ALTAZIMUTH SOLAR OBSERVATIONS. 347 
altitudes before and after noon, the mean of these would be to the 
east of the direction of the astronomical meridian! of the observer 
by the whole amount of the diurnal-aberration-correction for 
either position, provided the refraction and declination were 
identical for each position. From (21) §7 it will be seen that 
the correction for diurnal aberration is very small, since A and ¢ 
are generally not very far from 90°. If applied, however, the 
deduced meridian must be shifted westerly, because the mean 
direction would be either north-east or south-east, the former if 
the sun be observed when north-east and north-west, the latter 
when south-east and south-west. 
The refraction is almost certain to be different at the two 
observations, because the barometric pressure and the temperature 
will have changed in the interval between them. The latter 
element should be measured in the same way for each, say by 
slowly whirling a thermometer in the air. The difference of 
refraction may be found with sufficient precision by multiplying 
the mean refraction for the zenith distance employed, by the 
differences of the products of the barometric and thermometric 
. factors for the two pressures and temperatures recorded. 
Let r denote the mean refraction, £8 and 6 the correcting factors,” 
the suttixes indicating the observation to which they apply, then 
the difference of the refractions is 
mr — 17, = 7 (Bz 9, - B, 91) (47) 
: 1 Determined by the intersection with the horizon, of the vertical 
circle containing the pole of the heavens, and the observer’s zenith, and 
Subject therefore to the deflections due to the rugosity and heterogeneity 
of the earth’s crust. This direction is of course quite distinct from the 
line perpendicular to the curve of the observer's latitude ; this last may 
be called the « geodetic ” meridian. 
? For a table of mean refractions computed for 29°6 in. pressure at 32° 
Fabr. and 48-°75 Fahr. air temperature, the factors will be those in (n) 
§ 6, or in (29) § 11. Bessel’s refractions and correction factors pp. 430, 
431, in Chamber’s Mathematical Tables, 1885 Edit., may be used: the 
Pressure and temperature of no correction is as before stated, those 
mentioned, 
