636 
ME. J. PKESTWICII ON SUBMARINE TEMPERATURES. 
approaching to the freezing-point or to that of the maximum density of ordinary 
sea-water. Moving in the same direction as the great body of colder water which it 
overlies, the warmer surface-water has a greater velocity than it, and moves over it 
as a surface-current — the causes which effect its impulsion being of a more energetic 
character than those which originate during the colder months of the year with the 
descent of the dense waters and their slow outward propulsion in a deep undercurrent. 
In the Pacific Ocean the great breadth of open sea, and the almost entire exclu- 
sion of the waters of the north polar seas, have produced conditions very different from 
those which obtain in the Atlantic. The temperature-soundings are too few to lead to 
any certain conclusion ; but, so far as they go, they seem to show that there is no uprising 
of cold undercurrents at the equator. The observations referred to by Lexz are so scat- 
tered and at such small depths, that they may have been affected by the action of the 
great cold current which passes northward up the west coast of South America, and is 
deflected westward at the equator, and by various other surface-currents. 
In any case, the remarkable rise of the bathymetrical isotherms in the North 
Pacific, which cannot be accounted for by any current passing through Behring Strait, 
leads me to infer that the Antarctic waters pass under the whole length of the Pacific, 
and are thrown up by the barrier presented at its northern extremity by the American 
and Asiatic coasts. Some of the great currents of the North Pacific may owe their 
origin to, while others seem to be strengthened by, these distantly derived waters. 
Nor is it easy to account in any other way for the rise of the isotherms of 35° and 
40° E. at the head of the Arabian Sea after traversing the deep bed of the Indian Ocean. 
The high temperature of the surface-waters, however, prevents the effects being so 
apparent in the upper strata of that sea. Again, the causes which influence the great 
currents of the North -Indian Ocean appear to correspond with the area of surging-up, 
as they approach the Asiatic continent, of the south-polar undercurrents. 
The cause of these phenomena in both hemispheres is, in all probability, connected 
with the intense cold of the polar regions, — the mean annual difference of from 
7 0° to 80° F. between the polar and the intertropical regions forming a permanent 
disturbing cause, owing to the alteration of density to which the affluent waters are 
unceasingly subjected*. It is a cause, also, which, from the variation in the density of 
the surface-water in winter and summer, must materially influence the operation of 
the currents generally, both at the Arctic and Antarctic regions, during the different 
seasons of the year, increasing the outflow from the polar seas in the cold months, 
and the influx in the warmer, whence the outflowing current through Behring Strait in 
the winter or spring, and the inflowing current in the summer. For the same reason 
we should expect to find the general circulation more active in the one season than in 
the other. But the discussion of these interesting questions is not our object. 
In no way are the effects of the remarkable interchange between the polar and equa- 
torial waters in the great oceans more conspicuous than in the comparison of the 
* According to Dove the mean temperature of the equator is 79°*8 and of the pole 2 0, 2. 
