676 AMERICA AND S. AMERICA. Mazatlan, in latitude 23 N., is well watered, because, jirst, the continent here is narrow; secondly, the table-land of Mexico, which is much lower than the Andes of Chili, is not so effectual a screen to intercept the moisture; and, thirdly, there is reason to believe that a branch of the trade-wind, which crosses the low part of the continent at Panama and Nicaragua, sweeps along the west coast during part of the year, and transports humidity with it. But beyond the point we have mentioned drought prevails. Sonora, though visited occasionally by rains, consists of sandy plains without herbage, where the streams lose them selves in the parched soil without reaching the sea ; and even Old California, which has the ocean on one side, and a broad gulf on the other, and ought apparently to be ex cessively humid, is covered with sterile rocks and sandy hills, where the vegetation is scanty, and no timber is seen except brushwood. This dry region extends as far as 33 or 34 N. ; but immediately beyond this we have another change of scene. New California is in all respects a contrast to the Old. It is rich, fertile, and humid, abounding in luxuriant forests and fine pastures; and the other American possessions to the northward preserve the same character. How can we account for this singular diversity of climate, except upon the principle which has been explained, namely, that in all regions where ranges of mountains intersect the course of the constant or pre dominant winds, the country on the windward side of the mountains will be moist, and that on the leeward dry; and hence parched deserts will generally be found on the west side of countries within the tropics, and on the east side of those beyond them 1 Our hypothesis applies equally to the country east of the Rocky Mountains. For the space of about 3000 miles along the foot of this chain the surface consists of dry sands or gravel, sometimes covered with saline incrustations, almost destitute of trees and herbage, and watered by streams flowing from the mountains, which are sometimes entirely absorbed by the arid soil. The central and eastern part of the basin of the Mississippi would in all probability have been equally barren had the configuration of the land been a little different in the south. A tract of country extremely low and level extends along both sides of this river; and a portion of the trade- wind blowing from the Mexican Gulf, finding its motion westward obstructed by the high table-land of the Cordil lera, is deflected to the right, and ascends the valley of the Mississippi and Ohio. This wind, whose course was first traced by Volney, bears with it the humidity of the torrid zone, and scatters fertility over a wide region that would otherwise be a barren waste. The views on the subject of climate we have been un folding will enable us to throw some light on an interest ing point the distribution of forests. We are induced to think, that in all countries having a summer heat exceed ing 70, the presence or absence of natural woods, and their greater or less luxuriance, may be taken as a measure of the amount of humidity, and of the fertility of the soil. Short and heavy rains in a warm country will produce grass, which, having its roots near the surface, springs up in a few days, and withers when the moisture is exhausted ; but transitory rains, however heavy, will not nourish trees, because after the surface is saturated with water, the rest runs off, and the moisture lodged in the soil neither sinks deep enough, nor is in sufficient quantity to furnish the giants of the forest with the necessary sustenance. It may be assumed that 20 inches of rain falling moderately, or at intervals, will leave a greater permanent supply in the soil than 40 inches falling, as it sometimes does in the torrid zone, in as many hours. It is only necessary to qualify this conclusion by stating, that something depends on the subsoil. If that is gravel, or a rock full of fissures, the water imbedded will soon drain ofl ; if it is clay or a com pact rock, the water will remain in the soil. It must be remembered, also, that both heat and moisture diminish as we ascend in the at mosphere, while eva poration increases ; and hence that trees will not grow on very high ground, though its position in reference to the sea and the prevail ing winds should be favourable in other respects. In speak ing of the region of forests, we neither restrict the term to those districts where the natural woods present an unbroken continuity, nor ex tend it to every place where a few trees grow in open plains. Sketch Map showing the Forest Regions of America. It is not easy to give a definition that will be always appropriate ; but in using the expression, we wish to be understood as applying it to ground where the natural woods cover more than one-fourth of the surface. The small map of America prefixed will enable the reader to follow our statements with ease. The long black lines show the positions of the chains of mountains ; the shading represents the regions of forests ; the dense forests being marked by the double shading, and the thinner ones by the open lines. The white spaces represent the lands on which little or no wood grows. The equator and the parallel of 30 on each side are indicated by the horizontal lines marked and 30. The arrows show the direction of the prevailing winds; but it must be remembered that, though the intertropical wind is assumed to have its course right from the east, this is only true at the equator, its direction inclining to north-east as we approach the northern tropic, and to the south-east as we approach the soiithern. In North America A is the woody region on the west coast, extending from latitude 35 to about 58, and of unknown breadth. B, the region on the east side of the Ilocky Mountains, partly a bare desert, partly covered with grass and dotted with trees. C, the forests of the Alleghany chain, thick on the east and south, and thin on the west; bounded by a curved line passing from St Luis, in Mexico, through Lake Huron, to the mouth of the St Lawrence, in latitude 50 N. The arrow at M points out the direction of the wind, which ascends the valley of the Mississippi, and nourishes the western part of these forests ; and the arrow at E, that which blows across the isthmus of Panama. D is the table-land of Mexico, graduating on the north west into the dry plains of Sonora and California, all bare, or nearly bare, of wood. E is the Llanos or bare plains of Caraccas, nearly fenced round with mountains. F G is the lovig strip of bare dry sands on the west side of the Andes which constitutes Lower Peru and the north part of Chili ; and N is Amatapu, its northern boundary. H is the great region of forests which constitutes the basin of the Amazon, and occupies all the rest of Brazil. Near the equator the moisture is so excessive, that after 150 or 200 inches of rain have fallen on the east coast, there is still sufficient humidity in the atmosphere to afford copious
showers to all the country up to the Andes. Here, there-
