to two or three levels, and the whole circle of thirty feet finely hammered. So great are the difficulties of working on this rock, that the pit will probably cost as much as all the rest of the work. Sometimes it is impossible to land for two months in succession, and even then it is only during the extreme low water of spring tides that work can be continued from one to two hours. On two occasions, work has continued for four hours at a tide. During 1855, about 130 hours of work were made, and with a small force one-fourth of the rock was cut down. It is probable that, with an increased force, building may begin in 1857, when three or four years will still be necessary for completion. The difficulties to be overcome are greater than at Bell Rock, or Skerryvore, as the foundation is to be lower, and the sea is rougher. An iron scaffold, now in progress, will greatly facilitate operations. Wharves, shops, store-rooms, etc., are built on an adjacent island purchased for the purpose. When the lighthouse is being actually built, they will be in preparation. All hands work on this, except a boat's crew who stay by the rock, when there is a probability of being able to land. When occasion favors the landing, the master signals the workmen in time for them to reach the rock as soon as it would be of service. This interesting operation, we are pretty confident, will succeed, and, despite its perils, we hope that no disaster will mark its progress. Its cost cannot be very precisely estimated, as so many circumstances are likely to influence it; but we may safely assume that the Bell Rock and Skerryvore precedents are not likely to equal this case in economy of proceeding. We know that it is in good hands, and that there will be neither foolish economy nor causeless expenditure.

From stone towers, we pass to iron. It is believed that the design of an iron light-house for Bell Rock, by Capt. Brodie, R. N., was the first formal proposal to build structures of this class with iron. In 1800, Robert Stevenson prepared a design for the same site, composed of a pyramid of eight castiron columns, with braces and ties. In 1821, he erected the Carr Rock beacon, at a cost of $25,000-the lower portion being of stone, and the upper of castiron. Very many instances might now be quoted, in which iron, both cast and

wrought, has been used in Europe for light-houses and beacons.

There are virtually three distinct systems of iron construction for towers. In the first, Mitchell's screw-pile, having a broad helicoidal flange, like an augerpod, is, by simply turning, bored into sand, mud, or other penetrable bottom, so as to form a foundation with a broad bearing, on which the weight of a columnar superstructure may be safely diffused, and to which this is firmly fastened. In the second system, the columnar piles are sunk into the solid rock, by drilling, or are set in an artificial foundation bed, or through broad iron discs. In the third system, the towers are composed of cast-iron plates, and are loaded with masonry at the base for stability. These towers are only fit for sites, either naturally or artificially, dry or out of water. The first two classes are adapted to submerged sites, and can frequently be erected where no other plan would be deemed practicable. The clustre of piles, with the requisite ties and braces, offer but a slight resistance to the waves, compared with that of a solid structure of equal base. The first class is applicable to sandbanks, on which the waves and currents would soon undermine any solid mass. There can be no question that Mitchell's screw-pile has made it possible to place secure lights and beacons where, without them, no durable construction could at all be established. Hence, the invention has peculiar value as one of the chief instruments for superseding light-vessels by permanent light-houses. Mr. Alexander Mitchell, its inventor and patentee, prepared with it a foundation for the Maplin sands light, near the Thames entrance; and, in 1841, Mr. Walker, the admiralty engineer, brought to a successful conclusion the superstructure placed thereon, according to his own plans. In 1839'40, Mr. Mitchell built the Fleetwood light, twenty-eight feet above ordinary tides, on a site subject to tides of thirtytwo feet, at a cost of only £3,500. In 1844, the Mitchells erected a lighthouse in Carrickfergus bay, in water never less than ten feet; and, in 1843, a beacon on Kish bank, in water never less than fifteen feet. Mr. Mitchell stated, before the parliamentary committee of 1845, that he was ready to undertake the replacement of the two light-ships between Dover and

Harwick, by permanent screw-pile lights, at an average rate of £10,000 each. We need not dwell further upon European pile-lights and beacons, as they involve no special principles not equally illustrated in our own like constructions.

The application of iron piles to light-house construction, in the United States, has been chiefly made under officers of the Topographical Engineers, who have given a greater development to the system than it has elsewhere received. To them are due numerous improvements in the combination of the frame work, appropriate arrangements of the elevated keepers' houses, the disc-pile foundation for coral or encrusted bottoms, and improved plans for the foundation story. They have successfully built on a variety of novel sub-marine foundations; and we owe to them the gratifying fact, that the finest specimens of this species of construction are, in every sense, American. A law of 1847, by assigning six difficult light-house constructions to the Topographical Bureau, gave the stimulus which has led to this important result.

