ing the constant export of countless tons to all parts of the world for the manufacture of Portland Cement.'

But besides these normal movements some seas Boulder clay. carry beach out for other seas to bring in again. Thus I was told by fishermen at Abbotsbury that their crab-pots were sometimes filled or buried by shingle at a considerable distance at sea, and at Slapton sands that smugglers dared not sink their tubs there for fear of losing them in that way. It is probably in this way that shingle passes round headlands, and quantities of beach are carried out too far to be washed in again. This is the origin of the rounded pebble, the travelled pebble, the sand and the Druid-sandstone in our 'plastic clay and sand.'

natural groin

the Chesil

Now, I think that the law of the sugar-basin Portland a and the laws of the groin apply universally to all which catches beaches and that the peninsula of Giens and the beach. peninsula of Portland may be viewed as natural groins. At St. Salvadour on the coast of France, opposite to Giens, I found what I conceive to be a raised beach. Raised beaches exist along our south coast, and I think that it is the shallows caused by the rising of the land which has allowed the accumulation of double beaches between Giens and the land as well as between Portland and the land (Chesil beach and Smallmouth sands).

Dover built on

years old.

Pebbles never accumulate where the water is deep and the cliff steep, not because they cannot travel there, but because they cannot stop there. A boulder lifted there travels with the wave almost as freely as the water which composes the wave. Boulders can no more accumulate where the water is deep and the cliff steep than soil can accumulate on the face of a precipice though the soil is perpetually forming there from disintegration. It is the friction of a sloping shore and the constant tendency to be landed high and dry which stops the boulder on his journey, and they only stop when they have got out of the way of the sea. They cannot rest within reach of the waves. A storm at a high spring tide may put them out of route for years, centuries, nay, thousands of years, and again set them at liberty.

6

Speaking liberally, we know that part of the a beach 2,000 beach at Dover has been stationary for 2,000 years; since Lyon (page 149, vol. i.) says the Romans erected some of their first buildings upon it at this place and a part of the ruins have remained to our time,' and these banks must not only have been formed, but must have assumed a permanent character before they were built on by the Romans. From these times to the present day, and especially from the times of Edward the Confessor, William the Conqueror, Henry VII.,

and Henry VIII., the case has been perpetually litigated, engineers versus travelling beaches, or Act of Parliament against act of ocean. The engineer's art, however, has not been called on to prevent ocean from destroying its apparently most ephemeral land together with the town on it, but to prevent ocean from consolidating its work to such a degree as to stop up the harbour. Lyons says (page 146) the houses at the foot of the high and almost perpendicular cliff, are considered by strangers in a perilous situation; but use has familiarised it to the inhabitants, and experience has taught them that there is not much danger to be apprehended by the falling of the chalk above them.' This fact is quite in accordance with the theories which I fight for. The sea undermining the foot of a cliff keeps it a precipice and induces heavy falls. Remove the action of the sea and atmosphere disintegration and denudation work most mildly in softening down the precipice, and the accumulation from its gradual decay is easily prevented. But for this it would long ago have formed a smooth slope of grass. But in parts now the foot both of the east cliff and west cliff is terraced and cut into somewhat smoky hanging gardens.

Every headland projecting from a line of coast forms a boulder-parting or division of a boulder

Submerged forests.

flow, as the crest of a hill forms a water-parting or division of the rain-flow and soil-flow, and the foot of each headland is bare of beach as the crest of the hill is bare of soil. The ruins of the headland, however, furnish a constant supply of land boulders for the bay on each side, which accumulate there as soil accumulates in the valley.

If in the centre of the bay rain and rivers have denuded the land to the sea level, the sea cannot farther erode it except by first shovelling the whole beach on to the land and then eroding the ground on which the beach stood. Thus the hard headland while it faces the attack of the eternal eroder at the same time throws a shield of boulders before its softer neighbours.

Where the upright stems of trees, with their roots in their place as they grew, are found on the shore below high water mark, ‘a catastrophe,' or a grand convulsion,' or a subsidence of the land,' or an oscillation of level' is inferred. Mantell calls these submerged forests,' or collections of stumps subterranean forests,' and he compares them to the silicified forest of the dirt-bed' of Portland, and the along-shore boatmen are quite angry if you will not allow that where the sea is now a forest was before the flood.' But such 'catastrophes' are, as I have stated in the Athenæum, simply the result of the most gradual opera

6

tion of rain, rivers and the sea.

In former days

the stream or the rain valley cuts its estuary far deeper even than low water mark, and forms what is called an arm of the sea. In later days the sea throws up a bank of shingle across the mouth of the deep-cut estuary, completely dams itself out and partially dams the streams in, though these often soak through the shingle at low water so as never to rise near the height of high water. Thousands of such cases exist in England. These sea beaches thrown up by storms frequently stand not only very much higher than the high water of the sea which throws them up, but the land behind them is often much lower than the high water of the sea, and thus, according to circumstances, peat, pasture or wood grows below the high water mark. The rapidity of the growth of alluvial deposit from periodical inland floods is then much increased. For all the alluvial wash of the entire valley or waterslope is here at once stopped short, none of it can percolate the shingle into the sea; deposit is rapidly accumulated on deposit, and rooted trees are found under peat, in peat, and above peat, not only on the shore outside the shingle-bank, but in cutting the sluices inside the shingle-bank, and by degrees the land which was below low water mark may be raised by alluvial deposit far above high water mark. When man