cavates the softer stratum at the base of the cliff, and the superincumbent rock, being left unsupported, gives way.

Now, might not this inroad of the sea upon the dry land be prevented, and that, too, at a trivial cost, by building up with concrete masonry the excavated part of the stratum at the base of the cliff, which is often not more than one or two inches in thickness? and thus the further abrasion of the softer stratum, upon which the cliff is recumbent, would be kept in abeyance, and the further encroachment of the sea effectually prevented.

93. A mural cliff of sandstone occurs upon the sea-coast near the city of St. Andrew's. The cliff extends from the baths of that city westward, and rests upon a thin stratum of coal. The sea excavates the coal at the base of the cliff, and the superincumbent rock gives way. It is said, that at one period the sea made such rapid encroachments at this place that the inhabitants contemplated the construction of a breakwater, with a view to the protection of the cliff. Now, might not the further demolition of the rock be easily and effectually prevented by building up, with solid masonry, the exposed part of the coal-bed, which is not more, as far as my recollection goes, now fifty years ago,

than six inches in thickness? The coal would thus be protected from further abrasion, and the further demolition of the cliff prevented.

94. A mural cliff consisting of basalt, called the King's Craig, is situated between the towns of Burntisland and Kinghorn, and is considerably elevated above the present level of the sea. At the bottom of the craig, a thin stratum of coal, about an inch in thickness, extends the whole length of the base. Now, whence came the perpendicularity of that rock? and whence its mural aspect? Was that rock, at a former but remote period, exposed to the action of the waves? and because of this, the stratum of coal at the base of the cliff would undergo a much more rapid abrasion than the rock above; and, therefore, the exposed part of the superincumbent rock would, from time to time, be left unsupported, which, giving way, a mural cliff presenting a perpendicular face would be formed?

ON CAVES.

95. What is the origin of those caves which occur in the cliffs upon the sea-shore, and which present somewhat the form of a hollow sphere? Are they formed thus-In a cave which presents

such a form, I have observed, that in the rock in which it occurs, a rent or cleft extends along the roof of the cave from the mouth inwards into the rock above; and when the sand and gravel at the bottom of the cave is removed, the same cleft is found to extend from the mouth of the cave inwards into the rock? If, therefore, this cleft existed in the rock before the cave was formed, the waves, as they dashed at random upon the sea-shore, would carry along with them particles of sand, which, as they penetrated the cleft in the rock, would widen it by their attrition, and would continue to enlarge it, until at length gravel as well as sand would be dashed in by the waves; and while the particles of sand were

FIG. 16.

penetrating still further into the cleft of the rock, the

gravel which had entered would be left behind, to give additional width to that part of the cavity that had already been formed. The breach in the rock would, at length, became so large that boulders as well as gravel would be dashed in by the waves, which, during their continued action, would excavate a cavern, the form of which would be that of a hollow sphere.

ON ERRATIC BLOCKS.

96. It would appear that the theory which Sir Charles Lyell has promulgated with the view to account for the transport of erratic blocks—namely, by the agency of icebergs-has not been fully adopted by geologists, as we still read of currents of water and waves of translation as the agents employed in the transport of those blocks.

97. When erratic blocks have been removed to a distance remote from the parent rock, and have, notwithstanding, preserved their angular parts sharp and entire, does not this argue that those massive fragments have been carried to the place that they now occupy, and there deposited? that as their edges, during their transport, had not been subjected to attrition, therefore, neither currents of water nor waves of translation were the agents by which those

blocks of stone were impelled forwards to occupy their present position, inasmuch as currents of water, or waves of translation, adequate to the transmission of such masses, must have swept before them all the loose sand and gravel and earthy matters which they met with in their course, and have made that part of the bed of the ocean bare over which they travelled? Those blocks of stone would have, therefore, been impelled forward over a rocky bottom; and because of this, their edges would have been subjected to attrition, and their angular parts rounded off.

The

In the island of Arran, about a mile east from the village of Lamlash, I found upon the sea-shore blocks of granite of several tons weight. upper part of each had its angular parts rounded off; but the base of each, which rested upon the level surface of a sandstone rock, was flat, as is here represented.

FIG. 17.