Temperature of the Globe. 343 



Pole nearly as much as America and Asia do ; and that, on the 

 contrary, it lies opposite the greatest extent of sea- water, free from 

 ice, which is known in the whole polar zone. The coldest points 

 of the earth, which have lately l)een improperly called Poles of 

 Cold, do not coincide with the magnetic poles, as Dr Brewster has 

 endeavoured to prove in the English version of my paper on the 

 Isothermal Lines. According to Captain Sabine's researches, 

 the minimum of the annual mean temperature on the surface of 

 the earth, is to the NW. of Melville's Island, in the meridian of 

 Behring's Straits, probably in 8^ to 83° north Lat. The sum- 

 mer boundary of the ice, which, between Spitzl)ergen and East 

 Greenland, recedes to 80° and 81° north Lat., is in about 75° 

 N. Lat., every where between Nova Zembla, the Bone Islands of 

 New Sil)eria and Icey Cape, the most western cape of America. 

 Even the winter boundary of ice, the line on which the ice ap- 

 proaches the nearest to our continent, scarcely surrounds Bear 

 Island. From the North Cape, which is heated by a south- 

 western current of the sea, the navigation to the most southern 

 promontory of Spitzbergen is never interrupted, not even in the 

 most severe winters. The polar ice diminishes in quantity 

 wherever it finds an opening to flow out, as in Baffin's Bay, and 

 between Iceland and Spitzbergen. The situation of the Atlan- 

 tic Ocean exerts a most beneficial influence on the existence of 

 that sea-water, free from ice, in the meridian of East Green- 

 land and Spitzbergen, which has so important an influence upon 

 the climate of the north of Europe. 



On the other hand, the icebergs, which are driven from Baf- 

 fin's Bay and Barrow's Straits to the south, accumulate in that 

 large mediterranean sea, which geographers designate by the 

 name of Hudson's Bay. This accumulation of ice increases the 

 cold of the neighlK)uring continent so much, that, as reported 

 by Captain Franklin in his latest MS., in York Factory, and 

 at the mouth of Hayes River, which lie in the same latitudes 

 as the north of Prussia and Courland, in digging wells, ice is 

 found everywhere at the depth of four feet. The most northern 

 and mc^t southern boundaries of the fixed polar ice, that is, the 

 summer and winter boundaries, on the situation of which the 

 temperature of the northern continents depends, seem to have 

 changed but little, as far as historical records go ; which fact 



z2 



