Take, for example, the isothermal line of 0 or zero—that is, the line where the mean or average height of the thermometer for the year is at zero. At Behring’s Straits this line is a little below the Arctic circle, or the parallel of 66.30 north latitude. Passing east over North America, it descends into Canada, almost to Lake Superior, and to about the 50th parallel: that is to say, it is on an average during the year as cold on our continent at the 50th parallel as it is near Behring’s Straits at the 65th parallel. Passing east, the line of zero rises again over the Atlantic Ocean until, in the meridian of Spitzbergen, it reaches, within the Arctic circle, up almost to the 75th parallel. So, too, the isothermal of 5° below zero, which is below the 60th parallel in Siberia, rises in the North Sea, above Behring’s Straits, to the parallel of 75°, descending on the continent in North America to the 55th parallel, and rising again almost to the pole at Spitzbergen, to descend again in Siberia, while the isothermals of 10° and 15° below zero, which in North America are but just above the latitude of 60° and 75° respectively, ascend abruptly surrounding the magnetic pole, and falling short of the geographical one. Let this projection of the lines of equal temperature, and particularly the situation of the magnetic poles, be studied well, for we shall recur to it hereafter in illustration of many important portions of our subject.

It is apparent from these facts, and were it necessary might be rendered still more so by referring to others, that other causes operate in the distribution of heat over the earth besides the direct action of the sun’s rays upon it. Doubtless very considerable allowance is to be made for the difference of seasons, and difference during the same season upon the land and upon the ocean; in mountainous countries and level ones. But making every allowance for them, the fact that other causes have a controlling influence in producing the deviations still remains most obvious. Neither the difference of temperature between the land and the ocean, or land surfaces of unequal elevations, will account for the elevation of the isothermal lines on different portions of the ocean, or their extension around the magnetic poles.

Returning to a consideration of the arrangements for the diffusion of heat, we observe: First, that the earth itself is intensely heated in its interior. This is inferred, and justly, from the fact that the thermometer is found to rise about one degree for every fifty-five feet of descent—whether in boring artesian wells, exploring caves, or sinking shafts in mines. It is demonstrated, also, by the existence of hot springs and the action of volcanoes. Heat is supposed to be conducted from the center toward the surface every where, but with difficulty and slowly. It is also supposed to be conducted from the tropical regions toward the poles. Such is the opinion of Humboldt. (Cosmos, vol. i. p. 167.)

Probably it reaches the surface and exerts an influence, also, upon the weather through the ocean, and by heating it in its greatest depths. Little attention has been paid, so far as I am informed, to the question how far the ocean is thus heated in tropical latitudes. Doubtless a portion of the warmth of the ocean there is derived from that source, and it has its influence in changing the temperature of the deep-seated cold polar currents of, the great oceans. Perhaps it may yet be found that the icebergs are detached by it in the polar seas—the observations of Dr. Kane point to such a result. (Grinnell Expedition, p. 113, and also chap. 48.)

Little need be said of the inconsiderable quantities of heat supposed to be derived by radiation from the stars, the planets, and from space. If any such are derived they are too inconsiderable to be of importance in this inquiry.

Heat is also carried, and in quantities which exert very considerable influence upon the weather, from the tropics to the poles by the great oceanic currents which flow unceasingly from one to the other.

The most important of these with which we are acquainted is the Gulf Stream of the Atlantic. Gathering in the South Atlantic, and passing north through the Caribbean Sea and the Gulf of Mexico, it issues out through the Bahama Channel, and flows north along the eastern coast of the United States, but some distance from it, to Newfoundland, and from thence continuing to the north-east and spreading out over the surface of the ocean—a portion of it mingling with the waters of the North Atlantic in passing—it flows up on the western coast of Europe, around the Faroe Islands, and Spitzbergen, to the polar sea; passing around Greenland, and perhaps through its Fiords, it descends again through the sounds and channels of the Arctic regions into Baffin’s Bay, and through Davis’s Straits, burdened with the icebergs and floes of the polar waters, to return again to the South Atlantic. For reasons which will appear in the sequel, it has comparatively little influence upon the weather of the United States. Western Europe, however, Greenland, the islands which lie in its course, and the polar seas, are most materially influenced. Although not the only cause, it has very much to do with the remarkable elevation of the isothermal lines over the Northern Atlantic, and upon Western Europe, as seen upon the map.

A like oceanic current exists in the Pacific Ocean, the influence of which may also be traced upon the map by the elevation of the isothermal lines at the northern extremity of that ocean, and upon the north-west coast of North America. A vast amount of heat is transported from the tropical to the temperate and frozen regions of the earth by these great oceanic currents.

Another supply is derived from aerial currents which flow from the tropics toward the poles. These currents exist every where over the entire surface of the earth, but in more concentrated volumes along the great “lines of no variation,” and greater magnetic intensity, on the western side of the great oceans, over the eastern portions of the two continents of North America and Asia. Not, as meteorological writers suppose, in the upper portions of the atmosphere, having risen in the trade-wind region and run off at the top toward the poles by force of gravity, but near, and sometimes in contact with the earth. The influence of these aerial currents upon the temperature of the atmosphere, and in producing the phenomena we are to consider, is exceedingly important. We shall have occasion to examine them with great care and minuteness under another head, for upon them, more than any other portion of the arrangements, depend not only the diffusion of heat, but also the distribution of moisture.

Still another supply of heat, during the sudden changes, at least, is produced by the action of terrestrial magnetism and electricity. Very great progress has been made within a short period, in the investigation of the nature of these agents. The identity, or at least intimate association or connection of heat, light, electricity, and magnetism, always suspected, has been in various ways, and by a variety of experiments demonstrated. The influence of magnetism if distinct from gravitation, is second only to that; and its agency in producing the phenomena we are considering is primary and controlling. We will only, in this connection, ask the reader to note the situation of the north magnetic poles (for there are two of them); the manner in which the isothermal lines surround them; the fact that they are poles of cold, i. e., that it is colder there than even to the north of them. We shall recur to this part of the subject again.