The great flowering hibiscus is here a conspicuous and highly ornamental plant among the scattering trees in the low grounds. The occurrence of the black walnut, for the first time since we left the Missouri, indicates a soil somewhat adapted to the purposes of agriculture. Portions of the river valley, which are not covered with loose sands, have a red soil, resulting from the disintegration of the prevailing rocks (red sandstone and gypsum) intermixed with clay, and are covered with a dense growth of fine and nutritious grasses. Extensive tracts of the great woodless plain, at a distance from the river, appear to be based upon a more compact variety of sandstone, which is usually of a dark gray colour, and less pervious to water than the red. For this reason some copious springs are found upon it, and a soil by no means destitute of fertility, yielding sustenance to inconceivable numbers of herbivorous animals, and through them to innumerable birds and beasts of prey. It must be supposed, however, that the herds of bisons daily seen about the river, range over a much greater extent of country than was comprised within our limited views. The want of water in many places may compel them to resort {318} frequently to the river in dry weather; though at other times they may be dispersed in the high plains. August 18th. In speaking of a country whose geography is so little known as that of the region S. W. of the Arkansa, we feel very sensibly the want of ascertained and fixed points of reference. Were we to designate the locality of a mineral, or any other interesting object, as found twenty or thirty days' journey from the Rocky Mountains, we should do nearly all in our power; yet this sort of information would probably be thought vague and useless. The smaller rivers of this region have as yet received no names from white hunters; if they have names among the Indians, these are unknown to us. There are no mountains, hills, or other remarkable objects to serve as points of departure, nearer than the Rocky Mountains and the Arkansa. The river itself, which we supposed to be the Red river of Natchitoches, is a permanent landmark; but it is a line and not a point; and aids us only in one direction, in our attempts to designate locality. The map accompanying this work was projected in conformity to the results of numerous astronomical observations for latitude and longitude; but many of these observations were made at places which are not, and at present cannot be known by any names we might attempt to fix upon them.[64] More extensive and minute examination than we have been able to bestow might establish something like a sectional division, founded on the distribution of certain remarkable plants. The great cylindric cactus, the ligneous rooted cucumis, the small-leaved elm, might be used in such an attempt; but it is easy to see that the advantages resulting from it, would be for the most part imaginary.