THE GULF STEEAM. 381 



that, theoretically, the stream should rush through the openings between 

 the islands with redoubled velocity, but no such phenomenon occurs. 

 The Currents here are constant, but not more violent than in the Ocean 

 to the East. Moreover, it is shown on pages 358 to 368 (327) to (343), 

 that the current has not sufficient power in the Caribbean Sea to maintain 

 a persistent Westwardly course, but that it is liable to frequent interrup- 

 tions, and is sometimes even reversed. 



(366.) That part of the stream which fails to enter the passages is now 

 considered to be deflected to the North- Westward, and to join the Gulf 

 Stream after it has issued from the Straits of Florida. It is, however, but 

 ill-defined and of small velocity, and from the evidence of floating bottles 

 and other observations it has been considered that part of this drift forms 

 a recurving current to the Southward and S.E., along the N.E. side of the 

 Bahamas. So indefinite is this drift, that in the previous edition of this 

 work it was stated that one half of this great drift therefore appears to die 

 out on reaching the obstacle formed by the islands, and no clue has as yet 

 been given to its extinction, but it shows one thing, that the movement 

 must be very superficial, or it would be impossible for a great body of water 

 to be thus quietly arrested.* The only outlet it appears to have is the 

 comparatively puny Gulf Stream with a breadth of not more than one six- 

 hundredth part of the parent source. 



Lieutenant Pillsbury remarks that it seems evident there are two con- 

 tributions to be found which tend to force the Stream onward, after it has 

 left the Straits of Florida. One the current outside the islands, and the 

 other the prevailing Westerly direction of the Anti-Trades. We have 

 direct evidence of the flow outside the islands in the observations taken 

 between Great Abaco Island and Cape Hatteras. At anchor about 60 miles 

 from the Bahamas, the currents were all to the Northward and Westward, 

 at the surface N.W., and at a depth of 200 fathoms N.W. by N. Midway 

 between it was found that a slow movement reached to a great depth ; the 

 directions here were nearly all between N.N.E. and N.E. by E. 



The meeting of the current outside the Bahamas with the Gulf Stream 

 varies in poeition. We have no means of knowing the width of this out- 

 side stream, except by temperature observations, and these give over 200 

 miles. It is slower in movement, but of much higher temperature than the 

 Gulf Stream. 



We have previously given (325), pages 355 — 356, Sir Eobert Schom- 

 burgk's remarks on this outer current. 



Having thus endeavoured to trace the origin of the water of the Gulf 

 Stream, we now proceed to give some account of its peculiarities in the 

 various regions it traverses, premising, however, that in no sense can it be 

 considered as constant, and it is fairly open to doubt whether Lieutenant 

 Pillsbury, TJ.S.N., is to be considered as infallible when he affirms : — 

 " That this wonderful Stream is governed by law in all its motions there 



• There is an exactly analogous case in the abrupt termination to the Guinea Cur- 

 rent in the Bight of Biafra. It is traced up to the coast itself, and with a mean rate 

 of 15 or 16 miles per day, and then disappears; it does not apparently unite with the 

 South African Current setting to the Northward. 



