a conduit or channel for the conveyance of water. It is derived from aqua, water, and ductus, a conduit. It is applied more particularly to those structures of masonry which have been erected for the conveyance of water across valleys, to which, however, we would rather give the name of aqueduct bridges, extending the term aqueduct to the whole conduit or channel by which the water is conveyed from one place to another. The conveyance of water for the supply of large cities has in all ages formed a very important object of public economy; and aqueducts of various kinds have been in use for this purpose from the earliest times, the remains of which have been examined by travellers in different parts of the East. Pococke describes a work of this kind erected by Solomon, for conveying water from the pools and fountains near Bethlehem to Jerusalem. "The aqueduct," he says, "is built on a foundation of stone; the water runs in round earthen pipes about 10 inches diameter, which are cased with two stones, hewn out so as to fit them, and they are covered over with rough stones well cemented together; and the whole is so sunk into the ground on the side of the hills, that in many places nothing is to be seen of it." But it was in the luxurious capital of Rome that the system of aqueducts was brought to the greatest perfection, and carried to an extent which has never been equalled even in modern times, and has justly excited admiration both from the number and magnificence of the works themselves, and the prodigious quantities of water which by these means were continually poured into the city. These aqueducts extended, some of them 30, 40, and even 60 miles from the city, in one continued covered channel of stone, carried by arcades over the widest and deepest valleys, and by tunnels running in many parts for miles through mountains and through the solid rock. "If we consider attentively," says Pliny, "the quantities of water brought into the city for the use of the public, for baths, for fish-ponds, for private houses, for artificial lakes, for gardens in the neighbourhood of the city, and for villas; if we look also at the works which have been constructed for forming a regular channel for the waters—arches raised up, mountains pierced with tunnels, and valleys filled up to a level; it must be acknowledged that there is nothing in the whole world more wonderful."

For about 400 years after the building of the city the Romans were contented with the waters of the Tiber, or what was drawn from wells or from fountains in the city and its neighbourhood. But the great increase of the population rendering a more ample supply desirable, the censor Appius Claudius was the first to introduce an aqueduct to convey the waters of distant springs into the city. About thirty-nine years after this, M. Curius Dentatus brought in an additional supply from the neighbourhood of Tibur. These examples were afterwards followed by various other public men, as the wants of the city rendered new supplies necessary. Among these were Papirius, Crassus, Marcius, Agrippa, and Augustus; and most of the succeeding emperors, even Tiberius, Claudius, Caligula, Nero, and Caracalla, esteemed it an honour to connect their names with such great and useful works.

Frontinus, who was appointed curator of the aqueducts by the emperor Nerva, has left the most ample account of them. According to him, there were nine great aqueducts by which the city was supplied. Five more were added by Nerva, and the number was afterwards augmented by succeeding emperors to twenty. Of these, the most remarkable were the Aqua Appia; the Old and New Anio; the Aqua Martia, which also conveyed the Aqua Julia and the Aqua Tepula; the Aqua Virginia; and the Aqua Claudia. The Aqua Appia was so named from the censor Appius Claudius, by whom it was constructed in the 442d year of Rome. It commenced in a field near the Via Praenestina, between the 6th and 8th mile-stones, made a circuit of 780 paces to the left, and then proceeding by a deep subterranean channel of more than 11 miles, entered the city at the Appian Way by the Porta Capena, and delivered the main body of its waters into the Campus Martius. The Old and New Anio were so called from their bringing into Rome the waters of that river. The former began above the Tiber, at the 30th mile-stone, and consisted mostly of a winding channel, carried through an extent of about 43 miles. The latter, constructed under Nero, took a higher level, running 7543 paces above ground, and then through a subterranean passage of 54,267 paces in length. The Aqua Martia, which owed its formation to Quintus Martius, rose from a spring distant 33 miles from Rome, made a circuit of three miles, and afterwards forming a vault of 16 feet diameter, it ran 38 miles along a series of arcades at an elevation of 70 feet. It had openings perforated at certain distances for discharging the collected air, and at different places deep cisterns, in which the water settled and deposited its sediment. On this account it was remarkable for its clear green colour, and is celebrated by Pliny for its coolness and salubrity. The Aqua Julia and the Aqua Tepula were brought to Rome by the same aqueduct as the Aqua Martia, but on higher levels. The whole aqueduct above the arcades was divided into three stories or channels. In the uppermost flowed the Aqua Julia, in the second the Aqua Tepula, and in the lowest the Aqua Martia. From the ruins of this combined fabric, which still subsist, it appears to have been a very superb structure. The Aqua Virginia was constructed by Agrippa, who laboured to improve and beautify Rome, and who, according to Pliny, formed in one year 70 pools, 105 fountains, and 130 reservoirs. It commenced at a very copious spring, in the middle of a marsh, at the distance of eight miles from the city, and ran about 12 miles, passing through a tunnel of 800 paces in length. The Aqua Claudia, begun by Nero and completed by Claudius, took its rise 38 miles from Rome. It formed a subterranean stream 36½ miles in length, ran 10½ miles along the surface of the ground, was vaulted for the space of three miles, and supported on arcades through the extent of seven miles, being carried along as high a level as to supply all the hills of Rome. It was built of hewn stone, and still continues to furnish the modern city with water of the best quality, which has hence procured it the name of Aqua Felice.

