September 9, 1904.] 



SCIENCE. 



339 



continuation of the gr.eat Gulf Stream of 

 Japan or Kuro Siwo. To tliat solution of 

 the problem there are two objections which 

 appear insuperable. 



One is that the Japan stream must nec- 

 essarily pass out and dissipate itself in the 

 vast breadth of the Pacific Ocean long 

 before reaching Alaska. Even if enough of 

 the Kuro Siwo could hold together to form 

 a noticeable current on the Alaskan coast, 

 it could be only a wasted and meager 

 stream, wholly unlike the mighty current 

 that skirts the continent. 



The other objection is the extreme cold- 

 ness of the American coast-current. This 

 could not possibly proceed from water of 

 such high temperature as the Kuro Siwo, 

 if the latter had held together as a compact 

 stream. So compacted, it must have re- 

 tained most of its warmth, even in travers- 

 ing the higher latitudes. It has been 

 alleged that it has been chilled by passing 

 the vicinity of the great Alaskan glaciers. 

 But the Muir glacier, the Malaspina and 

 all combined are of comparatively limited 

 extent. They send no bergs out to sea; 

 they are totally inadequate materially to 

 cool the adjacent ocean. 



What then is the real source of the Pa- 

 cific coast cold current? Leaving for a 

 little this problem, turn we now to a second 

 one, which may help us to find the answer 

 to the first. The depths of the Pacific 

 Ocean are occupied throughout by an im- 

 mense stratum of ice-cold water at 35°, 

 which is slowly creeping northward. The 

 existence of this cold stratum is well estab- 

 lished by all deep-sea temperature sound- 

 ings throughovit the Pacific. That it is 

 creeping northward is inferred from an 

 analogous condition under the Atlantic 

 Ocean. It is a fact familiar to hydrog- 

 raphers that a similar stratum of cold 

 water underlies the Atlantic at a depth of 

 2,000 fathoms. Near the equator, however, 

 the 35° stratum is found to rise to within a 



few hundred fathoms of the surface. The 

 necessary inference is that two opposite 

 currents are meeting and pushing each 

 other upwards, the one derived from the 

 Arctic ice, the other from the Antarctic. 

 Nearing the surface they gather warmth 

 and commingle with the warm equatorial 

 current running westward. 



But in the Pacific Ocean, while the 

 depths are at the same 35° as in the Atlan- 

 tic, there is no rising of the cold water at 

 the equator. The reason is evident. The 

 supply of cold is confined AvhoUy to the 

 Antarctic ice. The Pacific is closed at the 

 north, and the Arctic ice exerts no in- 

 fluence on its waters. There is no northern 

 deep-sea current to meet and push up the 

 southern one at the equator, and the latter 

 creeps on unobstructed to the northern 

 bound of the ocean. 



This southern current is evidently of 

 great magnitude. It proceeds from a Pa- 

 cific frontage of the vast Antarctic conti- 

 nental glacier extending 4,000 miles in 

 length along the Antarctic Circle. The 

 ocean waters from the north encountering 

 this massive ice-front and its bergs are 

 chilled and sink to the bottom. Thence 

 they creep north in a wide and deep 

 stratum underlying the whole breadth of 

 the Pacific. Reinforced and pushed on by 

 constantly chilled supplies behind, they 

 press on up the whole 8,000 miles' length 

 of the ocean, encountering no adverse cur- 

 rent from the Arctic Circle. 



Here comes our second problem. Where 

 does this vast creep of cold water find out- 

 let 1 Somewhere it must escape to the sur- 

 face. Where does it emerge to unite with 

 the upper ocean? Before indicating the 

 answer to this second problem, we may note 

 two conditions encountered by this vast 

 creep of water, which may assist in guiding 

 ITS to the answer. 



The first condition is that after passing 

 the Tropic of Cancer, the Pacific Ocean 



