HARBOURS

Deep Water Harbours.

Harbours of refuge are distinguished from tidal harbours mainly by the superior depth of water which they possess, and the larger area which they inclose. The requisites are shelter during storms, and easy access for shipping at any time of tide. There has been much discussion as to whether piers for harbours of refuge should be vertical or sloping. Col. Jones, R.E., has especially advocated the superior merits of the vertical wall; and the discussions on his plan at the Institution of Civil Engineers, and the able

Harbours. protest by Sir Howard Douglas, will be found, from their interest and importance, to merit a careful perusal.

The principle which is asserted is, that oceanic waves in deep water are purely oscillatory, and would occasion no impact against vertical barriers, which would be the most eligible, as they would only have to encounter the simple hydrostatic pressure due to the height of the advancing billow, and would reflect the waves without causing them to break.

Were it even admitted that the waves were purely oscillatory, and were reflected by a vertical barrier, would no force, it may be asked, be expended when the motion of the particles was reversed? The reflection of a wave is equivalent to the nearly instantaneous creation of a wave in the opposite direction, for which a very considerable force must surely be required.

We believe, however, that from the effect of tide currents, to which we have already referred, and perhaps from other causes whose action seems to have been overlooked by the advocates of the upright wall, any form of barrier, in whatever depth it may have been erected, must be occasionally subjected to heavy impact. We conceive that the possibility of waves of translation being generated in the deepest water has been already established, if we succeeded in satisfying the reader of the truth of the following assertions:—First, That waves break in deep water during calm weather; a fact which is apparent to the eye and familiar to all sailors; and, secondly, and negatively, That to leeward of those races or portions of broken water, which certainly do not reflect the incoming waves, there is comparatively smooth water both at sea and on the adjoining shores, until such time as the strength of the tide is exhausted, and the roost has disappeared, when violent action is again fully manifested.

It may be argued that these are extreme cases, and that such high velocities in the current of the tide are seldom met with. This objection has, no doubt, truth in it; but still the tendency is shown, and though the velocities may be less in other quarters, there may yet be quite enough to destroy the condition of stagnation which the oscillatory theory assumes. The breaking of waves at sea, and the existence of races, seem to prove beyond question that waves of translation are possible in the deepest water. Is it not also a probable case that waves which have been reflected by a vertical wall, and have (irrespective of the question of tide currents) combined with the advancing waves, may then become waves of translation, possessing all the elements which endanger the stability of a sea work? Or, again, how much more damage would result to a vertical wall than to a slope of loose stones, from the sinking of the foundations, or from their getting underwashed by the reaction of the waves? It therefore appears that the method generally resorted to of forming deep water harbours of masses of rubble stone with long slopes, so as to form an artificial beach for the waves to spend on, is, in most circumstances, the best and cheapest kind of construction. We incline, however, to the adoption of an upright wall, founded on the rubble as a basis (similar to that at Cherbourg, about to be described), in preference to long paved slopes, as there is always experienced a great difficulty in founding the toe of such talus walls among the loose rubble. When pitched slopes are adopted, great benefit will be found to accrue from leaving at the bottom or toe of the slope a wide foreshore. Much, however, depends on local peculiarities in selecting the best design for any work; and the nature of the bottom is all-important. Where the bottom is soft, a vertical wall can hardly, if ever, be attempted.

In making these remarks, we must not be understood as condemning the adoption of vertical walls in cases where the foundation is good. All that we assert is the opinion, that waves of translation do exist in deep water, and therefore that harbours of refuge will prove failures unless they are built in such a manner as to resist the impact of those waves

of translation. The Cherbourg breakwater has been often referred to as a successful instance of the application of a vertical wall, and has been contrasted with the Plymouth breakwater, which has a long slope. But this appeal is quite fallacious, as the profile of that work is, as already hinted, of a composite character, consisting of a talus wall sloping at the rate of 10 horizontal to 1 perpendicular, surmounted by a plumb wall; so that whatever merit may be supposed to belong to the vertical profile is entirely nullified at Cherbourg by the long talus wall in front, on which the violence of the waves is much broken. Moreover, the heaviest waves at Cherbourg come from the N.W., and do not assail the breakwater at right angles to its direction, but come more nearly end on to the work, so as to a great extent to run along the outer wall. The N.W. waves are propagated from the Atlantic, while the waves which are most trying to the work come from the N., in which direction the line of exposure is only about 21 leagues. These facts we obtained during a recent visit to Cherbourg, undertaken for the special purpose of ascertaining the physical characteristics of the place. The attempt to make out a parallelism between Plymouth, which faces the Atlantic directly, and Cherbourg, which is comparatively land-locked, cannot, in our opinion, stand the test of a candid inquiry.

Other comparisons may be referred to which have been advanced on equally untenable grounds. Thus, the old pier of Dunleary, which is vertical, and has stood well, has been compared with the talus walls of Kingstown Harbour, which now protect Dunleary, and which have often received much damage. The all-important element of depth of water has been in this instance entirely overlooked; for at Kingstown there is a depth of 27 feet, while Dunleary is all but dry. An able writer on the same questio verata, in comparing different sea walls in the Firth of Forth, has, in like manner, not sufficiently adverted to the great differences in the depths opposite the works to which he refers.

