cial. This is locally the case on the Norfolk coast, where large basins or pockets filled with Middle Glacial sand and gravel have been excavated in the Contorted Drift. Some of these basins or pockets, however, are evidently due to contortion.

Mr. Wood considers that the Middle Glacial deposits are older than any of the shell-bearing glacial beds in other parts of Britain.

UPPER GLACIAL BEDS.

The Great Chalky or Upper Boulder Clay is a clayey deposit full of pellets or pebbles of Chalk, containing also Chalk-flints, blocks of various rocks, many of them scratched or grooved by ice-action.

In thickness it varies from a few feet to 150 feet. It contains numerous fossils derived from various formations, including Gryphaa incurva, G. dilatata, Belemnites, &c.

It is much used for marling fields, whence the number of marl pits. The surface is often very loamy, and in this respect hard to distinguish from the London Clay in Essex:' some of these surface beds have been used for brick-making. This surface is frequently piped' in the same way as the Chalk, a feature evidently due to a similar cause, namely, the dissolution of the chalky matter by carbonated water. In the pit at Mackie's Nursery near Norwich, the Middle Glacial sands have by this agent been cemented into a comparatively hard rock which has, indeed, been used for building-purposes.2

This Upper Glacial clay was, in the opinion of Mr. S. V. Wood, jun., for the most part formed underneath a great

1 Sometimes called the Brown Clay' in the Eastern Counties. 2 Partly used in the construction of Norwich Castle. (S. Woodward.)

sheet of ice, and gradually extruded therefrom upon the floor of the sea, which was burdened with floe-ice.

It is well developed in Norfolk, Suffolk and Essex, in the country around Tivetshall, Eye, Framlingham, Halstead, Dunmow, Braintree, and the Rodings, but it has not been met FIG. 22.-Section at Writtle, near Chelmsford.

with south of the Thames, nor in the Thames Valley. It occurs at Finchley, at Bricket Wood near Watford, and other places north of London, and around Buntingford, Biggleswade, Huntingdon, Horncastle, &c.

Overlying this Boulder Clay, there are in Yorkshire beds of a purplish-brown colour termed the 'purple clay' by Messrs. Wood and Rome. This clay has the most extensive development of any bed superior to the Chalk in this county, not only overlapping the basement clay in all directions, but extending far beyond the north scarp of the Wolds in an irregular belt along the coast northwards. In its lower por

tion it abounds with boulders of lower Secondary, Paleozoic, and Metamorphic rocks, and at Holderness it contains small quantities of Chalk. Messrs. Wood and Rome state that in the upper portion the small fragments disappear, and the large blocks also become far less frequent, so that the uppermost part (which at Dimlington can be well contrasted with the lower) is scarcely entitled to the distinction of Boulder Clay.

Bridlington Crag.-In the lower and central portion of the purple clay are some beds of sand and gravel, one of which at Bridlington Harbour (according to Messrs. Wood and Rome) yielded the mollusca first made known by Mr. Bean as from the Bridlington Crag. This Crag is rarely seen now save at low tide. Its thickness is estimated at not more than 10 feet, and its fossils include Saxicava rugosa, Trophon scalariformis, Astarte borealis, Nucula Cobboldia, &c. More than one-third of the species are arctic in character.

Drifts of South Wales and the South-west of England.

In South Wales there are deposits of drift gravel and boulders, apparently of local derivation, which may yet be connected with the Glacial period. Such are the scattered deposits near Cardiff, Cowbridge, Bridgend, Llantrissant, &c.

There is very little drift on the Cotteswold Hills, but in the neighbouring Liassic vales there is much gravel; this has been divided by Mr. W. C. Lucy and Mr. E. Witchell, into the Angular Gravel of the slopes, the rolled Oolitic Gravel of the river valleys, and the Northern Drift.

Near Bath, on Hampton Down and other hills, fragments of flint and pebbles of quartz are met with.

The Mendip Hills are remarkably free from drift, but there are some deposits of loam and clay, with here and there

a boulder of some local rock, whether Old Red Sandstone or Millstone Grit, whose position cannot well be accounted for by the action of rain, rivers, or sea.

Quartz, granite, and porphyry are found on the high grounds east of Poole in Dorsetshire, and also in the Isle of Wight. (See fig. 20, p. 274.)

In Devonshire and Dorsetshire, on the summits of the hills and high lands formed of the Chalk and Greensand there is generally found a deposit of very variable nature composed of re-assorted materials chiefly from the Chalk and Greensand themselves. The Chalk itself, as is well known, is generally capped by a deposit called Clay-with-flints, this being in great measure due to the dissolution by carbonated water of the Chalk leaving any insoluble portions and the flints behind. These are generally mixed with a considerable proportion of clay, due either to the former presence of Eocene deposits or of Drift clays.' This Clay-with-flints is found in many places, but it is associated frequently with deposits of Chert detritus, the chert fragments being sometimes rolled, and often transported to a distance (though not necessarily a great one) from the spots where they originally were formed. The flints from the Chalk are often rolled and transported to positions where they could not have been worked out in situ. Mingled with these deposits there is an abundance of small quartz pebbles, and not a few as large as a hen's egg: there are also pebbles of hard quartzose grit and of quartzite which are evidently foreign to the immediate neighbourhood.

Looking at these deposits in a large way, it seems that they may be partly due to the Denudation of the Chalk and Greensand, which was perhaps originally caused by the sea before the escarpments were formed; and they may also have

1 For farther account of Clay-with-flints, see chap. XI.

been worked up and some of the foreign pebbles introduced during submergence at the Glacial period.

The Gravels and Sands bordering the Bovey Basin near Newton Abbot seem to be connected with the Drifts which cap the Haldon and Blackdown Hills. They appear to have been deposited in this Basin, and perhaps to have filled it, when many of the neighbouring valleys had not been excavated. (See fig. 12, p. 20.)

Mr. Pengelly has described some scratched stones found. at Englebourne, and some boulders met with at Waddeton in South Devon, which are suggestive of Glacial action.

In describing the disturbances or terminal curvature' of the laminæ of slate in South Devon, Mr. Godwin-Austen observed (more than 30 years ago) that by the action of frost on the exposed edges of inclined slate rocks the laminæ separate, earthy matter is carried in between them, and thus the space they required is gradually increased. He added, that though it would be hazardous to say that these appearances may not have resulted from long-continued action of the actual frosts, yet when we consider the great extent to which this separation of the leaves of the slate has been carried, and the very inconsiderable depth to which frost at present penetrates in this part of England, we seem to require a period with a lower temperature and the action of deeper searching cold.

There are a number of drifts (brickearth and gravel) which have been referred by Mr. Wood to the denudation of the Glacial beds during their emergence from the great depression.

Amongst these are extensive deposits of coarse gravel in the central and eastern part of Norfolk, such as those, on

1 These sands have usually been termed Greensand. The occurrence of the arctic and alpine Betula nana in a superficial deposit at Bovey Heathfield, has been noted by Mr. Pengelly. (See p. 331.)