282 2 I. P. Косн. 
most cases be of no importance; but an error as great as that will 
only occur rarely, even in very unfavourable circumstances. The 
previously indicated manner of pointing by touching the two limbs 
of the sun is, therefore, not necessary; one may very well take the 
liberty to bisect the centre of the sun, as 
— indicated in Fig.40, which manner of pointing 
er was almost exclusively used on the Dan- 
A | mark-Ekspedition. 
| As the observation, as seen above, may 
be made in a very cursory way, it is not 
necessary to take more than two pointings one with “telescope left” 
and one with “telescope right”. The error of the azimuth due to 
the inclination of the horizontal axis is trifling in this connection; 
hence the reading of the level of the axis is superfluous. 
In case the azimuth is to be determined by the zenith distance, 
the greatest accuracy is attained by performing the observation in 
the prime vertical; on the other hand, in case the azimuth is to be 
computed by time, the best result is attained, when the hour angle 
is near 0° or 180°. In high latitudes where the inclination of the 
day circle towards the horizon — 90° +g — is inconsiderable, the 
hour angle may, however, be used through all the twenty-four hours 
for the determination of the azimuth; on the other hand a determina- 
tion of the azimuth by the zenith distance becomes extremely uncertain, 
when the heavenly body is near the meridian. This fact plays a part 
in the valuation as to which of the two results one should ascribe the 
greater accuracy to. In case the observation is taken near the meri- 
dian, the determination by the zenith distance may be used as a 
sort of rough and ready control. 
One might, from the above general remarks, perhaps be tempted 
to believe that it would be practicable to use the determination of 
azimuth promiscuously with the observation of time and latitude. 
The observation being a zenith distance observation, only presenting 
the particular feature that the vertical thread is to pass through the 
centre of the sun (to touch upon the limb of the sun) and that the 
horizontal circle is to be read, it would only be necessary, in the 
course of the determination of time and latitude, to observe these two 
things, in order to be able to compute both azimuth and time or 
latitude by the observation. In this manner one could also dispense 
with an observation, which in the difficult conditions, under which 
the work is often done, would mean a not inconsiderable alleviation. 
However, I warn people against permitting themselves an alleviation 
of this kind, because the observations of time and latitude thus be- 
come more complicated; the continuity of a series of observations 
Fig. 40. 