An important advantage of the sloping wall is the small resistance which it offers to the impinging wave, but it should also be borne in mind that the weight resting on the face stones in a talus wall is decreased in proportion to the sine of the angle of the slope. If we suppose the waves which assail a sloping wall to act in the horizontal plane, their direct impulse, when resolved into the force acting at right angles to the sloping surface of the talus wall, will be proportional to the sine of the angle of incidence. The effective force when estimated in the horizontal plane, will be proportional to the square of the sine of the angle of incidence. But if we assume the motion of the impinging particles to be horizontal, the number of them which will be intercepted by the sloping surface will be also reduced in the ratio of the sine of the angle of incidence, or of elevation of the talus wall. Hence the tendency of the waves to produce horizontal displacement of the wall, on the assumption that the direction of the impinging particles is horizontal, will be proportional to the cube of the sine of angle of elevation of the wall.

If it farther happens that there is obliquity of action in the azimuthal as well as in the vertical plane arising from the relative direction of the pier and of the waves, there will be another similar reduction in the ratio of the squares or cubes of the angle of incidence according as the force is resolved into that at right angles to the line of the pier, or to that of the direction of the waves.

Let $\backslash phi$ = vertical angle of incidence or angle of elevation of wall;

$\backslash phi\text{'}$ = azimuthal angle of incidence;

$f$ = horizontal force exerted on unit of surface at right angles to the line of harbour wall;

$h$ = height of greatest assaulting waves;

The above expression assigns, we think, too great a reduction, as the motion of the particles may not be horizontal, and no account is taken of the effects of friction against the rough surface of the masonry. Experiments are therefore wanting to determine the constant for correcting the theoretical results due to this expression. For further information on this subject, we refer the reader to the article on HYDRODYNAMICS.

Mr Scott Russell recommends the parabolic curve as that best suited for the profile where the object is to break the waves, and not to reflect them, as is the case in sloping breakwaters. This curve possesses, according to Mr Russell, the advantages of superior strength, of economy in the materials, of breaking the wave early, and of continuing an uniform action over the longest period of time. When the tide is low, the toe of the slope, which springs out of the foreshore and forms the vertex of the parabola, would, we fear, be found rather weak, and perhaps difficult to form. On the whole, we rather incline in such cases simply to throw in the materials, and to allow the sea to form its own slope.

According to Sir John Rennie (Account of Plymouth Breakwater), rubble breakwaters with slopes formed at the angle of repose, were adopted by the Greeks in the moles of Tyre and Carthage, and by the Romans at Athens and Halicarnassus. The same design was also followed at Venice, Genoa, Rochelle, Barcelona, and other places. In this kingdom the first example on a large scale which we find is at Howth. Kingstown, Holyhead, and the noble breakwater at Plymouth, were afterwards carried out on the same principle, and chiefly under the directions of the late Mr Rennie. The great national harbours of refuge at present in progress in this country, according to Mr Rennie's designs, at Holyhead and Portland, are on a similar principle; while those under Messrs Walker, Burgess, & Cooper, at Dover, Alderney, and Jersey, are more nearly vertical.

On the best Forms of Walls for Tidal Harbours.

Having now considered the few facts of which we are in possession regarding the disputed nature of the impulse of the waves in deep waters, we shall direct the reader's attention to their effects in shallow water. Those in deep water were chiefly whole waves, and regarded by many as being purely oscillatory, while those in shoal waters are breaking waves, and therefore regarded by all as waves of translation. We have hitherto been considering breakwaters erected in deep water, and which were constantly exposed to the waves; we now turn to piers and sea-walls which are placed within the range of the surf, and which are exposed to its force for a limited period only, being sometimes left nearly, or altogether dry by the receding tide.

The impulse of the waves against a sea-wall or pier may be resolved practically into four directions:—1st, The direct horizontal force which tends to shake loose, or carry before it, the blocks of which the opposing masonry consists. This force may also blow up the pitching, or overturn the inner or quay-wall by condensing the air, or pressing upon the water which occupies the interstices of the rubble. We know two cases in the German Ocean where, in consequence of want of width in the pier, coupled, in one instance, with insufficient workmanship, the inner or quay-walls were observed first to bulge and fall, before the sea-wall was injured. One of these piers measured 26 ft. 4 in., and the other 24 ft., on the roadway. 2d, The vertical upward force which may act on any projecting stone or protuberance. 3d, The vertical downward force of the water which results either from the wave breaking upon the toe of a talus wall, or from the wave passing over the parapet, and falling upon the pitching behind, so as to plough it up. 4th, The back-draught which tends by reaction from the wall to plough up the soft bottom, and thus to undermine the lower courses of the work, or perhaps by suction to pull out the face-work. We

may conclude from the above that the points which require to be carefully attended to are—1st, The contour and quality of masonry of the wall itself; 2d, The parapet, which, if not of sufficient height, or built in a proper direction, leads to damage in the pitching behind it; and 3d, The foundation-courses, in the design and construction of which, if similar precautions be not attended to, underwashing of the bottom may in some situations take place, so as to leave the lowest courses without protection.

