The lake formerly called “Lacus Alsietina,” now called Lago di Martignano, is situated on the hills on the western side of Rome, between the Via Aurelia and the Via Claudia, (not far from the old carriage-road from Florence to Rome). It is about 679 ft. above the level of the sea; and as Rome is only 204 ft. at the Porta Maggiore, the lake is 475 ft. above the level of Rome at that high point. There is another small lake about a quarter of a mile from, and a little above the level of, the Alsietina, called Strachia Capra, which has recently been drained by a tunnel in imitation of the emissario from the lake of Albano. The water in the Lacus Alsietina has also been very much lowered in the same manner[105]. In consequence of this reducing of the level of the water, the specus of the aqueducts are now brought to view, being on the bank and above the present level of the water, so that they can distinctly be seen and entered into[106].

To begin with the highest, the specus of the Aqua Paola begins in the upper lake, and forms a junction with the water from the Alsietina Lake, near the bank of the latter, at the south end of that lake. The specus of the Alsietina of Augustus is a tunnel cut in the rock[107]. The Lacus Sabatina[108] is at the elevation of about 523 ft., and therefore 156 ft. below the Alsietina. Augustus drew an additional supply of water from this lake, and this portion only of the aqueduct of Augustus was restored by Trajan. The water for this was not drawn from the lake itself, but from the springs that supply the lake on the western shore, the side most distant from Rome. Some of the work of the time of Trajan is visible there. The specus of the Alsietina has been traced to the point of junction with the Aqua Paola (this water having been restored to use by Pope Paul III., A.D. 1540), and it still supplies the district of the Trastevere. It does not appear that Trajan used the water of the Alsietina at all, but the engineers of Pope Paul evidently did so; a tunnel, or specus, from that lake having been made in the time of Pope Paul, the entrance to which remains, with the grooves for a flood-gate[109]. After the junction with the Aqua Paola, the old wells can no longer be traced; their place is supplied by the constructions called respirators, each of which is a small square structure surmounted by a pyramid. These are evidently built over the wells, and very rudely constructed of the stone of the country, worked rough. When we arrive at the Osteria Nuova,—which is near the site of the ancient city of Cariæ[110], mentioned by Frontinus, and the old village of S. Maria in Celsano, the point of junction of the Alsietina and Sabatina,—there are evident marks of another junction of aqueducts there. The house itself, now used as an osteria or hostelry (auberge), is made out of an ancient Castellum Aquæ, as is frequently the case with similar houses in the country round Rome, and sometimes even within the walls. At about half a mile from this osteria, in a hollow, is a white house called Casale Bianco; and close to this is a fountain, supplied by a spring which runs into a specus. This water, being one of the sources of the Aqua Paola, passes through the hill at a considerable depth. At about half the distance between this and the osteria is a remarkable passage for the Aquarii into the tunnel by a very steep descent, passing sideways across the specus of the Aqua Paola, and going on to a much greater depth to the Aqua Alsietina[111]. This passage is 150 ft. in length from the surface of the hill, and 70 ft. in perpendicular depth, with ninety steps, of which only a few at the top are visible; the others are covered with earth. The water still runs through the upper specus, and is still in use, but not through the lower one[112].

From the valley beneath the sloping passage for the Aquarii, near the Osteria Nuova, Nibby[113] traced the Alsietina as a tunnel cut in the tufa rock, a part of which he saw near an oil mill. The specus then followed the low ground from the tenements of S. Niccola, Porcareccina, Maglianella, and from the Villa Panfili, to the principal gate of the monastery of S. Cosimato (or SS. Cosmas and Damian in the Trastevere), where it is stated by Cassio[114] that the specus was found in 1720, about thirty feet underground. The Naumachia of Augustus is said to have been near this monastery.

The complaints of Frontinus appear to have been listened to by the Emperor, and great changes and improvements were made in this aqueduct under his direction, in the time of Trajan; this is in fact the same as the Aqua Trajana (see X.), and the water still comes from the Lacus Sabatina, but not from the Alsietina. In going from Rome, the respirators can be followed across Monte Mario, and near the high road that passes over Ponte Molli, to the point of junction about ten miles on the road to the Cariæ [now Osteria Nuova]. Here the respirators cease; but their place is supplied by a line of old wells descending into the subterranean specus, which follows the line of the old road. This is not always the same as the new one. The other branch, which supplied the Naumachia, was the only one made in the time of Augustus. Paul V. repaired this aqueduct along the whole line, restored it to use, and put up in various places inscriptions recording this, in which he calls it the Aqua Alsietina[115]. The last of these is on his fountain at the mouth of the aqueduct, where the water still gushes out in great abundance, as it did in the sixth century, when it was observed by Procopius; this is on the Janiculum, above S. Pietro in Montorio, the highest ground in Rome.

