250 
THE PHYSICAL GEOGRAPHY OF THE SEA. 
Antarctic, and it is upon these waters that that sea is to be pene- 
trated, if ever. The North Pacific, except in the narrow passage 
between Asia and America, is closed to the escape of these w^arm 
waters into the Arctic Ocean. The only outlet for them is to the 
south. They go down toward the Antarctic regions to dispense 
their heat and get cool ; and the cold of the Antarctic, therefore, 
it may be inferred, is not so bitter as is the extreme cold of the 
Frozen Ocean of the north. 
543. The warm flow to the south from the middle of the In- 
dian Ocean is remarkable. Masters who return their abstract logs 
to me mention sea M^eed, which I suppose to be brought down by 
this current, as far as 45° south. There it is generally, but not 
cdways, about 5 degrees warmer than the ocean along the same 
parallel on either side. 
544. But the most unexpected discovery of all is that of the 
warm flow along the west coast of South Africa, its junction with 
the Lagullas current, called, higher up, the Mozambique, and then 
their starting off* as one stream to the southward. The prevalent 
opinion used to be that the Lagullas current, which has its genesis 
in the Red Sea 55), doubled the Cape of Good Hope, and then 
joined the great equatorial current of the Atlantic to feed the Gulf 
Stream. But my excellent friend Lieutenant Marin Jansen, of the 
Dutch Navy, suggested to me a few months ago that this was 
probably not the case. This induced a special investigation, and 
I found as he suggested, and as is represented on Plate IX. Cap- 
tain N. B. Grant, in the admirably well-kept abstract log of his 
voyage from New York to Australia, found this current remarka- 
bly developed. He was astonished at the temperature of its wa- 
ters, and did not know how to account for such a body of warm 
water in such a place. Being in longitude 14° east and latitude 
39° south, he thus writes in his abstract log ; 
" That there is a current setting to the eastward across the 
South Atlantic and Indian Oceans is, I believe, admitted by all 
navigators. The prevailing westerly winds seem to offer a su:^- 
cient reason for the existence of such a current, and the almost 
constant southwest swell would naturally give it a northerly direc- 
tion. But why the water should be warmer here (38° 4^ south) 
than between the parallels of 35° and 37° south is a problem that, 
in my mind, admits not of so easy solution, especially if my sus- 
