128 The Agricultural Relations of the Western Portion 
worn, and of the wreck of the eocene tertiaries. They occur at 
heiglits varyint; from the sea level to 600 feet above it. The 
<;ravel varies in depth from 2 and 3 feet to more than 20. It is 
found on the summits of some of the chalk hills (Purbeck Hill), 
on the denuded surface of the lower eocenes (Bere Hill), at lower 
levels than these, on the tabular hills of the lower and middle 
eocenes through the whole of the tertiary district under consi- 
deration ; and it forms terraces in the valleys of the principal 
rivers, from 20 to 30 feet above the level of the alluvial flats 
which border their streams. In the Isle of Wight this gravel 
occurs on the summit of Headon Hill and on the high grounds 
near Osborne House. 
The flints of which it is composed are of the largest size when 
found at the greatest elevation. There is a diminution of size 
with every stage of descent. Little increase, however, of 
abrasion accompanies this diminution of size, and the slight 
degree of wear seen in the gravel of the terraces bordering the 
rivers contrasts strongly with the rolled and smooth condition of 
those flint pebbles which form occasional beds in the eocene 
tertiaries. 
This slight abrasion of the newest tertiary gravel of the south 
is an important fact difficult of explanation, if we suppose, which 
seems probable, that the detached masses which cap the highest 
hills once formed a continuous stratum over the whole area ; and 
that its materials have been removed by denudation to lower 
levels. It would seem to indicate that the movements of eleva- 
tion to which the denudation is to be attributed took place 
rapidlv. The gravel is covered, and frequently replaced, by a 
loamy deposit, analogous to that which, in a paper on the geology 
of Norfolk, 1 have called the warp-of-the-drift.* 
Its depth and composition vary with the contour of the sur- 
face, and with the nature of the strata exposed in the vicinity. 
On the chalk it is much more common than beds of jrravel. It 
is evident, from the presence of outliers, or detached masses, of 
the eocene tertiaries, that they once extended far beyond the pre- 
sent limits of the main body. 
To the wreck of these strata and of the chalk we may refer 
much of the matter of which the warp-drift is composed. Beds 
of clay, loam, and sand, mixed with large flints, are met with 
researches -^vhich I have since made in other localities, I am now able, ivith some 
degree of confidence, to identify the gravel south of the Thames, with the latter 
portion of the upper Erratic period. See Quarterly Journal of the Geological 
Society, vol. viii. p. 28G. 
* For more recent investigations respecting these deposits see papers by the 
author in Quarterly Journal of the Geological Society, vols. vii. p. 19,31,20; 
viii. p. 273, 282, and 286 ; also Sir R. Murchison, ibid. vol. vii. p. ,349. 
