water.

Thus between the Canary and Cape Verde Islands, for about fifty miles off land, the wind blows almost constantly in shore, or in a direction exactly contrary to the trade winds: this is due to the intense heat of the Sahara, which causes an ascending current strong enough to draw in the air from the cooler Atlantic to supply its place.

A similar interruption occurs with uncertain winds on the west coast of America, which may be attributed to the vast chain of the Andes sheltering the coast beneath them like a wall. Other variations occur which cannot be enumerated in a sketch like the present. It may, however, be observed, that the central region of calms and that of the trade winds on either side move northwards or southwards, according as the sun is north or south of the equator. The variation is, however, comparatively small; for, while the sun travels as much as 23° north and an equal distance southward, the limits of the above-mentioned winds travel only through 5° or 6o, and of the central calm only through 2° or 3°. The southern limit of the calms, however, is always 1° or 2° north of the equator.

There are certain parts of the globe where this course of nature is interrupted. A large area from the east coast of Africa, north of Mozambique, to the Chinese Seas, and some parts of the Pacific, are subject to the Monsoons. These are alternate winds, blowing during part of the year, constantly in a direction uniform with that of the trade winds, and, during another part, as constantly in a direction exactly the reverse. The change is accompanied by violent storms, and the interval between the monsoons lasts about a month. The reverse monsoons, or those which blow from the south-west in the Indian Ocean and from the north-west on the coasts of Australia, have been attributed to the rarefaction of the air over Hindostan and the Birman Empire in the one case, and over the great plains of Australia in the other, and occur severally at the periods when those portions of the globe are most

heated, namely, the south-west monsoon from April to September in the Indian Ocean, or when the sun is in the northern hemisphere, and the north-west monsoon, from October to April on the coasts of Australia, when the sun has passed to the southward of the equator.

A similar cause may serve to explain other instances of periodical winds, as those on the east coast of South America, the north winds of the Mediterranean, the Khamsin of Egypt. It is probable that other causes combine with those above enumerated, as the daily and monthly fluctuations caused by the attraction of the sun and moon upon the great ocean of our atmosphere, the currents of heated water in the sea, &c. Some portions of the globe are peculiarly exposed to violent gales of wind called Typhoons or Hurricanes: these are of the nature of vast whirlwinds, and are found uniformly to revolve in the northern hemisphere the reverse way to that of the hands of a clock, i. e. from north to west, then south, then east. In the southern hemisphere, their revolution is from north to east, then south and west. While they have a rotatory motion of terrific violence, they have also a progressive movement from the region of calms northward, to the north of that neutral ground, and southward, to the south of it. The West Indian hurricanes, of which we know most, have a progressive track, first north-west, then north-eastward, perhaps deflected by the warm strata accompanying the gulf stream, and losing their fury in the coasts of the United States, or in the Atlantic. They are accompanied by violent electric phenomena, and we have but recently ascertained their laws and Their causes still remain obscure.

nature.

We proceed to mention some of the more remarkable ocean currents. First near the equator, throughout the ocean, there is found a current flowing from east to west, which is supposed to be owing to the trade winds in conjunction with the revolution of the earth in a contrary direction. The centre of this current strikes the north-east angle of South

America, which divides it, sending part of the water southward and part north-westward along its shores. The southern current forms a counter stream to the equatorial current, and returns eastward, a little south of the Cape of Good Hope and South Australia. The northern branch flows into the Gulf of Mexico, where it is forced up against the shores of Central America, so as to raise the surface of the ocean in those parts, and the accumulated waters escape northward into the Atlantic round the coast of Florida, and flow north-eastward to the coasts of Norway and Spitzbergen. Its waters are greatly heated before they leave the Mexican Gulf, even to a temperature of 85°-88°, and carry no small portion of their warmth far northward, thus rendering the climate of the shores they wash much milder than the latitude would lead us to expect. This is the reason of the great upward bend of the isothermal lines about North Cape, and to this cause also we must attribute the circumstance of the maximum of temperature lying so far north of the equator in South America, just as the plains of Hindostan, Arabia, and the Sahara cause it to rise northward in Asia and Africa. A cold current on the other hand, from Baffin's Bay, sets southward by the coasts of Labrador and Newfoundland, causing a colder temperature than might have been expected in the former, and continual fogs arising from the meeting of the warm air of the gulf stream with the cold air of the Arctic current, about the latter. An eastern current of considerable importance flows from the Antarctic regions to the coasts of South America about Chiloe, where it divides, part rushing northward to supply the equatorial current, and part southward, escaping round Cape Horn. In like manner it will be seen, by inspection of the map, how the general western stream, when checked by Borneo and the Philippines, partly escapes through the Straits of Sunda, and partly is deflected from the Chinese coasts, forming a strong current north-eastward along the coasts of the Japanese Islands, while on striking the east coast

