Survey of Northeast Greenland. 155 
performed, within the period from July 23rd 1907 to July 9th 1908, 
from the Observatory fifty-three levellings of each of the following 
four points: Сати” №, arret IX, Cairn V “and? the “Skerry” The 
object of this was to investigate the variations of the co-efficient of 
refraction in the various strata of air at different periods. 
Cairn X. The distance to the cairn from the Observatory is 
7494.5 m; the altitude of the centre of the wooden disc of the cairn 
is 429m. The sight thus traverses the strongly marked boundary 
between the strata at 200 m, but is does not reach so far as to strata 
of air the temperature of which is lower than the average tempera- 
ture at the Observatory”). The sight exclusively traversed land 
through the valley which extends east of the Varderyggen (the map, 
1: 100,000, of Danmarks Havn and Environs, PI. II) at a distance of 
about 3km from the coast and divided from the latter by a ridge. 
Cairn IX. The distance from the Observatory is 2648.4 m. 
The altitude of the sighting point on the cairn is 180m. Thus the 
sight lies practically altogether in the lower stratum of air, which 
is characterized by a marked inversion of temperature of about 0.4 
degrees as the mean of a year. It passes exclusively across land, 
but immediately along the coast towards west. 
Cairn V is situated in the southern part of Lille Koldewey at 
a distance of 157804 m. The altitude of the sighting mark on the 
cairn is 90m. For the last three km nearest the cairn, the sight 
traverses a low island; otherwise it lies exclusively across water (ice) 
at an average altitude of 52 m. 
The “Skerry”. The distance is 12286.0 m. The altitude of the 
sighting mark 16m. The sight nearly exclusively traverses water 
(ice) at an altitude of 14—16 m. 
Whenever observations were made, two sets were taken, thus in 
all 106 sets at every point. The two sets measured within the same 
series of observations were epitomized into a mean set, which is 
given under column z in the table below. On the strength of all of 
these fifty-three mean sets a mean was computed, and the individual 
Ad 
deviations from it are given under the column 42(42 ==: +) 
By differentiation of 
. R . о 
1— k — Do (71 + 2, — 180°) 
and putting dz, — dz, we get 
2R 
dk = — —= dz 
De 
1) As regards alterations in the meteorological elements with the altitude at Dan- 
marks Havn see ALFRED WEGENER: Drachen- und Fesselballonaufstiege, Medd. 
om Grenl. XLII. 
