The Gulf Stream has recently attracted a large share of the attention of our men of science. The strange weather which we experienced last winter (see date of essay) has had something to do with this. The influence of the Gulf Stream upon our climate, and the special influence which it is assumed to exercise in mitigating the severity of our winters, have been so long recognised that meteorologists began to inquire what changes could be supposed to have taken place in the great current to account for so remarkable a winter as the last. But it happened also that at a meeting of the Royal Geographical Society early in the present year the very existence of the Gulf Stream was called in question, just when meteorologists were disposed to assign to it effects of unusual importance. And in the course of the discussion whether there is in truth a Gulf Stream—or rather whether our shores are visited by a current which merits such a name—a variety of interesting facts were adduced, which were either before unknown or had attracted little attention. As at a recent meeting of the same society these doubts have been renewed, I propose to examine briefly, in115 the first place, a few of the considerations which have been urged against the existence of a current from the Gulf of Mexico to the neighbourhood of our shores; and then, having rehabilitated the reputation of this celebrated ocean river—as I believe I shall be able to do—I shall proceed to give a brief sketch of the processes by which the current-system of the North Atlantic is set and maintained in motion.

In reality the Gulf Stream is only a part of a system of oceanic circulation; but in dealing with the arguments which have been urged against its very existence, we may confine our attention to the fact that, according to the views which had been accepted for more than a century, there is a stream of water which, running out of the Gulf Stream through the Narrows of Bemini, flows along the shores of the United States to Newfoundland, and thence right across the Atlantic to the shores of Great Britain. It is this last fact which is now called in question. The existence of a current as far as the neighbourhood of Newfoundland is conceded, but the fact that the stream flows onward to our shores is denied.

The point on which most stress is placed is the shallowness of the passage called the ‘Bemini Narrows,’ through which it is assumed that the whole of the Gulf current must pass. This passage has a width of about forty miles, and a depth of little more than six hundred yards. The current which flows through it is perhaps little more than thirty miles in width, and a quarter of a mile in depth. It is asked with some appearance of116 reason, how this narrow current can be looked upon as the parent of that wide stream which is supposed to traverse the Atlantic with a mean width of some five or six hundred miles. Indeed, a much greater width has been assigned to it, though on mistaken grounds; for it has been remarked that since waifs and strays from the tropics are found upon the shores of Portugal, as well as upon those of Greenland, we must ascribe to the current a span equal to the enormous space separating these places. But the circumstance here dwelt upon can clearly be explained in another way. We know that of two pieces of wood thrown into the Thames at Richmond, one might be picked up at Putney, and the other at Gravesend. Yet we do not conclude that the width of the Thames is equal to the distance separating Putney from Gravesend. And doubtless the tropical waifs which have been picked up on the shores of Greenland and of Portugal have found their way thither by circuitous courses, and not by direct transmission along opposite edges of the great Gulf current.

But certainly the difficulty associated with the narrowness of the Bemini current is one deserving of careful attention. Are we free to identify a current six hundred miles in width with one which is but thirty miles wide, and not very deep? An increase of width certainly not less than thirtyfold would appear to correspond to a proportionate diminution of depth. And remembering that it is only near the middle of the Narrows that the Gulf Stream has a depth of four117 hundred yards, we could scarcely assign to the wide current in the mid-Atlantic a greater depth than ten or twelve yards. This depth seems altogether out of proportion to the enormous lateral extension of the current.

But besides that even this consideration would not suffice to disprove the existence of a current in the mid-Atlantic, an important circumstance remains to be mentioned. The current in the Narrows flows with great velocity,—certainly not less than four or five miles an hour. As the current grows wider it flows more sedately; and opposite Cape Hatteras its velocity is already reduced to little more than three miles an hour. In the mid-Atlantic the current may be assumed to flow at a rate little exceeding a mile per hour, at the outside. Here, then, we have a circumstance which suffices to remove a large part of the difficulty arising from the narrowness of the Bemini current, and we can at once increase our estimate of the depth of the mid-Atlantic current fivefold.

But this is not all. It has long been understood that the current which passes out through the Narrows of Bemini corresponds to the portion of the great equatorial current which passes into the Gulf of Mexico between the West Indian Islands. We cannot doubt that the barrier formed by those islands serves to divert a large portion of the equatorial current. The portion thus diverted finds its way, we may assume, along the outside of the West Indian Archipelago, and thus joins the other portion—which has in the118 meantime made the circuit of the Gulf—as it issues from the Bemini Straits. All the maps in which the Atlantic currents are depicted present precisely such an outside current as I have here spoken of, and most of them assign to it a width exceeding that of the Bemini current. Indeed, were it not for the doubts which the recent discussions have thrown upon all the currents charted by seamen, I should have been content to point to this outside current as shown in the maps. As it is, I have thought is necessary to show that such a current must necessarily have an existence, since we cannot lose sight of the influence of the West Indian Isles in partially damming up the passage along which the equatorial current would otherwise find its way into the Gulf of Mexico. Whatever portion of the great current is thus diverted must find a passage elsewhere, and no passage exists for it save along the outside of the West Indian Isles.

The possibility that the wide current which has been assumed to traverse the mid-Atlantic may be associated with the waters which flow from the Gulf of Mexico, either through the Narrows or round the outside of the barrier formed by the West Indies, has thus been satisfactorily established. But we now have to consider difficulties which have been supposed to encounter our current on its passage from the Gulf to the mid-Atlantic.

