8 F. L. Ekman, 



area to move through, tlie diftereuces of pressure, which drove the sur- 

 face-stream, would require to be about equal to those, which drove the 

 under-current, or the position of the warm stratum would be about in the 

 middle between the extremes above mentioned. But, as the case really is in 

 the ocean, the section of the surface-stream is far less than that of the 

 under-stream, and accordiug'ly the differences of pressure, which drive 

 the superficial current, must be very considerably greater than those, 

 which set in motion the under-current. Or in other words: The position 

 of the warm stratum must be much nearer to that, which its specific gravity 

 trould assign to it, than to the other extreme. Here no allowance has been 

 made for the change in the position of the strata, which may arise from 

 the vis viva of the streams produced. 



If the water at the surface of the ocean is cooled to a lower 

 temperature than the subjacent strata, the water at the surface becomes 

 heavier, and supposing the saltness of the lower strata not to be greater, 

 it must, like every other heavier body, sink towards the bottom. If the 

 cooling over the whole locality be uniform and constant, there will re- 

 sult a vertical circulation, which will continue till all the water has rea- 

 ched the same temperature. If on the other hand only a portion of the 

 surface is exposed to cooling influence from above, the level of that 

 portion will sink below that of the rest and there will result an external 

 stream towards ■ the chilled place, arising from the greater hydrostatic 

 pressure of the surrounding warmer water. But in this case there arises 

 at the same time a new current, independent of the change of level; for 

 the chilled water, in consequence of its increased specific gravity, sinks, 

 displaces the water which previousl}^ formed the bottom-strata and diffu- 

 ses itself under them. These motions, if the water in the unchilled 

 locality were unable to receive heat, would continue, though with less 

 and less velocity, untill the whole mass of water had reached the tem- 

 perature of the chilled place. 



The changes of volume, which the superficial water undergoes 

 from raised or lowered temperature are however accompanied by the 

 following phœnomena. When the superficial water is anywhere warmed, 

 its dilatation in horizontal dimensions experiences a resistance from the 

 adjoining strata, the effect of which must be to cause the chief dilata- 

 tion to be in the vertical direction; this increases the diff'erences of 

 pressure, which give rise to superficial current, while the resistance to 

 the advancing under-current is diminished. And vice versa, when the 

 surface is cooled, its horizontal contraction is accelerated by the pressure 



