268 THE ATLANTIC. [chap. v. 



Other two, for there is high and extensive land between tlie me- 

 ridians of 55° and 65° west, in 65° south latitude ; and the warm 

 current, already led far to the southward by the American coast, 

 appears to bifurcate upon Graham Land, and to produce another 

 bight in 90° west longitude, a little to the west of the southern 

 point of Soutli America. In this bight, Cook, in 1771, and Bel- 

 lingshausen, in 1821, pushed nearly to the seventieth parallel of 

 south latitude. 



I have already referred (vol. ii., p. 75) to the principal tem- 

 perature phenomenon of the eastern portion of the Soutli At- 

 lantic — the equatorial counter-current, and its extension as the 

 Guinea Current, The cause of the counter-current to the east- 

 ward in the zone of calms is somewhat obscure, as the only ob- 

 vious explanation — that it is a current in an opposite direction 

 induced in the space between the current of the north-east and 

 south-east trades to supply the water removed by them — seems 

 scarcely sufficient to account for its volume and permanence. 



The comparative thinness of the belt of warm surface-water 

 in the equatorial region is at first sight remarkable, and has 

 given rise to a good deal of speculation ; but it will be seen 

 by comparing the distribution of temperature at Station CXII. 

 (Fig. 54), nearly on the line, with that at Station CCCXXYIL, 

 (Fig. 56), in the latitude of Tristan d'Acunha, that the positions 

 of the isothermobaths of 4° and 5° C. are nearly the same : the 

 slight difference apparently depends upon the latter station be- 

 ing within the influence of the Brazil Current. The phenome- 

 non is thus essentially a continuation to the north of the equa- 

 tor of southern conditions, and the small effect of the vertical 

 sun in raising the temperature to any depth below the surface 

 is doubtless due to the removal of the heated layer as soon as it 

 is formed by the trade-winds and their counter-currents, and to 

 the rapid abstraction of heat in the formation of watery vapor. 



One of the best-marked and most important phenomena of 

 the distribution of temperature in the upper layers of the 