Northwards, along the shores of the United States, the current has been traced by the singular blueness of its waters until it has reached the neighbourhood119 of Newfoundland. Over a part of this course, indeed, the waters of the current are of indigo blue, and so clearly marked that their line of junction with the ordinary sea-water can be traced by the eye. ‘Often,’ says Captain Maury, ‘one half of a vessel may be perceived floating in Gulf Stream water, while the other half is in common water of the sea—so sharp is the line, and such the want of affinity between the waters, and such, too, the reluctance, so to speak, on the part of those of the Gulf Stream, to mingle with the littoral waters of the sea.’

But it is now denied that there is any current beyond the neighbourhood of Newfoundland—or that the warm temperature, which has characterised the waters of the current up to this point, can be detected farther out.

It is first noticed that, as the Gulf current must reach the neighbourhood of Newfoundland with a north-easterly motion, and, if it ever reached the shores of the British Isles, would have to travel thither with an almost due easterly motion, there is a change of direction to be accounted for. This, however, is an old, and I had supposed exploded, fallacy. The course of the Gulf Stream from the Bemini Straits to the British Isles corresponds exactly with that which is due to the combined effects of the motion of the water and that of the earth upon its axis. Florida being much nearer than Ireland to the equator, has a much more rapid easterly motion. Therefore, as the current gets farther and farther north, the effect of the easterly120 motion thus imparted to it begins to show itself more and more, until the current is gradually changed from a north-easterly to an almost easterly stream. The process is the exact converse of that by which the air-currents from the north gradually change into the north-westerly trade-winds as they get farther south.

But it is further remarked that as the current passes out beyond the shelter of Newfoundland, it is impinged upon by those cold currents from the arctic seas which are known to be continually flowing out of Baffin’s Bay and down the eastern shores of Greenland; and it is contended that these currents suffice, not merely to break up the Gulf current, but so to cool its waters that these could produce no effect upon the climate of Great Britain if they ever reached its neighbourhood.

Here, again, I must remark that we are dealing with no new discovery. Captain Maury has already remarked upon this peculiarity. ‘At the very season of the year,’ he says, ‘when the Gulf Stream is rushing in greatest volume through the Straits of Florida, and hastening to the north with the greatest rapidity, there is a cold stream from Baffin’s Bay, Labrador, and the coasts of the north, running south with equal velocity.... One part of it underruns the Gulf Stream, as is shown by the icebergs, which are carried in a direction tending across its course.’ There can be no doubt, in fact, that this last circumstance indicates the manner in which the main contest between the two currents is settled. A portion of the arctic current finds its way between the Gulf Stream and the continent of121 America; and this portion, though narrow, has a very remarkable effect in increasing the coldness of the American winters. But the main part, (heavier, by reason of its coldness, than the surrounding water,) sinks beneath the surface. And the well-known fact mentioned by Maury, that icebergs have been seen stemming the Gulf Stream, suffices to show how comparatively shallow that current is at this distance from its source, and thus aids to remove a difficulty which we have already had occasion to deal with.

Doubtless the cooling influence of the arctic currents is appreciable; but it would be a mistake to suppose that this influence can suffice to deprive the Gulf current of its distinctive warmth. If all the effect of the cold current were operative on the Gulf Stream alone we might suppose that, despite the enormous quantity of comparatively warm water which is continually being carried northwards, the current would be reduced to the temperature of the surrounding water. But this is not so. The arctic current not only cools the Gulf current, but the surrounding water also—possibly to a greater extent, for it is commonly supposed that a bed of ordinary sea-water separates the two main currents from each other. Thus the characteristic difference of temperature remains unaffected. But in reality we may assume that the cooling effect actually exercised by the arctic current upon the neighbouring sea is altogether disproportionate to the immense amount of heat continually being carried northwards by the Gulf Stream. It is astonishing how122 unreadily two sea-currents exchange their temperatures—to use a somewhat inexact mode of expression. The very fact that the littoral current of the United States is so cold—a fact thoroughly established—shows how little warmth this current has drawn from the neighbouring seas. Another fact, mentioned by Captain Maury, bears in a very interesting manner upon this peculiarity. He says: ‘If any vessel will take up her position a little to the northward of Bermuda, and steering thence for the capes of Virginia, will try the water-thermometer all the way at short intervals, she will find its reading to be now higher, now lower; and the observer will discover that he has been crossing streak after streak of warm and cool water in regular alternations.’ Each portion maintains its own temperature, even in the case of such warm streaks as these, all belonging to one current.

Similar considerations dispose of the arguments which have been founded on the temperature of the sea-bottom. It has been proved that the living creatures which people the lower depths of the sea exist under circumstances which evidence a perfect uniformity of temperature; and arguments on the subject of the Gulf Stream have been derived from the evidence of what is termed a minimum thermometer—that is, a thermometer which will indicate the lowest temperature it has been exposed to—let down into the depths of the sea. All such arguments, whether adduced against or in favour of the Gulf Stream theory, must be held,123 to be futile, since the thermometer in its descent may pass through several submarine currents of different temperature.

Lastly, an argument has been urged against the warming effects of the Gulf Stream upon our climate which requires to be considered with some attention. It is urged that the warmth derived from so shallow a current as the Gulf Stream must be, by the time it has reached our shores, could not provide an amount of heat sufficient to affect our climate to any appreciable extent. The mere neighbourhood of this water at a temperature slightly higher than that due to the latitude could not, it is urged, affect the temperature of the inland counties at all.

This argument is founded on a misapprehension of the beautiful arrangement by which Nature carries heat from one region to distribute it over another. Over the surface of the whole current the process of evaporation is going on at a greater rate than over the neighbouring seas, because the waters of the current are warmer than ............