The first notable operation of this kind was the rebuilding of the Black Rock beacon, some four and a half miles southwest of Bridgeport, Conn., by Capt. W. H. Swift, Top. Engs. Three successive stone beacons, costing $21,000 in the aggregate, had, in twelve years, been overthrown by the sea at this point. Capt. Swift, at a cost of $4,600, prepared a durable foundation, and erected a pile beacon, thirty-four feet high above low water, and three feet higher than its predecessor. As the detritus of the stone-beacon wrecks was spread over the site, an artificial foundation was made by excavating and bedding six twelve-ton stones, properly placed, and thence concreted into one solid platform. Five wrought-iron periphery piles, and one centre one, from five and a quarter to three inches in diameter, were sunk some distance through holes drilled in the bed-stones, so as to hold firmly to the platform and the mass underneath. These rise in a conic frustum, and are solidly joined together, and duly capped at the top. Several beacons, having similar superstructures, have since been erected on rocks, and on screwpile foundations; besides a number in

which the centre pile, conspicuously surmounted, is made the main fabric, which the surrounding or sloping shafts simply serve to brace. There is, in fact, great latitude for variations in combining iron beacons for different sites.

The Minot's ledge iron light-house, to which we have already alluded, was based on a solid rock, by drilling holes about five feet deep, in which the wrought iron foundation piles, eight inches in diameter were directly fastened by wedging. There were one centre and eight periphery shafts placed on an octagon of twenty-five feet diameter, and the top diameter, at the height of sixty feet from the base, was fourteen feet. The whole height, to the top of the lantern, was about seventy feet. A complex system of diagonal bracing connected the shafts, to give stiffness to the structure. The failure of this edifice was apparently due primarily to the violent breaking of the waves on the rock, rising boldly against their progress-a violence which, in severe storms, much exceeded the anticipated vehemence, and which caused a destructive tossing upward. Secondarily, the formation of sheet-ice, by the freezing of spray on the numerous ties and braces, exposed to wave-action, a much greater surface than was calculated-a fact which the contorted state of these rods strikingly exhibited. This led to great vibrations and loosening of joints, by which the stiffness of the structure was radically impaired. The great height to which the wave-crests rose, especially in the fatal storm, probably subjected the inclosed rooms to their effective action, and this having so long a leverage was a highly destructive force. Other minor causes conspired, but the final overturn seems due mainly to these, and to the remarkable severity of the final storm. The rock-fastening held perfectly, and the piles were broken from four to six feet above the rocks, leaving stumps all bent from the storm. It should be an extreme case, in which, after this experience, such a structure is ventured where spray can freeze in masses on the ties, or where, in the severest storms, the upshooting waves can strike the inclosed portion. The lessons from disasters such as those at Minot's ledge, Bishop's rock, the Skerryvore barrack, and the Bo-Pheg rock beacons, ought to be kept fresh in professional memory, not absolutely to

prohibit analogous structures, or to inspire unreasoning timidity, but to indicate their special liabilities, and as landmarks of that boundary line which cannot safely be crossed.

The Brandywine shoal light-house, in Delaware bay, is one of the finest specimens of a structure supported on screw-piles. Its focal plane is forty-six feet above sea water, and a convenient keeper's house is arranged just below the lantern. Its most remarkable feature, is its exposure to the drift of the Delaware ice, which made it necessary, or at least prudent, to establish an icebreaker for its protection. This is a hexagonal pier, seventy-five by fortyfive feet, composed of thirty screw-piles, twenty-three feet long, and five inches in diameter. These are connected by horizontal or spider-web braces, at their heads and near low water, by which a shock on one pile is diffused to all. The violence of the ice-concussions has required a strengthening of the icebreaker bracing, but otherwise this construction has been very successful. was begun in 1848, and lighted in 1850. The cost was $53,317 for the lighthouse, and $11,485, for the ice-breaker. Its design and execution were due to Major Hartman Bache, Topograpical Engineer, who is a thorough inaster of this class of operations.

It

The structure designed by the Light House Board, for the Seven-Feet Knoll, at the mouth of the Patapsco river, is a very good combination for a site but moderately exposed. Eight eight-and-a-half inch screw piles, twenty-three feet long, are bored twelve feet in the ground at the angles and middle points of the sides of a square of thirty-two feet, whose centre is marked by a similar pile. The design assumes seven feet of water at low tide, and thirteen feet at high tide. Cast-iron tubes (we would prefer wrought shafts) rise from the foundation-piles to twelve feet above high water, being duly braced and tied. From this level, the square pyramid, forming the keeper's house and watchroom, rises in three stories to the lantern-the tubes converging to a twelvefeet square at top. The total height above low water is about sixty-five feet. Some peculiarities of foundation required this plan to be modified in construction, and the height is in fact reduced, so that the focal plane is only

forty-three feet above sea level. We need not further specify constructions of this character, though they are growing numerous along our shores, especially where running ice is not to be feared.