We shall in the first place consider how far those remarks are applicable where the bottom is solid rock. Such a supposition will render unnecessary any precautions arising from the wasting of the bottom, and, ceteris paribus, there does not seem to be any reason for preferring a talus to a vertical wall. The question of preference in such a case will in the main depend upon the kind of material which can be obtained. Should the stone be scarce or costly, and the quality such as to warrant the introduction of masonry of the best description, the vertical wall may be found to be the most economical. Where freestone is to be used, it is not only desirable that it should be got in large blocks, but that the face stones should possess considerable hardness. This precaution is particularly necessary in selecting the stones for the lower courses, and especially where the beach consists of hard gravel. For the same reason, it is highly important that all stones which are subject to decay from atmospheric influence should be either entirely rejected or assembled in the upper courses of the parapet.

Where the materials are abundant, but of an unworkable nature, a long talus wall will generally be found most economical. For such walls the rate of slope must depend very much upon the exposure of the place, and upon the plentifulness of rubble-stone hearting. The easily-dressed and naturally flat-bedded materials, which the stratified rocks of the secondary formation very often furnish, are especially applicable for the construction of vertical walls; while the uncouth blocks of the primary and igneous formations are better suited for talus walls. Such rocks as gneiss, the schists, basalts, greenstones, amygdaloids, and the tougher kinds of granite, are best fitted for this purpose. With some of those rocks the angularity of the pieces, and the excessive difficulty of dressing, render it necessary to assemble them without almost any alteration of their shape, by an adaptation of their salient and re-entrant angles, so as to make a kind of random rubble face-work. In this kind of work, mortar is very seldom employed. The parapet generally consists of squared masonry, surmounted by a heavy cope, and it should in every case be set in good lime mortar.

Where the materials are light and of small sizes it is desirable to equalize the action of the sea over the whole work, and not to concentrate it against any particular place. Mr Russell states that the cycloidal form was recommended for this purpose by Franz Gerstner of Bohemia. The only instance with which we are acquainted of the adoption of this curve was in a sea-wall erected at Trinity, near Edinburgh, by the late Mr Robert Stevenson, in 1822.

It has been already stated that, irrespective of the quality of the masonry, the two points in the structure which are weak or dangerous are the top and bottom of the wall. With a rocky bottom the risk of failure at the foundations is removed; on the other hand, where the shore consists of rotten rock, moving shingle or sand, it is obvious that provision must be made for both those sources of evil. In fact, if we consult the history of harbours, we shall find that by far the most frequent cause of damage is the reaction of the sea against the shore.

The general slope of a fragmentary beach must depend upon the size and nature of the particles and the force of the sea. The dissimilarity between the slopes of a beach near the levels of high and low water, arises from a decrease

Harbours. in the force of the waves, owing to their being broken before they reach the high-water mark. The great object, therefore, is to design the profile of our wall so as to alter as little as possible the symmetry of the beach. Where isolated rocks or large boulders are seen projecting above the surface of a sandy beach, there will generally be formed around them hollows, corresponding in depth to the kind of obstruction which the rocks present. The principal point in the design, therefore, must be to avoid great and sudden obstructions to the movement of the water. The best form which could be adopted in any situation would of course be the same as the cross section of the beach itself, but this would answer no possible purpose; and, as the wall is to consist of heavy blocks of stone instead of minute particles of sand, it is clear that a much steeper slope may be adopted than the profile of conservancy of the coast, provided the lower part of the slope be flattened out so as to meet the sand at a low angle. The action of a bulwark is to arrest the waves before they reach the general high water mark, and to change the horizontal motion of the fluid particles to the vertical plane, or to compel the waves to destroy themselves on an artificial beach consisting of heavy stones. To prevent underwashing, the two following requisites should therefore be as far as possible secured:—1st, The foundation courses or bottom of the wall should rise at a very small angle with the beach, so that their top surfaces may be coincident with the profile of conservancy of that portion of the beach out of which the wall springs; 2d, The outline of the wall should be such as to allow the wave to pass onwards without any sudden check till it shall have reached the strongest part of the wall, which should be as far from the foundation as possible.

Those two requisites show clearly how inapplicable a vertical wall must in most cases be for a sandy beach. Instead of altering the direction of the wave at a distance from its foundation, the whole change is produced at that very point, and unless the wall be founded at a great depth, its destruction is all but certain. Where the materials are costly, but admit of being easily dressed, we are disposed to think that horizontal, or nearly horizontal wall connected with a vertical one by a quadrant of a circle may be found suitable. Such a form will prevent to a considerable extent the danger of reaction by causing an alteration in the form of the wave at that part where the wall is strongest and at the greatest distance from the toe or curb course. Where the materials are abundant and of a rougher nature, a cycloidal wall, with vertical and horizontal tangents somewhat similar to that erected at Trinity, to which we have already referred, may be adopted with advantage.