In all these aqueducts the channel for the waters was carried with a regular declivity from the one end to the Aqueduct other of the aqueduct, and such as was sufficient to carry the water easily along. In some cases the declivity, had the line been carried straight forward, would have been too great; it was therefore made to take a circuitous route, winding along the sides of the hills, and prolonging the length of the channel so as to reduce the degree of descent, and cause the water to run gently along. The whole of the channel was regularly built of stone or brick, and arched above to cover in the channel, excepting in those places where it was cut out of the solid rock. Along all the valleys and water-courses which lay in the way it was elevated by a series of arches, all raised to the level of the conduit, resting on massive pillars, and all built in the most solid and substantial manner, with brick, and often with hewn stone. The sketches of Roman aqueducts in Plate XLIV., taken from Fabretti, will give a better idea than any description, of the manner in which the work was executed. Sometimes there was only a single arch, as in the two middle figures; sometimes, again, where the conduit was to be elevated higher, as in the right-hand figure, a double row of arches was raised, one above the other, for greater strength and security. The figure on the left hand shows two conduits or aqueducts carried in different levels along the same building. The upper one is the New Anio, the lower is the Aqua Claudia. When we consider the labour and difficulties attending the construction of such arches and arched channels of masonry, the spirit and enterprise which could have undertaken such works as are above enumerated, undaunted by the expense, or any of the other obstacles which lay in the way, appear astonishing. In this country, where bridges, canals, and other water-works have been carried to a great extent and perfection, we consider an aqueduct of six or eight arches a work of no small extent and importance. What would we think, then, of the aqueduct of the New Anio, extending 6½ miles in one continued series of arches, many of them upwards of 100 feet high! If we allow a similar number of arches in the length, we shall have in all more than 600; and yet this is nothing compared with the aqueduct of the Aqua Martia, extending 38 miles, and containing in all nearly 7000 arches. Even an aqueduct bridge of five or six miles in length appears incredible; and yet how can we otherwise translate Frontinus, where he states the lengths of all the aqueducts, and how much was above or under ground, and how much was built in arches? Of the New Anio, for instance, he says its conduit was 63 miles 700 paces in length. Of this 49 miles 200 paces consisted of a subterranean stream, and 9 miles 400 paces were above ground, of which last the higher part consisted of "substructionibus aut opere arcuato," in several places of great length; and nearer the city, at the 7th mile-stone, consisted of "substructione" 609 paces, and "opere arcuato" 6 miles 491 paces; and he adds, "These arches are the highest of any, being raised in some parts 109 feet." The term substructure probably means a conduit built by opening up the surface of the ground and then covering over the building with earth, as we do in such works at this day. But the term opere arcuato can only refer to a continued series of arches, and certainly conveys a vast idea of the extent and magnitude of such works.

The total quantity of water delivered into Rome by these aqueducts was altogether astonishing, and quite beyond what we have any conception of now, for either comfort or luxury. Strabo said truly, that whole rivers flowed through the streets of Rome. According to Frontinus, the nine earlier aqueducts delivered daily 14,018 quinaria, which corresponds to nearly 28 millions of cubic feet; and when all the aqueducts were in operation, the quantity must have amounted to 50,000,000 cubic feet; Aqueduct which, reckoning the population of the city at that time at a million, would give 50 feet daily, or 7 hogsheads to each individual. This is more than ten times the supply of London, which is now reckoned to be quite profuse.

Of the modern aqueducts in Rome the principal are the Aqua Felice, the Aqua Virginia, and the Aqua Paulina, of Modern Roma. The first was constructed by Sixtus V. It commences at Palestrina, about twenty-two miles from the city, and discharges itself at the Fontana di Termini. The Aqua Paulina was repaired by Pope Paul V. in the year 1612. It divides itself into two principal channels, one of which supplies Monti Janiculum, and the other the Vatican and its neighbourhood. It is conveyed from the district of Bracciano, about twenty miles distant; and three of its five streams are not inferior to small rivers. According to the calculation of Prouy, these three aqueducts, with some additional sources, deliver in twenty-four hours 5,305,000 cubic feet. This, among a population of 130,000, gives about 40 cubic feet for each individual, which is nearly equal in proportion to the supply of ancient Rome in the period of its utmost splendour.