VIII., IX. The Claudia and Anio Novus (A.D. 52).

VIII. The Claudia.

IX. The Anio Novus.

The Neronian Arches.

The River Anio.

The highest source of the river Anio, or Aniene, is in a gorge in the highest part of Mount Cantaro, about 63 miles from Rome. This spring never fails in the hottest and driest weather, and is always cool; it is situated in the close or (serra) of St. Antonio, and is called La Canala, or the canal at the gate of the small castle or fortified village of Filettino. Pliny says[132] that it rises in the territory of Treba[133]. In going from Filettino towards Trevi the road has an ancient pavement, and on the left-hand side is a magnificent substructure of Cyclopean character called Mura Saracine, against the cliff. In 1857 Signor Gori saw some of the large stones of this fine ancient substructure thrown down to the banks of the stream, to make the foundations of a small bridge across it. The stream receives several accessions from other springs in its course, two from a cave called Pertusu, the small limpid stream called Suria, on the hill of Trevi, and another called Capo d’Acqua. The Anio thus augmented falls in two cataracts, or cascades, near Pertusu, and passes under the Ponte delle Tartare with much violence. This bridge is a natural arch of rock formed by the force of the water piercing it. The river then passes on through a gorge in the mountain-pass in small waterfalls, especially at a place called Pendema, where in 1855 an ancient mosaic pavement and a wall of reticulated masonry were found.

At the two bridges of Communacchio (comune acqua), where an amphitheatre of mountains and the castle of Valle-pietra are situated, another stream coming from the mountains of the Trinitá and Autore, the highest in that district, joins the Anio.

In the basin of Valle-Pietra, two fine cascades fall over the massive rocks and unite in a single stream. From the bridge of Communacchio a road leads to Arcinazzo, situated in a large plain, surrounded by the mountains, with several roads leading into it. One of these from Palestrina, passing by Piglio, is an ancient paved road, and some persons consider that an arcade of Cyclopean masonry, near Guarcino, carried an aqueduct to the thermæ there. On the south-west side of this plain are the ruins of an imperial villa, commonly called of Nero; but two inscriptions found there upon leaden pipes in 1860, shew that it was of Trajan[134]. Great quantities of marble were dug up here in the time of Pius VI., A.D. 1795.

The river passes below the monastery of S. Benedict, called the Sacro Speco, and a little lower down, the monastery of S. Scholastica, both of which stand on the brink of a precipice, over which the Anio makes a large and picturesque waterfall, and passes under the bridge of S. Mauro or Piedilago, where the rocks approach so closely as to leave only a narrow passage, which was formerly closed by a gigantic wall, forming a long lake or lock. The stream now passes through the ruins of the wall, the demolition of which was caused in 1305 by two ignorant monks, who pierced a hole at the foot of the wall, the result being that the whole country was inundated up to the walls of Rome, and serious mischief done, as recorded in the anonymous chronicle of the monastery[135].

On the banks of the lake are the ruins of a villa of early character, called Casa de’ Saraceni and Carceri, with a nymphæum and baths. At Pianigliu, are other ruins on an enormous scale.

