reducing Lunar Observations. 383 
that the error of the estimated longitude affects the elements 
entering into the computed altitudes taken from the Nautical 
Ephemeris, of which the observed altitudes are independent. 
Immediately before or after new moons the faint image of the 
moon when she is high may be difficult to bring down to the 
horizon, and occupy such a position in respect of it, that 
neither upper nor lower limb can be correctly observed. The 
computed altitude of a fixed star is also more to be depended 
on than its observed one; in these cases it is better to com- 
pute, particularly the high altitudes, with the reduced geo- 
centric latitude, as the refraction corresponds thereto, which, 
by rights, ought to have been taken from the tables for the 
angle of the ray of light with the normal or with its comple- 
ment; the altitude above the sensible horizon then requires only 
a trifling correction. 
When the lower object is a star, and the moon’s altitude is 
not too small, the usual methods are sufficiently correct, pro- 
vided allowance is made for barometer and thermometer. 
Here follows a specimen of the table of the contraction of 
the vertical diameter on account of the refraction, which is 
calculated for the mean diameter of the sun, and for mean re- 
fraction. 
Correction for Correction for Correction for 
Altitude. Tower 
Limb. | 
4°50"! — 
4 45 
40 
35 
Lower | Upper i c Upper 
Limb. | Limb. i Limb. 
fe} 
=r 
= OW Ph or 
cooceccos 
9 
9 
9 
, 
9 
9 
8 
8 
8 
8 
8 
8 
BOO NIA ARAAaNNS 
mm Ro 
i — i — i — i — i — i — 
WCwOWWWAEL EE RE PR 
v 
NNN 
c & or 
—i——) 
or 