Foundations in clay. A special caution may not be out of place regarding clayey bottoms. Many are apt to suppose that there can be no better foundation than clay; and it is indeed true that some kinds of hard clay form a satisfactory subsoil. But there are others of a softer kind, and permeated by sandy beds, which are extremely treacherous. If there be the slightest dip seawards, there is always a risk of any pier that may be built on such a base slipping bodily into the sea. This holds especially true of inland lochs, where the sides very often slope suddenly. In one instance, the particulars of which we got on the spot shortly after the accident, a pier built on a clayey beach, sloping below low water at the rate of 1 in 12½, suddenly began to move, and after two hours it had slipped seawards 150 feet, and had by that time descended bodily a height of 34 feet, the top of the pier being then no less than 23 feet below low-water spring tides.

Construction of Harbours.

Our space will not admit of our going much farther into the subject of the construction of harbours than the few remarks we have already made. A knowledge of such matters may to some extent be acquired by a careful perusal of the pub-

lished histories of marine works; but, after all, it must be confessed that the only valuable teacher in this wide practical field is experience. It is, in truth, impossible to lay down any general rules of guidance as to matters of this kind. All that can be done within our space is to notice very briefly some of the more important methods of working. And first, with regard to that invaluable piece of apparatus, the diving-bell, we would refer to the article on the subject Diving-bell, in this work, and to Smeaton's Account of Ramsgate Harbour, published in 1721, where it was first applied by him to harbour works. The diving-helmet is a most useful and convenient modification of the diving-bell, and is now very generally employed.

Of late years Mr Walker has introduced from France the use of beton as a substitute for backing. This artificial concrete is sometimes used in enormous masses. We have seen at Cherbourg blocks of 50 tons prepared in boxes, whose sides and tops are removed after the concrete has set, in order to be again similarly employed. The proportions used at Cherbourg by M. Rebeille were two of sand or fine gravel, to one of Portland cement.

We may also mention that the method of assembling stones on their edges, instead of on their beds, which formerly was in use in some old Scottish harbours and sea-walls, as at St Andrews, Prestonpans, &c., deserves to be more generally known and adopted, from its superior strength.

The proposal of Mr Bremner, of Wick, for putting in the foundations of low-water piers also merits notice. Mr Bremner proposes to construct, in some adjoining place of shelter, enormous pontoons of timber, on which the under parts of the work are built, and afterwards floated to the desired spot in favourable weather, and carefully grounded. Such a plan might, we have no doubt, be found economical and useful in some situations.

Mr Rendel has introduced an improved method of assembling the pierres perdues or rubble used in the construction of large breakwaters; this method he employed at Millbay Pier, near Plymouth, in 1838, in a depth of 38 feet; and he is at present carrying out the same principle on a still larger scale, in the construction of the breakwaters at Holyhead and Portland. The improvement consists in depositing the rough materials from stagings of timber elevated a considerable height above high water. The stones are brought on the staging in waggons, through the bottoms of which they are discharged into the sea. The principle on which the stagings are designed is that of offering the smallest possible resistance to the sea, the under structure consisting of nothing more than single upright piles, there being only one line of piles for each roadway.

Mr Rendel, in a letter kindly communicated to us, states, "I use no timber braces of any kind, as these offer more resistance to the sea than strength to the staging. At Portland, however, where any accident would be a serious evil, owing to our employing convicts in the quarries, we stay the piles with iron guys, fixed to Mitchell's screw moorings, and also truss the outer piles in each row with iron rods. We also fix the piles in the ground with a screw."

"At Holyhead, however, we only attach to each pile boxes filled with small stones, for the purpose of getting them into a vertical position, and use no stays or guys of any kind."

"The superstructure consists simply of balks of timber, with rails laid on them to carry the waggons. The piles are placed in rows 30 feet apart, and the ease and certainty with which the staging is constructed is such that a length of 30 feet, including the screwing in of the piles, the laying down of the roadways, and all minor works necessary to make them fit to carry the waggons, never occupies more than one working day and a half, and often less. The length of the piles that we are now using varies from 84 to 90 feet, the depth of water at both Holyhead and Portland being about 11 fathoms."

Harbours. "Of the strength of the stage you may judge from its carrying on each roadway as much as three waggons, weighing in the gross 12 tons each.

The advantages of the staging are obvious. It contributes greatly to the consolidation of the stone, it makes a greater length of breakwater to be under construction at the same time, and it enables the deposits to be carried on without interruption, almost in the heaviest weather. As an instance of this, I may remark that my resident at Portland informs me that the waggons and locomotives were engaged yesterday at a time when such a sea was running that large bodies of spray were thrown 55 feet above the water level. As a proof of the facilities which the stage affords for rapidity of construction, I should state that we have deposited this year at Holyhead, where free labour is employed, nearly one million tons of stones. The loss from accidents to the stage is comparatively small on its first cost, and when spread over the cost of the whole works it is a mere trifle. I find the sea-slopes are, in the deep water and exposed parts, from 5½ or 6 to 1 between 6 feet above high-water and from 12 to 15 below low-water, from which point they rapidly become about 1 to 1. The inside slopes are never more than 1¼ to 1, and seldom more than 1 to 1. The materials are excellent for our purpose."