Subiaco is more than two thousand feet above the level of Rome, and all the foregoing description applies to the part of the river above Subiaco. In all this upper part of the stream, the river Anio has very much the character of a large mountain-torrent rushing through the rocks with great violence. It usually appears to be a clear stream of beautiful water, excepting in time of floods, when the clayey soil is brought down into the stream and makes it muddy. To guard against this evil, Nero made his great lakes, which do not correspond at all to the usual English idea of a lake, by which is commonly understood a considerable sheet of water, like the Swiss lakes or the lakes of Westmoreland, Scotland, and Wales. The Roman word lacus, and the modern Italian lago, may mean a lake of this description also; but it includes a reservoir of water of any kind. We are expressly told in the Breviarium, at the end of the Catalogue of the Regionaries, that a lacus is a well, puteus, and the numerous lacus of that Catalogue are the same as the castella aquarum of Frontinus, within the walls of Rome. These justly-celebrated lakes of Nero are in fact portions of the river Anio, intercepted in a gorge of the rocks about a mile above Subiaco, and are formed by cutting away some large pieces of the rock on each side in large masses, and with these building a great high and massive wall across the stream, forming an effectual barricade or dam to stop the water and raise it to the level of the top of this great wall. There were three of these walls across the stream, over each of which the river fell in tremendous cascades. The first loch (lacus) commenced at the Mola di Ienne, where the first great wall of enclosure is situated. The second at the cascade of the river, under the great monastery of S. Benedict, and called “the Sacred Specus,” from a cave in which the saint is said to have lived, extends to the bridge called Ponte di S. Mauro, where the specus of Trajan commenced. The third is from the Ponte di S. Mauro, through a gorge, where the wall of enclosure is visible. Here was the Piscina Limaria of Claudius, and the specus of the Anio Novus originally commenced at the Emissarium, restored by Cardinal Barberini. This was made by Claudius, but was abandoned by Trajan, because the earth from the adjoining fields had fallen between the river and the specus.

The Villa Sublacensis of Nero was below the level of the upper lake or loch, as mentioned by Frontinus. This was just above the present bridge called S. Mauro or Piè-di-lago, which seems to be made upon the two ends of the dam. This dam, wall, or barricade, was quite 150 feet above the surface of the water in ordinary times. The bridge is still 144 feet above the water by measurement, and the specus is nearly ten feet higher. This specus or conduit is cut in the rock of the cliff on the side of the valley, at the level at which the water originally stood in this lake or loch; the wall was a few feet higher, in order to force a portion of the water to pass through the specus before the rest fell over the cascade. There are ruins of the piscinæ on the bank where the specus began, and of the villa of Trajan on both sides of the loch, with the well-known brickwork and Opus Reticulatum as facings for the walls. These magnificent cascades still remain, being a natural formation; but as the bed of the river is very deep, they are much concealed by the banks, and the shrubs upon them. The great walls or dams of Nero and Trajan being brought out to the edge, the cascades falling over them must have had a much finer effect, although the natural site is extremely grand and most picturesque; in fact it is celebrated among all the landscape painters of Italy, the scenery about Subiaco being among the finest of its kind of river and mountain scenery that is known. The enormous reservoirs or lakes, or lochs of Caligula and Claudius (commonly called of Nero), cover the space between the natural cascade and the outer wall of rock artificially constructed, over which the water was made to fall. When the ignorant monks in the fourteenth century made an aperture in the great wall or dam, the force of the water soon enlarged it, and washed the whole structure away, leaving the great masses of stone or rock, of which it had been built, scattered in the stream below as if they were natural rocks, where they still remain. The object of the monks was to release their fields adjoining to the monastery above the falls from a temporary flood.

The specus is nearly six feet high, and only sixteen inches wide; the men must have cut it standing sideways. There are apertures into it at intervals now open, and there probably always were such openings for the use of the aquarii to keep the course clear. The specus which Frontinus calls subterraneus, although that is literally true, does not mean exactly what we now call a tunnel; but this specus is cut in the rock of the cliff, with a few feet of stone only as an outer wall to it, and in this manner it is continued along the edge of the valley of the river Anio for many miles, always on the left side of the river in going towards Rome. An old road runs by the side of it, not now used for carriages, but remaining as a cart-road only; this must be the Via Sublacensis of Frontinus. Another road runs along the right-hand side of the valley, and is the one now in use: this is the Via Sublacensis Neroniana of the time of the Empire. The valley varies in width very much, in some parts it is three or four miles wide; this is the case where the springs of the Marcia and the Cerulean Lake gush out from the rock under the diverticulum of the Via Valeria or present carriage-road, on the right-hand side of the valley.