But the system of aqueducts was not confined to the capital of Rome. It was gradually extended throughout the provinces of that vast empire; and every city and considerable town had its conduits and aqueducts for supplying it with water, many of which still remain to attest the magnificence with which these works were carried on. In Plates XLIV. and XLV. we have given a view of some of these. The first is the remains of one of the principal aqueduct bridges in the aqueduct of Antioch. It is a work of great magnitude and height, but a very rude structure. The lower part consists almost entirely of solid wall, and the upper part of a series of arches with very massive pillars. It appears, however, to be a Roman work; and there are remains of another aqueduct on a lower level, and of an older date. The water to Antioch was brought from a distance of four or five miles, from a place called Battelma, which Pococke thinks was the very spot where Daphne stood. Several springs, one of which was so large as to turn several mills, were conveyed in channels of hewn stone, and united in one main stream, which was thence conveyed along the surface of the ground in a similar channel. Across all the rivulets and valleys it was raised on arches or aqueduct bridges, some of which are very lofty, and the principal is the one exhibited in the plate, extending upwards of 700 feet in length, and upwards of 200 feet high in the deepest part. From the remains of the aqueduct in the island of Mytilene, represented from Pococke in the same plate, this appears to have been also a magnificent work. It was built of gray marble rusticated. It is much superior in point of skill to the aqueduct of Antioch, the arches being carried in two ranges throughout the building. It extended about 500 feet in length, and was about 70 or 80 feet high at the deepest part.

The aqueduct or aqueduct bridge of Pyrgos, near Constantinople, forms a portion of the extensive hydraulic works with which that capital was supplied with water after it became the seat of empire. They are described by Andreossy in his voyage to the Black Sea, and account of the Thracian Bosphorus. It is a grand work, very remarkable both in design and execution, and affords a fair specimen of the style of such structures among the Romans in the middle ages. It consists of two branches, one of which only is seen in elevation in the plate; the other stood nearly at right angles to this, and is seen partly on the plan; it was hence called the Bended or Crooked Aqueduct, to distinguish it from another termed the Long Aqueduct, which was situated near the sources of the Aqueduct waters. The branch seen in elevation extends 670 feet in length, and is 106 feet in height at the deepest part. It is composed of three rows of arches, those in each row increasing in width from the bottom to the top—an arrangement very properly introduced with the view of saving materials without diminishing the strength of the work. The two upper rows consisted of arches of semicircles, the lower of Gothic arches; and this circumstance serves to fix the date of the structure, as these last were not introduced until the 10th century. The breadth of the building at the base was 21 feet, and it diminished with a regular batter on each side to the top, where it was only 11 feet. The base also was protected by strong buttresses or counterforts, erected against each of the pillars. The other branch of the aqueduct was 300 feet long, and consisted of 12 semicircular arches.

This aqueduct serves to convey to Constantinople the waters of the valley of Belgrade, one of the principal sources from which the city is supplied. These are situated on the heights of Mount Haramus, the extremity of the Balkan Mountains, which overhangs the Black Sea. The water rises about 15 miles from the city, and between three and four miles west of the village of Belgrade, in three sources, which run in three deep and very confined valleys. These unite a little below the village, and then are collected into a large reservoir. After flowing a mile or two from this reservoir, the waters are augmented by two other streams, and conveyed by a channel of stone to the Crooked Aqueduct. From this they are conveyed to another, which is the Long Aqueduct; and then, with various accessions, into a third, termed the Aqueduct of Justinian. From this they enter a vaulted conduit, which skirts the hills on the left side of the valley, and crosses a broad valley two miles below the Aqueduct of Justinian, by means of an aqueduct with a double row of arcades of a very beautiful construction. The conduit then proceeds onward in a circuitous route, till it reaches the reservoir of Egri Kapan, situated just without and on the walls of the city. From this they are conducted to the various quarters of the city, and also to the reservoir of St Sophia, which supplies the seraglio of the grand signior. The Long Aqueduct is more imposing by its extent than the Crooked one, but is far inferior in the regularity of design and disposition of the materials. It is evidently a work of the Turks. It consists of two rows of arcades, the lower being 48 in number, and the upper 50. The whole length was about 2200 feet, and the height 80 feet. The Aqueduct of Justinian is a very excellent work, and without doubt one of the finest monuments which remain to us of the middle ages. It consists of two rows of large arcades in the pointed style, with four arches in each. Those of the lower story have 52 feet of span, the upper ones 40 feet. The piers are supported by strong buttresses, and at different heights they have little arches passing through them, which relieve the deadness of the solid pillar. The length of this aqueduct is 720 feet, and the height 109 feet. This aqueduct, though it bears the name of Justinian, was probably erected in the time of Constantine.

