The wonderful sameness and equability of the temperature of the tropical ocean over spaces covering thousands of square miles, particularly between 10° N. and 10° S. lat., far from the coasts, and where it is not intersected by pelagic streams, affords, according to Arago, the best means of solving a very important, and as yet unanswered question, concerning the physics of the globe. "Without troubling itself," says that great natural philosopher, "about mere local influences, each century might leave to succeeding generations, by a few easy thermometrical measurements, the means of ascertaining whether the sun, at present almost the only source of warmth upon the surface of the earth, changes his physical constitution, and varies in his splendour like most stars, or whether he has attained a permanent condition. Great and lasting revolutions in his shining orb would reflect themselves more accurately in the altered mean temperature of those ocean plains than in the changed medium warmth of the dry land." The warmest part of the ocean does not coincide with the Equator, but seems to form two not quite parallel bands to the north and south. In the northern Atlantic, the line of greatest temperature (87° F.) which on the African coast is found but a little to the north of the Equator, rises on the north coast of South America as high as 12° N. lat., and in the Gulf of Mexico ranges even beyond the tropic. The influence of the warmth-radiating land on inclosed waters is still more remarkable in the Mediterranean (between 30° and 44° N. lat.) where during the summer months a temperature of 84° and 85° is found, three degrees higher than the medium warmth of the open tropical seas. While in the torrid zone the temperature of the ocean is generally inferior to that of the atmosphere, the contrary takes place in the Polar seas. Near Spitzbergen, even under 80° N. lat., Gaimard never found the temperature of the water below +33°. Between Norway and Spitzbergen the mean warmth of the water in summer was +39°, while that of the air only attained +37°. In the enclosed seas of the Arctic Ocean, the enormous accumulation of ice, which the warmth of a short summer is unable totally to dissolve, naturally produces a very low temperature of the waters. Thus, in Baffin's Bay, Sir John Ross found during the summer months only thirty-one days on which the temperature of the water rose above freezing point. In the depths of the sea, even in the tropical zone, the water is found of a frigid temperature, and this circumstance first led to the knowledge of the submarine polar ocean currents; "for without these, the deep sea temperature in the tropics could never have been lower than the maximum of cold, which the heat-radiating particles attain at the surface."[B] [Footnote B: Humboldt's "Kosmos."]