Survey of Northeast Greenland. 271 
from the series of observations — one ought, before forming an 
: IV 
estimate of the mean error, first to have formed the means = , 
П-+У III + V 
a and ——. 
From the preceding it appears that on sledge trips one may 
count on being able to determine ones latitude by the sun — by 
means of the small Hildebrand universal instrument — with a mean 
error of about ten seconds, which accuracy is quite satisfactory for 
ordinary geographical purposes, and which in conditions, so difficult 
as those we are here dealing with, can hardly be attained by any 
other instrument. 
The greatest source of error, as far as the universal instrument 
is concerned, is to be found in the level, which when travelling 
cannot be protected from the effect of radiation, and which conse- 
quently becomes sluggish. At star observations one may, when the 
temperature is not too low, carry the accuracy further, which the 
expedition has done in many cases. To conclude, I shall here give 
an observation of this kind, performed under particularly favourable 
conditions. 
Pustervig, 6/XI 1907. Determination of latitude by 
Polaris (Observer: Косн; Assistant: WEGENER). b = 749 mm; Ё = 
— 99,6; 1 division of the level = 20”; correction of sidereal chrono- 
meter 57203:s = 1513™41s; | = 1h24m19s w. of Grw.; declination of 
Polaris д = 88°48'54”. 
Telescope right 
Telescope left 
(ae ar 
| 
Уегшег | Level | Vernier | Level 
57203 = Sy =. 51209 > | ЕЕ 
| В | Obs. Obj. | | A IE Obs. ОБ}. 
| | | | | | | 
Oh10m12s _ 11°59'.0 | 59.7 | 4.3 | 14.0 | Qh]4m]2s | 348911'.1 | 11.82, 1.0288 
17 28 59.0 | 59.8 | 4.2 | 13.9 | 20 35 11.0 | ЕО 11.2.1 37 
23 23 59.2 | 59.7 | 40 | 13.8 | Я | 10.8 | JEUN |MEG 
For the computation of the latitude by Polaris there are special 
tables in the Nautical Almanac. Here we are going to use the formula: 
Dh 
ф = h=pcost+3—p sin” ttan В 
where h is the altitude and p the polar distance. The observations 
are grouped in pairs, as in the earlier examples: 
XLVI. 18 
