The trade-winds are only met with on the sea; but iu'sorne coun tries of the globe between the tropics, or near them, regular and constant easterly winds occur, which may owe their origin to the same cause. These winds only occur in extensive level plains, where there is nothing to break their force or to change their direction ; for if the wind comes in contact with high land or mountains, its regular pro gress is obstructed. But over a considerable tract of low level land the wind primes without being much changed in its direction and velocity, particularly if the land be barren and destitute of moisture. In the Sabel, or western part of the Sahara, an easterly wind blows all the year round with great force, but In the eastern district of the Great Desert It is leas constant and less violent, so that in all respects it may be compared with a trade-wind. An easterly wind is also always found on the plain drained by time Amazonas; and by its assistance the voyage against the strong current of the river may be accomplished nearly in the sante time as the voyage downwards by means of time current. Humboldt found that this easterly wind, which, near time mouth of the Amazonas is moderate, has acquired such a force at the base of the Andes, that it is almost impossible to keep one's footing against it. A similar easterly wind, though of less strength, is found in the great plain which is traversed by the lower course of the Orinoco.
Time countries just mentioned, io which these easterly winds blow constantly, are contiguous to those parts of the Atlantic Ocean where trade-winds in general are regular all the year round. But the trade winds of the ocean and the land-winds of the plaina do not come into contact with one another. They are aeparated by a tract of the globe in which other winds prevail. This tract lies within the ocean, and extends along the coasts of the continonta; its width varies greatly. Where it lies east of the trade-wind it is usually a hundred miles wide, but it is of inconsiderable breadth when the land lies to the west of the trade-winds. The continuity of the easterly winds is evidently interrupted by the difference of the temperature of the air incumbent on the sea and on the land. This difference changes with the seasons, the air over the land being hotter than that of the sea when the sun Is near, and colder when it is far off. Hence it follows that during the first period the wind blows from the sea to the land, and in the sweat from the land to the sea. Thus a kind of monsoon is produced along the coasts of the continents, oven within the region of the tmade•wiuda. A large island in such a situation is therefore surrounded by winds blowing from all quarters. When the land of Australia is heated by the presence of the sun in the southern hemisphere, the wind generally blows from the westward upon the north-western coast, and from the south-west upon the western coast ; from the south-west, south, and south-cast on the southern coast; and from the south-east and east upon the east coast. In the opposite season, however, the winds are less regular, because the greatest part of the island is then without the reach of the trade-winds.
The trade-winds occur on both sides of the equator in the Atlantic and Pacific oceans, but they vary considerably iu extent and force in both oceans. Some account of this difference is given under the heads of ATLANTIC OCEAN and PACIFIC OCEAN, in Grroo. Div. We shall
here add, respecting the lastaucutioned sea, that the trade-winds in the southern parts appear to be subject to great changes in direction and force, and that they properly occur only along the coasts of South America, where a constant south-easterly mind is met with at the dis tance of 500 to 600 miles from the coast : but in the middle of the Southern Pacific the trade-wind seems by no means regular aud con stant. Admiral Fitzroy, in speaking of the Paantuto Islands, or the Dangerous Archipelago of the Low Islands, says that among them a steady south-easterly trade-wind prevails from March to October. but that in the rainy season, from October to March, westerly winds, squalls, and rains are frequent ; and in the abstract of his meteoro logical journal we find that in runuing from the Galapagos to Otalleite he experienced only south-eastern winds near the equator ; and that in the remainder of his voyage the wind blew almost constantly from the north-east, north-north-east, or north .east by east. Kotzebue, on his first voyage, observed it as a remarkable circumstance, that between Easter Island and 51' S. lat. he met only with winds blowing front north, north-east, and east-north-east. This anomaly in the trade winds of the Southern Pacific is probably produced by the innumerable islands and comp rocks which cover that ocean between the equator and the southern tropic, and extend from 130° W. long. to the coast of Australia. Horsburgh says, probably from his own observation, that " where shoal coral-banks shoot np out of the deep water in many places between the tropics, a decrease of the prevailing wind is fre quently experienced; for when a sternly wind is lulowiug over the sur face of the deep water, no sooner does a ship get upon the verge of a shoal coral-bank than a sudden decrease of the wind is often perceived. This is, in his opinion, occasioned by the atmosphere over these banks being less rarefied by the increased evaporation than that over the deep water, and consequently not requiring so great a supply of air to restore the equilibrium as the circumjacent parts, which are more rarefied and heated." When such effects, according to the statement of this intelligent hydrographer, are produced by single coral-banks in the midst of the ocean, we may easily comprehend that their number aud immense extent in the Southern Pacific not only diminish their force, but change the direction of the trade•winds, and that these reefs and islands affect them nearly to the same extent as a large continent. It appears that in the Southern Pacific the trade-winds are replaced by the north-eastern, northern, and western winds only during the period when the sun is in the southern hemisphere. A south-eastern trade wind prevails also in the Indian Ocean from within a few degrees of the eastern side of Madagascar nearly to the coasts of Australia, between the parallels of 10' and 23° S. lat.; but in this ocean front 10° S. lat. to the coasts of 11 indostan the winds are periodical ; the influence of the land issuing in a complete reversal of the north-mat trade during rt considerable portion of the year, aud the production of MoNsooxs, that is of winds which blow half the year in one, and the other half in the coutrary direction, as explained at length in the article devoted to them.