pound, as sulphuric acid and the four basic principles, soda, magnesia, lime, and potash. Chlorine is by far the most abundant principle, and should be credited with more than half the weight of the saline matter. Open any book on chemistry or the physics of the globe, and you will find that sea-water contains, by the litre, so much chloride of sodium, so much sulphate of magnesia, so much chloride of magnesium, etc. These affirmations are wholly hypothetical, for our acquaintance with chemistry is not sufficient to permit such conclusions. Analysis shows that there are in a litre of sea-water so much chlorine, so much sulphuric acid, and so much magnesia, but does not reveal to us how the radicals are united, for the combinations we get in analysis are not probably precisely the ones that existed in the water. We might say that the numerous simple bodies entering into the composition of seawater are all the time contracting new and incessantly variable alliances, according to the temperature or the concentration of the liquid. It is by intelligently utilizing these laws that they succeed at the salt works in forcing the mother-waters to deposit at one time cooking-salt, at another time some other combination useful in industry, or which it is desirable to get rid of.
In evaporating to dryness a known quantity of sea-water, under certain precautions, we obtain a residue which, well dried and weighed, furnishes the weight of the total quantity of salts originally dissolved. It is then easy, by a simple calculation, to estimate the proportion of solid substances contained in a litre. Salt-water is denser than fresh water of the same volume and temperature, and this excess of density is evidently proportional to its richness in saline matters. This can be obtained by multiplying the excess of density by 1·32. We may thus replace the chemical operation by a determination of density, an easier experiment, and one that can be made on board ships.
The different oceanic regions are not equally rich in salts. What we have said respecting variations of specific weight shows this very clearly. But, if we always draw the water from a sufficient depth, the variations become much less, as Forchharamer has proved. The figures in his tables oscillate between thirty-four and thirty-five grammes per litre. The relative proportion of the different elements is still more invariable, and we can establish a few slight differences only by taking the means of a large number of estimations. This fixity of relation might have been foreseen, because evaporation concentrates, without taking away an atom of salt, while fresh waters dilute without furnishing any. It follows, then, that the composition of a specimen of seawater can be estimated with a fair degree of accuracy by ascertaining the proportion of one of its constituents, the chlorine, for instance, and that element is much used as a standard. The amount of chlorine in a litre of liquid collected along the shore diminishes obviously when the ship is approaching ice, or if it is cruising near the mouth of a great river.
