Much stress has been laid on the fact that no works have been found on the lowest-formed of the river terraces which mark the subsidence of the western rivers. “And as there is no good reason,” remarks Mr. Baldwin, “why their builders should have avoided erecting them on that terrace while they raised them promiscuously on all the others, it follows, not unreasonably, that this terrace has been formed since the works were erected.”[87] To any one familiar with the great rise and fall which takes place annually in the water-level of the Ohio and Mississippi and all of their tributaries, the fallacy of such an argument is at once apparent. We must at least allow that the Mound-builders learned by experience, just as animals do, even if we could deny them a very high order of intelligence. Little time could have elapsed after their advent to these valleys before they observed the impracticability of erecting mounds or enclosures on most of the alluvial bottoms bordering these streams. The raging torrents which sometimes sweep through the valleys of the central basin, uprooting the largest trees, carrying away natural embankments, forming immense deposits of new alluvium, submerging miles of adjacent country, and in many ways changing its physical conformation, would in a few years obliterate any traces of earthworks built within their reach.[88] Far more certain data, however, is furnished in the arborescent vegetation which covers many of the works, with which to measure part of the period during which they have remained unoccupied, though we are left in uncertainty as to the remoteness of their abandonment. The annular rings of a tree present us indisputable evidence as to its age.[89] It is evident that the forests which cover these remains have grown up since they were vacated, as no difference exists between them and the surrounding vegetation—no break exists in the density of the forests in the immediate vicinity of the works. The oldest of the trees found upon the works present eight hundred annual rings, indicating as many years of growth.[90] This cannot, however, be set down as the limit of the period of their abandonment, since, as it seems that this country was open and mostly unwooded in the sections thickly settled by the Mound-builders, a considerable time would be requisite for the slow encroachments of a forest, even when the trees which now stand upon the mounds may have been preceded by trees of other species or by two or three generations of their own.[91] The age of the trees on the mound-works in the Ohio Valley or farther north, rarely exceeds five hundred or six hundred years, and such cases as that cited by Sir Charles Lyell are the exceptions. Farther south, in the Mississippi Valley and near the Gulf, they are still younger than those at the north.[92] So noticeable is this that we are led to think the Gulf coast may have been occupied by the Mound-builders for a couple of