ANTARCTIC OCEANOGRAPHY 277 



Ekman, which flowed parallel to the coast and to the edge of the shelf 

 ice and affected nearly the whole thickness of the polar water. The 

 dependence of both the drift and the current on the direction of the 

 wind has been demonstrated conclusively by W. Brennecke^^ for 

 Weddell Sea also. J. M. Wordie^^ also emphasizes their dependence 

 on the winds and the direction of the coast line and makes this de- 

 duction from a comparison of the drifts of the Belgica, Deutschland, 

 Endurance, and Aurora, the first in the South Pacific, the second and 

 third in Weddell Sea, and the last in Ross Sea. Thus we now know 

 that, below the superficial drifts, there is a strong westward-flowing 

 current system completely encircling the Antarctic Continent, a 

 system which owes its existence to the predominant winds and flows 

 parallel to the coast and to the inland-ice and shelf-ice walls. Its 

 thickness is variously estimated. Where I observed it it was 100 meters 

 to over 300 meters. This current system under the influence of the 

 earth's rotation tends toward the left, i. e. toward the coast, and it 

 holds the drift ice together. Over it lie those purely surface and 

 fluctuating currents due to variable winds which carry the drift ice 

 hither and thither and project irregularly but not far into the west- 

 wind drift. 



The Cold Bottom Water and Its Origin 



Under the intermediate, or tropical, water lies the cold bottom 

 water which, as has long been known, is characterized by great uni- 

 formity and occupies the bottom of all the oceans to beyond the 

 equator. Inasmuch as the Arctic Sea is segregated from the oceans, 

 whereas the Antarctic waters everywhere connect with them, as fur- 

 thermore the cold bottom water slowly increases in warmth towards 

 the north, and as a number of deep-sea basins which are closed to the 

 south seem to be cut off thermically from this bottom water, it has long 

 been assumed that this bottom water comes from the Antarctic. The 

 Antarctic expeditions of the last three decades have confirmed this 

 view. It was in addition possible both in the Gauss region and in 

 Weddell Sea to show that the bottom, water arises through mixture of 

 the shelf-sea water, or polar water, and the intermediate, or tropical, 

 water. The former is brought to the latter by the above-mentioned 

 flow (Fig, i); where they meet, undulation and mixing take place 

 which create the bottom water. This is a mixture because its tem- 

 perature lies between those of the two constituent waters. Its mass, 

 however, is primarily derived from the tropical water because this is 

 quantitatively greater at the shelf edge than the polar water and 

 because its salinity is exactly or almost exactly the same as that of 



16 Brennecke, op. cit., PI. 15. 



" J. M. Wordie: The Ross Sea Drift of the "Aurora" in 1915-1916, Geogr. Journ., Vol. 58, 1921, 

 pp. 219-224; reference on pp. 223-224. 



