THE GULF STREAM. 255 



trade winds in originating the oceanic circidation of the Atlan- 

 tic. That winds blowing steaddy from one quarter give rise to 

 powerful currents is well known, and it is not difficult to imagine 

 the prominent part the trades must play in setting in motion, 

 in a southwesterly and a northwesterly direction, the mass of 

 water over which they SAveep so persistently on each side of the 

 equator. 



The chano-e of currents in the Indian Ocean due to the shift- 

 ing of the monsoons is well known. How far below the surface 

 this action of the winds reaches, is another question.^ Theo- 

 retically it has been calculated by Zoeppritz that one hundred 

 thousand years is ample time to allow the friction of the parti- 

 cles to extend from the surface to the bottom, say to two thou- 

 sand fathoms, were the winds to blow without intermission in 

 one direction during that time, with the average power they are 

 known to possess.' 



We may imagine the whole of the mass of the Atlantic within 

 the belt of the tradewinds to be moving in a westerly direction, 

 and impinging upon the continental slope of South America ^ 

 and upon the Windward Islands ; at which point it is deflected 

 either in a southerly or northerly direction, or forces its way 

 into the Caribbean. In our present state of knowledge it is 

 difficult to trace the path of the equatorial water as it is forced 

 into the eastern Caribbean. Commander Bartlett supposes that 

 it is warmed in the Caribbean by circulating round the whole 

 basin. The water which is swept into the Caribbean by the 

 tradewinds through the passages between the Windward Islands, 

 and being then driven into the Old Bahama Channel funnel, 



^ The movement arising from the ac- the conditions producing them (acting 



tion of the winds on the surface is trans- from the surface) have ceased to be effec- 



mitted by friction from one layer to an- tive by any break of continuity due to the 



other, and communicates tlie velocity of interposition of islands or of banks in 



the upper particles to the underlying the track of oceanic currents, 



layers in succession. If this is continued ^ Did the Gidf Stream not meet conti- 



long enough, the velocity of the lowest nental masses, it would simply expand 



layers wUl equal within a fraction that of north and south, losing its initial velocity, 



the upper layer. and gradually cool down towards the 



^ It is therefore possible that currents, poles ; the cold penetrating all the deeper 



which owe their existence to causes that portions of the ocean, just as we find it 



have been modified to a certain extent, reaching the higher summits that rise 



should still exist in the ocean long after above the line of perpetual snow. 



