126 THE PHYSICAL GEOGRAPHY OF THE SEA. 
a lower level. So far from this being the case, some currents of 
the sea actually run up hill, while others run on a level. 
The Gulf Stream is of the first class (§ 10). 
238. The currents which run from the Atlantic into the Medi- 
terranean, and from the Indian Ocean into the Red Sea, are the 
reverse of this. Here the bottom of the current is probably a 
water-level, and the top an inclined plane, running down hill. 
Take the Red Sea current as an illustration. That sea hes, for 
the most part, within a rainless and riverless district. It may be 
compared to a long and narrow trough. Being in a rainless dis- 
trict, the evaporation from it is immense ; none of the water thus 
taken up is returned to it either by rivers or rains. It is about 
one thousand miles long ; it lies nearly north and south, and ex- 
tends from latitude 13° to the parallel of 30° north. 
239. From May to October, the water in the upper part of this 
sea is said to be two feet low^er than it is near the mouth.* This 
change or difference of level is ascribed to the effect of the wind, 
which, prevaihng from the north at that season, is supposed to 
blow the water out. 
But from May to October is also the hot season ; it is the sea- 
son when evaporation is going on most rapidly ; and when we 
consider how dry and how hot the winds are which blow upon 
this sea at this season of the year, we may suppose the daily evap- 
oration to be immense ; not less, certainly, than half an inch, and 
probably twice that amount. We know that the waste from ca- 
nals by evaporation, in the summer time, is an element which the 
engineer, when taking the capacity of his feeders into calculation, 
has to consider. With him it is an important element ; how much 
more so must the waste by evaporation from this sea be, when we 
consider the physical conditions under which it is placed. Its feed- 
er, the Arabian Sea, is a thousand miles from its head ; its shores 
are burning sands ; the evaporation is ceaseless ; and none of the 
vapors, which the scorching winds that blow over it carry away, 
are returned to it again in the shape of rains. 
240. The Red Sea vapors are carried off and precipitated else- 
where. The depressibn in the level of its head waters in the sum- 
mer time, therefore, it appears, is owing quite as much to the effect 
of evaporation as to that of the wind blowing the waters back. 
* Johnston's Physical Atlas, 
