420 SECOND VOYAGE FOR THE DISCOVERY 



1823. distant objects may be seen over them, though there is near them always 



Marcli. . . •" , . , , , ,, . ^ . . . , , , , ^ 



w^-^^ a mistiness to which they perhaps owe their origin. Although however 

 the winter atmosphere of these regions is seldom free from numberless mi- 

 nute particles of snow, which are abundantly deposited upon any thing 

 left in the open air, yet it was not observable, except in some cases 

 of snow-drift, that parhelia were more frequent or distinct when this 

 de])Osit was the greatest, than when the atmosphere was comparatively 

 clear, though in the latter case they are always to appearance most 

 distant. Parhelia occur most frequently, and exhibit the greatest inten- 

 sity of light, at low altitudes of the sun. This is often particularly ob- 

 servable in the short days, when these phenomena assume a very brilliant 

 appearance soon after sunrise, decrease in splendour towards noon, and 

 resume their brightness as the sun descends towards the horizon ; con- 

 tinuing however distinctly visible the whole time, and being sometimes 

 accompanied by a more or less perfect halo undergoing corresponding 

 variations. 



Another peculiarity observed in this winter was the rare occurrence of 

 the Aurora Borealis, and the extraordinary poorness of its display when- 

 ever it did make its appearance. It was almost invariably seen to the 

 southward, between an E.S.E. and a W.S.W. bearing, generally low, the 

 stationary patches of it having a tendency to form an irregular arch, and 

 not unfrequcntly with coruscations shooting towards the zenith. When 

 more diffused it still kept, in general, on the southern side of the zenith ; 

 but never exhibited any of those rapid and complicated movements observed 

 in the course of the preceding winter, nor indeed any feature that renders 

 it necessary to attempt a particular description. The electrometer was fre- 

 quently tried by Mr. Fisher, at times M'hen the state of the atmosphere 

 appeared the most favourable, but always without any sensible effect being 

 j)roduced on the gold leaf. 



The difference in the temperature of the day and night began to be 

 sensible as early as the first week in March, and the daily range of the ther- 

 mometer increased considerably from that time. The increase in the average 

 temperature of the atmosphere, however, is extremely slow in these regions, 

 long after the sun has attained a considerable meridian altitude ; but this is in 

 some degree compensated by the inconceivable rapidity with which the days 

 seem to lengthen when once the sun has re-appeared. There is indeed no 

 change which continues to excite so much surprise as that from almost con- 



