484 SECOND VOYAGE OF DISCOVERY. 



1823. passed Button's Isles, from which wc took our final departure. Icebergs of 

 v-*w/ large dimensions occurred from about the seventy-third degree of longitude 

 downwards to the entrance of the Strait, and we remarked that below the 

 sixty-third degree of latitude the land was still comparatively clear of snow. 



From the time of our qiiifHiig the ice we began occasionally lo notice 

 flocks of dovekies, and soon afterwards kittiwake»i and mallemucks with 

 their young became numerous, especially after leaving Hudson's Strait. In 

 passing Cape Farewell we saw only one or two shearwaters, probably in 

 consequence of our being too far to the southward of that head-land. A 

 very gradual increase took place in the temperature of thc' sea-water as well 

 as in that of the atmosphere as we advanced to the eastward, which changes 

 will best be shown by reference to the Meteorological Abstract for the month 

 of September. The Aurora Borealis was visible more or less almost every 

 night during our passage across the Atlantic ; it occurred generally in large 

 detached and irregular patches of yellowish light indifferently in all parts 

 of the heavens, and frequently afforded as much light as the moon in her 

 quarters. In a single instance, when the light hap|)cned to be confined to 

 one portion of the heavens, it was so vivid as to make (he shadows of objects 

 distinctly visible on the deck. On the same night, the phenomenon assumed 

 the form of a brilliant arch extending across the heavens through the zenith 

 from true east to west. It often happened also in cloudy weather, that the 

 Aurora produced the same kind of general light at niglit as the moon does 

 under similar circumstances ; the compasses were never perceptibly affected 

 by this phenomenon. 



On the 24th of September, in kit. 60" 30' and long. 01° 30', wc i)icked 

 up a piece of ycllov/ pine-tree, ten feet long from thc root and a foot in 

 diameter ; it was quite sound, not at all water-logged, and had no appear- 

 ance of being worm-eaten. On the 30tli, in hit. 57° 35', long. 39° 30', we 

 October, passed another tree of considerably larger size ; and on the 2d of October, 

 Thur.2. i,^ 1^^ 580 ^Q,^ j^j^g 3Q0 Qr^,^ observed a spar from twenty to thirty feet 

 in length. 



We were now generally favoured by strong westerly winds, and nothing 

 Tues. 7. worthy of notice occurred till the 7th, when being in lat. .59° 20', and long. 

 10° .5.5', a Six's thermometer was sent down to a depth of three hundred 

 and fifty fathoms, and indicated a temperature of 50|°, that of the surface 

 being the same, and of the air .53°. A solan goose was seen on this and 

 the {^receding day, and these birds became more numerous as we ap- 



