UNDISCOVERED LAND IN THE ARCTIC OCEAN 59 



this island by a broad and presumably shallow strait can be inferred with 

 considerable certainty from a comparison between the drifting of the 

 "Jeannette,' especially the later portion, and that of the "Fram." The 

 "Jeannette" was frozen in at a point lying northeasterly from Wrangel 

 Island and her geneial subsequent course lay approximately toward Cape 

 Chelyuskin; the "Fram" was frozen in at a point to the northwestward of 

 the New Siberian Islands, and her general subsequent course lay toward 

 the Greenland Sea. 



The change in the direction of the drifting ice seems to occur in about 

 the longitude reached by the "Jeannette" when she sank. This indicates 

 that one corner of the unknown land lies not very far to the northward of 

 this point and so probably extends westward to about the meridian 150° E. 

 With this strait and land mass assumed, it is easy to see that the semidaily 

 tide transmitted through this strait will be greatly reduced in range upon 

 entering the broad expanse of water to the eastward through a portion 

 of which the "Jeannette" drifted. Hence the small range of the semidaily 

 tides at Pitlekaj, Siberia, and at Point Barrow and other places along the 

 northern and northwestern coast of Alaska. The strait and land mass can 

 also account for the observed fact that the semidaily flood stream at Point 

 Barrow comes from the west and not from the north as the absence of the 

 land mass would imply, especially if the soundings or known depths around 

 Point Barrow be taken into consideration. 



An obstruction of the kind already established, probably implies that 

 the land mass extends nearly continuously from the Bennett Island corner- 

 to within a short distance of the Arctic Archipelago. 



The fact that the tide comes from the west at Point Barrow indicates 

 the existence of a cape or corner of the unknown land lying to the northward 

 of this point. But the fact that the tide is nearly simultaneous all along 

 the northern coast of Alaska implies, not only that Beaufort Sea is a deep 

 body of water, but also that this corner in question lies at a considerable 

 distance from Point Barrow, say five or six degrees of latitude. Such a 

 position of the cape or corner would permit the ice off the northern coast 

 of Alaska to set west -northwest when driven by an easterly or east-north- 

 easterly wind, as was noted on numerous occasions by Mikkelsen and 

 Leffingwell in April and May, 1907, in longitudes varying from 148° to 

 151° W. 



That there is a northern coast to Beaufort Sea in some such position as 

 that shown in the diagram and extending from north of Point Barrow 

 nearly to Banks Island, can be inferred from the following considerations: 



1. The ice in Beaufort Sea does not drift freely to the northward, and 

 is remarkable for its thickness and age. Northeasterly winds drive the 

 ice westward. West of Banks Island large and old ice floes probably 

 always extend down to the 72d parallel. 



2. The observations just referred to indicate not only a considerable 



