1

amount, while its discharge-the quantity of water moving eastwards-is doubled. It advances farther, to the meridian of the Azores, in 25° W., where, about the latitude of 47°, it bifurcates, one part turning to the right, along the coast of Portugal and southwards towards the Cape Verdes, while the other moves onwards past our own coasts, and makes its influence felt even within the Arctic Circle at Hammerfest, in Norway.

The Gulf Stream is essentially a current of hot water. It is generally intensely blue in colour, and is described as a vast river flowing past the Americar coast and spreading out across the Atlantic like a great pennant waving in the wind. The contrast of temperature between it and the American Arctic current inside it is so great, and the change in passing from one to the other so sudden, that off Halifax a difference of 20°, or even 30°, has been reported between the temperatures recorded during the same day, and a change of 8° or 10° in as many minutes. In fact,

along the northern edge, what Bache called the 'cold wall,' the line of demarcation is perfectly definite, owing to the change of colour of the water, and the simultaneous temperatures taken at the bow and stern of the ship, when crossing that line, have been known to differ by several degrees.

The Gulf Stream flows through the Straits of Bemini with a temperature of upwards of 80°; as it moves on it loses much of this extreme heat, but still its effect on the temperature of the North Atlantic is very considerable. Various estimates of the amount of this effect have been made. Among those most commonly quoted is that of Dr. James Croll, F.R.S., who

puts it at one-fifth of the total amount of heat possessed by that ocean; so that, were the Gulf Stream nonexistent, the average surface temperature of the whole ocean would be -3° F. The details of his calculation will be found in a note.1

However, in a recent paper, read before the British Association in 1881 by the Rev. Dr. Haughton, F.R.S., and printed in extenso in the 'Report' (p. 451), a very different estimate of the effect of the Gulf Stream will be found. Using Ferrel's tables of temperature, for all latitudes, in both hemispheres, Dr. Haughton gives the following figures for the influence of the Gulf Stream on the climates affected by it in July and January respectively :

On this principle we see that the Gulf Stream actually depresses the temperature in summer, while it certainly raises it in winter. The total annual effect is about one-half the winter effect. Dr. Haughton goes on to show that the result of the stoppage of the Gulf

1 According to Dr. Croll, the Gulf Stream, issuing from the Gulf of Mexico, and flowing in a broad current into the Atlantic, is estimated to convey to that ocean one-fifth of the total heat which is found in it. If we take the temperature of its surface, on the whole, as 56°, and the temperature of space, with Sir J. Herschel and others, at about -239°, the total heat of the Atlantic Ocean would correspond to a temperature of 295° (239 +56). One-fifth of this value is 59°, and if, therefore, we subtract from the assumed temperature 56°, the amount (59°) of one-fifth of the total heat, the residue is 3°. In other words, on this hypothesis, the surface temperature of the North Atlantic, without the Gulf Stream, would be thirty-five degrees below the freezing-point.

Stream in July would be nil; in January it would produce a general fall of temperature of about 4°.

On our own west coasts the water moves very slowly -not more than an inch or so per second-but that it does move, and carry foreign bodies with it to the coasts of Europe, partly owing to the influence of the prevailing Westerly winds, is proved by West Indian products, such as hard-shelled fruits, beans, &c., being picked up on the western coasts of the British Isles and of Norway.

The current, which, as we have said, turns to the right at the dividing point on the meridian of the Azores, flows south-eastward outside the Bay of Biscay, turns westwards off the coast of Portugal, and is continued as a south-west current past the Canaries and the Cape Verdes to join the northern equatorial drift, and so complete the circuit.

This circulation is naturally right-handed, owing to the influence of the earth's rotation, and as the winds circulate round the anti-cyclonic region near the tropic of Cancer, so this water flows round and round a broad area where no current is perceptible, and where all drift wood and foreign bodies, which are sloughed off the main current to the right, collect. This region is known as the Sargasso Sea, from the peculiar Gulfweed, the fucus natans, or sargassum bacciferum, which grows without a root, covering leagues and leagues of water surface, and affording a home to millions of fish, crustaceans, and molluscs. This weed is not met with in the main current outside.

The water of the Gulf Stream has been estimated to take about five and a half months to reach the coast of Europe from its starting at Cape Florida, and, on this

hypothesis, the time required for a particle to make the ntire circuit round the Sargasso Sea into the Gulf of Mexico again would be about two years and ten months.

One portion of the southerly current off the Straits of Gibraltar moves close along the African coast, and eventually, near Cape Palmas, joins the equatorial counter-current, and flows along into the Bight of Benin, as the Guinea current-an easterly current, along the coast, about 100 miles wide-while outside the equatorial current flows to the westward.

Sir Edward Sabine describes how, in the old slave trade days on the Gold Coast, Captain Clavering, in the Pheasant,' in which vessel Sir Edward was a passenger, on his pendulum expedition in 1822, used to lie out in the equatorial current to the eastward of the slave ports, knowing that the inshore current must bring the slavers to the spot where he was waiting for them.

The southerly current off Cape Verde is still a cold current, but its continuation, the Guinea current, is warm, at least at the surface, while a little underneath the water is cold; and in a paper read before the Meteorological Society (Quarterly Journal,' vol. iv. p. 32) Commander E. G. Bourke, R.N., states that in the month of October the temperature at the surface was 80°, and at the bottom, at the depth of eight fathoms, it was 65°. His attention was attracted to this remarkable difference by the fact of his finding the water used for washing the decks, drawn from beneath the surface by a pump, to be much colder than the water on the surface itself.

The presence of cold water at no great distance from the surface near the equator has also been proved by various observers, who have found that in mid

ocean, close to the Line, the cold stratum wells up, so that the navigator meets with patches of water giving a temperature much below that which might be expected for the season. The charts of Sea Temperature for the equatorial region of the Atlantic, published by the Meteorological Office, show for August the isotherm of 73° in 15° W. and 1° S., with the curves for 75° on each side of it; and on August 3, 1863, Captain Toynbee, between 1° 49′ N. and the equator, and between 15° and 17° W., took seven observations of the surface temperature, all below 71°, their mean being 70°3. This cold water undoubtedly comes up the west coast of Africa, where the low temperature of the surface inshore is a well-known phenomenon, which will now be explained.

The circulation of the South Atlantic is more simple than that of the north, as the coasts are not indented by great gulfs and inland seas, and the islands are so few and so small as to be quite insignificant. The cold water comes up from the Antarctic Ocean and strikes on the African coast near Cape Town, where, as already stated, it makes the temperature about 20° lower than at the corresponding latitude on the east coast, and fills Table Bay with a profusion of fish, for almost all the fishes extensively used for food inhabit cold water. The current flows close along the coast up to the Line, where it turns westward and crosses over to Cape St. Roque, the cold water having gradually sunk below the surface, and only betraying its presence in such spots as those in which Captain Toynbee met with it.

I have already described the course of the northern branch of this stream, which turns off from Cape St. Roque. The southern branch flows down along the