Besides the waters of Belgrade, Constantinople was supplied from several other principal sources, one of which took its rise on the heights of the same mountains, three or four miles east of Belgrade. This was conveyed in a similar manner by an arched channel, elevated when it was necessary on aqueduct bridges, till it reached the northern parts of the city. It was in the course of this aqueduct that was constructed the contrivance of the souterrai or hydraulic obelisks described by Andreossy, and which has excited some attention, as being an improvement on the method of conducting water by aqueduct bridges. "The souterrai," says Andreossy, "are masses of masonry, having generally the form of a truncated pyramid or an Egyptian obelisk. To form a conduit with souterrai, we choose sources of water, the level of which is several feet higher than the reservoir by which it is to be distributed over the city. We bring the water from its sources in subterranean canals, slightly declining until we come to the borders of a valley or broken ground. We there raise on each side a souterrai, to which we adapt vertically leaden pipes of determinate diameters, placed parallel to the two opposite sides of the building. These pipes are disjoined at the upper part of the obelisk, which forms a sort of basin, with which the pipes are connected. The one permits the water to rise to the level from whence it had descended; by the other, the water descends from this level to the foot of the souterrai, where it enters another canal under ground, which conducts it to a second and to a third souterrai, where it rises and again descends, as at the last station. Here a reservoir receives it and distributes it in different directions by orifices of which the discharge is known." Again he says, "it requires but little attention to perceive that this system of conducting tubes is nothing but a series of syphons open at their upper part, and communicating with each other. The expense of a conduit by souterrai is estimated at only one fifth of that of an aqueduct with arcades." We really cannot perceive any advantage in these pyramids, further than as they serve the purpose of discharging the air which collects in the pipes. For if the water is to be conveyed in pipes across the valley, what other purpose can these columns possibly serve? They are in themselves an evident obstruction, and the water would flow more freely without any interruption of the kind. In regard to the leaden pipes, again, they would have required, with so little head pressure as is stated, to be used of very extraordinary dimensions to pass the same quantity of water as was discharged along the arched conduits. There is something, therefore, which would require explanation in this account of Andreossy regarding these pyramids, or else he has misunderstood the nature of them when he says that they supply advantageously the place of the aqueducts or arches. A train of pipes properly laid, and of proper dimensions, might do this; but what advantage the pyramids possess further than to answer the purpose of air-cocks, is not very apparent.

The other principal source from which Constantinople is supplied, is from the high grounds six or eight miles west of the town, from which it is conducted by conduits and arches, in the same manner as the others. The supply drawn from all these sources amounts, according to Andreossy, to 400,000 cubic feet per day; about two thirds of a foot to each person of a population of 600,000. The charge of the water-works at Constantinople belongs to a body of 300 Turks and 100 Albanese Greeks, who form almost an hereditary profession.

Of the aqueducts which still remain as relics of Roman Aqueducts grandeur, the most remarkable are, the aqueduct of Metz; of Metz, the aqueduct of Nismes, or the Pont du Gard; and the aqueduct of Segovia in Spain. "The aqueducts of Rome," says Montfaucon, "were without doubt wonderful on account of their great length—arcades continued over the space of 40 or 50 miles; their great number, with which the Campagna of Rome was filled on every side: all this surprises us. But it must be confessed that if, without considering the total extent, we only look at any of the parts which remain round Rome, there is nothing that approached the aqueducts of Metz, of Nismes, or of Segovia." The aqueduct of Metz is represented in Plate XLV. Nearly the half of it, it will be observed, has been carried away; but there still remain a great num- Aqueducts, her of arches, and enough to give an idea of the extent of the whole. It extended across the Moselle, a very considerable river, and very broad in this place; and served to convey the delicious waters of the Gorge to the city of Metz. These waters, according to Meurice, in his history of the bishops of Metz, printed in 1634, were so abundant that they furnished water for floating the vessels every time that a naval fight was to be exhibited. They were collected into a reservoir, and from thence conducted by subterraneous canals constructed of hewn stone, and so spacious that a man could easily walk in them upright. They then passed the Moselle by means of the aqueduct, which was situated about six miles from Metz, and from thence were conducted underground in stone channels, similar to the others, to the city, to the baths, to the place of the sea-fight, and all over the city. Judging from the drawing, this aqueduct seems to have been nearly 1000 feet in length, the arches 50 feet high at the deepest part, and 50 in number. They formed only one series, the height not requiring a double row. They were so well built and cemented together, that, excepting the middle part, which the descent of ice down the river has in the lapse of ages carried away, they have resisted, and will continue to resist, the effects of time and of the most violent seasons.

The Pont du Gard was executed by the Romans in the reign of Augustus, and was then merely an aqueduct for bringing the waters of the fountain of Hure to Nismes. It was composed of three rows of arches filling up the valley between two mountains, between which ran the river Gardon. The first row comprehended six arches, each 60 feet span, excepting one, which was the largest, and was 75 feet span; the second row contained 12 arches of the same span as the first; and the third had 36 little arches, on the top of which was the channel for conducting the water. This bridge exhibits a decided improvement and superiority over all the other Roman aqueducts, in the lightness and striking boldness of its design. The arches are wider, and the piers in proportion lighter, than any other structure of the kind previously constructed; and had the same principle been extended so as to have formed only a single row from top to bottom, it would have equalled in the skill and disposition of its materials (circumstances in which the Roman works were almost universally wanting) any of the more judicious and elegant structures of modern times. About the year 1749, this bridge, being of no more use as an aqueduct, was converted into a road-way, by widening it; or rather building in a manner another bridge to the side of it, having all the arches of the same span and dimensions. The execution of the work was attended with considerable difficulties, but these were all successfully overcome by the French engineer Pitot.

The aqueduct of Segovia, according to Culmenares, who travelled in Spain, and has written the history of Segovia, may be compared with the most wonderful works which antiquity has transmitted to us. There still remain of it 159 arches, all built with large stones, and without any cement. There are two rows of arches, one above the other, and the whole height of the edifice is 102 feet. It runs quite across the town, and passes over the greater part of the houses which lie in the hollow.