The lowest of the three lakes of Claudius above Subiaco was circular, the rock being cut away to a half circle on each side of the stream; into this great basin the grand cascade fell from the second lake. The lowest lake or loch was comparatively not very deep. It seems most probable that the lowest reservoir was intended to serve for the Aqua Claudia. The specus has not at present been traced quite so far; but it is found a little lower down, above the modern paper-mill. This is more than a hundred feet below the level of the Anio Novus. It may be that this was one of the springs that fell into the Anio, and was intercepted for the aqueduct. It is probable that the same was the case with the Anio Vetus, as we know it was with the Marcia; but as the water from the springs sometimes ran short in dry seasons, the Anio Novus was taken from the river itself, a part of which was turned into it from the great lake or loch. For this reason that water was always more abundant than all the rest, and was used to supply the deficiency in case of need, as Frontinus tells us. This specus can be entered and examined; it is here a tunnel made in a rock of soft stone, with fissures filled with clay. The specus is lined with brick, and covered with large flat tiles, placed at an angle, so as to form a roof sloping down to the two sides from the ridge in the middle. There are inscriptions recording repairs by Cardinal Barberini, nephew of Urban VIII. This specus is that of the Anio Novus, constructed by Claudius; that of Trajan is about half-a-mile above the town of Subiaco, and on the right-hand side of the river Anio, not on the left, as the Anio Novus is.

The thirty-eighth milestone on the Via Sublacensis was found by Fabretti in situ, and the thirty-eighth milestone in the Via Valeria, now at Arsoli, is said by Gruter[136] to have been formerly at La Sonnoletta, or ad fontem Somnulæ: so that the two sources of the Claudian water, called Cæruleus and Curtius, were at the lake now called S. Lucia, in the territory of Arsoli; and the source called Albudinus is the first of the four springs now called Acque Serene, while the other springs of the same name formed the Aqua Marcia.

The Piscina Limaria, referred to by Frontinus as erected by Claudius for filtering the stream, at the entrance of the Anio Novus, at the forty-second mile[137] on the Via Sublacensis, is visible at the Parata della Cartiera of Subiaco. This piscina is not covered over, but open at the top, and is excavated in the rock of the bed of the river Anio, with a great declivity and with four cascades to throw down the sand, wood, and weeds brought down the river, and so to purify the water; but as it was still liable to become muddy in the time of floods, Trajan, according to Frontinus, excavated another specus in the rock of the mountain near the Ponte di S. Mauro, where the great wall of the lake was situated[138].

The specus of the Claudia seems to have been carried on the right-hand side of the valley as far as Vicovaro, about half-way to Tivoli, and then across the valley on an arcade, and carried on under the Anio Novus, and above the Marcia and Anio Vetus. All of these seem to be at different elevations on the side of the hill to the left of the valley, until they arrive at another valley crossing this, called the Valley of the Arches, about two miles from Tivoli, where they were carried across upon arcades. On these ancient arcades, or out of the materials of them, modern bridges have been made, both at Vicovaro and in the Valley of the Arches. The ruins of these splendid arcades are among the finest and most picturesque ruins of the Roman empire.

After passing the bridges across the valley of the Anio, the Anio Novus, Marcia, and Anio Vetus are continued along the side of the hill in what may still be called the cliff of the valley of the Anio, to near the cascades at Tivoli, at different levels, the Anio Novus considerably higher than the others. To avoid the cascades here the aqueducts wind round the end of the hill. In going out of Tivoli to the promenade of Carciano, on the side towards Rome, there is a large college of the Jesuits on the left hand, the specus of the Anio Novus passing under this, and through a wine-cellar. About a quarter of a mile out of the town, two specus are visible on the side of the hill above the road or promenade, and these may be traced at short intervals for miles, with openings into them in several places. The lower one is the Marcian, passing in a more direct line, and further from Tivoli, and is generally cut in the rock as a tunnel; the upper one is the Anio Novus, and is in some places faced with brick or with Opus Reticulatum, where it has to cross an opening. The Aqua Marcia here passes at a lower level below the road, with reservoirs at intervals, as already described.