Mr Walker has also kindly contributed some facts connected with the construction of the great works now going on under the direction of Messrs Walker, Burgess, and Cooper at Jersey, Alderney, and Dover. At Alderney, which is a very exposed place, the base, up to 12 feet below low-water, is formed by stones thrown, or rather dropped from barges. Up to low-water the work is all done by diving-helmets. The wall is faced with granite, backed with blocks of beton made of sand, shingle, and Portland cement. Above low-water it is faced with stone of the island, a kind of millstone-grit, and is backed with blocks of rubble set in Roman cement. The millstone-grit is raised in very large blocks. The profile is to consist of a quay, an esplanade, and a parapet.

Jersey is much the same as Alderney, but the pell-mell work is carried to low-water, having nearly vertical walls of conglomerate built above. Dover has nearly vertical walls, faced with granite from the very bottom, which is now 45 feet below low-water. This work was done with diving-bells.

Sir J. Rennie, in his Account of the Plymouth Breakwater, says, "From the bottom to within 8 feet of low-water springs, we find that the slope is 2½ or 3 to 1. Here the effect of the waves is comparatively small, being neutralized by the mass of water. From thence to low-water of spring-tides the slope increases from 3 or 4 to 1, but between low-water of spring-tides and high-water, when the effect of the waves is greatest, there we found that the rubble would not lie at less than 5 to 1, whilst, on the inside, the slope stands generally at from 1¼ or 2 to 1."

The above interesting details regarding these national works show, from the variety which they exhibit, how difficult it is to lay down any general rules for the construction of harbours, and confirm the principle that each work must be judged of per se.

Miscellaneous Observations.

The ultimate object of constructing harbours is, by lowering the height of the waves, to preserve the tranquillity of the area of water which is inclosed by the piers; and this property is variously possessed by harbours of different forms, and depends much upon the relative widths of the entrance, and of the interior, the depth of water, the shape of the entrance, and the relation between the direction of its opening, and that of the line of maximum exposure.

The only formula of which we are aware is that by the writer of this article (Edin. New Phil. Journal, 1853), which gives an approximation to the reductive power, or

is, in other words, a numerical form of expressing how much a wave of given height becomes reduced, after it has entered a harbour. Though the results obtained by the formula may not be absolutely correct, this will be no objection where the object is merely to obtain a comparative value, as, for example, in comparing one design for a harbour with another.

When the piers are high enough to screen the inner area from the wind, where the depth is uniform, the width of entrance not very great in comparison with the width of the wave, and when the quay walls are vertical, and the distance not less than 50 feet,—let

H = height in feet of waves in the open sea.

x = reduced height of waves in feet at place of observation in the interior of the harbour.

b = breadth of entrance to harbour in feet.

B = breadth of harbour at place of observation in feet.

D = distance from mouth of harbour to place of observation in feet.

This formula has been found to give good approximations at several harbours where the heights of the waves were registered. When H is assumed as unity, x will represent the reductive power of the harbour.

In situations where the highest waves cross the harbour mouth at an oblique angle, a further reduction is due to height of waves from lateral deflection. We have been unable to find any observations that have been made on this subject by others, and for want of better, we shall give three observations made under our directions at Latheronwheel harbour:—

These must, however, be regarded as but approximations. It is obvious that as the wave may be deflected through more than 360°, the curve representing the reduction must be a spiral; but more observations are wanted to determine of what kind.

Booms are logs of timber placed across the mouth of a harbour or the entrance to an inner basin or dock, having their ends secured by projecting into grooves cut in the masonry on each side of the entrance. The booms are dropped into those grooves to the number of from 10 to 20, or as many more as will insure close contact of the lowest one with a sill-piece placed in the bottom of the harbour, without which precaution the swell is found to enter the harbour from below the booms. By this contrivance, which forms a temporary wall, the waves are completely checked and prevented from spreading into the interior basin. The longest booms we have seen are about 45 feet, and in some places, as at Hartlepool and Seaham in Durhamshire, they are taken out and in by steam power.

Though perfectly successful in their tranquillizing effect (provided they are kept in contact with the sill piece at the bottom), booms are not suited for the mouths of harbours where there is much traffic, as the shipping and unshipping of so many logs of timber can hardly take less than a quarter of an hour—a delay which might in many cases be attended with serious consequences.

It is very desirable, and in some cases essential, that there be either a considerable internal area, or else a separate basin opposite the entrance for the waves to destroy or spend themselves. Such a basin should, if possible, be made so as to preserve a portion of the original shore for the waves to break upon, and when circumstances render this impossible, there should at least be a flat talus of 2 or 3 to 1. Talus walls of 1 to 1, or steeper, will not allow the waves to break fully, but will reflect them in such

Harbours. a manner as might in some cases make the entrance difficult or even dangerous of access, and the berthage within unsafe. There are many instances of harbours being materially injured by the erection of a quay wall across a beach where the waves were formerly allowed to expend their force.