In modern times various aqueducts have been formed after the manner of the Romans, particularly in France. The most remarkable are those which were constructed in the reign of Louis XIV., at vast expense, for conducting water from Marly to Versailles. Of these the famous aqueduct bridge of Maintenon, which was erected for conveying the waters of the river Eure to Versailles, is without doubt, in point of magnitude and height, the most magnificent structure of the kind in the world. In Plate XLV. we have given a view of a portion of this work, on the same scale as the other aqueducts here represented. Had the whole been delineated on the same scale, it would have extended to four times the breadth of the plate. It extends about 4400 feet in length, being nearly seven eighths of a mile, and upwards of 200 feet in height, and contains 242 arcades, each divided into three rows, forming in all 726 arches about 50 feet span. Of the subterranean aqueducts in France the finest is that of Arcueil, which serves to conduct water to that village. It is 44,300 feet in length, or upwards of eight miles, extending from the valley of Arcueil to the castle at the gate of St Jaques, all built of hewn stone. It is about six feet in height, and has on each side a foot-path 18 inches wide; it has a declivity of one foot in 1300. Another aqueduct of this kind is that of Roquencourt, part of the system which brings water to Versailles; it is 11,760 feet in length, or upwards of two miles, and a declivity in its whole course of only three feet. In some parts of its course it was necessary to make excavations 80 or 90 feet deep, which rendered the execution very difficult.

The great waterworks that supply the city of Marseilles with the water of the Durance, by a canal about 60 miles in length, are among the boldest undertakings of the kind in modern times. This canal, begun in 1830, and not yet completed (1852), has already cost above L2,000,000 sterling. It is conveyed through three chains of limestone mountains by forty-five tunnels, forming an aggregate length of 8½ miles, and across numerous valleys by aqueducts; the largest of which, the Aqueduct of Roquefavour, over the ravine of the River Arc, about 5 miles from Aix, surpasses in size and altitude the ancient Pont du Gard. The immense volume of water, which passes at the rate of 198,000 gallons per minute, is carried across as in the old Roman aqueducts by a channel of masonrywork. The height of this aqueduct is 262 feet, and its length 1287. The number of cubic yards of masonry contained in it is 57,000; the total cost has been L151,394.

One other aqueduct of recent construction is worthy of notice. In those parts of British India where the fall of the rain is scanty and uncertain, recourse is had to artificial irrigation, and the waters of many of the rivers of the country have been rendered available for this purpose by means of public works constructed by the government. Of these the most important is the Ganges Canal, which traverses the north-western provinces of Bengal, and distributes over their vast area nearly the whole volume of the waters of the Ganges. See Ganges. The canal begins at the point where the river issues from the mountains and enters the plains of Bengal. About twenty miles from its source, the line of the canal crosses the valley of the Solani River, and the works for effecting the transit are designed on a scale worthy of the undertaking. The valley is between two and three miles in width. An earthen embankment is carried across, raised on an average between 16 and 17 feet above the surrounding country, and having a width of 350 feet at its base, and 290 feet in the upper part. This embankment forms the bed of the canal, which is protected by banks 12 feet in depth and 30 feet wide at the top. To preserve these banks from the effects of the action of the water, lines of masonry formed into steps extend on each side throughout their entire length. The Solani River is crossed by an aqueduct 920 feet long, having side walls 8 feet thick and 12 deep, the depth of the water being 10 feet. The water of the canal passes through two separate channels. That of the River Solani flows under fifteen arches, having a span of 50 feet each, constructed in the most substantial manner and springing from piers resting on blocks of masonry sunk into the bed of the river. The cost of the aqueduct was upwards Within the last century, the invention and improvement of the manufacture of cast iron has completely changed the mode of conducting water into cities, by the introduction of cast-iron pipes instead of the stone conduits of former times. These pipes can now be formed of almost any dimensions, and united together into a continued series, so closely as to prevent the escape of the water, even under a violent pressure arising from the altitude of the fountainhead. They enable us, therefore, to take advantage of and give effect to that grand principle in hydrostatics, that the fluid element tends continually to a level, even though it be confined in the smallest or most complicated system of pipes; so that however low it be carried in any valley, or to whatever distance, still it will rise on the opposite side to the original altitude of the fountainhead—a principle which is most important indeed in such works, seeing that by it we are not restricted, as the Romans were, almost to a perfect level in the line of the conduit. We have seen that, for the purpose of attaining this level or very gentle declivity all along the conduit, they were under the necessity of raising it by arcades continued in one unbroken series, frequently 30 or 40 miles in extent; and, in addition to this, often prolonging the length of the track by a circuitous route, turning and winding for miles out of its course, for the very purpose of increasing its length.

