On the general Causes of the Ocean-Currents. 13 



in' comparison with the decrease, produced by the diminution of the 

 waterpiUar's height. Now as evaporation takes place with very different 

 rapidity in different locaHties and attains its maximum in the regions, 

 where the rainfall is least, viz: the region of the trade-winds, this cir- 

 cumstance must necessarily have a considerable influence on the oceanic 

 circulation. 



By evaporation the equilibrium of the water-particles is disturbed 

 down to the bottom, and accordingly the motion, which takes place in 

 order to fill up the vacuum, may extend to the whole mass, although 

 the velocity will be different at different depths. If we suppose a cup- 

 shaped hollow to be formed by evaporation in the surface of a mass of 

 water, the difference of pressure in the level, that passes through the 

 centre of the hollow, will be greater than in any level above it, and the 

 motion of the water-particles in that plane will therefore be more rapid 

 than in any plane above it. But in all lower levels the difference of 

 pressure is the same as in that level, and it would consequently be un- 

 reasonable to suppose that the hollow would be filled only by superficial 

 streams of its own depth. Motion must also arise in the lower strata, 

 and this motion under the centre of the hollow must be vertically upward. 

 It will accordingly depend upon the relative dimensions of the water- 

 mass and the hollow, and on the inclinations of the latter's surface, 

 whether the filling of the hollow be owing more to the afflux of the 

 water from the sides or the rise of that beneath. 



The depression of the ocean's surface through evaporation is ex- 

 tremely small in comparison to the extent of the sunken surface, and 

 the volume of the hollow is extremely small in comparison with that of 

 the ocean. The motion therefore of the approaching particles under the 

 surface of the hollow will be chiefly upward, but will gradually approach 

 to horizontal, as the distance from the locality of the hollow increases. 

 Thus the great evaporation within the torrid zone ought to produce two 

 slow, deep motions in the sea, one in each hemisphere, which, origina- 

 ting in the higher latitudes, meet in the neighbourhood of the equator. 

 As evaporation is greater in the regions near the equator than in those 

 more distant — if we except the narrow equatorial belt that includes 

 the region of the calms and rain — the quantity of water, which in a 

 unit of time must flow to the former region, will be relatively greater than 

 that, which moves towards the latter; but the former must pass through 

 a greater distance reckoned from the origin of the stream. In conse- 

 quence of these causes the under-streams, caused by evaporation, ought to 



