INTRODUCTION. if 



the whole auroral station, as already stated, was moved to Bossekop, in the vicinity of the polar station 

 of 1882 83, on the I3th March, 1903. Before many days had passed, all the instruments were again 

 in operation. 



The Thermometer-Hut. 



This was built like an English hut of wood, and large enough to contain the thermometer-screen 

 and the thermograph. The arrangement was the ordinary one. By the thermometer-hut was placed a 

 weather-vane, with which measurements were taken 3 times daily of the velocity of the wind, with the 

 aid of an anemometer Richard. 



The barograph was placed in an unused room in the dwelling-house. Near it stood the cloud- 

 measuring apparatus, especially for use in determinations respecting polar bands and cirrus clouds. 



The electric measurements with Elster and Geitel's apparatus, were also made in the vicinity of the 

 dwelling-house, in order that wind and weather should not have too disturbing an influence. 



5. During our stay at the stations Haldde and Kaafjord, a journal was kept of the meteorological 

 elements, and of the aurora and cirrus-bands observed. These observations cover a period extending from 

 the 28th August to the end of February. For the last month, March, there are no records of this descrip- 

 tion, as the entire day was taken up with registering, especially rapid registering with changing of the 

 photographic paper on the instruments every two hours. 



The meteorological observations were made regularly 3 times a day - - at 8 a. m., 2 p. m., and 

 8 p. m. 



These observations show that the weather, as is usual in these regions at this time, has been very 

 variable. The sky has very seldom been quite clear, but was as a rule covered with clouds, a circum- 

 stance which has to some extent hindered us in our observations of aurora. 



Some idea of the weather-conditions at this time may be obtained by looking at the table below, 

 in which the highest and lowest temperatures and barometer-readings, and the highest wind-velocity ob- 

 served at the above-mentioned hours are given for each month. 



In August and the first half of September, the atmospheric pressure was fairly low, but with little 

 precipitation to speak of. The temperature remained, on an average, at about 3C. In the latter half of 

 September, there was high pressure with rain. On the 27 th September, the first snow fell, the temperature 

 at the time being about 2'2. 



Birkeland, The Norwegian Aurora Polaris Expedition, 1002 1903. 



