agitated. The progress of the fire is so slow, and the heat so great, that every combustible object in its course is consumed. Wo to the farmer whose ripe cornfields extend into the prairie, and who suffers the tall grass to grow in contact with his fences! The whole labor of the year is swept away in a few hours. But such accidents are comparatively unfrequent, as the preventive is simple, and easily applied.

It will be readily seen, that as soon as these fires commenced, all the young timber within their range must have been destroyed. The whole state of Illinois, being one vast plain, the fires kindled in different places, would sweep over the whole surface, with a few exceptions, of which we are now to speak. In the bottom-lands, and along the margins of streams, the grass and herbage remain green until late in the autumn, owing to the moisture of the soil. Here the fire would stop for want of fuel, and the shrubs would thus escape from year to year, and the outer bark acquire sufficient hardness to protect the inner and more vital parts of the tree. The margins of the streams would thus become fringed with thickets, which, by shading the ground, would destroy the grass, while it would prevent the moisture of the soil from being rapidly evaporated, so that even the fallen leaves would never become so thoroughly dry as the grass of the prairies, and the fire here would find comparatively little fuel. These thickets grow up into strips of forests, which continue to extend until they reach the high table-land of the prairie; and so true is this, in fact, that we see the timber now, not only covering all the bottom-lands and hill sides, skirting the streams, but wherever a ravine or hollow extends from the low grounds up into the plain, these are filled with young timber of more recent growth. But the moment we leave the level plane of the country, we see the evidences of a continual struggle between the forest and the prairie. At one place, where the fire has on some occasion burned with greater fierceness than usual, it has successfully assailed the edges of the forest, and made deep inroads; at another, the forest has pushed out long points or capes into the prairie.

It has been suggested that the prairies were caused by hurricanes, which had blown down the timber and left it in a condition to be consumed by fire, after it was dried by laying on the ground. A single glance at the immense region in which the prairie surface predominates, must refute this idea. Hurricanes are quite limited in their sphere of action. Although they sometimes extend for miles in length, their track is always narrow, and often but a few hundred yards in breadth. It is a well known fact, that wherever the timber has been thus prostrated, a dense and tangled thicket shoots up immediately, and, protected by the fallen trees, grows with uncommon vigor.

'Some have imagined that our prairies have been lakes; but this hypothesis is not tenable. If the whole state of Illinois is imagined to have been one lake, it ought to be shown that it has a general concavity of surface. But so far from this being true, the contrary is the fact; the highest parts of the state are in its centre. If we suppose, as some assert, that each prairie was once a lake, we are met by the same objection; as a general rule, the prairies are highest in the middle, and have a gradual declivity towards the sides; and when we reach the timber, instead of finding banks corresponding with the shores of a lake, we almost invariably find valleys, ravines, and water-courses depressed considerably below the general level of the plain.

'Wherever hills are found rising above the common plane of the country, they are clothed with timber; and the same fact is true of all broken lands. This fact affords additional evidence in support of our theory. Most of the land in such situations is poor; the grass would be short, and if burned at all, would occasion but little heat. In other spots, the progress of the fire would be checked by rocks and ravines; and in no case would there be that accumulation of dry material which is found on the fertile plain, nor that broad, unbroken surface, and free exposure, which are necessary to afford full scope to the devouring element.

'By those who have never seen this region, a very tolerable idea may be formed of the manner in which the prairie and forest alternate, by drawing a colored line of irregular thickness, along the edges of all the water-courses laid down on the map. This border would generally vary from one to five or six miles, and often extend to twelve. As the streams approach each other, these borders would approach or come in contact; and ull the intermediate spaces not thus colored would be prairie. It would be seen that in the point formed by the junction of the Ohio and Mississippi, the forest would cover all the ground; and that, as these rivers diverge, and their tributaries spread out, the prairies would predominate.'

Between the Platte river, and the head-waters of the Colorado and Sabine rivers, there is an extensive desert tract, which has been called the Great American Desert, stretching from the Ozark Mountains to the Chippewan. Over this desert the members of Long's expedition travelled nearly a thousand miles. The intense reflection of light and heat, from this tract, added much to the fatigue and suffering of their journey. We often met with extensive districts covered entirely with loose and fine sand, blown from the adjacent hills. In the low plains along the river where the soil is permanent, it is highly impregnated with saline substances, and too sterile to produce any thing except a few stinted carices and rushes.' As we approached the mountains, we felt or fancied a very manifest change in the character of the weather, and the temperature of the air. Mornings and evenings were usually calm, and the heat more oppressive than in the middle of the day. Early in the forenoon, a light and refreshing breeze often sprung up, blowing from the west or south-west, which again subsided on the approach of night. This phenomenon was so often observed, that we were induced to attribute it to the operation of the same local cause, which in the neighborhood of the sea produces a diurnal change in the winds, which blow alternately to and from the shore. The Rocky Mountains may be considered as forming the shore of that sea of sand, which is traversed by the Platte, and extends northward to the Missouri above the great bend. The rarefaction of the air over this great plain, by the reverberation of the sun's rays during the day, causes an ascending current, which is supplied by the rushing down of the condensed air from the mountains. **** For several days the sky had been clear, and in the morning we had observed an unusual degree of transparency in every part of the atmosphere. As the day advanced, and the heat of the sun began to be felt, such quantities of vapor were seen to ascend from every part of the plain, that all objects at a little distance appeared magnified, and variously distorted. An undulating and tremulous motion in ascending lines was manifest over every part of the surface. Commencing soon after sunrise it continued to increase in quantity until the afternoon, when it