It may be observed that when there is an inner harbour or stilling basin, the elliptical form seems to be the most promising. Let one focus be supposed to be on the middle line of the entrance and to coincide with the point from which the waves in expanding into the interior radiate as from a centre (which they do approximately), and if the other focus is situated inland of high-water mark, the waves will tend to reassemble at the landward focus, and on their way will be destroyed by breaking on the beach. This appears from the well-known property of the ellipse, that if two radii vectores be drawn from the two foci to any point in the curve they will make equal angles with the tangent at that point; and as the angles of incidence and reflection of a wave from any obstacle are practically equal, each wave will be nearly concentrated at the focus opposite to that from which it emanated.

Another cause of disturbance in harbours, which is often not sufficiently considered, is the indiscriminate deepening of the entrance without a proportionate enlargement of the internal area, or the execution of other works for counteracting the effect. As the depth of the water is more and more increased, waves of greater height become possible at the entrance, so that larger waves gain admission to the interior. The writer has had repeated proofs of this in the course of his practice. At the port of Sunderland Mr D. Stevenson recommended the removal of nearly the whole of the south stone pier, and the substitution of works of open framework in order to tranquillize the interior. These works, which have been quite successful, were rendered necessary by the frequent dredging of the channel at and near the entrance.

The preservation of the depth of harbours where there is a tendency to deposit is often attended with great difficulty and expense. Where the deposit of silt is confined to the space between high and low water marks, the scouring by means of salt or fresh water is in general comparatively easy, but where there is a bar outside of the entrance the case becomes most materially changed. The efficacy of the scour, so long as it is not impeded by encountering stagnant water, is kept up for great distances but soon comes to an end on its meeting the sea. Probably the only way in which this difficulty might to some extent be obviated would be by conducting the water in iron pipes to the bar, a plan which the author proposed in 1843 for Hynish harbour, but the expense was considerable and the success doubtful. The same plan was proposed by Mr Alexander Swan for Kirkcaldy some years later. When the volume of water liberated is great compared with the alveus or channel through which it has to pass, the objection based on the stagnancy of the water originally occupying the channel does not hold to the same extent as when the scouring is to be produced by a sudden finite momentum. In the one case the scouring power depends simply on the quantity liberated in a given space of time, while in the other it depends on the propelling head and the direction in which the water leaves the sluice. Mr Rendel's scheme for Birkenhead was on the former principle. The first example of artificial scouring in this country seems to be due to Smeaton who used it effectually at Ramsgate in 1779.

At Bute Docks, Cardiff, designed by Sir W. Cubitt, the access to the outer basin is kept open most successfully by means of artificial scouring on a gigantic scale. The entrance was cut through mud banks for a distance of about three-fourths of a mile seaward of high-water mark. The initial discharge when the reservoir is full, is stated to be 2500 tons per minute. The writer has known even so limited a discharge for an hour to two as one ton a minute, pro-

duce very useful effects in keeping a small tidal harbour clear of sand.

Many proposals have from time to time been made for floating mooring in the open sea floating frameworks of timber with break-the view of sheltering the space inclosed by them. The objections to floating breakwaters are so great and obvious that there seems little chance of their ever being much used. From what was stated on the subject of booms, it will be recollected that it is a requisite that they should fit closely to a sill piece at the bottom, otherwise the run is found to extend into the harbour. From what will be afterwards stated regarding the liability of timber to speedy destruction from the marine worm, and to iron by chemical action, it is obvious that floating structures of wood, connected by iron and moored by iron chains, cannot possibly be of long duration. If to all these sources of evil we add the risk of their being broken by the sea we think the case may be almost regarded as hopeless. No doubt green-heart might be employed so as to resist the ravages of the worm, but its high specific gravity and its great expense would prove bars to its employment.

In some situations where there is a long shallow beach, a harbour or pier of timber or masonry may be made at or near the low-water mark, which may be connected with the shore by means of a suspension bridge. The inducements to adopt the suspension principle are its economy, and the free passage it affords to the currents which in this way are prevented from forming accumulations of sand, silt, or gravel. These advantages are, however, much reduced by the great wear and tear consequent upon the perishable nature of the structure. The late Sir Samuel Brown erected two chain piers, the one at Brighton, and the other at Newhaven, near Edinburgh, both of which are still in existence.

In every situation where it is easily practicable to make two entrances to a harbour, it will be found well worth the extra expense, provided they can be so placed that the one shall be available when the other has become difficult of access. In harbours which have but one mouth, vessels are often detained for a great length of time by the continuance of the wind in the direction which throws a heavy sea into the entrance. Whereas if there are two entrances situated as we have supposed, vessels are at once able to take their departure by the sheltered side. At the port of Peterhead, the north and south harbours were some years ago united by a canal, according to the writer's plans, and there the advantage has been of the most marked description. Vessels can now clear out as soon as loaded, either by the north or south mouth, according to the state of the sea. Some caution is necessary, however, as the run is apt to extend from the one harbour to the other unless there be a considerable area.

