August 15th. Extensive tracts of loose sand, so destitute of plants, and so fine as to be driven with the wind, occur in every part of the saline sandstone formation we had as yet seen. They are perhaps invariably the detritus of the sand-rock, deposited in valleys and depressions, where the rapidity of the abrading currents has been checked by permanent obstacles. This loose sand differs in colour from the sandstone, which is almost invariably red; the difference may, however, have been produced simply by the operation of water suspending and removing the light colouring matter, no longer retained by the aggregation of the sandstone. These fields of sand have most frequently an undulating surface, occasioned probably not less by the operation of winds than by the currents of water; a few plum bushes, almost the only woody plants found on them, wherever they take root, form points, about which the sand accumulates, and in this manner permanent elevations are produced. The yucca angustifolia and the shrubby cactus are here rarely seen; the argemone and the night-flowering bartonia have not entirely disappeared, but are not of frequent occurrence. Our horses had broken loose, and a part of them strayed from camp. This occasioned some unusual delay, and the morning was somewhat advanced when we commenced our ride. The day was bright and cool, comparatively so at least, the mercury in {310} the extreme heat rising to only 95° in the shade of our tent, whereas on several of the preceding days, it had stood at or above 100° in a fairer exposure. A light breeze sprung up from the south-west, and continued during the remainder of the day. Our course led us twice across the bed of the river, which we found one thousand and four hundred paces in width; and without water, except in a few small pools, where it was stagnant. This wide and shallow bed is included between low banks, sometimes sloped gradually, and sometimes, though rarely, perpendicular, and rising scarcely more than four feet from the common level of the bottom of the channel. Drift wood is occasionally seen without these banks, affording evidence that they are at times not only full, but overflowed. If they are ever but partially filled, it is easy to see that what, for a great part of the year, is a naked sand beach, then becomes a broad and majestic river. It must flow with a rapid current, and in floods; its waters cannot be otherwise than of an intense red colour. The immediate valley of the river had now become little less than two miles in width; and had, in some places, a fertile soil. This happens wherever there occur spots having little elevation above the bed of the river, and which have not recently been covered with sand.