South of the Arkansas are rocks of basaltic origin, overlaying the red sandstone. By the vastness and broken character of their masses, and their dark color, they present a striking contrast to the light, smooth and fissile sandstone on which they rest. Sometimes they are compact and apparently homogeneous in their composition, and in many particulars of structure, form and hardness, more analogous to the primitive rock than to those recent secondary aggregates with which they are associated. In other instances, dark and irregular masses of porous and amygdaloidal substances are seen scattered about the plain, or gathered in conical heaps, but having no immediate connection with the strata on which they rest. Most of the rocks of this class were observed in the neighborhood of the sources of the Canadian; and may be distinguished into two kinds, referable to the two divisions called greenstone and amygdaloid.

The valley immediately east of the Rocky mountain range is composed of an extensive accumulation of sand, seemingly the debris of the mountains.To an unknown depth, the soil is made up of rounded fragments of granite, varying in dimension from a grain of sand to a six pound shot. This accumulation has evidently been washed from the mountains, and slopes gradually from their base. The small particles derived from the quartzose portions of the primitive aggregates, being least liable to decomposition, have been borne to the greatest distance, and of these the eastern margin of the great sandy desert is almost entirely composed; the central portions are of coarser sand, intermixed with particles of mica and feldspar; nearer the mountains, boulders and pebbles occur abundantly, and at length cover almost the entire surface of the country.

In many other respects besides geological structure, the Apalachian range of mountains differs from that we have just been considering. The whole of their eastern front is composed of primitive rocks, comprehending both the granitic family and its associated strata of clay-slate and limestone. In New England, rocks of this class constitute the seacoast, and with some exceptions extend inwards towards the St. Lawrence. South of the Hudson, the edge of the primitive follows the general contour of the mountains, at a variable distance from the sea to their termination, and until it meets more recent deposits at the extremity of the mountain range. The breadth of this primitive belt is very unequal. In passing through the states of Pennsylvania and Maryland, it occupies but a small part of the country; in Virginia it increases in breadth, and proportionably in height, composing the greatest mass as well as the most elevated points of the mountains in Georgia and North Carolina. Besides this range, there is a great mass of primitive on the west side of lake Champlain.

In general, the primitive rocks run from a north and south to a north-east and south-west direction, and dip generally to the south-east at an angle of more than forty-five degrees with the horizon; their highest elevation is towards their north-western limit. The mountains of this formation consist generally of detached masses, with rounded flat tops and a circular waving outline. Granite in large masses constitutes but a small part of this formation, and is found indifferently in the plains and on the tops of mountains. Gneiss extends perhaps over a half of this formation, and includes in a great many places beds from three to three hundred feet thick. These beds are mixed, and alternate occasionally in the same gneiss with the primitive limestone, the beds of hornblende and hornblende slate, serpentine, magnetic iron ore, and feldspar rocks. In short, there are scarcely any of the primitive rocks that may not occasionally be found included in the gneiss formation.

The breadth of the transition district, like that of the primitive, is variable. Narrow towards the gulf of Mexico, it gradually widens towards the north-east, till it reaches the river Hudson. From its upper portion it sends off a considerable arm, which penetrates for several hundred miles into the granitic region, overlaying it, but running parallel with the principal body. After the primitive, it forms some of the highest mountains in the range, and seems to be both higher and wider to the west in Pennsylvania, Maryland and part of Virginia, where the primitive is least extended and lowest in height. It contains all the varieties of rocks found in the same formation in Europe.

It varies in breadth from twenty to one hundred miles. In the limestone of this formation there are many and extensive caves, some of whichextend for miles under ground, and contain the bones of animals. It is the lowest, and is considered the most ancient of the rocks containing organized remains, which are those of cryptogamous plants, and animals without sight. The graywacke has been observed to contain impressions of organized remains, but they are usually those of zoophytic animals, and are exceedingly unlike those found so abundantly in the coal formations. Its colors are variable; it is, however, most commonly bluish, black, or dark brown. The graywacke seems to form the connecting link between the clay-slate and a rock which has been called the old red sandstone, and is usually found intimately blended either with the one or the other. This sandstone occurs throughout the whole extent of the transition formation, and evidently belongs to the oldest depositions of that rock. It is for the most part distinctly stratified, and in all cases its stratification is inclined.

