THK AKCTIC SKAS. 117 



winter residence there beincj dreaded as one of the 

 direst cahamities that can befal them ; and therefore, 

 until lately, our knowledge of winter phenomena was 

 very meagre, and mainly derived from the reports of 

 a few unhappy men, by accident compelled to remain 

 in a clime so inhospitable. By the experience of the 

 officers and crews engaged in the recent voyages of 

 discovery, we have become nearly as familiar with 

 the phenomena of the long winter's night as with 

 those of the short summer's day. In Spitzbergen the 

 day is rather more than four months long : the night 

 is of the same duration, and in the two months 

 which intervene between the sun's constant presence 

 and his constant absence, that luminary rises and sets 

 as with us. But the appearance of the sun in spring is 

 accelerated, and its disappearance in autumn retarded, 

 a few days by the influence of refraction : so that 

 it is actually seen somewhat longer than it is in- 

 visible. Thus Captain Parry, at jMelville Island, 

 saw the sun on the first of February, which was 

 about four days earlier than its actual elevation 

 above the horizon ; in like manner it remained 

 visible until the 11th of November, whereas it had 

 actually sunk beneath the horizon on the 7th. Then 

 the darkness of the Arctic winter is not total and 

 incessant ; even in the depth of the season, at Spitz- 

 bergen, there is a faint twilight for six hours each day, 

 and this is longer and brighter in proportion to the 

 distance from mid-winter on either hand. The moon 

 also shines in those clear skies with peculiar bril- 

 liance, and is often visible twelve or fourteen days 

 without setting. There is, moreover, a large proportion 



