temps, pour 1792, p. 251. 