But the use of pipes enables us to dispense with these long arcades all raised nearly to the same level with the fountainhead; because the conduit may be varied in its level to any extent, and still will rise at last to its original altitude. The pipes, therefore, are merely laid all along the surface of the ground, with a cover of 2 or 3 feet of soil to place them beyond the reach of frost. To prevent, however, the frequent or abrupt alternations of rise and fall, any sudden inequalities in the ground are equalized by cuttings and embankments, but not to anything like the extent that would be required to raise the whole to a level. This, therefore, forms a capital improvement in the method of conducting water, and the greatest indeed which has ever been made in this important branch of practical mechanics. That it was not introduced by the Romans, is not to be ascribed, as many have done, to their ignorance of the hydrostatic principle, that the fluid would rise to a level in the opposite branches of the same train of pipes. Professor Leslie has shown that they were well acquainted with this principle, and has moreover obtained from Italy a portion of a leaden pipe, supposed to have been used in the baths of Caracalla, which sets this matter at rest. But, from the low state of the arts at that period, they were unable to give effect to the principle. They had not the means of fabricating pipes of such a magnitude as would have been required for the enormous quantities of water consumed in Rome, and at the same time of strength sufficient to withstand the pressure from the fountainhead. Lead was the only material that could be used by them for the purpose; and besides the enormous thickness that so weak a material would have required, and the impracticability of their forming them, and uniting them together endwise, they were too well acquainted with the tendency of lead to render the water unwholesome by its poisonous impregnation. The use of cast iron was quite unknown. There remained, therefore, no resource but in the aqueducts, which, though attended no doubt with vast expense, and requiring great enterprise, as well as both skill and patience, were yet attainable by these means, and formed when completed a simple and very perfect mode of effecting the object. Hence arose all those works above described which have since excited such astonishment. Now, however, when the manufacture of cast iron has been brought to such perfection, and methods contrived for uniting perfectly together all the Aqueduct pipes into one connected train, this improved system has been universally adopted.

The Croton aqueduct, by which the city of New York Croton is supplied with water, may be regarded as the most magnificent work of the kind executed in modern times. It was commenced in 1837, and completed in 1842, at an expense of $8,575,000 dollars, the distribution pipes costing $1,800,000 dollars additional. Its length from the Croton lake to the receiving reservoir is 38½ miles. The Croton lake, which is formed by the Croton Creek, a small stream of wholesome water falling into the Hudson, covers 400 acres, and contains a body of water of about 500,000,000 gallons. To the valley of the Harlem river, a distance of 33 miles, the aqueduct is built of stone, brick, and cement, arched over and under, 6 feet 3 inches wide at the bottom, 7 feet 8 inches at the top, and 8 feet 5 inches high; and capable of discharging 60,000,000 gallons per day. It is carried over the Harlem valley in iron pipes laid upon a magnificent bridge 1460 feet long, constructed of arches 114 feet above high-water mark at Yorkville. These pipes pass into the receiving reservoir, which is 1826 feet long and 836 feet wide, covering an area of 37 acres, and capable of containing 150,000,000 gallons. Hence, to the distributing reservoir, a distance of 2½ miles, the water is conveyed by a double line of iron pipes 3 feet in diameter. This second reservoir is 420 feet square and 44 feet above the streets, with a capacity of 20,000,000 gallons,—whence the water is conveyed through the city by about 170 miles of pipe principally from 6 to 12 inches in diameter.

The works undertaken by the Edinburgh Water Company in 1819 were probably the most complete and perfect waterworks of the time. They were designed by Mr Jardine, the then works engineer of the company, and carried out under his superintendence in a style quite worthy of the city, and offering, both in the general design and in all the details, a model of propriety and skill in this species of hydraulic architecture. The Crawley springs were conducted by an aqueduct into a covered cistern at a point about 7 miles distant from Edinburgh, and a supply from the stream called the Glencorse Burn, conveyed by an open-work tunnel from about a mile and a half westward. This tunnel is in some places upwards of 30 feet deep, and the valley through which it passes, consisting entirely of gravel, acts as a filter through which the water descends and percolates, all solid matter being intercepted in its passage to the tunnel from whence it is delivered into the cistern, and conveyed to Edinburgh by a chain of pipes varying from 20 to 15 inches of interior diameter, without being exposed to the light of day. From the numerous undulations of the surface, the fall of the pipe is not uniform. Abrupt inequalities, however, were removed by cutting and embanking. Towards the northern termination of the line the pipe is carried through a tunnel of 2160 feet in length and about 70 or 80 feet under the surface of Heriot's Green. In crossing the Grassmarket it forks off by one branch to a reservoir in the Castlehill, and by another about 120 feet under the reservoir, through a tunnel 740 feet in length, cut through the rock of which the ridge leading to the Castle is composed. Branches were laid through all the principal streets.

The pipes are in lengths of 9½ feet each, and were tested before being laid by a pressure equal to a vertical column of 800 feet of water. The joints are what are termed spigot and socket. Cocks for the discharge of air accumulating in the pipes are placed at the summits of all the considerable elevations; and in the hollows are placed sluice cocks for the purpose of running off sand or other solid matter which may collect in the pipe. It is capable of delivering 253-56 cubic feet of water per minute into the reservoir at the Castlehill.

