that are light, but which, like the trade winds, are continually acting on the whole of a zone, cause a real movement of transition, which we do not observe in the heaviest tempests, because these last are circumscribed within a small space. When, in a great mass of water, the particles placed at the surface acquire a different specific gravity, a superficial current is formed, which takes it's direction towards the point where the water is coldest, or that which is most saturated with muriat of soda, sulphat of lime, and with muriat or sulphat of magnesia. In the seas of the tropics we find, that at great depths the thermometer marks 7 or 8 centesimal degrees. Such is the result of the numerous experiments of Commodore Ellis and of Mr. Peron. The temperature of the air in those latitudes being never below 19 or 20 degrees, it is not at the surface that the waters can have acquired a degree of cold so near the point of congelation, and of the maximum of the density of water. The existence of this cold strata in the low latitudes is an evident proof of the existence of an inferior current, which runs from the poles towards the equator: it also proves, that the saline substances, which alter the specific gravity of the water, are distributed in the ocean, so as not to annihilate the effect produced by the differences of temperature *.

* In fact, if the mean saltness of the sea was 0.005 greater

Considering the velocity of the molecules, which, on account of the rotatory motion of the globe, vary with the parallels, we may be tempted to admit that every current, in the direction

under the equator than in the temperate zone, as several naturalists pretend, a current at the bottom, from the equator towards the pole, would be the result: for 0·005 produce a difference of density of 0.0017, while, according to the tables of Hallstrom, a refrigeration of 16 centesimal degrees, between the 20th and 4th of temperature, causes only a change of 0.00015 in the specific gravity. After attentive examination of the results of the experiments of Black, reduced by Mr. Kirwan to the temperature of 16°, I find on the average the density of the water of the sea,

The proportion of salt corresponding to these four zones are, according to Bishop Watson, 0 0374; 0·0394; 0·0386; and 0-0372. Those numbers sufficiently prove, that the experiments hitherto published do not in any way justify the renewed opinion, that the sea is salter under the equator than under the 30th and 44th degrees of latitude. It is not therefore a greater quantity of saline substance held in solution, which opposes itself to this inferior current, by which the equinoctial ocean receives particles of water, which during the winter of the temperate zones have sunk towards the bottom of the sea, from the 30th to the 44th degree of southern and northern latitude. Baumé has analysed the sea-water collected by Mr. Pagès in different latitudes, and found in this water 0-005 less salt at 1° 16' of latitude than between the 25th and 40th degrees. (Kirwan's Geol, Essays, p. 350. Pagès Voyage round the Word, vol. ii; p. 6 and 275.)

from south to north, tends at the same time to ward the east, while the waters, which run from the pole toward the equator, have a tendency to deviate toward the west. We may also be led to think, that these tendencies diminish to a certain point the speed of the tropical current, in the same manner as they change the direction of the polar current, which in July and August, is regularly perceived during the melting of the ice, on the parallel of the bank of Newfoundland, and farther north. Very old nautical observations, which I have had occasion to confirm by comparing the longitude given by the chronometer with that which the pilots obtained by their reckoning, are contrary to these theoretical id as. In both hemispheres, the polar currents, when they are perceived, decline a little to the east; and we think that the cause of this phenomenon should be sought in the constancy of the westerly winds which prevail in the high latitudes. Besides, the particles of water do not move with the same rapidity as the particles of air; and the currents of the ocean, which we consider as the most rapid, have only a swiftness of eight or nine feet a second it is consequently very probable, that the water, in passing through different parallels, gradually acquires a velocity correspondent to those parallels, and that the rotation of the Earth does not change the direction of the currents.

The variable pressures, which the surface of the sea undergoes by the changes in the weight of the air, are another cause of motion which deserves particular attention. It is well known, that the barometric variations do not in general take place at the same moment on two distant points, which are on the same level. If in one' of these points the barometer stands a few lines lower than in the other, the water will rise where it finds the least pressure of the air, and this local intumescence will continue, till, from the effect of the wind, the equilibrium of the air is restored. Mr. Vaucher thinks that the tides in the Lake of Geneva, known by the name of the seiches, arise from the same cause. Under the torrid zone, the horary variations of the barometer may produce small oscillations at the surface of the seas, the meridian of 4", which corresponds to the minimum of the pressure of the air, being situate between the meridian of 21h and 11h upon which the height of the mercury is the greatest; but these oscillations, if even they were perceptible, will be accompanied by no change of place *.

When this last movement is produced by the inequality of the specific weight of the particles, a double current is formed, the upper of which has a contrary direction to the lower. Thus in the greatest part of the straits, as in the seas of * Mouvement de translation.

the tropics, which receive the cold waters of the northern regions, the whole mass of water is agitated to a very great depth. We are ignorant if it be the same, when the movement of progression, which must not be confounded with the oscillation of the waves, is the effect of an external impulse. Mr. de Fleurieu, in his narrative of the voyage of the Isis *, cites several facts which render it probable that thesea is much less still at the bottom than naturalists generally admit. Without entering here into a discussion which we shall treat hereafter, we shall only observe, that if the external impulse is constant in it's action, like that of the trade winds, the friction of the particles of water on each other must necessarily propagate the motion of the surface of the ocean even to the inferior strata; and in fact this propagation in the Gulfstream has long been admitted by navigators, who think they discover the effects in the great depth of the sea wherever it is traversed by the current of Florida, even amidst the sand-banks which surround the northern coasts of the United States. This immense river of hot waters, after a course of fifty days, from the 24th to the 45th degree of latitude, or 450 leagues, does not lose, amidst the rigors of winter in the temperate zone, more than 3 or 4 degrees of the tem

* Voyage made by order of the king, in 1768 and 1769, to try the marine time-pieces. Vol. i, p. 513.