so soon as fresh water: so that, in consequence of all these circumstances, the ocean is not subject to such variations of heat as the land; and, by imparting its temperature to the winds, it diminishes the intensity of climate on the coasts and in the islands, which are never subject to such extremes of heat and cold as are experienced in the interior of continents, though they are liable to fogs and rain from the evaporation of the adjacent seas. On each side of the equator, to the 48th degree of latitude, the surface of the ocean is in general warmer than the air above it; the mean of the difference of temperature at noon and midnight is about 10.37, the greatest deviation never exceeding form 0°.36 to 2o.16, which is much cooler than the air over the land.

On land the temperature depends upon the nature of the soil and its products, its habitual moisture or dryness. From the eastern extremity of the Sahara desert quite across Africa, the soil is almost entirely barren sand, and the Sahara desert itself, without including Dafour or Dongola, extends over an area of 194000 square leagues, equal to twice the area of the Mediterranean sea, and raises the temperature of the air by radiation from 90° to 100°, which must have a most extensive influence. On the contrary, vegetation cools the air by evaporation and the apparent radiation of cold from the leaves of plants, because they absorb more caloric than they give out. The graminiferous plains of South America cover an extent ten times greater than France, occupying no less than about 50000 square leagues, which is more than the whole chain of the Andes, and all the scattered mountain groups of Brazil: these, together with the plains of North America and the steppes of Europe and Asia, must have an extensive cool

ing effect on the atmosphere, if it be considered that, in calm and serene nights, they cause the thermometer to descend 12° or 14°, and that, in the meadows and heaths in England, the absorption of heat by the grass is sufficient to cause the temperature to sink to the point of congelation during the night for ten months in the year. Forests cool the air also by shading the ground from the rays of the sun, and by evaporation from the boughs. Hales found that the leaves of a single plant of helianthus, three feet high, exposed nearly forty feet of surface; and if it be considered that the woody regions of the river Amazons, and the higher part of the Oroonoko, occupy an area of 260000 square leagues, some idea may be formed of the torrents of vapor which arise from the leaves of the forests all over the globe. However, the frigorific effects of their evaporation are counteracted in some measure by the perfect calm which reins in the tropical wildernesses. The innumerable rivers, lakes, pools, and marshes interspersed through the continents absorb caloric, and cool the air by evaporation; but on account of the chilled and dense particles sinking to the bottom, deep water diminishes the cold of winter, so long as ice is not formed.

In consequence of the difference in the radiating and absorbing powers of the sea and land, their configuration greatly modifies the distribution of heat over the surface of the globe. Under the equator only one-sixth part of the circumference is land; and the superficial extent of land in the northern and southern hemispheres is in the proportion of three to one: the effect of this unequal division is greater in the temperate, than in the torrid zones, for the area of land in the northern temperate zone

is to that in the southern as thirteen to one, whereas the proportion of land between the equator and each tropic is as five to four; and it is a curious fact, noticed by Mr. Gardner, that only one twenty-seventh part of the land of the globe has land diametrically opposite to it. This disproportionate arrangement of the solid part of the globe has a powerful influence on the temperature of the southern hemisphere. But, besides these greater modifications, the peninsulas, promontories, and capes, running out into the ocean, together with bays and internal seas, all affect the temperature: to these may be added, the position of continental masses with regard to the cardinal points. All these diversities of land and water affect the temperature by the agency of the winds. On this account the temperature is lower on the eastern coasts both of the New and Old World, than on the western; for, considering Europe as an island, the general temperature is mild in proportion as the aspect is open to the western ocean, the superficial temperature of which, as far north as the 45° and 50° of latitude, does not fall below 48° or 51° of Fahrenheit even in middle of winter. On the contrary, the cold of Russia arises from its exposure to the northern and eastern winds; but the European part of that empire has a less rigorous climate than the Asiatic, because the whole northern extremity of Europe is separated from the polar ice by a zone of open sea, whose winter temperature is much above that of a continental country under the same latitude.

The interposition of the atmosphere modifies all the effects of the sun's heat; but the earth communicates its temperature so slowly, that M. Arago has occasionally found as much as from 15° to 18° of difference between

the heat of the soil and that of the air two or three inches

above it.

The circumstances which have been enumerated, and many more, concur in disturbing the regular distribution of heat over the globe, and occasion numberless local irregularities nevertheless the mean annual temperature becomes gradually lower from the equator to the poles; but the diminution of mean heat is most rapid between the 40° and 45° of latitude both in Europe and America, which accords perfectly with theory, whence it appears that the variation in the square of the cosine of the latitude which expresses the law of the change of temperature, is a maximum towards the 45° of latitude. The mean annual temperature under the line in Asia and America is about 81° of Fahrenheit; in Africa it is said to be nearly 83°. The difference probably arises from the winds of Siberia and Canada, whose chilly influence is sensibly felt in Asia and America, even within 18° of the equator.

The isothermal lines are parallel to the equator, till about the 22° of latitude on each side of it, where they begin to lose their parallelism, and continue to do so more and more as the latitude augments. With regard to the northern hemisphere, the isothermal line of 59° of Fahrenheit passes between Rome and Florence, in latitude 43°; and near Raleigh, in North Carolina, latitude 36°; that of 50° of equal annual temperature runs through the Netherlands, latitude 51°; and near Boston, in the United States, latitude 4210; that of 41° passes near Stockholm, latitude 52°; and St. George's Bay, Newfoundland, latitude 48°; and lastly, the line of 329, the freezing point of water, passes between Ulea, in Lapland, latitude 66o, and Table Bay, on the coast of Labradore, latitude 54°.

Thus it appears, that the isothermal lines which are parallel to the equator for nearly 22o, afterwards deviate more and more; and from the observations of Sir Charles Giesecke in Greenland, of Mr. Scoresby in the Arctic seas, and also from those of Sir Edward Parry and Sir John Franklin, it is found that the isothermal lines of Europe and America entirely separate in the high latitudes, and surround two poles of maximum cold, one in America and the other in the north of Asia, neither of which coincides with the pole of the earth's rotation. These poles are both situate in about the eightieth parallel of north latitude; the Transatlantic pole is in the 100° of west longitude, about 5° to the north of Sir Graham Moore's Bay, in the Polar Seas, and the Asiatic pole is in the 95° of east longitude, a little to the north of the Bay of Taimura, near the Northeast Cape. According to the estimation of Sir David Brewster, from the observations of M. de Humboldt and Captains Parry and Scoresby, the mean annual temperature of the Asiatic pole is nearly 1o of Fahrenheit's thermometer, and that of the Transatlantic pole about 310 below zero, whereas he supposes the mean annual temperature of the pole of rotation to be 4° or 5°. It is believed that two corresponding poles of maximum cold exist in the southern hemisphere, though observations are wanting to trace the course of the southern isothermal lines with the same accuracy as the northern.

The isothermal lines, or such as pass through places where the mean annual temperature of the air is the same do not always coincide with the isogeothermal lines, which are those passing through places where the mean temperature of the ground is the same. The mean heat of the earth is determined from that of the springs, and if the