302 



Alf Trolle. 



The first winter 1906 — 07 both reversible thermometers were 

 broken during transport on the sledge and several water-samples 

 were lost. 



1 was unwilling therefore to risk using the two remaining good 

 thermometers for the Pettersson-Nansen water-bottles in the winter 

 of 1907/08, and contented myself with taking samples with Ekman's 

 water-bottle, which was the most convenient to handle in the cold. 



Fig. .5. Sledge for a journey to make hydrographical observation; 

 С В. Thostrup phot. 



It is a study to work with instruments at temperatures down to 

 — 30°C., where everything is frozen hard the moment the water- 

 bottle comes out of the water and the fingers become numb as soon 

 as they are taken out of the mittens, and no little time is required 

 to gain experience in the conditions of working from a sledge. 



I may, therefore, quite briefly note some points which in my 

 opinion should be taken into consideration in case of future similar 

 investigations. 



I 