At about three miles below Tivoli, and half-a-mile after passing one of the great reservoirs of the Marcia at the lower level, the Anio Novus is carried on an arcade across a valley and a small stream, at a place called after the arcade Arcinelli. The great Villa of Hadrian is nearly under this at the foot of the hill, perhaps a mile lower down. This was supplied with water by branches from the aqueducts, but without interfering with the main streams, which went on at the high level still at least five hundred feet above the level of Rome. From this high level the aqueducts led gradually down in a serpentine course, crossing several narrow valleys or gullies through the hills, along which some mountain-stream flows far below. At two important points, the aqueducts have to be carried across such valleys or gullies on fine arcades or bridges. The one nearest to Tivoli is called the bridge of S. Antony, from a small chapel made in the Middle Ages, probably in the fourteenth century, out of a portion of the specus of the Marcia, in one of the chambers of a castellum aquæ belonging to it, the rest of which has been destroyed. The specus of the Anio Vetus under it has also been destroyed, but the arcade is preserved for the use of horses and foot-passengers. This bridge is about eight miles from Tivoli. There is no carriage-road to it, but a tolerable path for horses or donkeys; it is one of the finest and most picturesque objects on the whole line of the Aqueducts, and is about 100 ft. above the water in the stream below which is called the fosse of S. Antony. This valley or gorge in the mountain, with the bridge across it, is not easily seen from any distance; but the site may be indicated by a medieval castle with a tall square brick tower, which is a conspicuous object for miles. This is distinctly seen from the bridge, and as it stands at the mouth of the valley, that object must be seen from thence on looking up the valley. This bridge is a really grand work; it is 373 ft. long, 16 ft. wide, and about 104 English feet high. The dome of S. Peter’s at Rome is visible at a long distance from the hill above the medieval castle.

Following the direction of the aqueducts towards Rome for about a mile, along a very rough cross-country path, we arrive at a good carriage-road from Rome to Poli, and about a mile beyond that is another magnificent structure, as fine and more perfect than the last, called Ponte Lupo, one of the greatest works of all the Roman aqueducts. There are not so many open arches as at the bridge of S. Antony, but a greater extent of substructure and wall at each end, and the height above the mountain-stream is even greater. Several of the arches seem to have been filled up for greater strength, at an early period; it is all work of the first century. Here the two specus of the Claudian and the Anio Novus are perfect, one upon the other, and serve as a lofty parapet to the road for horses, which passes across the valley along the side of them, the specus occupying about one-third of the width of the bridge. The two specus seem to have been first brought together at this point, and afterwards continue one upon the other for the rest of their course into Rome, interrupted only by the piscinæ and castella aquarum at intervals.

There are ruins of another arcade across the valley at a lower level by the side of this grand bridge, only a few yards from it, which can only be that of the Marcia. The Anio Vetus appears to be carried across under the Claudia and Anio Novus at a much lower level, and it is probably for this reason that the arches are closed instead of being left open. They pass upon four other bridges over the streams of S. Gregory and of the Inferno, before they arrive at the piscinæ. They all four meet at this point, and then diverge again for miles. This splendid work is about ten miles from Tivoli, half-a-mile from the carriage-road, five or six miles from the Villa of Hadrian, and out of the line of the direct road to Rome, but very near the road from Rome to Poli. After passing these great works, the aqueducts are continued at their respective levels in the direction of La Colonna and of the piscinæ, and are carried in tunnels through the hill, but these tunnels do not appear to be of any great length; they merely pass through part of one side of the hill. They bring us to the piscinæ, which are about six miles from Rome, where the arcades begin, and from this point the Marrana follows the same direction in the bed of the Almo, winding about but never very distant from the arcades, and always receiving the surplus water. This stream is here divided into two branches, one of which goes through Rome and into the Tiber at the Pulchrum Littus; the other passes through the valley of the Caffarella, and falls into the Tiber near the church of S. Paul beyond the walls, as is explained under the head of the Aqua Crabra and the Marrana.

The specus of the Anio Novus is always faced with brick or with Opus Reticulatum; it is carried upon the Claudian arcade, and the one specus always rests upon the other: what applies to one applies to both after the first junction. The aqueduct of Claudius has a stone specus carried on a stone arcade for the last five miles into the city. It was begun by Caligula, and carried on by his successor; this portion was completed in fourteen years, a very short period for so enormous an undertaking. The Anio Novus was the most abundant of all the aqueducts, as stated by Frontinus[139].

The arches of aqueducts, stretching for miles across the open country and entering Rome at the eastern angle, are the first objects to attract the attention of strangers on approaching the city. These are the arches of the Claudian aqueduct, built of large square stones, with the angles chamfered off, and carrying the streams of the Aqua Claudia and of the Anio Novus.

Of the Piscinæ mentioned by Frontinus, the ruins of one remain at the place now called Porta Furba, where the aqueducts just mentioned meet again about two miles from the city. The lofty Claudian arcade passes over the Marcian arcade at this point, at one of the numerous angles, as it had previously done at the Torre Fiscale. From the Porta Furba to Rome, the Aqua Felice is carried against the side of the Claudian, or on the piers of the Marcian, as most convenient, frequently crossing from one to the other on arches over the road. The specus of the Felice is built in the roughest manner of old materials taken from the other two arcades. The Anio Vetus also passes underground at the Porta Furba. Immediately after the entrance of the Claudian specus into Rome at the extreme east end, in the gardens of the Sessorian Palace (now S. Croce), there are remains of at least four reservoirs or castella aquarum before it reaches the Porta Maggiore[140].

