HYDEOGEAPHY. 19 



connecting with the 2,000-fathom basin to the northeast of the Commander Islands, 

 or are these islands situated on a more or less elevated ridge extending from Cape 

 Kamchatka to Attn? 



The latter alternative seems for the present the more probable, though the depth 

 of the ridge between Cape Kamchatka and Northwest Cape of Bering Island can of 

 necessity only be vaguely guessed at. I have also selected the alternative of the 

 3,000-fathom channel parallel with the Kamchatkan coast and introduced these 

 features in the map (pi. 87), bat I wish to reiterate what I said with regard to the 

 first edition of it, viz, that — 



The curves of the various depths from 100 fathoms down to 2,000 fathoms and 

 over are, as a matter of necessity, highly conjectural. In the northeastern section of 

 the map they appear even somewhat problematical, in view of the fact that a series 

 of shallow soundings, running southwest from Cape Olintorski on the charts of the 

 United States Hydrographic Office, have been left out of consideration altogether. 

 The reason is, that the series is crossed by the deep soundings of the Albatross on her 

 return passage from the Commander Islands in 1895 in such a manner that it is 

 impossible to reconcile them. They may possibly belong farther west' — a not 

 unreasonable supposition, since the determination of the longitude of the various 

 coasts and promontories in that part of the world is in such utter confusion^ that a 

 resurvey of the whole coast from Petropaulski to Providence or Plover Bay is 

 imperatively demanded.' 



In all this uncertainty only a few points can be regarded as fairly well 

 established, viz: , 



(1) The Commander Islands are situated upon a small and narrow plateau, which 

 may possibly be connected with the Kamchatka mainland to the northwest by a ridge 

 of greater or less elevation. 



(2) This plateau rises very abruptly from an ocean floor between 2,000 to 3,000 

 fathoms, so that the islands themselves on nearly all sides rise almost perpendicularly 

 out of this depth. 



(3) Between the Commander Islands and Attu, the nearest of the American 

 Aleutian Islands, there is a gap certainly more than 1,900 fathoms deep. Whether 

 the Albatross maximum sounding of 1,996 fathoms, only a short distance from the 

 south end of Copper Island, is really the maximum depth, thus indicating a slightly 

 elevated ridge between the floor of the Bering Sea and the so-called Tuscarora deep, 

 or whether there may not be a channel of 2,100 fathoms, or thereabouts, on one side 

 of the sounding in question, remains to be seen; 



1 1 find on Berghaus's "Chart of the World on Mercator's Projection" a, sounding of 2,700 fathoms 

 indicated in (approx.) latitude 56° 40' north and longitude 168° 20' east, the authority for which I am 

 ignorant of. It is situated almost in a line between the 1895 Albatross soundings of 2,137 and 1,866 

 fathoms, and if correct would indicate a depression below the general level of about 2,100 fathom's in 

 that part of Bering Sea. 



' Witness the fact that the various charts of the region for more than ten years have borne the 

 following inscription: "The coast of Kamchatka north of Cape Koslof is reported to be charted 15 

 miles too far east." Yet nothing has been done to clear up the doubt. 



3 This conjecture has in a measure been confirmed by a statement of Capt. E. N. Crowell, ma8);er 

 of the British sealing schooner Brenda (Venning's Eeport, 1893, p. 90), to the effect that the only 

 "bank" he knew of in the vicinity, of the Commander Islands is north of Bering Islandy ofi: Cape 

 Olintorski, the "center of the 'bank' being in about latitude 58° north, longitude 170° east. 

 Soundings can be had from 40 fathoms up." 