A remarkable iron-pile light-house was begun for the Carysfort reef, Florida, in 1848, and finished in 1852. This is founded on a coral bank near the edge of the Gulf stream, four and a-half feet below low water, and its focal plane is one hundred and twelve feet above the rock. An elevated keeper's house forms part of the design. The whole was so made, framed, and tied together in Philadelphia, as completely to obviate the difficulties of fitting on the spot. Its entire cost was $105,069. A careful examination in 1854, proved that the work had remained without alteration appreciable by test instruments. The Carysfort foundation, was so peculiar as to lead to a novel construction. A hard exterior coral crust covers a softer mass of calcareous sand, so that screw-piles which would pierce the crust would have an insufficient bearing underneath it. This led to the use of large iron footplates, to diffuse the pressure over a large surface of crust (one hundred and thirteen square feet in all), and the piles, passing through centre eyes in the plates, were driven about ten feet into the foundation, till brought up by the lodging of under shoulders on the bedplates. Nine piles, eight inches in diameter, mark the centre and angles of an octagon, and a carefully studied system of cross-ties and braces gives rigidity to the aggregate column. The erection of this preeminently useful light-house was begun by Capt. Stansbury, and completed by Major Lionard, of the Topographical Engineers, ably assisted throughout by the late Mr. I. W. P. Lewis.

The Land Key-light is constructed on a plan analogous to the Carysfort, though it is founded in deeper water and on screw-piles. Its focal plane is 121 feet above the foundation, and 110 above the sea level. It was completed in 1853, at a cost of $101,520, by Lieutenant Meade, Topographical Engineer, and has been found to answer every expectation for stability and usefulness. The same officer has erected an iron screw-pile lighthouse on the flats near the N. W. channel, at Key West harbor; also

This

an important iron pile beacon on Rebecca shoals. He is now engaged in building a first class light-house on Coffin's Patches, Florida reef, about fifty miles east of Key West. stands in about eight feet water, and the focal plane will be 140 feet above the water, giving a range of over twenty statute miles. The foundationpiles are twelve inches in diameter, resting centrally on cast-iron discs eight feet in diameter, and penetrating the rock to a depth of ten feet. They are placed at the angles and centre of an octagon fifty-six feet in diameter, and are braced by horizontal, radial, and periphery ties of five-inch round iron. The pyramidal frame rises from this foundation in six sections, and converges from fifty-six to fifteen feet diameter-all the shafts, except those of the lower section, being of hollow castiron. The keeper's house, in the secoud section, is thirty feet square, of boiler-iron. lined with wood, giving ample accommodations and stowage. The ascent to the lantern is by a circular stairway, in a cylinder of boiler-iron, lined with wood. The entire cost of this noble structure, including illuminating apparatus, is estimated at $118,405, and when complete it will probably be quite unequaled in stability, range, and durability among towers of this character. It is a grateful spectacle to observe how rapidly the Florida reefs are being exorcised of their long-endured terrors by these benevolent towers, and by the signal beacon piles and charts of the coast survey.

We need no longer feel a sense of national shame in comparing our Florida lights with those on the British West India islands just opposite.

Light-houses built on dry foundations, and composed of cast-iron plates, are reputed to have been first suggested by Capt. Samuel Brown, R. N., and they have been successfully designed and executed by Mr. Alexander Gordon, at Gibbs Hill, Berinudas, for £7,689; Point de Galle, Ceylon, for £3.300; Morant Point, Jamaica, for £11,608; Grand Turk for £3,500; Barbadoes for £5,400: and Cape Pine for £6,800. The Morant Point tower was first completely erected in London, then taken down, shipped to Jamaica, and permanently erected in 1842. Its total height is about 108 feet, and it is filled in, or loaded with concrete, for twenty

