turned it every spring into a roaring torrent. These floods were probably more destructive to animals even than man himself; while, however rude they may have been, our predecessors can hardly be supposed to have been incapable of foreseeing and consequently escaping the danger. While the water, at an elevation of 160 feet above its present level, as for instance at Liercourt, had sufficient force to deposit coarse gravel; at a still higher level it would part with finer particles, and would thus form the loëss which, at the same time, would here and there receive angular flints and shells brought down from the hills in a more or less transverse direction by the rivulets after heavy rains.
As the valley became deeper and deeper the gravel would be deposited at lower and lower levels, the loëss always following it;[1] thus we must not consider the loëss as a distinct bed, but as one which was being formed during the same time, though never at the same place as the beds of gravel. Fig. 3, I have given an imaginary diagram, the better to illustrate my meaning; the loëss is indicated by letters with a dash and is dotted, while the gravels are represented as rudely stratified. In this case I suppose the river to have run originally on the level (a), and to have deposited the gravel (a) and the loëss (a′); after a certain amount of erosion which would reduce the level to (b), the gravel would be spread out at b, and loess at (b′), Similarly the loëss (c′) would be contemporaneous with the gravel (c).
Thus while in each section the lower beds would of course be the oldest, still the upper-level gravels as a whole would be the most ancient, and the beds lying on the lower parts of the valley the most modern.
For convenience I have represented the sides of the valley as forming a series of terraces; and though this is not actually the case, there are several places in which such terraces do occur.[2]
It is, however, well known that rivers continually tend to shift their courses; nor is the Somme any exception to the rule; the valley itself indeed may be comparatively straight, but within it the river winds considerably, and when in one of its curves, the current crosses "its general line of descent, it eats out a curve in the opposite bank, or in the side of the hills bounding the valley, from which curve it is turned back again at an equal angle, so that it recrosses the line of descent, and gradually hollows out another curve lower down in the opposite bank, till the whole sides of the valley, or river-bed,
- ↑ See Mr. Prestwich's paper read before the Royal Society, June 19th, 1862.
- ↑ While attributing the excavation of these valleys to the action of the existing rivers, Mr. Prestwich doubts whether they could have produced such an effect without an elevation of the land. Marine shells occur at Abbeville about 25 feet above the sea-level; this bed Mr. Prestwich correlates with some of the raised beaches round our coasts, and with the lower level valley gravels. The higher level valley gravels correspond in his opinion with the raised beaches which occur at a higher level.
