TIME DIVISIONS OF THE ICE AGE. 4.03 
river, or even, in the vicinity of Abbeville, beneath the sea 
level. During many centuries the peat has been excavated 
for fuel, and many small ponds occupy these hollows, and in 
other places mark abandoned parts of the earlier river 
channel. Since the close of the Glacial period, the Somme 
valley has received too little alluvium to keep pace with the 
slight epeirogenic depression which has been in progress ; 
but at the mouth it has been filled by the coastwise drift of 
sand from the marine shore erosion, and by the muddy 
sediments deposited from inflowing tides. 
On each side of the valley the upland extends far away in 
a great plainlike expanse, as seen in any wide view, though 
everywhere somewhat undulating, with an elevation of 150 
to 250 feet above the river and its bottomland. Rounded 
outlines descend to the Somme and to the ramifying 
tributary streams, betokening a prolonged period of subaerial 
denudation by rains, rills, brooks, and the main river, .'The 
Somme, lying south of the European glaciated area, has 
in its ovavel deposits only materials derived from its own 
drainage basin, which consists of approximately horizontal 
Cretaceous strata of chalk, with concretionary flint nodules, 
and here and there overlying remnants of Eocene sand and 
clay, locally hardened to sandstone and shale. Residuary clay 
and loam, left in the process of denudation, covers the 
upland surface to a depth varying generally from 2 or 3 
to 6 or 8 feet. 
The erosion of the Somme valley to essentially its present 
width and depth seems to me attributable to the work of the 
river during the very long Miocene and Pliocene periods, 
and to hae been completed nearly as now before the great 
epeirogenic uplift causing the Ice age, which probably 
raised the British Isles and northern France at least 1 900 to 
2,000 feet higher than now, while south-western France and 
the Spanish peninsula are known to have been elevated much 
more. The record of the Glacial period here appears, in my 
view, to be almost wholly represented by deposition of the 
gravel and sand on the lower flanks of the valley slopes, 
chiefly adjoining the mouths of tributaries; and it is in the 
older of these gravel beds that the flint implements occur, 
indicating the sojourn of men there at the beginning and 
through the early part of the Ice age. 
In Teferring the valley erosion thus to middle and late 
Tertiary er an I agree with Alfred Tylor, whose careful 
discussions of this vall ley and those of southern Britain have 
2D2 
