When concentrated by any means, sea-water deposits, first, carbonate of lime, next gypsum, or sulphate of lime, and then salt; and, lastly, the salts of magnesia and the bromides. The phenomena are not quite so simple in practice, and the deposits of salt-marshes are rarely composed of a single substance; but we have only intended to indicate the general course of the operation. The salt of commerce is rich in magnesia, or chloride of magnesium, in proportion to the strength of the concentration. Some have even sought to ascribe the enormous deposits of gypsum found in certain regions to ancient seas which, in drying up, deposited that substance among the first.
Potash and the bromides, substances that are relatively little abundant, accumulate in the mother-waters till they become so condensed as to make the industrial working-out of them remunerative. Bromine is less abundant in the Mediterranean than in the waters of the Dead Sea, which may some day become a source of production. Eighteen centuries ago the Romans, according to Pliny, brought to Italy at great expense the water of the Asphaltine Lake, the curative properties of which were held in high esteem. The excess of bromine in this water, however, corresponds exactly with its greater total saltness, so that, except for a few qualifications to which we shall refer again, the relative composition of the dry residue of the Dead Sea is the same as of that from the ocean. In other words, any marine water evaporated to the same degree of density as that of the Dead Sea would be as deleterious to living beings.
Marine ice was formerly regarded as formed of solidified pure water retaining by mechanical adhesion traces of the saline liquid. These traces could be expelled by energetic pressure, when acids and bases would be found in the residue of desiccation in invariable proportions as in the sea. The question of chemical composition of the ice of the Arctic Ocean is complicated in other ways, but it gains in interest what it loses in simplicity. When salt-water is cooled artificially, a small part escapes solidification. The uncongealed residue is insupportably bitter to the taste, and analysis shows that nearly all the magnesia is concentrated in it. The solid block, if it is homogeneous and is not full of holes, and if previously drained, may furnish a passable drink. The natural ices of the Northern Sea are frequently moistened with a kind of brine, which sometimes embodies crystals of special character, easy to distinguish from the ice around them. According to Otto Petterssen, the relative proportions of chlorine and magnesia are much stronger in these exudations than in the water at the expense of which the ice is formed. The liquid can not then have been mechanically absorbed. On the other hand, there is a deficiency of sulphates; and the conclusion that sea-water ice retains the sulphates more abundantly is confirmed by analysis. With congelation, a sorting of matters takes place; most of the sulphuric acid passes into the part that solidifies, while magnesia and chlorine prevail in the part
