l8 B1RKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, IQO2 1903. 



During the first half of October there was low atmospheric pressure with frequent falls of snow, 

 often accompanied by high wind. Throughout the latter half of the month the pressure was higher, with 

 sleet and snow, the latter sometimes very thick. 



In November the weather was variable, without much precipitation, but sometimes with high winds. 

 The temperature was not very low, having kept at about o. 



During the first half of December, the sky was alternately clear and overcast, but there was little 

 precipitation. Towards the end of the month, the pressure was lower. High winds were frequent, though 

 they did not attain a higher velocity than 15 metres per second. 



During the first week of January, the weather was cold and calm, the lowest temperature being 

 20 '3. Later on a lower pressure supervened, with mild weather and high wind. 



From the 8th to the i5th February, we had the lowest pressures that were observed. It went right 

 down to 711.0 mm. and remained at about that height for several days. With the exception of a couple 

 of days in the middle and end of the month, the atmospheric pressure throughout February was unusually 

 low, with a cloudy sky and some snow. 



In the course of the autumn and winter, 27 auroral phenomena, some of them very well developed 

 and of long duration, were observed and described. It appears that almost without exception, they make 

 their appearance in the afternoon and during the evening, generally disappearing soon after midnight. 



They usually develope from the northern sky, but not infrequently, especially during a bright mani- 

 festation, they appear on the southern sky. This was observed in the cases of the bright, exceedingly 

 beautiful and long-lasting auroras of the nth, 24th and 3ist October, and 24th November, which took 

 place simultaneously with some of the very greatest magnetic storms that were observed during that period. 



The aurora of the 24th November in particular was one of extreme beauty. It developed into an 

 auroral corona, which lasted some minutes, and then dissolved into a great number of intensely brilliant, 

 red streamers. These moved backwards and forwards across the heavens for some time, making the 

 sky glow with red. 



Considering that there was so much cloudy weather in October, it must be admitted that we were 

 exceptionally fortunate in being able to observe these beautiful auroral phenomena. On the other hand, 

 it is not improbable that the overcast sky from the 8th to the I5th February may have caused some 

 auroral phenomena to escape our attention, as at that time, owing to magnetic conditions, bright aurora 

 might have been expected. 



The weather on the whole must be said to have been not unfavorable. The violent storms experi- 

 enced on former occasions up at the mountain observatories, we that winter escaped by keeping down in 

 the valley at Kaafjord. The greatest wind-velocity measured was not more than 19 metres per second. 



AURORAL STATION IN DYRAFJORD, ICELAND. 



6. The station was situated upon a promontory, Hofdaodden, on the north side of Dyra Fjord (see 

 Fig. 17). Its latitude was 66 15' N., and longitude 22 30' W., equivalent to i hour and 30 minutes 

 before Greenwich time. 



The members of the expedition were SEM S^ELAND (leader), amanuensis to the University Physical 

 Institute, and LARUS BJORNSSON (assistant). Saeland left Christiania with his equipment on the roth October, 

 and arrived in Iceland on the loth November, 1902. The voyage was satisfactorily accomplished, but 

 the vessel was delayed a fortnight by snow-storms. 