diminished gradually, keeping an even pace with the intensity of the sun's heat. The density of the vapor was often such as to produce the perfect image of a pool of water in every valley upon which we could look down at an angle of about ten degrees. This aspect was several times seen so perfect and beautiful as to deceive almost every one of our party. A herd of bisons, at the distance of a mile, seemed to be standing in a pool of water, and what appeared to us the reflected image was as distinctly seen as the animal itself.* Illusions of this kind are common in the African and Asiatic deserts, as we learn from travellers and from the language of poets.'

The Pine Plains are a district of sandy alluvion, bounded by the gravelly soil of Guilderland and Duanesburgh on the south-west, and by the river alluvions of Niskayuna and Watervliet, on the north-east, and covering an area of about seventy square miles. This tract is included in a triangle formed by the junction of the Mohawk with the Hudson, and of which the Helleberg, a lofty chain of highlands, visible from the plains at the distance of twenty miles, forms the south-western boundary. Situated near the centre of a state, computed at forty thousand square miles, and containing a population of nearly two million souls, this tract presents the topographical novelty of an unreclaimed desert, in the heart of one of the oldest counties in the state, and in the midst of a people characterized for enterprise and public spirit. Several attempts have lately been made to bring this tract into cultivation, and from the success which has attended the introduction of gypsum, and other improved modes of agriculture, it is probable the whole will, at some future period, be devoted to the cultivation of the various species of grasses, fruit trees, and esculent roots; three branches of agriculture to which its sandy soil seems admirably adapted.

GENERAL REMARKS ON PLAINS AND PRAIRIES.

Plains like valleys are of two classes; the high plains, which are found between two chains of mountains, are frequently of great extent, and are placed as it were upon the shoulders of secondary mountains; such are the elevated plains of Tartary, of Persia, and probably of the interior of Africa. The plains of Quito are twelve thousand feet above the level of the sea; those of Karakorum, in Chinese Mongolia, are probably as elevated. The low plains, whose soil is composed of sand, gravel and shells, seem formerly to have been the basins of interior seas. Such are the plains on the north side of the Caspian, the large plain to the south of the Baltic, and that through which the river of the Amazon flows; the Tehama of Arabia, the Delta of Egypt, and others of a similar nature, which seem to have been once covered by the waters of the ocean and its gulfs. The immense plains covered with grass, called prairies in the United States, are the steppes of Asia, and the pampas of South America.

It is common in our own country, says the London Monthly Review, for ground mists to assume the appearance of water, to make a meadow seem inundated, and to change a valley into a lake; but these mists never reflect the surrounding trees and hills. Hence the mirage must consist of a peculiar gas, of which the particles are combined by a stronger attraction of cohesion than the vapors of real water; the liquor silicum of the alchemists is described as exhibiting in some circumstances this glossy surface, yet as being equally evanescent.

CHAPTER IV.-RIVERS.

ALL the rivers of the United States, of the first magnitude, have their sources, either in the Rocky Mountains, or in elevated spurs projecting from the sides of that range. Many of the rivers which descend from the western sides of the Alleghanies are of inconsiderable volume, and by no means remarkable for the rapidity or the directness of their course. Those which flow from the eastern and southern sides of these mountains are worthy of extended description, even in the same pages with the great tributaries of the Mississippi. They afford the advantages of a good inland navigation to most parts of the states.