The formation of the Compensation Reservoir, was un- doubtedly the greatest work of hydraulic engineering of its day. It was designed and completed by Mr Jardine, and with the then limited experience of contractors and workmen in the construction of similar works, its successful completion does great honour to the genius and perseverance of the engineer. It has been twice enlarged, and now forms an artificial lake extending over an area of 46 imperial acres.

In 1843 the Company obtained the authority of Parliament to bring in the Black Springs and those of Listonshields and Bavelaw. These springs are situated on the northern margin of the Pentlands; the Black Springs fully 9 miles, those of Bavelaw upwards of 10, and the Listonshields between 12 and 13 miles distant from Edinburgh, in a direction nearly south-west by west. The Black Springs vary from ten cubic feet per minute to as high as 240. The Listonshields and Bavelaw from 135 to 170. The quality of these waters is nearly uniform, and of great purity and excellence.

In the autumn of 1847, the springs of Bavelaw and Listonshields, about 40 in number, were made available by being conveyed in clay pipes into a stone cistern at a place called Westrigg, about 12 miles distant from Edinburgh. They are thence conveyed for nearly five miles through an aqueduct to a point at the south-west end of Torphin Hill, and thence to Edinburgh by an iron pipe of 16 inches interior diameter. Compensation to the millowners on Bavelaw Burn and Water of Leith was provided by the construction of two reservoirs at Threepmuir and Harlaw, capable of containing about 68,000,000 of cubic feet. These reservoirs form one continuous chain, with not more than 30 feet of difference of level between them when full, and together occupy an area of fully 200 acres.

In the same year the Company constructed a new reservoir in the Glencorse valley, at a place called Loganlea; about a mile to the westward of the Compensation reservoir already referred to; two more on the northern side of the Pentland ridge at Clubbiedean and Torduff; another on the site of two old ones which had been long disused, in a flat and hollow part of the hill to the south of Bonally.

Glencorse reservoir increased its capacity from 46,242,045 to 55,160,409 cubic feet. The Loganlea reservoir contains 18,801,858; the Clubbiedean, 9,836,000; the Torduff, 17,821,459, and the Bonally, 8,000,000 cubic feet. The total storage in these reservoirs is 109,619,786 cubic feet, and their capacity is estimated to be equal to a supply of 350 cubic feet per minute for a period of four months without rain.

The delivery of water over a great part of the years 1851 and 1852 has been upwards of 520 cubic feet per minute, and estimating the population dependent on them for a supply, including the town of Leith, to which a supply was extended in 1852, and Portobello, which was supplied in 1851, at 190,000, even the present limited supply is equal to nearly 25 gallons per head per day of that population. When the whole works are completed and in full operation, this supply will not only be considerably augmented, but will be made secure.

Filters on a large scale, and on the most modern principle, have been provided at Glencorse for the purification of the water that may be taken out of the burn.

The works, under the Act of 1847, were designed by Messrs Rendel and Beardmore of London, and carried out under the superintendence of Mr Leslie of Edinburgh, the Company's engineer. The capital expended in these works amounts to £410,000, while it is understood that a sum of £20,000 or £30,000 more will be required for their completion.

In all places such as the above, where there is a deficiency of level to carry the water naturally to the highest parts of the town, there is no resource but in the employment of machinery. Steam power is most commonly applied to this purpose, and the water is generally either forced into a reservoir of sufficient height to supply the houses therefrom by natural gravitation, as in the city of York; or, it may be forced over upright pipes called stand-pipes, as is the practice in some of the water-works which supply London; or it may be propelled to the houses through a train of street-pipes, as is partially the practice in Glasgow.