Nero was the immediate successor of Claudius, A.D. 54-69, and carried on the conduit into the city, on what are still called the arches of Nero, which are faced entirely with brick, and some of the most beautiful brickwork in the world. So gigantic a work as these aqueducts could not have been completed in the lifetime of a single emperor, however large the number of slaves he may have employed upon it.

The Neronian Arcade.

The arcade of Caligula and Claudius, which is entirely of stone, terminates at the Porta Maggiore, the Esquilina of Frontinus. The work of Nero includes the arches within the city from the wall close to the Porta Maggiore to the great castellum on the Cœlian, over the arch of Dolabella. The specus of the Anio Novus is easily distinguished from the specus of the Claudia, as the latter is of squared stone.

The arcade of the Claudia has been considerably repaired with brick, and the arches filled up in several places with brickwork of the time of Septimius Severus and Caracalla, A.D. 193-217. This may be distinctly seen in that part of the arcade which is between two and three miles from the city, near the Porta Furba. It was again repaired by Pope Hadrian I. in 780, and several times by other Popes.

The remains of these two aqueducts, one above the other, are admirably seen in their course along the top of the Porta Maggiore; and at this spot their relative levels with regard to the Marcia, Tepula, and Julia, are also clearly exhibited, the latter entering the wall close to the gate, and almost at right angles to it. Thus, at one view, we are able to see the specus of five out of the nine aqueducts, mentioned by Frontinus, actually remaining. A sixth, the Anio Vetus, also passes under the Walls of Rome at the same point; the specus is half underground, and now concealed by the restorations made in 1869.

An inscription upon the face of the specus itself in the wall of the city over the archway, records that Claudius the son of Drusus, caused to be brought into the city the water of the Claudian conduit, from the springs called Cæruleus and Curtius, from 45 miles distance; also the water of the Anio Novus, from 62 miles, at his own expense[141]. These dates here given correspond with the year 798 of Rome, or A.D. 45. Great repairs were made by Vespasian and Titus to the Claudian aqueduct. The distance from which the Anio Novus is brought, according to this inscription[142], is 62 miles, which agrees with Frontinus[143].

The architect of the Claudian is believed to have been Claudius Annius Bassus, mentioned by Tacitus as chief engineer at Carthage under Marcus Silanus, the father-in-law of Caligula. The remains of the tomb of Bassus may be seen near Vicovaro. The tomb of another architect and his family, on the side of the agger upon which the arcade of Nero stands, near the Porta Maggiore, was excavated in 1865. From its situation immediately under the Neronian Arcade against the bank on which it stands, there seems no doubt that this Tiberius Claudius Vitalis, whose name is inserted on the front of the tomb, was the architect of the arcade; and it does infinite credit to him, for it is one of the finest pieces of brickwork in the world.

This arcade, commonly called the Arches of Nero, carried the specus of the Claudia[144] and Anio Novus combined. It first crosses the foss on a double arcade, one upon the other, to give more height across this wide and deep inner trench; it is then carried on a high bank, and therefore on a single arcade only to the Lateran, and on another bank across the foss between the Lateran and the City to the Cœlian, and along the north side of S. Stefano Rotondo to the reservoir and piscina over the arch of Dolabella. This arch was the principal entrance to that part of the Cœlian in which the Claudium, or courts and temple of Claudius, were situated. There are magnificent ruins of a large castellum aquæ over this arch, faced with the beautiful brickwork of Nero; in part of this the small church of S. Thomas in Formis has been made.

Another large subterranean reservoir remains perfect on the west side of this lofty brick castellum of Nero. This consists of three large vaulted chambers under the garden between the church of the twelfth century and the small monastery of the Redemptorists, now (1872) all belonging to the Villa Mattei, or Celimontana. There is, through the crown of each of the vaults, a circular opening, or well, closed by a stone, which can be moved at pleasure for letting down buckets into the water; it is still used for the purpose of irrigation. From the low level of this reservoir, it must have belonged to the Aqua Appia. By the side of the garden over this, and above ground, in a line with the Arch of Dolabella, is a wall of the first century, faced with reticulated masonry, probably part of the castellum of that period, as new reservoirs have been built there for each successive aqueduct that came to this point. The highest, being that of Nero, was 50 feet above the level of the ground.

