On the general Causes of the Ocean-Cuerents. 45 



esseutial condition for the circulation, caused by the united effect of 

 evaporation and rainfall, taking the form here described, is the tempera- 

 ture of the water in the less evaporating region. I we supposed e. g. 

 that the .warm surface-current preserved its once assumed temperature, 

 instead of cooling in the higher latitudes, the whole mass of water must 

 ultimately be raised to a high temperature, in which case it seems pos- 

 sible that the molecules concentrated by evaporation might sink down, 

 so that the circulation went out from the most heated place into the 

 deep. The afÛux would then take place from the sui-face, as is the case 

 in the Mediterranean, where the course of the stream is confined to 

 about the same latitudes, instead of crossing them as in the ocean. This 

 mode of circulation in the Mediterranean is however also especially fa- 

 voured by the small dimensions of the opening for the admission of 

 the inflowing water relatively to the surface exposed to evaporation. 



I have endeavoured to show, that in the hotter part of the ocean 

 in consequence of evaporation, rainfall and heat, there ought to be for- 

 med a broad and deep hollow filled with warm and salter water; that 

 this hollow is continually filled up from below by two deep streams of 

 cold water, one from each pole; that the greatest j^art of the confluent 

 water evaporates between the tropics, but that another part flows back 

 again over the edges of the hollow in the form of a shallow broad 

 stream of warm water, and that this warm current ought at a certain 

 latitude to be met by another superficial current of cold, but less salt 

 water, arising from the perturbation of the superficial stratas equilibrium, 

 which the rainfall in the high latitudes had left as one of its consequen- 

 ces. These results show, as regards the course of the surface-streams, 

 a by no means unimportant deviation from the phœuomena, which the 

 ocean actually presents. Let us therefore see, whether we have taken 

 account of all the most essential forces, the combination of which pro- 

 duces these currents. 



By the rotation of the Earth the streams proceeding towards the 

 poles receive an impulse towards the East, those descending from the 

 poles an impulse towards the West, and even in this a reason is appa- 

 rent, why the ascending warm and descending cold streams should be 

 able to pass by each other, each on its own side of the channel, in- 

 stead of coming in collision; but there remains an other and probably 

 still more active force to be taken into consideration. The broad belt 

 of warmed water between the tropics is attacked by the trade-winds in 

 two concurrent directions, viz: from N. E. and S. E. The watei-'s ten- 



