of the South Atlantic, and of both to that of the equatorial depth of about 700 fathoms, gradually rises as the equator is approached and it is between the equator and 7 S., where the surface-temperature rises to nearly 80, that cold water is soonest reached, the isotherm of 40 rising to within 300 fathoms of the surface, while that of 55, which in lat. 38 N. lies at nearly 400 fathoms depth, and in lat. 22 N. at about 250 fathoms, actually comes up under the equator within 100 fathoms of the surface. At the same time, while the bottom-temperature under the equator is the lowest anywhere met with, namely, 32 4, 1 the thickness of the stratum beneath the isotherm of 35 is not less than GOO fathoms. In passing southwards, the superficial iso therms are observed to separate again from each other, partly by the reduction of the surface-temperature, and partly by the descent of the isotherm of 40 to a depth of something less than 400 fathoms, which it keeps with little reduction as far south as the Cape of Good Hope. The significance of these facts becomes more remarkable, when we consider that if a portion of the oceanic area under the
[chart]
belt. The isotherm of 40, which in lat, 22 N. lies at a equator were to be secluded, like the Mediterranean or the
[chart]
Red Sea, from all but local influences, the temperature of its water from the sub-surface stratum downwards to the bottom whatever its depth would be its isochcimal or mean winter-temperature, which, in the equatorial zone, would be certainly not below 75. Nothing, Dr Carpenter contends, could more conclu sively support the general doctrine of a Vertical Oceanic Circulation sustained by opposition of temperature, than the precise conformity of the facts thus determined by observation to the predictions which his confidence in the theory had led him to put forth. These predictions were essentially as follows : "1. That instead of the local depressions of bottom-temperature imputed by previous writers to polar currents, the temperature of every part of the deep sea-bed in communication with either of the polar areas would be not many degrees above that of the polar areas themselves. " 2. That this general depression of bottom-temperature would be found to depend, not upon such a shallow glacial stream as might be maintained to be a return from the polar areas of water propelled towards them by wind-currents, but upon a creeping flow of the whole under-stratum, having a thickness of from 1000 to 2000 fathoms. "3. That as the depression of bottom-temperature in any part of the general oceanic basin would be proportional to the freedom of communication between its deeper portion and that of one or other of the polar areas, the bottom-temperature of the South Atlan tic would probably range downwards to 32, while that of the North Atlantic would not be below 35, except where it first receives the Arctic flow, or comes under the influence of the Antarctic underflow, which would very probably extend itself to the north of the equator. "4. That as the Arctic and Antarctic underflows must meet at or near the equator, whilst the surface-stratum is there continually being draughted off thence towards either pole, there would be a continual ascent of glacial water under the line, showing itself by a nearer approach of cold water to the surface in the ZHfcr-tropical than in the eatfra-tropical zone." It was further pointed out by Lenz, and more recently (in ignorance of his doctrine) by Carpenter, that additional evidence of such ascent is furnished by the low salinity of the surface-water of the equatorial belt corresponding with that of polar water. For, as was originally observed by Humboldt, then by Lenz himself, and subsequently by many other voyagers, the specific gravity of the surface- water of the Atlantic gradually increases as either tropic is approached from the polar side of its own. hemisphere, reaches its maximum a little nearer the equator, and then rapidly diminishes, coming down under the equator to the standard of polar water. Thus a mean of eight observations taken in the "Challenger" expedition between Bermuda (32 N.) and St Thomas (18| N.) gave 1027-2 as the sp. gr. of surface-water, whilst a mean of seventeen observa tions between the Cape Verd Islands (16| N.) and Bahia (13 S.) gave a sp. gr." of only 1026 3. Now, since between St Thomas and Bermuda the eight " Challenger" observa tions of bottom (polar) water gave a mean sp. gr. of 102 6 3, whilst between Cape Verd and Bahia the mean sp. gr. of the bottom-water was even slightly lower (the results being 1 That the bottom-temperature beneath the equator was lower than any that was met with in the South Atlantic, is attributable to the cir cumstance that, in consequence of unfavourable weather, the tempera ture-soundings were taken at intervals too wide to detect the deep channel through which the coldest Antarctic water doubtless flowed to wards the equator.