There is generally much prudence required in the alteration or repairs of existing marine works. The risk of having the whole structure destroyed by a gale coming suddenly on while there is an open breach in the works, must be obvious; and in one instance, where the exposure of the place was great, and the evil was a hidden one, the writer could not recommend the face-work being disturbed. The cause of failure in this instance was supposed to be the decay of the backing, which having deprived the face-stones of support allowed them to be driven inwards by the force of the waves. Instead of removing the face-work, the only recommendation that could be given was to inject the whole pier with fluid cement, so as, if possible, to render the mass monolithic. An alternative of this kind is obviously of very doubtful success, and can be regarded as nothing short of a last resort, for there is but a small chance of getting the injected fluid to permeate the whole mass of the pier. The system of permeating the masonry with fluid matter could, however, be employed with more chance of success in the

Harbours. formation of a pier, while each course lies open to view. In 1844, at a harbour that had stood for very many years, two or three faulty stones had been incautiously taken out

of the face-work by a mason who intended to replace them by others, when a sudden gale came on, and nearly the whole of the work was levelled with the beach.

Register of Height of Waves for 1852 observed at Lybster, Caithness-shire.

As an example of the suddenness with which our eastern coast is visited by gales, and as indicating graphically the relative eligibility of the summer and winter months for carrying on harbour works, we give the accompanying diagram of the heights of waves, as observed for the writer, by Mr William Middlemiss, resident engineer at Lybster harbour.

Timber piers.

In landlocked bays, where a deep-water landing-place is all that is required, and where the bottom is sandy or soft, timber may be employed with great advantage. Even in exposed situations, timber can also be used, but the fatal disadvantage attending its employment in most places where there is no admixture of fresh water, is the rapid destruction occasioned by marine worms.

The damage occasioned to harbours in this way is noticed by Semple in his treatise On Building in Water in 1776, and very probably by much earlier writers. Indeed, the ravages of the Teredo navalis are very ludicrously described by Hector Boece in his Croniklis of Scotland, printed at Edinburgh circa 1536. In the Atlantic Ocean the Teredo navalis, and at many places in the German Ocean the Limnoria terebrans, are the animals which are found to destroy any structure of timber which is exposed to the water. They are found to eat most rapidly between the bottom and low-water mark, but above low-water the damage is not so great; and what is singular, they do not appear to exist at all below the bottom where the pile is covered with sand. These observations do not, however, quadruple with Mr Hartley's at Liverpool, for he found the parts which were alternately wet and dry to decay faster than the parts which were constantly immersed. Even solid limestone is often destroyed by the persevering efforts of another marine animal called the Pholas.

The late Mr R. Stevenson made several experiments on the ravages of the Limnoria terebrans at the Bell Rock in 1814, 1821, 1837, and 1843, by fixing pieces of different kinds of timber to the rock, and getting regular reports on their decay. From those experiments it appeared that green-heart, beef-wood, and bullet-tree, were not attacked by the worms, while teak stood remarkably well, although suffering at last. The kyanizing fluid and other preparations have been tried, but were not found to be of permanent service. In addition to these experiments on timber, no fewer than 25 different kinds and combinations of iron were tried, including specimens of galvanized iron. Although separate specimens of each were tried in places where they were always under water, and also in places where they were alternately wet and dry, yet all the ungalvanized specimens were found to oxidize with much the same readiness. The galvanized specimens resisted oxidation for three or four years, after which the chemical action went on as quickly as in the others.

The following Table shows the different kinds of Wood which were made the subject of experiment at the Bell Rock in 1814, 1821, 1837, 1843, with their relative durability.

Harbours. Green-heart timber is now generally had recourse to in places where the worms are destructive. It appears to have been first used by Mr J. Hartley of Liverpool, who published in the Minutes of Institution of Civil Engineers an account of its virtues in 1840, as ascertained at the Liverpool Docks. Its cost is considerably greater than memel or than most of the other timbers generally used. Memel logs for the inner piles of piers might perhaps, from their not being exposed to abrasion from ships, be clad with green-heart planking at those parts which are exposed to the worm. Copper sheathing and scupper nailing are often and successfully employed as protections for piles in exposed situations. Breaming or scorching the wood, and afterwards saturating it with train oil, also forms a partial protection.

It is much to be regretted that timber is so expensive in this country, and that some simple and economical specific against the worm has not been discovered for protecting memel and the cheaper kinds of pine. The grand desideratum in harbour works, which is the want of continuity in the structure, would then be supplied. It follows, from the known laws of fluids, that each individual stone in a pier which is equally exposed throughout its whole length, is subjected to a force which it can only resist by its own inertia, and the friction due to its contact with the adjoining stones. The stability of a whole hydraulic work may therefore be perilled by the use of small stones in one part of the fabric, while it is in no way increased by the introduction of heavier stones into other parts. By the use of long logs of timber carefully bolted together a new element of strength is obviously obtained. A pier could be erected almost free of sea risk if constructed of rectangular or other shaped prisms, consisting of logs of timber treenailed and bolted together, so as to form boxes, say 10 feet square and 30 or 40 feet long. The interior of the boxes would be filled with rubble or beton. The first layer would be arranged across the pier, so as to fit the irregularities of the bottom, and above that, they might be arranged lengthways of the pier, so as to form its outer and inner walls, the space between being filled with common rubble or beton.

