FIG. 2.-High Shore Dunes on the Northern Part of the Diluvial High Shore of the Island of Hiddense, near Rügen, Baltic Sea. Photograph by the author.

ported partly by the waves, partly by the inundation of the rivers, partly through the sandbanks under the immediate influence of the air, has been heaped up into dunes by the wind. Consequently water and wind are the two main factors in the creation of hydroæolian dunes; the one prepares the material, the other forwards it.

The value of the proposed genetic classification, it is suggested, is that it does away with the difficulties of the geneto-topographical system and also gives a proper place to the river dune, which hitherto has occupied only an indefinite middle ground between the seashore and the continental dune.

The relation of these two systems is shown in the following table:

SOUTHAMPTON ISLAND

On his recent return from a prolonged whaling voyage in the waters about Southampton Island, at the northern end of Hudson's Bay, Captain George Comer kindly communicated to the Society various minor changes and additions to his previous map of Southampton Island published in the Bulletin (Vol. 42, 1910, facing p. 84), which resulted from observations taken on this trip.

Map of Southampton Island with Additions and Corrections by Captain Comer to his Previous Map (see Bulletin, Vol. 42, 1910, facing p. 84). Scale, 1:4,000,000.

These changes and additions have been incorporated in the accompanying map. Where possible, they are distinguished by a heavier line. Specifically, they refer to the following features.

On Southampton Island itself the upper course of the main river emptying into the Bay of God's Mercy has been added, as well as a shorter stream emptying into the same bay and one paralleling the western shore of South Bay, which

has its source in a lake and empties into a cove lying in 63° 35′ N. The two latter abound in salmon, while the first does not. Salmon also occur in the lake-fed stream, already shown in the first edition of the map, which drains into the northern end of South Bay. Attention may here be called to the fact that some of the lakes lying near the foothills (see the original map), which contain salmon, do not always freeze over in winter. Even in the coldest weather the ice, if it forms at all, can be broken through with a stone. The lake in the northernmost peninsula of the island, marked "Salmon Pond" on the original map, should be extended much farther north, as shown on the accompanying map. A few short coastal streams, a range of hills on the western side of South Bay and a permanent snow-bank on the westernmost prong of the island, north of Cape Kendall, constitute the remaining additions made with regard to Southampton Island itself. With regard to the adjacent waters and their navigation the following additions to the original map have been made: In Ne Ultra Strait various islands and reefs have been added, mainly in 652° N. (Desdemona Island; so named by Captain Comer after a whaling bark wrecked here in the autumn of 1896, whose crew he saved), 65° N. and 644° N. (just north of Whale Point). All of the soundings shown (figures represent fathoms) are new. Changes in the shore-line were made by the addition of coves north of the mouth of Wager Inlet, on its southern shore and north of Whale Point. An island south of Beach Point has been changed to a promontory: there is a good anchorage on its southern side. A cape in 6434° N. has been given the Eskimo name of Nuvack Point. Good anchorages, which were not shown on the first map, are also to be found in the land-locked cove just mentioned, to the north of the mouth of Wager Inlet and on the western side of Duke of York Bay at the northern end of Southampton Island.

The tidal currents in Ne Ultra Strait are as follows: The flood tide enters from Frozen Strait and follows the western shore of Ne Ultra Strait to Wager Inlet (the tide rips are strong here). The strong ebb tide coming out of Wager Inlet impinges on this current and forces it over towards Southampton Island, as shown on the map, until it is forced back again to the western shore by the southwestern "nose" of the island. Therefore, in navigating Ne Ultra Strait from south to north, a ship should keep the western shore of the strait in sight up to Wager Inlet and then cross over to its eastern shore, as shown on the map.

The tide range, where observed, was: at Whale Point, on the western shore of Ne Ultra Strait, 8 feet as a minimum, 16 feet as a maximum; at Coral Harbor, the inner cove of South Bay on the south side of Southampton Island, 8 feet as a maximum.

The voyage on which Captain Comer made these observations lasted more than two years. He arrived at Cape Fullerton on the western side of the entrance to Ne Ultra Strait (shown on the inset map) on August 14, 1910, and left Repulse Bay, north of its northern end, on September 14, 1912. During the twenty-five months spent in these waters only five small whales were killed, a pronounced decline from previous catches in the same region. Compensation was afforded by barter with the Eskimos, white fox skins being the chief article of exchange.

In this connection it may be mentioned that Captain Comer's original map was reproduced from the Bulletin in Petermanns Mitteilungen in October, 1910 (Vol. 56, II, Taf. 36), accompanied by a note by H. Wichmann (pp. 191–192). Captain Comer's delineation has also been incorporated on British Admiralty charts Nos. 863 and 2177 and, for the southern part of the island, on the

standard base map of Canada on the scale of 1:2,217,600 published by the Department of the Interior, Ottawa.

Such maps as Captain Comer's, although tentative, are of great value for regions not adequately surveyed. Until superseded by systematic surveys they represent the only material available. Not their least merit is, precisely because of their preliminary nature, to call attention to the necessity for adequate surveys, all the more urgent in the region under consideration in view of the probable establishment of the Hudson Bay-Liverpool route as an outlet for the wheat crop of western Canada.

NOTES ON THE DESCRIPTION OF LAND FORMS.-X

Fault-line Scarps in Central Sweden.

EXPLANATION OF MAP OF LAND-FORMS IN THE SURROUNDINGS OF THE GREAT SWEDISH LAKES. By S. de Geer. Sveriges Geol. Undersök., Ser. Ba, No. 7, 1910. 30 pp.

ÜBER DIE BEZIEHUNG ZWISCHEN GEOLOGIE UND LANDSCHAFTSFORMEN IN MITTELSCHWEDEN. Führer einer morphologischen Exkursion. By O. Nordenskjöld und S. de Geer. XI Internat. Geologkongressen, Stockholm, 1910.

There is a district of curious form in central Sweden, which is roughly represented in the foreground section of the accompanying diagram (looking northwest), taken from the writer's "Erklärende Beschreibung der Landformen." The district may be empirically described as possessing a central upland of crystalline rocks, having an equable altitude and being divided into irregular lozenge-shaped masses (of too uniform pattern in the diagram) by a series of narrow, criss-cross valleys, and being bordered on the north and south by broad depressions holding lakes, to which the uplands descend by rather abrupt escarpments of more or less zigzag pattern. The lacustrine areas possess here and there small patches of nearly horizontal limestones, which lie on an even floor of crystalline rocks, and this floor rises slowly north and south from the depressions, and thus again forms uplands of lozenge-shaped masses divided by narrow criss-cross valleys.

If the same district were described in explanatory fashion, the description would vary according to the explanation adopted, but all explanations agree that the district has been glaciated. If we follow the Swedish authors above cited, or