PROCEEDINGS OE GEOLOGICAL SOCIETIES. 



191 



Oise, which is only separated by a watershed six miles broad from that of 

 the Somme, passes from it into the latter valley. 



2. The presence of freshwater shells in some of the intercalated beds, 

 many such as live in clear and rapid streams, indicates a probable fluviatile 

 origin for these deposits. 



3. The mammalian remains and land shells give evidence of dry land. 

 The occasional occurrence of bones in the position they hold during life 

 shows that the carcases and limbs of animals were dropped into the old 

 shingle before they were freed from their integuments, or within a short 

 time after death, whilst the perfect state of preservation of the land shells 

 is an indication of their not having been transported far. 



All these characters tend to prove that these beds are to be referred to 

 old river action. This, however, must have taken place when the river 

 occupied a level about 100 feet higher than it does now. It is true that 

 similar gravels, containing similar mammalian remains and also flint im- 

 plements, occur at lower levels (forty feet) in the valley, whence it is in- 

 ferred that similar causes were in operation when these also were de- 

 posited. But it is plain that the two could not have been deposited at the 

 same time. For the deposition of the high-level gravels on the supposi- 

 tion that the valley had been previously excavated, would have required a 

 river at some times filling a channel more than a mile wide and 100 feet 

 deep— a state of things not to be accounted for under any circumstances. 

 The alternative therefore of a river flow ing at the higher level and gradu- 

 ally excavating its channel is adopted. 



The character of the climate may be inferred from the fauna. The 

 land and freshwater shells are of species now living in France, but they 

 also range as far north as Russia, Finland, and Siberia. They are there- 

 fore such as, though occurring in temperate climates, are capable of exist- 

 ing in high northern latitudes. The animal remains furnish more positive 

 testimony. The woolly mammoth and rhinoceros were fitted by their 

 coating to endure the rigours of a cold climate, such a« Hussia and Sibe- 

 ria, where their remains abound, and where they seem to have fed on ve- 

 getation common to such latitndes. A species of tiger now lives in Central 

 Asia, and is often tracked and hunted down in the winter on the snow and 

 frozen lakes of that region. The reindeer, of which we have the remains 

 in the valley of the Somme, and the musk ox, which occurs in the same 

 deposits in the valley of the Thames, indicate still more clearly the north- 

 ern tendencies of this group. There is a difficulty about the hippopota- 

 mus, but the elephant and rhinoceros originally presented the same diffi- 

 culty ; and there seems no reason why in this case also the extinct species 

 should not be found to have been fitted to live in a severe climate. 



These conclusions are corroborated by the physical phenomena. Mr. 

 Prestwich pointed on the large section to numerous blocks of sandstone 

 but little worn, and varying in weight from half to five tons, which could 

 hardly have been carried and deposited, as now found, by water alone. 

 He also showed various contortions in the upper beds of gravel (whilst the 

 lower ones were hardly disturbed), and in the laminated sands overlying 

 them. These he attributed to ice-action. The blocks, to transport from 

 places higher up the valley on ice-floes at the breaking up of the ice in 

 the spring, and the contortions to the grounding of ice-floes on the soft 

 sand and loose gravel, impinging into them and piling up the gravel, as 

 now occurs on the banks of some of the Canadian rivers. He pointed es- 

 pecially to the pendent masses of brick-earth isolated in the upper part of 

 the sands, and which he attributed to angular masses of ice brought down 