But though the system of pipes has thus superseded the use of stone channels all raised to a level in the conveyance of water, there are still cases, such as those of canals, where the water must be kept on a perfect level, and where, therefore, aqueduct bridges are still necessary in conveying it over the valleys; and of these we have long had examples in France, on the Languedoc canal. The first aqueduct bridges for canals in this country were those made by the Duke of Bridgewater, under the direction of the celebrated Brindley, and which, being quite new here, excited no small degree of astonishment. The first and largest was the aqueduct at Barton Bridge for conveying the canal across the Irwell, 39 feet above the surface of the water. It consisted of three arches, the middle one 68 feet span, and admitting under it the largest barges navigating the Irwell with sails set. It was commenced in September 1760; and in July of the following year the spectacle was first presented in this country, of vessels floating and sailing across the course of the river, while others in the river itself were passing under them. Since that period canal aqueducts have become more common; and many excellent examples are to be found both in England and Scotland. Of these are the aqueducts over the river Lune, on the Lancaster canal, designed by Rennie, a very excellent and splendid work of five arches, each 72 feet span, and rising 65 feet above the level of the river; and the Kelvin aqueduct, near Glasgow, which conveys the Forth and Clyde canal over the valley of Kelvin, consisting of four arches, each 70 feet span, and rising 70 feet above the level of the river. In Plate XLV., we have given views of three other principal aqueducts, viz., the aqueducts of Pont-y-Cysylte, of Chirk, and of Slateford near Edinburgh. Of these the Pont-y-Cysylte by Mr Telford is justly celebrated for its magnitude, for the simplicity of the design, and the skilful disposition of the parts, combining lightness with strength in a degree seldom attempted. This aqueduct serves to convey the waters of the Ellesmere canal across the Dee and the vale of Llangollen, which it traverses. The channel for the water is made of cast iron, supported on cast-iron ribs or arches, and these resting on pillars of stone. The iron being much lighter than stone arches, this is one reason why the pillars have been reduced apparently to such slender dimensions. They are quite strong enough, however, as experience has proved. The whole length of the aqueduct is about 1000 feet, and consists of 19 arches, each 45 feet span. The breadth of the pillars at the top is 8 feet, and the height of the four middle ones is 115 feet to the springing. The pillars have a slight taper, the breadth of the middle ones at the base being 15 feet. The height from the surface of the water in the Dee to that in the canal was to be 126 feet eight inches. The channel for the water consists of cast-iron plates, cast with flanges, and these screwed together with bolts; they are represented in the drawing, between the arched ribs and the railing. The lines there show the joinings of the different plates. In order to preserve as much water-way as possible, the channel is made the full width of the canal and towing-path, and the latter projected over one side, and supported inside by posts resting on the bottom of the canal. The aqueduct of Chirk was designed by the same able engineer, and serves also to convey across a valley the waters of the same canal. This aqueduct was the first in which any iron was employed. Hitherto the channel for the waters had been constructed of stone, or partly of stone and partly of clay puddle, which it was generally found very difficult to Aqueduct keep water-tight for a length of time. It was determined, therefore, by Mr Telford, to try the effect of cast iron, and to lay it at first only on the bottom. The plates were accordingly laid directly over the spreader walls, which they served to bind together, and united by flanges and screws. The sides of the channel were built with stone facings and brick hearting laid in water-lime mortar. This plan has succeeded completely, and the quantity of masonry in the aqueduct was thereby greatly reduced. The aqueduct itself is 600 feet long, and 65 feet high above the river, consisting of ten arches, each 42 feet span. The piers are ten feet thick.

The aqueduct of Slateford conveys the waters of the Edinburgh and Glasgow Union Canal across the valley of the Water of Leith at Slateford. It is an elegant structure, similar in plan to that of Chirk, only that the water-channel is composed entirely of cast iron, which is moreover built in with masonry. It is about 500 feet in length, and consists of eight arches, each 45 feet span; and the height of the canal is about 70 feet above the level of the river. On this canal another aqueduct of the very same construction occurs in crossing the valley of the Almond, and having several more arches. There are, in different parts of the country, various other aqueducts, which might be described, did our limits permit. It is the less necessary, as, excepting the formation of the water-way, these structures differ in no respect from bridges, particularly those undertaken not so much with the view of crossing rivers as of raising the level of the road entirely out of the valley,—an object now become of great importance, from the improvements within the last half-century in our modes of conveyance. Formerly people were content to traverse slowly all the inequalities of the country through which the road might pass, descending into the valleys, and mounting the steepest acclivities. Now, however, a road is thought imperfect, and quite behind the standard of improvement, unless every rise greater than 1 in 15 or 1 in 20 feet be cut down. In crossing the valleys, therefore, it is not enough now that we build a bridge in all respect sufficient for crossing the stream itself; we must raise it nearly to a level with the ground on each side of the valley; and this gives rise to new and very extensive works of this kind. Of these we may just instance the splendid bridge of one arch of 140 feet span, built over the Den Burn at Aberdeen, to form a new access into that town; also the beautiful bridge of Cartland Craigs, built by Mr Telford, over the little stream of the Mouse, on the new road from Glasgow to Carlisle, consisting of three arches 50 feet span, and elevated 130 feet above the bed of the stream. More recently the introduction of railways has opened a new and still wider field for the skill and talents of the engineer in the erection of such works. This species of road, it is well known, must be kept still more nearly on a level than any of the roads of the ordinary construction. (See RAILWAY.) In this respect the railway is somewhat similar to the old Roman aqueducts, and, where the country is low, must in like manner be elevated on a series of arcades. These bridges have received the name of Viaducts; there is an extensive one on the Liverpool and Manchester railway, termed the Sankey Viaduct, of nine arches; and numerous others have since been constructed throughout the country. But for the principles and mode of construction of these works, as well as of the aqueduct bridges, so far as the arch is concerned, we refer to the articles ARCH and BRIDGE; and for further information on the subject of aqueducts, see Julius Frontinus, De Aqueductibus Urbis Romae; Raphaelis Fabretti De Aquis et Aqueductibus Veteris Romae Dissertatio; Famiani Nardini Roma Vetus, lib. viii, cap. iv.; Plini Hist. Nat. lib. xxxvi. cap. xv.; Montfaucon, Antiquité Expliquée, tome iv. tab. 128; Governor Powys's Notes and Description of Antiquities in the Provincia Romana of Gaul; Bellidor's Architecture Hydraulique, containing a drawing of the aqueduct of Maintenon; also Mem. Acad. Par.; Andrassy, Voyage à l'Embouchure de la Mer Noire, ou Essai sur le Bosphore; Philosophical Transactions Abridged, vol. i.; and Link's Travels in Portugal.