138 THE OCEAN. [Book V. 



division of the Sargasso weeds may be taken to indicate 

 the course of the theoretical stream ok; the central 

 part of that current making a clear way for itself by 

 throwing off the weeds on both sides into the more 

 slowly-moving portions of the rotating currents on 

 each side of it. If these floating weeds thus indicate 

 the course of the stream o r, it must then also be 

 inferred that the theoretical stream e lies west of the 

 smaller or western portion of the floating weeds ; 

 these weeds being encircled by the stream o r, and 

 the counter- streams e and h. And, in fact, numerous 

 observations show a setting of the water from the 

 border of the Gulf Stream south-eastwards towards 

 the Antilles ; though the connected course of the 

 counter- streams e and h, and the natural position of 

 the theoretical point y, appear to be interfered with 

 by the intruding under- current, which we have men- 

 tioned above as appearing to be forced to the surface 

 near the Bermudas. In this case, from the north of 

 the Antilles, the theoretical current h must continue 

 its course eastwards and south-eastwards towards 

 the equator, between the mid-ocean current and 

 the equatorial current, both running in the opposite 

 direction. 



This gives a circulation in accordance with the 

 theoretical action of vis-mertiae described in Book II. 

 in connection with Plate IX., and also in accord- 

 ance with recorded observations, excepting that the 

 course of the counter- current is not very satisfactorily 