Of the rocks thus described, the limestone occurs extensively all along the north-western side of the primitive strata. It is probable that transition limestone is the foundation through their whole extent of the Alleghany mountains of Pennsylvania, Maryland and the western parts of Virginia, on a level with the surface at the base of their eastern declivities. The clay-slate occurs in the central portions of that extensive field of transition, which skirts the western margin of the primitive of New York and New England, and forms the great body of the Catskill mountains. The old red sandstone in the transition district, along the whole range of mountains, is perhaps more abundant than any other aggregate. This region has also a considerable mixture of trap. Various large bodies of transition rock are thrown to a considerable distance into the primitive region; while in many instances, secondary rocks are found running along the valleys far into the bosom of the mountains.

With the edge of the transition strata, we approach the western summits of the Apalachian mountains, or the line from whence they begin to fall toward the Mississippi valley. Along this line commences a series of secondary rocks, stretching westward to an immense extent towards the Mississippi and the lakes, and constituting one of the most interesting and important geological formations in the United States. This secondary region extends unbroken across the whole country to the shores of the lakes, being bounded on the west probably by the river Wabash, and in descending the Mississippi by the more recent formations through which that river flows. It consists generally of various strata of sandstone, limestone and clay. Immense beds of secondary limestone, of all shades from light blue to black, sometimes intercepted by extensive tracts of sandstone and other secondary aggregates, appear to constitute the foundation of this formation, which extends from the head waters of the Ohio, with some interruptions, all the way to the waters of the Tombigbee, accompanied by slaty clay and freestone with vegetable impressions; but in no instance yet ascertained, covered by or alternating with any rock resembling basalt, or indeed any of those called the newest floetz trap formation. A grand peculiarity of this secondary region is the uniform, horizontal direction of the strata.

We will now briefly examine the region which occupies the centre of the Mississippi valley. The Ozark mountains consist chiefly of secondary and transition rocks; but there are two points at which the primitive makes its appearance. About fifteen miles south-east from the hot springs, nearthe Washita, granite is found in situ. It is very soft, and disintegrates rapidly when exposed to the air. It is compounded of greyish-white quartz, yellowish-white feldspar, and an unusually large proportion of mica in variously and brilliantly-colored masses. This granite, if of secondary formation, is much more extensive than any of the kind hitherto known. ‘We are ignorant,’ says Dr. James, ‘of the manner of its connection with any other rock, nor do we know of any formation of primitive granite from which it could, by the action of water, have been derived: one can have no hesitation, however, in considering the Ozark mountains as a separate system within themselves, and having no immediate connection with either the Apalachian or the Chippewayan mountains.’ Mr. Schoolcraft mentions another granite region as occurring in the north-eastern extremity of the Ozark range, in the mining district of Potosi.

In connection with the granite of the Washita is found a stratum of clay-slate, and another of transition sandstone, but neither of them of great extent. The hot springs of the Washita issue from the clay-slate, and it is supposed that a very large mass of clay-slate is interposed between the surface of the granite and the point at which the springs rise. The slate-rock about the hot springs is highly inclined, often flinty in its composition, and, as far as it has been hitherto examined, contains no organic remains. It is traversed by large upright veins, usually filled with white quartz. The mountains contain vast beds of secondary limestone, which from its peculiar crystalline appearance might be easily mistaken for the primitive. These vast beds of sparry limestone, almost exclusively made up of deposits from chemical solution, would seem to have been formed during periods of great tranquillity in the waters. The sandstones of this small group of mountains appear under almost every variety of character. A region similar in mineralogical character to the Ozark mountains extends northward from the confluence of the Missouri, to the Ouisconsin and Ontonagon rivers of lake Superior. The sandstones, limestones and other rocks have a striking resemblance. Of the Black mountains in the north-western part of the Mississippi valley, but little is known; they appear to be composed of sandstone lying horizontally, and to be destitute of valuable minerals. Between these mountains and the central district, is a wide alluvial tract containing the course of the Missouri. The same appellation has been given by Dr. James to a space between the Ozark mountains and the Chippewayan sands, and to the country on both sides of the lower Mississippi.