breaking heavily on an open beach and forbidding all approach, it is a common occurrence for boats to land easily on outlying rocks, whose sides rise nearly vertically from deep water. It seems to be conclusive, therefore, that the natural onward movement of the water at the lower part of the swell is retarded by contact with the bottom, and that the drag below, consequent thereon, causes a tumbling over, or rotary motion of the apex and fore part of the swell, in which is abruptly expended its dynamic force, as seen in the breaker.

In a somewhat similar manner the breaking wave is caused in deep water during a strong breeze, only the process is to a certain extent reversed. Instead of contact with the ground retarding and disrupting the lower part, its surface is disrupted and driven by violence of the wind faster than the part below, and the rotary motion culminating in the breaker is the consequence.

Reasoning on this theory and the before-mentioned facts, it seems probable that a breakwater built up vertically from deep water would have much less strain to bear than a sloping structure.

The Admiralty Pier at Dover is an example of the vertical breakwater wall, and it has stood well throughout the most violent gales: the Alderney Breakwater (although built in a more exposed position than Dover, and, therefore, not a fair comparison) being built on a pierre perdue foundation, will serve to illustrate the slope principle; and it gives way apparently as a matter of course whenever it has to encounter an unusually severe gale. The sloping sea-face of the pierre perdue bank on which this very massive work is built, appears to trip the waves on their approach, just so as to cause them to exert all their force against the vertical wall, and the effect is tremendous:* stones of many tons weight are thrown about like boys' marbles, and although the breakwater pier is all solid stone and concrete, breaches are made in the course of a few hours large enough to sail a vessel through at low water.

It may be asked, Would the Dover pier withstand the heavy seas to which that at Alderney is subject?

There is some reason to believe that it would, for the Dover pier has no pierre perdue bank to intensify the sea, and the Alderney breakwater always gives way near its foundation, at the apex of this bank, which is not more than five or six feet below low-water mark. It is not however intended to assert that the system of construction adopted in the Dover pier is the strongest possible, for it

*It is said that in one gale of wind, upwards of two thousand tons of stone were thrown from the outer to the inner side of Plymouth Breakwater.-ED.

is not; nor is it desired to condemn the use of pierre perdue banks altogether, for building up breakwaters vertically from great depths (as at Alderney, which is in twenty-three fathoms at low water) would be too expensive; but it is necessary to ascertain and fix safe levels for the foundation of the vertical walls, so that they may be below the reach of any violent action of the sea.

The height of the waves the work will have to resist must be known to determine this point; for this as a rough estimate we may assume that to resist a wave thirty feet high the foundation of the breakwater wall should be laid about that depth below low-water mark.

With regard to the best lines or plan for constructing breakwaters, they must of course vary to suit different localities, but there are certain general rules applicable to all such structures: as for instance, concave lines seaward are most objectionable, as causing the sea to concentrate and increase its powers of destruction, whilst convex lines deflect and lessen its force; but the latter might nevertheless be objectionable if the sea were thereby thrown into a ship channel; therefore, as a rule straight lines are best, and they are of course the least expensive.

The general system of construction adopted in most breakwaters appears objectionable, differing but little (as they generally do), except in magnitude, from ordinary landing quays. In very exposed places, as at Alderney, enormous blocks of hard stone are used for the outer wall, backed by large rectangular blocks of concrete about low-water mark; all above which is formed of a nearly solid mass of that material. The stones forming the outer wall are laid in regular horizontal courses, on their broadest sides, and lengthways alternately with headers; much in the same way as a common wall on shore is built, excepting that each course is laid within the one beneath it, or has a batter of several inches. The sea-face profile of such a wall presents somewhat the appearance of a very steep flight of steps. The plan is bad, for the ledges catch the falling sea, and, assisted by pressure of the wave from without, cause it to act with greater energy in penetrating the work; moreover, the whole system of construction here described is weak, as will be shown hereafter.

In order to arrive at some conclusions as to the best style of breakwater construction, let us examine the action of the sea on the rocks of an exposed sea-coast. Where the strata are horizontal it will be occasionally seen that a soft underlying vein exposed to the action of the waves wears away quickly and involves the destruction of the entire superincumbent mass, and that where vertically strati

fied rocks exist the destruction of a soft vein frequently takes place without the hard strata on either side being much affected by it. In the first case, the coast wears away quickly, in the latter it remains without perceptible change for ages. In applying these illustrative cases to Alderney breakwater the horizontally stratified rocks may be represented by the long horizontal courses of stone facing, and the soft vein by its foundation at the apex of the bank of pierre perdue; the practical effects being as above stated.

Another cause of weakness in this breakwater is the peculiar shape of the large stones of which the outer wall is built. They expose such a large portion of surface to the action of the sea in proportion to their size and weight that they are very liable to be shaken loose in their beds; after which their great magnitude only enables them the better to act the part of battering-rams in quickly destroying the smaller and softer material by which they are backed.

From the foregoing arguments and reasons proceeding therefrom, as well as others based on observation and actual facts, it would appear to be generally advisable to build nearly vertical walls up from the bottom to breakwaters where the depth does not exceed five fathoms; that the sea-face of such structures should be inclined inward for security, about one foot to six in height, but batter should not be formed in horizontal ledges as at Alderney.

That the weakest part of breakwaters being near low water of spring tides, this part requires first-rate material and peculiar care in construction: concrete is, therefore, out of place here; stone only should be used, and in the following manner :-The blocks of stone (the heavier their quality the better) should be quarried in great length in proportion to their size, and should be placed in the work end on to the sea, by which plan only a small portion of each stone (in comparison with its size and weight) will be subjected to the direct strokes of the sea, and the long tails of the stones being firmly gripped by others lodged in the body of the breakwater, cannot vibrate or become loose.

Stones for breakwaters need not be of extraordinary large size (although the larger the better); but for all the most exposed parts of the work (that is the sea-face near low-water level) it is imperative that they should be somewhat near certain proportions, which I should roughly estimate as in length about three or four times their breadth, and half their breadth in thickness. For the inner part of the work the same proportions should be adhered to, and although the stones need not be so large, the entire work from the bottom up to nearly half-tide level should be regularly built dry according to the same plan as the outer wall. The stones for the outer wall