of Africa, above Madagascar, the whole body of water is turned southward, by meeting with the current from the Indian Ocean which has that direction, and flowing round the Cape, between it and the counter stream mentioned above, as proceeding from the coasts of South America, no doubt contributes to the dangerous sea that is frequently met with in those parts.

We must not omit to mention the influence upon our globe of the amount of rain. In some districts, as in the Sahara, parts of Arabia and Persia, and the desert of Gobi, the interior of Mexico, and the west coasts of Peru, rain is almost unknown. Vegetation is either wanting, or of the most scanty character, and men and other animals can only live in the portions watered by feeble streams or wells, whose supply is very precarious. Under the equator, or a little north of it, the continual precipitation generates an excess of moisture, which renders the climate unhealthy, and produces fevers and agues hence the unhealthiness of Sierra Leone, the West Indies, and the coasts of Brazil. Other parts of the world enjoy a more moderate climate. In the Arctic and Antarctic regions, rain seldom or never falls, but its place is supplied by snow: this also occurs in high mountains, upon which it is supposed that no rain ever falls above a certain altitude varying with the latitude of the place. The abundant daily rains of the equatorial district, which is however, as before observed, wholly north of the equator, probably arise from the upward current loaded with vapour from the heated ocean becoming rapidly cooled as it ascends, and being then unable to retain its moisture, which is precipitated in deluges of rain. The upward current, on the other hand, from the heated and dry Sahara, even when cooled, is able to retain the little moisture it possessed. The same is true to a greater or less degree of all extensive plains, and the rainless district would extend unbroken from the edge of the Sahara to China, were it not that the Red Sea and Persian Gulf afford sources

for evaporation in two instances, and the Himalaya Mountains with their perpetual snow in a third. It may be asked, why then is not the great plain of Brazil a rainless district? But this arises from the perpetual east winds, which come loaded. with moisture from the sea, and blow across the continent to the Andes. The cold of those elevated summits precipitates the whole remaining moisture, which may have travelled so far inland; and the Pacific coasts are nearly without rain. The rainless district of Mexico is, probably, owing to its being without the influence of the trade winds, which are interrupted in the northern part of the Mexican gulf. It may be added, that the eastern part of a tropical country will receive most rain, for the trade winds carry the vapours of the ocean thither; but the western coast of a temperate region has the most rain, from the prevalence of west winds, which blow from the sea, and bring its vapours. Thus, Ireland and Devonshire have more rain than Norfolk and Kent; but the Brazilian shores and Mozambique more than the coasts of Congo and Chili. Snow is only frozen vesicles of vapour, and will always fall instead of rain, where the temperature is such that the atmosphere never rises above 32° in the regions of the sky from whence the vapour is precipitated.

Vegetation is greatly affected by the amount of warmth and moisture. Their joint operation produces the luxuriant forests of Brazil. The absence of moisture renders the Sahara a desert, the absence of heat renders Iceland and Labrador almost equally barren. Some soils, doubtless, are more fertile than others, but the finest soil, without these requisites, is fruitless, and the poorest soil, with them, will throw up a more or less luxuriant vegetation. It is impossible, in a sketch like this, to indicate even a general outline of the botanical features to which these considerations point, but the reader is referred to the map (M. G. Pl. II.) for the limits of the principal plants most important to mankind. The occasional variations in the degree north or south to which.

C