Deposit of silt, sand, &c., and sea-bars. In many ports the original depth has been decreased by the deposit of silt, sand, and gravel. This is, indeed, a great evil, and one which unfortunately is most difficult of cure. So obscure and apparently capricious are the causes which lead to the formation of shoals, that in the present state of our knowledge it would be little short of quackery to lay down any general rules for the guidance of the engineer. In fixing on the site for a harbour, all existing obstructions should be examined to ascertain whether there be a tendency to deposit, and the works should be kept as far as possible from places where the tendency is most strongly developed. The agents which occasion bars at the mouths of harbours are the waves, the tide currents, and land streams where they exist. Rivers are often more pernicious than beneficial in their effects, especially where they intersect a gravelly soil; but in some cases the descending gravel may be successfully intersected by the erection of weirs from which the accumulations must be from time to time removed. We agree with Sir H. De la Beche in believing that the bars at the mouths of rivers are most generally formed by the constant tendency of the waves to preserve the continuity of the beach profile. It is therefore not to be wondered at, that heavy gales should distort and fill up the narrow trench which the back waters cut in gravelly or sandy beaches. The erection of breakwaters on each side has undoubtedly a good effect in protecting the channel, but still a bar is very apt to form outside of the breakwaters. In some cases the depth of the track might probably be main-

tained by driving, on each side of the mid-channel, dwarf piles to which continuous walings should be attached so as to confine the current at low-water. The timber framework should not project more than a foot or two above the bottom, which in some cases might be planked. This, however, is but a hint, and has, so far as the author is aware, never been tried. The principle on which the proposal is based is that of contracting the low-water channel to a smaller width than that of the high-water channel, and thus by fixing the low-water track, to prevent a tortuous channel. The same principle was adopted by the writer with success in controlling and fixing the meanders of a gravelly river, which is subject to very sudden and heavy freshes.

The want of sufficient funds occasions a great national loss in the construction of our harbours. The history of a large majority of those ports which have been erected by private or local enterprise, presents but a record of the building of piers at one period when the funds were small, and of taking them down at another when the trade had increased and more room and accommodation were required. Want of funds often prevents the original works from being carried within deep water, and in consequence the most expensive part of the protecting breakwater is often put down just in the very place which has afterwards to be converted, at great expense, into a deep water access or berthage. Sometimes, indeed, a whole line of pier is, from motives of economy, placed in such a manner as to interfere most materially with what might have been by far the best and safest berths for shipping, so that in the further extension of the works a great part of the old harbour has to be demolished. Want of a proper marine survey has also often led to very serious errors in the position of the piers.

To such an extent has this system prevailed, that were an engineer called on to value many of our works as they exist at present, his estimate, however fairly and fully made out, would fall lamentably short of the actual cost. This estimate would proceed on a measurement of what he sees, while the actual cost would include the building of piers and jetties which had long since ceased to exist. For these reasons we conceive there could hardly be a more advisable expenditure of the public money than by a system of grants for supplementing the local funds on a liberal scale. With such aid the authorities on the spot would be enabled to protect and improve the existing physical advantages which the shores possess, by preventing the construction of proposed improvements on too narrow a scale. But a comparatively slight increase of the means would, in instances of which the writer is aware, have inclosed a great extra area, and secured a deeper access with superior internal tranquillity, the want of which now cripples the trade, and is the subject of lasting regret to all frequenting the harbours.

For other subjects connected with harbours vide articles on DOCKS, AND PORTS.

Reference may be made to Brit. Assoc. Rep. 1850 (Scoresby); Min. Inst. Civ. Eng. 1848 (Rankine); Do. 1847 (Scott Russell); Do. 1844 (Brewer); Smeaton's Reports, passim. Rep. Com. on Waves by Brit. Assoc., J. S. Russell, Lond. 1848. Researches on Hydrodynamics, J. S. Russell, Trans. Roy. Soc. Edin., vol. xiv. 1837. Account of Experiments on Force of Waves of Atlantic and German Oceans, Thomas Stevenson, Trans. Roy. Soc. Edin., vol. xvi. 1845. On Reduction of Height of Waves after passing into Harbour, T. Stevenson, Edin. New Phil. Journ., 1852. Account of the Plymouth Breakwater, by Sir J. Rennie, London 1848. Bellidor's Architecture Hydraulique, Paris. Semple's Treatise on Building in Water, Dublin, 1776. Royal Tidal Harbour Commissions Reports (Captain Washington), London 1845-6. The Article on Tides and Waves in the Encyclopaedia Metropolitana, by G. B. Airey, Astronomer Royal. Report by Commissioners of Harbours of Refuge, with the Protest, by Sir Howard Douglas. (x. 2.)