252 1. P. Косн. 
bottle, and of the discussion which ensued among the members of 
the expedition in connection with this piece of glass. 
The above research must give one the impression that it seems 
hopeless to try to find support for the hypothesis of the drift of 
Greenland towards the west in the determinations of longitude under- 
taken by SABINE and the Germaniaexpedition, as long as the site of 
the ruins of the houses and so also the position of SABINE's Ob- 
servatory in relation to that of the Germaniaexpedition are unknown. 
The difficulty which we have hitherto not been able to surmount is 
that CLAVERING in his map has clearly indicated the site of 
SABINE's Observatory in a place which according to PAYER's 
map must be about 260 metres or about 8”.4 farther north 
than the Observatory of the Germaniaexpedition. In this 
place, where the projecting elbow in the coastline is, the ruins of 
the houses should be found. This also seems the more probable 
explanation, as the Esquimaux, when the waves flung themselves 
against the peninsula from the open sea, only in the little bight 
north of the above-mentioned eibow might find favourable landing 
conditions for their kayaks. 
I have on a previous occasion gone into the possibility that the 
determination of latitude made by SABiNE might be encumbered by 
a systematic error, but the research did not lead to the desired 
harmony with CLAVERING's map; on the contrary it made the discre- 
pancy greater. There is, however, also a possibility that the systematic 
error is to be looked for in the observation of latitude of the Ger- 
maniaexpedition. 
Accurate observations of the sun are a difficult thing in arctic 
regions, because the sun’s rays here have an extremely powerful 
effect and easily induce systematic errors. (For fuller particulars as 
to this, see “The Work in the Observatory” pp. 92—93). The use of 
an observatory may further occasion an error which is caused by the 
special refraction of the meridian aperture; the latter may, in an 
arctic observatory, easily amount to a couple of seconds, particularly 
in the summer, where the observatory is brought up to a very high 
temperature owing to the influence of radiation. 
The walls of the Observatory of the Germaniaexpedilion were 
built of natural stone’); the roof which was made of boards and 
canvas could not be lifted off, so that the circulation of air only 
took place through the door and the ‘2 metre broad meridian 
aperture. There has, however, probably also been an aperture for 
observations in the prime vertical. That the conditions of a great 
1) Die zweite deutsche Nordpolfahrt, Bd. I, pp. 434 and 435. 