I. RIVERS WHICH FLOW INTO THE MISSISSIPPI, AND THE GULF OF MEXICO.

The Mississippi with its branches drains the great central basin which lies between the Alleghany and the Rocky Mountains. This river has its rise in the table-lands within the territories of the United States, in north latitude forty-seven degrees and forty-seven minutes, at an altitude of thirteen hundred and thirty feet above the Atlantic, though the country at its source appears like a vast marshy valley. Mr. Schoolcraft fixes it in Cassina Lake, which is situated seventeen degrees north of the Balize on the gulf of Mexico, and two thousand nine hundred and seventy-eight miles, pursuing the course of the river. Estimating the distance to Lake La Beesh, its extreme north-western inlet at sixty miles, we have a result of three thousand and thirty-eight miles as the entire length of this wonderful river. Mr. Schoolcraft, in his very interesting Journal of Travels, observed that he believed there was no one then living, beside himself, who had visited both the sources and the mouth of this celebrated stream. As the description furnished by this gentleman is the clearest and most complete that we find, we have taken the liberty to transfer it to our pages, without mutilation :

'In deciding upon the physical character of the Mississippi, it may be advantageously considered under four natural divisions, as indicated by the permanent differences in the color of its waters-the geological character of its bed and banks,-its forest trees and other vegetable productions,—its velocity, the difficulties it opposes to navigation, and other natural appearances and circumstances.

'Originating in a region of lakes, upon the table-lands, which throw their waters north into Hudson's Bay,-south into the gulf of Mexico,—and east into the gulf of St. Lawrence-it pursues its course to the falls of Peckagama, a distance of two hundred and thirty miles, through a low prairie, covered with wild rice, rushes, sword grass, and other aquatic plants. During this distance, it is extremely devious as to course and width, sometimes expanding into small lakes, at others, narrowing into a channel of about eighty feet. It is about sixty feet wide on its exit from Red Cedar or Cassina Lake, with an average depth of two feet; but from the junction of the Leech Lake fork, increases to a hundred feet in width, with a corresponding increase of depth Its current, during this distance

still and gentle; and its mean velocity may be estimated at a mile and a half per hour, with a descent of three inches per mile. This is the favorite resort of water-fowl, and amphibious quadrupeds.

'At the falls of Peckagama, the first rock stratum, and the first wooded island, is seen. Here the river has a fall of twenty feet; and from this to the falls of St. Anthony, a distance of six hundred and eighty-five miles, exhibits its second characteristic division. At the head of the falls of Peckagama, the prairies entirely cease; and below, a forest of elm, maple, birch, oak, and ash, overshadows the stream. The black walnut is first seen below Sandy Lake river, and the sycamore below the river De Corbeau. The river, in this distance, has innumerable well wooded islands, and receives a number of tributaries, the largest of which is the river De Corbeau, its great south-western fork. The Pine, Elk, Sac, and Crow rivers, also enter on the west, and the St. Francis and Missisawgaiegon, on the east. The course of the river, although serpentine, is less so, than above the falls of Peckagama, and its bends are not so short and abrupt. Its mean width may be estimated at three hundred feet until the junction of the De Corbeau, and below that at two hundred and fifty yards. Its navigation is impeded, agreeably to a memorandum which I have kept, by thirtyfive rapids, nineteen ripples, and two minor falls, called the Little and the Big Falls, in all of which the river has an aggregate descent of two hundred and twenty-four feet in fourteen thousand six hundred and forty yards, or about eight miles. The mean fall of the current, exclusive of the rapids, may be computed at six inches per mile, and its velocity at three miles per hour. In the course of this distance it receives several small turbid streams, and acquires a brownish hue, but still preserves its transpa rency, and is palatable drink-water. A few miles above the river Corbeau, on the east side, we observe the first dry prairies, or natural meadows, and they continue to the falls of St. Anthony. These prairies are the great resort of the buffalo, elk, and deer, and are the only parts of the banks of the Mississippi where the buffalo is now to be found. Granite rocks appear at several of the rapids, in rolled pieces, and in beds; and in some places attain an elevation of one or two hundred feet above the level of the water, but the banks of the river are generally alluvial.

At the falls of St. Anthony, the river has a perpendicular pitch of forty feet, and from this to its junction with the Missouri, a distance of eight hundred and forty-three miles, it is bounded by limestone bluffs, which attain various elevations from one to four hundred feet, and present a succession of the most sublime and picturesque views. This forms the third characteristic change of the Mississippi. The river prairies cease, and the rocky bluffs commence precisely at the falls of St. Anthony. Nine miles below it receives the St. Peter's from the west, and is successively swelled on that side by the Ocano, Iowa, Turkey, Desmoines, and Salt rivers, and on the east by the St. Croix, Chippeway, Black, Ouisconsin, Rock, and Illinois. One hundred miles below the falls of St. Anthony, the river expands into a lake, called Pepin, which is twenty-four miles long and four in width. It is, on issuing from this lake, that the river first exhibits, in a striking manner, those extensive and moving sandbars, innumerable islands and channels, and drifts and snags, which continue to characterize it to the ocean. Its bends from this point onward are larger, and its course more direct; and although its waters are adulterated