116 Г.Р. Косн. 
edition of the Nautical Almanac, and we had consequently great 
difficulty in finding suitable stars for the determinations for time. 
In order to prevent the determinations for time and the moon ob- 
servations from clashing we often had recourse to À Urs. min. as a 
pole star instead ofa Urs. min. The former star is nearer the Pole 
than Polaris, and besides it has the advantage of being only 6.6 in 
magnitude, which facilitates an accurate pointing. 
In order to avoid the rather circumstantial computation of the 
determination for time on the vertical of Polaris, several methods for an 
abbreviated calculation have been set up, which methods, how- 
ever, either do not represent any noticeable lessening of labour or 
become tantamount to a considerably less degree of accuracy. The 
latter circumstance would, however, be of no importance, if only the 
‘accuracy of computation corresponded to the accuracy of observation. 
An abbreviated form of computation which we used from time 
to time is the following: 
By means of the clock time U for the pointing of Polaris and 
the approximate clock correction du,, the azimuth of Polaris a’, was 
taken from the table mentioned in the preliminary computation 
(argument: The hour angle of Polaris Ё). By means of the above 
mentioned table for the factor (eee the hour angle of the 
equatorial star Ё is computed, after which the sidereal time = a+t 
is determined. Result: 
du = а + t — (U + isec $). 
As an example the two above-mentioned determinations for time 
will be computed according to the abbreviated manner of computation 
here mentioned. The clock correction is approximately estimated 
du, = + 18 19% 3085 this makes: 
U', = a! di, — 1» 26m 495 + 1h 19m 305 = + 7m 195. 
In this case there is no question of applying the correction for 
collimation. 
If a comparison is made with the values of clock corrections, 
arrived at by means of the earlier computation, it appears that the 
errors resulting from the abbreviated computation are of 05.2 and 
05.9 respectively. A rough estimate will show that the chief reason 
for this is to be found in the error of the approximated clock 
value du,. 
The example shows that it would not generally have been 
permissible to adopt this abbreviated manner in the case of the 
determinations for time of the Danmark-Ekspedition. 
