82. The Bonifaccio current.—That the winds do make currents in the sea no one will have the hardihood to deny: but currents that are born of the winds are as unstable as the winds; uncertain as to time, place, and direction, they are sporadic and ephemeral; they are not the constant currents such as have been already enumerated. Admiral Smyth, in his valuable memoir on the Mediterranean (p.162), mentions that a continuance in the Sea of Tuscany of "gusty gales" from the south-west has been known to raise its surface no less than twelve feet above its ordinary level. This, he says, occasions a strong surface drift through the Strait of Bonifaccio. But in this we have nothing like the Gulf Stream; no deep and narrow channel-way to conduct these waters off like a miniature river even in that sea, but a mere surface flow, such as usually follows the piling up of water in any pond or gulf above the ordinary level. The Bonifaccio current does not flow like a "river in the sea" across the Mediterranean, but it spreads itself out as soon as it passes the Straits, and, like a circle on the water, loses itself by broad spreading as soon as it finds sea-room. As soon as the force that begets it expends itself, the current is done.
83. The heel of the Gulf Stream an ascending plane.—Supposing with Franklin, and those of his school, that the pressure of the waters that are forced into the Caribbean Sea by the trade-winds is the sole cause of the Gulf Stream, that sea and the Mexican Gulf should have a much higher level than the Atlantic. Accordingly, the advocates of this theory require for its support "a great degree of elevation." Major Rennell likens the stream to "an immense river descending from a higher level into a plain." Now we know very nearly the average breadth and velocity of the Gulf Stream in the Florida Pass. We also know, with a like degree of approximation, the velocity and breadth of the same waters off Cape Hatteras. Their breadth here is about seventy-five miles against thirty-two in the "Narrows" of the Straits, and their mean velocity is three knots off Hatteras against four in the "Narrows." This being the case, it is easy to show that the depth of the Gulf Stream off Hatteras is not so great as it is in the "Narrows" of Bemini by nearly 50 per cent., and that, consequently, instead of descending, its bed represents the surface of an inclined plane—inclined downwards from the north towards the south—up which plane the lower depths of the stream must ascend. If we assume its depth
