July 26, 1895.] 



SCIENCE. 



89 



position of the moon ; and the influence of 

 the pressure of the atmosphere over large 

 conterminous areas of higli and low barom- 

 eter, in accelerating or retarding the cur- 

 rent. He also determined the presence of 

 a fringe of current on the east of the Gulf 

 Stream proper, proceeding at a slower rate 

 than it, and conformably in direction. 

 These and other points make a very inter- 

 esting series of discoveries, but they do not 

 touch the causes of the Gulf Stream. They 

 relate solely to the agencies which modify 

 it. Were all these conditions absent, the 

 Gulf Stream, merely as a bodj^ of water 

 flowing in the direction which it now pur- 

 sues, would still exist substantially as at 

 present. 



In the particular not relating to influ- 

 ences merely modifying the stream, but to 

 the question how such portion of its waters 

 as have passed into the Carribeau Sea and 

 the Gulf of Mexico, under the influence 

 of the trade winds, have been translated 

 thither, the present writer cannot agree 

 with Lieutenant Pillsbury. For, whereas 

 Lieutenant Pillsbury's observations at the 

 Windward Islands show, as their result, that 

 the water there entering the Caribbean Sea, 

 as a current, represents only about half of 

 the volume which he determined as flowing- 

 out of the Straits of Florida, as the Gulf 

 Stream, he accounts for the other half bj^ 

 supposing that, in addition to the flow into 

 ' the Caribbean Sea at the Windward Islands, 

 a great amount is also delivered there by 

 the toppling of the seas towards the west as 

 they roll on towards and through the fias- 

 sages between the islands. That any force 

 of wind, however great or small, could be 

 resolved, on the one hand into a current, 

 and on the other into waves, and the vol- 

 ume of the latter be available in helping to 

 fill an interior basin through openings of an 

 archipelago, does not agree with dynamic 

 laws known and admitted. Toppling seas 

 proceeding in any given direction resolve 



themselves into and quicken current; in 

 fact, current in water, as derived from wind, 

 is conditioned upon the rufiling of the sur- 

 face, whether in the form of a ripple or a 

 billow. Therefore current and wave are 

 not dynamically separable as agencies of 

 transportation. It follows that, if the 

 Windward Islands' observations show only 

 about half the volume of water delivered 

 into the Atlantic by tlie Gulf Stream, the 

 remainder is chieflj' to be accounted for by 

 a volume of the polar current u^uderrunning, 

 and running by the sides of the Straits of 

 Florida. Phis the other supplies already 

 mentioned, and minus evaporation, we have 

 then the sum-total volume of water with 

 which nature is dealing in the production 

 of the stream. 



The primary causes of the Gulf Stream 

 are the rotation of the earth, the difference 

 of densit}^ between its polar and equatorial 

 water, and the presence of the American 

 continent. For, if we will eliminate in 

 imagination every agency but these, includ- 

 ing the trade winds, we shall shall see that 

 we have, through the facts of the form of 

 the earth and this difference of density, the 

 resultant of two lines of force represented 

 by the tendency in the direction of waters 

 so constituted to flow. On one of these the 

 centrifugal force of the earth's rotation 

 draws the waters as a sub-marine flow from 

 the poles to the equator, resulting in a su- 

 pra-marine flow from the equator to the 

 poles; while, on the other, the rotation of 

 the earth on its axis, at right angles to 

 those directions, tends to make the waters 

 move directly towards the east. If there 

 were no continents at all on the surface of 

 the earth, the effect of this, as viewed from 

 the equator, would be to make the sub-ma- 

 rine flow from the pole, in the northern 

 hemisphere, assume a southwesterly curve 

 convex towards the equator; and to make 

 the sux^ra-marine flow, from the equator to 

 the same j)ole, assume a northeasterly 



