"If we examine the ocean soundings, we at once find a conclusive reason why the main outlet must lie between Spitzbergen and Greenland. The sea here, so far as we know it, is at all points very deep; there is, indeed, a channel of as much as 2500 fathoms depth; while south of Spitzbergen and Franz Josef Land it is remarkably shallow--not more than 160 fathoms. As has been stated, a current passes northward through Bering Strait and Smith Sound, and the sounds between the islands north of America, though here, indeed, there is a southward current, are far too small and narrow to form adequate outlets for the mass of water of which we are speaking. There is, therefore, no other assumption left than that this mass of water must find its outlet by the route actually followed by the polar current. The channel discovered by the Jeannette expedition between Wrangel Land and the New Siberian Islands may here be mentioned as a notable fact. It extended in a northerly direction, and was at some points more than 80 fathoms deep, while at the sides the soundings ran only to 40 or 50 fathoms. It is by no means impossible that this channel may be a continuation of the channel between Spitzbergen and Greenland, [6] in which case it would certainly influence, if not actually determine, the direction of the main current. "If we examine the conditions of wind and atmospheric pressure over the Polar Sea, as far as they are known, it would appear that they must tend to produce a current across the Pole in the direction indicated. From the Atlantic to the south of Spitzbergen and Franz Josef Land a belt of low atmospheric pressure (minimum belt) extends into the Siberian Arctic Sea. In accordance with well-known laws, the wind must have a preponderating direction from west to east on the south side of this belt, and this would promote an eastward-flowing current along the north coast of Siberia, such as has been found to exist there. [7] The winds on the north side of the minimum belt must, however, blow mainly in a direction from east to west, and will consequently produce a westerly current, passing across the Pole towards the Greenland Sea, exactly as we have seen to be the case. "It thus appears that, from whatever side we consider this question, even apart from the specially cogent evidences above cited, we cannot escape the conclusion that a current passes across or very near to the Pole into the sea between Greenland and Spitzbergen. "This being so, it seems to me that the plain thing for us to do is to make our way into the current on that side of the Pole where it flows northward, and by its help to penetrate into those regions which all who have hitherto worked against it have sought in vain to reach.