The alluvium in the neighbourhood of Bath contains in its lower portion a layer of peat, with bones of the Celtic short-horn (_Bos longifrons_), stag, roe, horse, goat, and pig; and in its upper part are old refuse heaps, proved to be Roman by the coins and ware, which are also met with at various points underneath the surface soil, and sometimes at considerable depths. It is, therefore, of prehistoric and historic age, and since it is found only in the valley bottoms, we may conclude that the present courses of the rivers along the sides of which it is found date back from the prehistoric age; while their ancient courses are marked by the fluviatile deposits with the extinct mammalia standing at various levels, the higher being the older. In the section at Fisherton we have evidence that the river flowed at a lower level in the pleistocene age than in the prehistoric, and in that at Freshford that the lower portion of the valley had been excavated after the pleistocene strata had been formed. One or other of these physical changes is to be traced in nearly all river valleys.[176] We may conclude that both imply a considerable lapse of time, because similar changes are now produced with extreme slowness. In the pleistocene river deposits, which lie scattered about at various heights on the valley sides, we seek in vain for neolithic implements, or domestic animals. In the low-lying alluvia, and accumulations of peat, we seek equally in vain for traces of palæolithic man, or of the extinct mammalia, except the Irish elk. We may also gather, from the localization of the prehistoric alluvia close to the present streams, that the time represented by its accumulation is insignificant in comparison with the long lapse of ages implied by the pleistocene gravels and brickearths, that were deposited at various heights during the excavation of the valleys. The general surface of the valleys has undergone but little change since history began, and the excavation by the rivers has been so small as to have escaped accurate measurement. The alluvia represent the principal work done since the close of the pleistocene period. The most important testimony that the interval between the two periods was very long, is offered by the climatal change, and the severance of Britain from the continent. The arctic severity of the pleistocene winter in these latitudes had passed away before the prehistoric age, and the pleistocene valleys of the North Sea, St. George’s Channel, the British, and Irish Channels had been depressed beneath the waves of the sea before any prehistoric strata yet known had been deposited. The evidence that these changes actually took place must be referred to the two following chapters. _Comparison of Deposits in Valleys with those in Caves._