ON THE PHYSICAL THEORY OF UyDEE-CTJRKENTS. 151 



salinity, the latter being increased by surface evaporation, and 

 lessoned by rain and fresh water. An important distinction was 

 drawn between a liorizontal and almost superficial circulation, 

 produced by winds or tides, and a vertical circulation, in which 

 there was an interchange between water at the surface and at the 

 bottom. 



The physical conditions of certain inland seas were then 

 noticed. In the Caspian (which, as Von Eaer had shown, 

 formerly covered a much greater area) the level being now nearly 

 constant, it was clear that the surface evaporation was exactly 

 balanced by the rain-fall and river supply. The evaporation 

 current and deposition of salt in the Xaraboghaz were shown to 

 be due solely to solar heat. In the Eed Sea the surface 

 evaporation was about 23 feet per year. Since no rivers flowed 

 into the Eed Sea, and there was a strong evaporation in-current 

 through the strait of Babel Mandeb, it was evident that, as the 

 salinity of the water remained the same, there must be an 

 under-current flowing outwards through the strait. In the 

 Mediterranean, the surface evaporation was greater than 50 inches 

 per year, and the rainfall about 23 inches, leaving a deficiency of 

 27 inches. Now 27 inches depth over the Mediterranean area 

 amounted to 508 cubic miles, or -^-^ part of the volume of water 

 supplied by the Nile. It was clear therefore, since the level was 

 constant, that there must be an inflow from the Atlantic ; and as 

 the salinity did not increase year by year, that there must also be 

 an outflow. 



A comparison was then instituted between the temperature of 

 the Mediterranean and the Atlantic at the same depths in the 

 same latitudes — the temperature of the former not falling below 

 55° at any depth, while that of the latter continued to fall as the 

 depth increased (2000 fms., 36°. 5.) A detailed explanation was 

 then given of the causes of, and the work done by, the upper 

 and under currents in the Strait of Gibraltar, in which, by its tem- 

 perature and specific gravity, Mediterranean water (sp. gr. 1-0292) 

 could be easily distinguished from Atlantic water (sp. gr. 1-0268.) 



