Spencer — Lnprobability of Land at the North Pole. 337 



Spain and Portugal the shelf is reduced to 6-20 miles. While 

 the Norwegian channel is much longer, yet it hardly forms an 

 exception to the rule, for the edge of the continental shelf, 

 even here, is within the line of the land between Scotland and 

 Norway. Even now we have sufficient data to show that the 

 edge of the lower platform at a depth of 1200 feet (like in 

 Barentz Sea) does not exceed a distance of 60 to 100 miles from 

 land off East Greenland and in Baffin Bay. From my familiarity 

 with the study of continental shelves, the margin of that beyond 

 even the lower part of Prince Patrick's Land should not be 

 expected to extend more than a hundred miles, if indeed so 

 much, with such a deep channel extending far within the line of 

 the islands. Neither the Mecham nor the McClintock expedi- 

 tions found land beyond the northern side of Prince Patrick's 

 Land. These fjords are in the western side of the archipelago. 

 Four hundred miles farther east (south of the western end of 

 North Devon) in the head of Lancaster Sound we find another 

 fjord with a depth of 780 feet deepening in 150 miles to over 

 4000 feet at a point 50 miles within the line of the islands, while 

 other soundings show the shelf to attain a breadth of only a 

 few miles in this part of Baffin Bay adjacent to the deep fjord, 

 but elsewhere it reaches to 100 miles. xAbout 160 miles north of 

 Lancaster Sound, in Cardigan Strait, on the Arctic side of 

 the col between North Devon and North Kent, the depth of 

 the fjord exceeds 2400 feet. This point is situated about 250 

 miles within the line of Sverdrup's New Land, showing that 

 here is another deep channel trending to the Arctic basin. Now 

 with even the few soundings before us (and there are many 

 more not cited here) the fjord-like character of the archipelago 

 cannot be questioned. It is a dissected plateau region (much of 

 which is now 2000 or 3000 feet high, subsequently submerged 

 in part — a region fashioned by atmospheric agents whether now 

 above or below sea level. As the depths mentioned are far within 

 the land line, and in the case of one sound a depth of over 4000 

 feet occurs even 50 miles within the land line, there is every 

 reason to expect similar depths nearer the mouths of the other 

 fjords mentioned. Where the submarine valleys have attained 

 considerable depths within the land line the shelf should be rela- 

 tively narrow ; and judging from the position of Sogne fjord 

 (nearly 4000 feet deep) I should not expect to find the 

 breadth of the shelf exceeding 50 miles in front of Sverdrup's 

 New Land, as much of the region is a weatherworn plateau. One 

 sounding north of Grant Land strengthens this view. At 

 about 30 miles off the land the sounding shows 432 feet, some- 

 what in excess of the submergence of the Arctic continental 

 shelf wherever known off the Siberian coast. It would seem 

 that the edge is being approached. Nansen has pointed 



