tributaries of the Arkansa, particularly the Canadian, the Ne-gracka, and the Ne-sew-ke-tonga, discharge red waters at the time of high freshets, in such quantity as to give a colouring to the Arkansa all the way to its confluence with the Mississippi; from this it is inferred that those rivers have their sources in a region of red sandstone, whose north-eastern limit is not very far removed from the bed of the Arkansa.[73] We attempted to take sets of equal altitudes, but failed on account of a trifling inaccuracy in our watch; the variation of the magnetic needle was found to be the same as on the 25th, namely 11° 30′ east. Our hunters had been sent out in quest of game, as, notwithstanding the plenty we had enjoyed, and the great number of animals we had killed, we found it impossible, on account of the heat of the weather, and the frequency of the blowing flies, to keep a supply of meat for more than one day. At evening they returned, having killed a large black bear; the animal finding himself wounded, had turned with great fury upon the hunter, who, being alone, was compelled to seek his safety by climbing into a tree. It is well known that the black bear will sometimes turn upon his pursuers, and this it is probable is more frequently the case at this season than at any other, as they are now unincumbered with that profusion of fat, which for a part of the year renders them clumsy and inactive, and the males are moreover {10} excited by that uncommon ferocity which belongs to the season of their loves. August 28th. The weather during the night had been stormy, a thunder-shower from the north-west on the preceding evening had been succeeded by rain and high winds; the morning was cool, the thermometer at 64°. We had observed, that the sand-drifts, extending along all that part of the river we had passed in the three last weeks, were piled almost exclusively along the northern bank. The country we were now passing is too fertile, and too closely covered with vegetation, to admit the drifting of the sand, except from the uncovered bed of the river; yet along the northern side of the valley we frequently saw naked piles of sand, which had been wafted to considerable distance by the winds. From the position of these sand-banks, as well as from our experience, we were induced to believe, that the high winds of this region are mostly from the south, at least during the dry season.