TRAVELS lis ICEIANB, 



145 



-towards the north-west; but suddenly it expanded over the 'whole 

 sky with an extraordinary brilliancy: it was admirable to see this 

 phenomenon disappear several times in a sort of twinkling, 

 so that there remained only a feeble light at the south-east in the 

 lower part of the horizon. A strong wind had blown all the 

 day; and in the evening the frost had set in so severely, that the 

 sea was frozen between the isles and the coast. On this and 

 eight following days the wind was north-easterly, and often gentle 

 except on the 23d and 24th of November. 



On the 18th December 1754, the aurora borealis appeared 

 in the evening after a very stormy day, accompanied with hail, 

 thunder, and dreadful lightning. The storm continued in the 

 evening, and all the night; but between the gusts the sky appeared 

 clear and the moon bright, the winds blowing from the S.S.W. 

 The aurora which then appeared, did not vacillate; it extended 

 from W.N.W. in a number of luminous and narrow streams, at 

 first as far as the zenith, and afterwards descended E.N.E. to- 

 wards the horizon. The storm and lightning continued the fol- 

 lowing day: the frost was more intense, but there was no thun- 

 der, and the' aurora again appeared in the evening. 



On the 9th December 1755, the weather was fine with a 

 gentle east wind: it froze at half past nine in the evening, and 

 the moon appeared in the W.S.W.; w hen the aurora borealis ap- 

 peared suddenly in the west, and divided into two great arches, 

 which extended over the horizon north and south at no consider- 

 able height till they met together. This spectacle continued 

 for about a quarter of an hour; and the streams did not penetrate 

 into the upper region. At ten o'clock there appeared an arch 

 in the east passing the zenith, and proceeding westwardly: it 

 continued to expand till half past eleven, when the upper ho- 

 rizon was covered by it, and threw out a vivid light like 

 flames. The s~y could, however, . be perceived between these, 

 and the ring still appeared to surround the horizon.— At length, 

 at midnight the aurora entirely disappeared, with, the eruption of 

 the ring; which shone about an hour longer, and whose strongest 

 fight was on the northern side. 



On the 1st October 1756, a gentle w 7 ind blew from the N.W. 

 the sky was clear, and the w eather was cold, when a thick arch, 

 not particularly luminous, rose from the W. to the zenith, whence 

 it darted in a northern direction a luminous point that formed a 

 rectangle, and which was again subdivided into two other points. 

 This was the only time that we ever saw the aurora borealis ex- 

 hibit such a singular spectacle. Shortly afterwards the light re- 

 tired towards the east of the horizon, shining with brilliancy, 

 increasing in size, and rising in undulatory arches, which con- 

 tinued to dart more rapidly towards the S. aud JS". The light 



OLAFSEN.] "K 