five feet above the base, leaving only a stairway. The Gibbs Hill light is 115 feet high from the base, and 133 total height; the lower twenty-two feet being loaded with concrete. Its base is twenty-five feet in diameter, and the top is twelve and a-half. All of Mr. Gordon's iron towers are similar in construction and physiognomy, being circular, composed of moderate sized plates joined by broad flanges all around, with numerous small windows, concrete-filled bases, leaving a well stairway and the care-keeper's rooms and store-rooms between the concrete base and the lantern, an ornamental cornice, and numerous details derived from the use of cast-iron as the main material. He claims for these structures great economy-especially for out-ofthe-way localities-facility of erection, slight expense of repairs and keeping, perfect security against lightning in tropical climates, solidity during earthquakes, and easy adaptation to various sites. Despite Mr. Gordon's confidence in their stability during hurricanes, we are not without distrust of such combinations as the Point de Galle and Barbadoes towers. We doubt if they would pass the ordeal of a close discussion of their stability according to the principles so admirably defined by Leonor Fresnel, in his paper on the stability of the Belle-Ile tower, especially when wind-constants, derived from West India hurricanes, are duly introduced in the formula. There is, doubtless, much to commend these towers for various localities, though where granito facings can be erected without disproportionate expense, they are, beyond question, far superior in durability and security. We regret the lack of fuller information concerning the results of Mr. Gordon's efforts to improve the system of British colonial lights, and to erect these towers on many unoccupied points of danger.

In general, the use of iron for lighthouse constructions is a problem of much interest, and one deeply involv ing the security of commerce. While we prefer masonry, when the site permits its economical application, we regard the use of iron as in many cases indispensable for any construction whatever, and in many others as offering a greatly superior economy, even though the resulting fabrics should prove as wanting in durability as the worst fears

and indications betoken. We can hardly question that some effective protection of iron against sea water corrosion and plumbagoizing will soon be found. In this event, if iron structures are only combined according to proper mechanical principles, and with parts of adequate dimensions, there would seem to be abundant guarantees of permanence. Active progress may still be long expected in this field, and the existing structures, regarded as experiments, were most judiciously undertaken as the sure means of improvement. The old preference of engineers for securing stability in constructions exFosed to sea-actions by weight or inertia, rather than by tensile strength, or by anchoring to the foundation stratum, is, we believe, perfectly well founded, and hence, we anticipate no sweeping overturn of present practice, by any probable improvements in iron combinations. As one means of reaching a truer measure of the reliableness of iron marine constructions, further good experiments are needed to ascertain the rate of corrosion of iron in sea water, for the different kinds of iron in different exposures. A rude approximation from the old spindles, etc., along our coasts, fixes a rate not exceeding onetenth of an inch per annum, or four-inch spindles last twenty years. The experiments of Mallet, reported to the British Association, are excellent, so far as they go; but we need a no less careful study of our own irons, subjected to a range of exposures from Maine to Texas. We need to know precisely how much protection zincing affords, how effective a thick coating of oxide may be in saving the inner mass, and what advantage there would be in various sheathings, coatings, paints, etc. We must no longer dwell on light-house engineering, though many important topics are left quite untouched.

The art of light-house illumination presents two distinct fundamental problems; each of which possesses an immense economic importance, and embodies special principles and researches not without general interest. The first problem is chemical and mechanical, and concerns the determination of the most effective and economical sources of a steady, powerful light, from a limited flame; and the best mechanism for maintaining that flame, unimpaired, through the night.

The

second problem is purely optical, and demands, for its solution, the best arrangements for gathering up the divergent rays, and throwing them exclusively on the area of waters which requires illumination. The first point is, how best to generate an appropriate light; the second, how best to render the generated light wholly effective.

We cannot easily appreciate how completely modern is the study of the economical production of a strong light. It is altogether probable that ancient light-towers were only elevated fire-places, or hearths on which wood or coal fires were kindled in chauffers. The Isle of May light, in the Frith of Forth, had shown only a coal fire during the 181 years prior to 1816; and even the Eddystone tower, for forty years after Smeaton's labors were crowned with success, down to 1800, exhibited only a beggarly display of tallow candles. The Corduan light was, for a time, only an oak wood fire, then a coal fire; and it was not till 1780 that Lenoir introduced lamps; and in 1784, the Argand lamps and mirrors. Through all ages some means of artificial lighting have been used; and the history of the lamp bears us back to the oldest records and monuments. Egypt, Greece, and Rome have left innumerable lamp models, exhibiting countless graces of design. Passeri had in his museum 322 antique lamps, and Portici published a voluine of ninety-three copper-plates, representing lamps from Pompeii and Herculaneum; yet, so little was the true theory of combustion understood, even practically, that it was reserved for Argand, of Geneva, to invent the only correctly composed lamp with large flame, in 1784. While Wedgewood could do no better than to copy antique lamp-designs, in all artistic features, ancient magnificence and luxury could only command a powerful artificial light by sheer multitude of burners.

We are accustomed to speak of burning solids and liquids; but, in fact, only gases or gaseous vapors are really burned. The solid must first be melted and vaporized, and the liquid must by heat be converted to vapor, or combustible gases, before they are in a condition to effect that combination with the oxygen of the air, in which all ordinary combustion, with or without flame, consists. Continuous combustion, therefore, only takes place at the surfaces of contact