three thousand five hundred miles across (maximum about five thousand); and is nine thousand miles long. Its area has been estimated as thirty-three million square miles, and its volume as seventy-five million cubic miles. It is a vast place, with many miles of coast, upon which much of civilization depends: considering its size, it has few islands. In comparison, the Indian Ocean is not quite as large (about twenty-eight million square miles); but the Pacific (about sixty-four million square miles) has nearly twice the area, and is ten thousand miles across its widest part. The Arctic Ocean (about five and a half million square miles) is small and nearly full of ice at all times of year; in spite of this it is at times very full of life. Finally, it is usual to describe the cold waters round the Antarctic Continent (itself the same size as the Arctic Ocean) as the Antarctic Ocean.
South of the normal steamship route from Britain to New York the Atlantic is almost everywhere over two miles deep, and in large areas more than three. But down mid-ocean, following the tropical kink in the zig-zag, runs a very long submarine ridge, above which is less than two miles of sea; it is only broken by deeps for a short distance on the Equator, and it rises to the surface in places—in the northern hemisphere at the Azores and St. Paul Rocks, and in the south at the lonely isles of Ascension, Tristan da Cunha and Gough. Other oceanic Atlantic islands, such as Bermuda in the north, and South Trinidad and St. Helena in the south, rise abruptly from very deep parts of the ocean. A sketch-chart will be found in Fig. 2c.
It will be seen that there are prominent shallows along the east coast of southern South America, north of the mouths of the Amazon and along the Guianas, in parts of the Caribbean Sea and the Gulf of Mexico (there are also marked deeps in these tropical waters), off the New England States, Nova Scotia and (most particularly) Newfoundland, and round Britain, the Channel and the North Sea, and round Iceland. A submarine ridge, over which the sea is five hundred fathoms or less, cuts the North Atlantic entirely from the Norwegian Sea and the waters of the Polar Basin; Shetland, the Faeroes and Iceland lie on this ridge. Davis Strait is shallow, and the waters of Labrador and Hudson’s Bay very shallow. Where the waters are less than a hundred fathoms deep, what they cover is usually described as the Continental shelf. This has its own particular community of birds.
For practical purposes, and because all charts and maps mark the Arctic Circle and the Tropics, we have classified the North Atlantic and its birds into arctic, temperate and tropical areas based simply on latitude. In our analysis of breeding-distribution, for instance (see here), we regard birds nesting north of the Arctic Circle as arctic, those nesting south of the Tropic of Cancer as tropical, and those nesting between as temperate. However, the temperature of neither air nor water arranges itself, in the Atlantic, according to latitude.* For instance, if we examine the July air isotherms over the world north of the Tropic of Cancer we see that that for 45° F. runs well south of the Arctic Circle in the areas Greenland-Baffin Island and Bering Strait, and well north of it off Scandinavia, avoiding Lapland altogether.
FIG. 1
Diagram of the Atlantic Ocean
If we examine a map of the world (showing particularly the lands between the Tropics), we see that the summer isotherm for 80°F. (July in the northern hemisphere, January in the southern) runs well north of the Tropic of Cancer in Mexico and the southern States, and in Africa and Asia, and south of the Tropic of Capricorn in Africa and Australia; yet large parts of the tropical Pacific and Atlantic Oceans never reach an average summer air temperature of 80°F.
In the North Atlantic there is not only relatively little direct correspondence between isotherms and latitude, but there is a good deal of difference in position between the same isotherms under the surface, on the water surface and in the air.
The primary cause of the ocean currents, and of the prevailing winds which are associated with them, is the rotation of the earth. The plot of the Atlantic currents and Atlantic winds is almost, though not quite, coincident. To a very large extent the distribution of Atlantic water temperatures, and to a large extent that of air temperatures, is a consequence of these currents and prevailing winds. However, in parts of the Atlantic evaporation and the melting of ice produces temperature and salinity gradients which themselves produce consequent currents. Hence the web of sequence and consequence, of cause and effect, becomes complex. We must examine the great equatorial current first, for almost every one of the more important sea masses in the Atlantic owes its existence to it. It is quite justifiable to write in terms of sea masses, for, as we shall see, the Atlantic waters are by no means homogeneous and can be divided, sometimes with strikingly sharp boundaries, into volumes possessing very diverse properties.
We need scarcely remind the reader that if he faces a globe, poised in the ordinary way with North at the top, and spins it as the earth naturally rotates, the points on its surface will travel, as they face him, from left to right. The points travelling with the greatest velocity will be those on the equator, and the two points represented by the Poles will travel with no velocity relative to the earth’s axis.
In general terms it is true that, as the earth rotates, its atmosphere rotates with it. However, there is a certain effect due to inertia or drag; and this effect, obviously, is greatest at the equator, where the surface velocity is greatest. The effect operates on all objects but can put only liquids and gases into a dynamic state. Upon these Corioli’s force—the deflecting force of the earth’s rotation—acts in a simple manner. It sets them in motion in a direction which, at the equator, is opposite that of the rotation of the earth. Thus if we examine a map of the prevailing winds and ocean currents of the world, we find pronounced positive east-to-west movements in all equatorial regions. The liquids and gases thus displaced circulate into the temperature regions and perform return movements in the higher latitudes where the Corioli’s force is less. Consequently, in the northern hemisphere water and wind currents tend to turn right-handed, whereas in the southern hemisphere they turn left-handed. (Exceptions to this rule are mostly found in minor seas, where the impact of the currents upon coasts may cause contra-rotation.) The main clockwise movement of the northern hemisphere wind and currents is very obvious.
The Atlantic equatorial current can be traced from the African coast south of the equator westwards as far as the sea reaches. Approaching the coast of Brazil it attains a remarkable speed. It sets past the isolated oceanic island of Ascension so that even in calm weather it leaves a wake of turbulence which must make that island unusually visible from far off by its numerous bird inhabitants.
Just north of the equator the lonely St. Paul rocks, which represent the pinnacles of a submerged, steep-sided mountain over thirteen thousand feet high, face the full strength of the great equatorial current, especially in August, when the associated south-east trades are blowing their hardest. During the cruise of the Challenger in 1860 H. N. Moseley saw the great ocean current “rushing past the rocks like a mill race.” A ship’s boat was quite unable to pull against the stream.
The equatorial current divides when it impinges on the corner of Brazil at Cape São Roque. The northern element—the Guiana coast current flows past the mouth of the Amazon with sufficient rapidity to displace the outgoing silt 100 miles or more in a northerly direction; and it continues steadily past the mouth of the Orinoco and Trinidad to flow with scarce-abated force into the Caribbean, mainly through the channel between Trinidad and Grenada in the Windward Islands.
Through the Caribbean the current flows from east to west, turning northerly and entering the Gulf of Mexico through the fairly narrow channel between Yucatan and Cuba. It is no doubt aided here by the climate, for this part of the world is very hot, and not excessively wet, and there is much evaporation of the waters of the Caribbean and the Gulf of Mexico, which has to be replaced. The current finally comes up against the coast of Louisiana and Texas and proceeds to mill right-handed, escaping finally through the narrow gap between Florida and Cuba, into the Bahama Seas.
Here the Gulf Stream is formed, not only by the waters escaping from the Gulf of Mexico but by more northerly elements of