388 THE PHYSICAL GEOGRAPHY OF THE SEA. 



and after the sources of the current which brought it down have 

 been bound in fetters of ice, it pauses in the long nights of the 

 northern winter, and scarcely commences its return till the sun 

 recrosses the equator, and increases its power as well in intensity 

 as in duration. Thus, in studying the physical geography of the 

 sea, we must take cognizance of its actinometry also, for here we 

 have the effects of night and day, of clouds and sunshine, upon 

 its currents and its climates, beautifully developed. These effects 

 are modified by the operations of certain powerful agents which 

 i-eside upon the land ; nevertheless, feeble though those of the 

 Ibrmer class may be, a close study of this plate will indicate that 

 they surely exist. 



725. Now, returning toward the south : we may, on the other 

 A belt of uniform hand, iufcr that the mean atmospherical tempera- 

 temperature at sea. ^^j,g £qj, ^YiQ parallels between which the isotherm 



of 80° fluctuates is below 80°, at least for the nine months of its 

 slow motion. This vibratory motion suggests the idea that there 

 is probably, somewhere between the isotherm of 80° in August 

 and the isotherm of 60° in January, a line or belt of invariable or 

 nearly invariable temperature, which extends on the surface of 

 the ocean from one side of the Atlantic to the other. This line 

 or band may have its cycles also, but they are probably of a long 

 and uncertain period. 



726. The fact has been pretty clearly established by the dis- 

 The western half of covcrics to which thc wiud and current charts have 



the Atlantic warmer -,, , i -in n ^ at •/-\ 



than the eastern, led, that the wcstcm hall 01 the Atlantic Ocean is 

 heated up, not by the Gulf Stream alone, as is generally supposed, 

 but by the great equatorial caldron to the west of longitude 85°, 

 and to the north of Cape St. Roque, in Brazil. The lowest reach 

 of the 80° isotherm for September — if we except the remarkable 

 equatorial flexure (Plate IV.) which actually extends from 40° 

 north to the line — to the west of the meridian of Cape St. Roque, 

 is above its highest reach to the east of that meridian. And, now 

 that we have the fact, how obvious, how beautiful, and striking is 

 the cause ! Cape St. Roque is in 5° 80' south. ISTow study the 

 configuration of the Southern American Continent from this cape 

 to the Windward Islands of the West Indies, and take into ac- 

 count also certain physical conditions of these regions: the Ama- 

 zon, always at a high temperature because it runs from west to 



