Page:Physical Geography of the Sea and its Meteorology.djvu/278

252 heavier waters below from the cold, for De Haven invariably found the temperature of the water under the ice 28°, which is the temperature that average sea water invariably assumes during the process of congelation (§ 442). Moreover, the specific gravity of the surface water which Rodgers measured in the Arctic Ocean was (§ 427) less than that of average sea water—a fact in confirmation of this conjecture as to the office of rain and river water in the polar sea's. The freezing-point of strong brine is 4°; consequently the freezing-point of water in the sea may vary according to the proportion of salts in it, from 4° all the way up to just below 32°. Thus the salts of the sea impart to its waters an elasticity, as it were, giving a law,—a sort of sliding-scale—both for the thermal dilatation and of congelation, which varies between that of fresh water and the saltest sea water according to the degree of its saltness.

480. Layers of water of different temperature in the Arctic Ocean.—Rodgers tried with his hydrometer and thermometer the waters of the Arctic Ocean at the surface, below, and at the bottom, and as often as he tried he found this arrangement: warm and light water on the top, cool in the middle, "hot and heavy" at the bottom. His experiments were made near Behring's Straits in August, 1855, between the parallels of 71°-2″, and are as per example following:

Assuming the surface water which Rodgers used for these experiments to be a fair average of arctic surface waters generally, this