104 I. P. Косн. 
general object necessitated the taking of observations for time at 
intervals during the whole year, and during the summer we could 
naturally do nothing but measure the zenith distance of the sun, 
whereas in the winter we almost exclusively used transits on the 
vertical of a circumpolar star near the Pole. On the other hand 
the transits on the meridian were only used on a few occasions for 
a purely preliminary determination of the clock correction, and the 
reason of this was first and foremost our lack of a meridian mark. 
There is presumably no reason to occupy ourselves at greater 
length with the measurements of the zenith distance of the sun; 
they do not deviate in principle from the determinations of time by 
means of the sun, mentioned under the geographical survey, where 
also examples of computation are entered; only the computation 
becomes a little more elaborate here in consequence of the greater 
accuracy — tenths of a second — which in this case, at least from 
the point of view of computation, ought to be aimed at. 
On the other hand I am going to enter into the determinations 
of time in the vertical of a polar star near the Pole; but before that 
I will give a short account of the comparison of the watches and the 
determination of the thread intervals. 
Clock comparisons. Before and after the observations for 
time we made comparisons of the watches, partly in order to be 
able to control the rate of the observation watch during the observa- 
tion, and partly to be able to transfer the exact time to the standard 
watch of the expedition, mean time chronometer 57310. Four com- 
parisons were made in the course of a minute, as appears from the 
example quoted below. 
The comparisons were made by two persons, each provided with 
a watch. The assistant announced “Ready” five seconds before the 
moment of comparison, and after that “Now”; the observer estim- 
ated the time corresponding to the moment of comparison on the 
observation watch in tenths of second. 
Only very little practice is required to attain considerable skill 
in the estimating of the tenth of the second. By means of coincident 
beats of the mean time chronometer and the sidereal chronometer 
we might perhaps have carried the accuracy of comparison some- 
what further; but, as will appear from the following, this would not 
have been of practical value to us. For the determination of the 
transit of a heavenly body we were besides obliged to use the same 
primitive method, as we had no chronograph with us. 
Comparison of watches before and after the determination of 
time 24/1X 1907. 
