In the first stage of our planet, before the watery vapours contained in the primitive atmosphere were condensed, and before they had begun to fall on the earth in the form of boiling rain, the shell of the earth contained an infinite variety of heterogeneous mineral substances, some soluble in water, others not. When rain fell on the burning surface for the first time, the waters became charged with all the soluble substances, which were reunited and afterwards deposited, accumulating in the large depressions of the soil. The seas of the primitive globe were thus formed of rain water, holding in solution all that the earth had given up, collected in large basins. Chloride of sodium, sulphates of soda, magnesia, potassium, lime, and silica, in the form of soluble silicate; in a word, every soluble matter that the primitive globe contained formed part of the mineral contingent of this water. If we reflect that through all time up to the present day none of the general laws of nature have changed--if we consider that the soluble substances contained in the water of the primitive seas have remained there, and that the fresh water of the rivers constantly replaces the water which disappears by evaporation--we have the true explanation of the saltness of sea water. "It is a very simple theory, it is true," adds M. Figuier, "but one that we have found nowhere, and the responsibility of which we therefore claim. The chloride of sodium is by no means the only substance dissolved in sea water. It contains, besides, many other mineral substances: in short, every _soluble salt_ on the face of the globe, and, along with them, portions of different metals in infinitely small quantities." The mean temperature of the surface of the sea is nearly the same as the atmosphere, so long as no currents of heat or cold interpose their perturbing influence. In the neighbourhood of the Tropics, it appears that the surface of the water is slightly warmer than the ambient air, but experiments on the temperature of the sea from the surface to the bottom reveal, according to our author,[4] "some evidence which establishes a curious law. In very deep water a perfectly uniform temperature of four degrees below zero prevails, which corresponds, as physics have established, to the maximum density of water. Under the Equator this temperature exists at the depth of seven thousand feet. In the Polar regions, where water is colder at the surface, this temperature is maintained at four thousand six hundred feet. The isothermal lines of four degrees form a line of demarcation between the Zones, where the surface of the sea is colder, and those where it is warmer than the bed of four degrees below zero." This is more clearly shown in Fig. 4, which represents a section of the ocean, the curved line which touches two points at the surface indicating the depths where the temperature is constantly fixed at four degrees.