476 Bibliographical Notice. 



see that only the smaller portion can be thus accounted for. The whole 

 area of Greenland is estimated at 16,000 Danish square miles (one 

 Danish square mile equalling about 23 English square miles), of 

 which 10,000 are supposed to lie on the western slope of the central 

 chain of mountains. If, now, a line is drawn along the heads of 

 all the great fjords or bays, this will cut off the islands and penin- 

 sulas (altogether about 1600 Danish square miles), leaving more 

 than 8000 square miles as the extent of inland ice west of the sup- 

 posed central chain. The number of ice-fjords, or bays in which the 

 inland ice reaches the sea and breaks up into icebergs, .does not ex- 

 ceed fifteen, of which many evidently correspond to inland valleys 

 of but moderate extent, whilst the large ice-fjords must correspond 

 to valleys so extensive that, in southern latitudes, they would feed 

 rivers as large as the Thames. Assuming that, in Greenland as 

 elsewhere, only one-third of the rain and snow would be carried into 

 the sea by the rivers, and assuming also the rain (snow)-fall to be, 

 at an average, all over Greenland only one-third of what it is in 

 Julianehaab (that is, about 12 inches), there would nevertheless be 

 water enough left to form considerable rivers. But, however difficult 

 it may be to calculate accurately the quantity of ice floating out of a 

 large ice-fjord in the course of the year, the calculation may be car- 

 ried far enough to show that it cannot represent more than a frac- 

 tion of the water which, somehow or other, must be carried into the 

 sea from the interior. From this consideration Mr. Rink concluded, 

 long ago, that there must be large rivers at the bottom of these val- 

 leys, draining the inland ice which fills up the valleys, hides these 

 rivers, and gives to the interior of Greenland the false aspect of a 

 plateau. These hidden rivers of course terminate at the true heads 

 of the fjords, which, however, cannot be observed closely, because the 

 glaciers or outrunners of the inland ice reach so far into the fjords 

 that theirfronts,though only about 100 feet above thewater, have never- 

 theless often 1000 feet real height, the remainder being concealed in the 

 deep water in which the head of the glacier is suspended, gradually 

 advancing, and breaking up. If this supposition is true, the fresh 

 water must be expected to rise in the fjord in front of the glaciers ; 

 and this is actually the case. At a little distance from the extremity 

 of the glaciers, there is invariably a place where the water is very 

 much disturbed, bubbling and whirling ; and clouds of gulls are 

 continually hovering over these places and diving for prey. The 

 Greenlanders call these places " springs ;" and that their true 

 nature is as supposed by Mr. Rink seems to be shown by the exist- 

 ence, at the Godthaabs Fjord, of a freshwater lake, an English mile 

 in diameter, at some distance inside the edge of the glacier, which 

 periodically, though not regularly, fills and empties itself. "When 

 it is filling, the nearest springs in the fjord are comparatively inactive, 

 but when it empties itself (generally very suddenly) the movement in 

 the sea is so much increased that it is perceptible as far down the 

 fjord as 13 English miles from the glacier. 



The same phenomenon of a lake periodically emptied through 

 canals in the ice has been observed at Sermiliarsuk and Nasarlik, 