Thence the water was distributed in different directions, one branch to the Claudium[145], and thence again to the stagnum or pool originally of Nero, but afterwards retained under the Colosseum; a second to the Palatine, passing down the western side of the Clivus Scauri opposite to the church of SS. John and Paul. Then, after making one of the usual angles, it was carried across the valley on the arches attributed to Nero, but in this part really after his time, the lower portions of some of which remain. The third branch was to the Aventine. The plan for this was not carried out until the time of Trajan, when a lofty arcade was made across the valley from the Cœlian to the Aventine, passing over the Porta Capena above the Aqua Appia, which had previously been made in the same line, and with new reservoirs for it, generally by the side of the old subterranean ones. One great reservoir under the Cœlian occupies a considerable part of the space between the cliff and the present road, and is now turned into a gardener’s house. The excavations made there in 1868 have been described in the account of the Appia; the piers of the tall brick arcade of Trajan remain on both sides of the road leading direct to the north end of the ruins of the Piscina Publica, on the other side of the Marrana, under S. Balbina; the upper part of this great building is of the time of Trajan. On the side of the Cœlian are remains of another large castellum, of two stories, of the same period. There are also remains of this aqueduct on the brow of the Aventine Hill, near the church of S. Prisca, in front of the monastery, now in a vineyard opposite to the Palatine, and overlooking the Circus Maximus.

The water both of the Anio Vetus and of the Anio Novus was of inferior quality, and was used chiefly for watering gardens, and for the more common purposes in the city; the Anio Novus, being higher than any of the others, and the water very abundant, assisted all the rest. Trajan endeavoured to improve the quality of the water by excluding the more turbid sources, and using only the most pure, as we learn from Frontinus, who was the person charged with the execution of the work[146]; and this inscription was put up, IMPERATOR CAESAR NERVA TRAJANVS AVGVSTVS.

The waters of the Claudian and Anio Novus are stated by Frontinus[147] to have been united after their entrance into the city, and then distributed to all the fourteen Regiones. Several subdivisions must therefore have been made at different points, and, wherever a division or a junction took place, a castellum aquæ was required. The union of the two streams was probably made in the large reservoir at the angle of the Sessorian gardens and City wall near the Porta Maggiore. Another great division was over the Arch of Dolabella; one branch was afterwards carried on by Domitian to the Palatine upon the lofty arches, some of which still remain, and into his reservoir at the south-west corner of the Palatine Hill, part of the baths or thermæ of his palace. The branch over the Via Appia to the Piscina Publica and the Aventine was not made until the time of Trajan and Hadrian. A great deal of work was done to the aqueducts at that period. The amazement of the people at seeing copious streams of water pouring over the arid heights and slopes of the Aventine, is recorded by a contemporary author. This was the branch of which a portion remains near Santa Prisca. The branch from the Palatine to the Capitol, of which two of the tall piers remain[148], was made by Caligula; this is sometimes called a bridge across the Forum. To convey the water to the other Regiones, the older conduits or specus were probably used.

In the fourth century, there were great complaints of the stealing of the Aqua Claudia by the farmers through whose lands it passed, and several strenuous decrees against this practice were issued by the Emperors Arcadius and Honorius, A.D. 400. Similar edicts were issued repeatedly by these Emperors, and by Constantine[149].

The springs called Cæruleus and Curtius, to which (when united in this aqueduct) the name of Claudia[150] was given, as Frontinus tells us, are situated in a valley on the south side of the river Anio, thirty-eight miles from Rome, and eight miles below Subiaco (which is forty-six miles from Rome), and not more than half a mile from the source of the Marcia. Several streams issue from the rock under the present carriage-road, a diverticulum or branch of the ancient Via Valeria, and form a beautiful small lake of very clear water having a distinctly bluish tint. Another of these springs, the Albudinus, is still considered as equally good with the Marcia, or nearly so. The Anio here flows at the opposite end of this small valley, and both the lakes are now emptied into it. This valley is in the territory of Arsoli, so called from a neighbouring village on a height overlooking it, and the lake of S. Lucia. The spring called Curtius is another of these streams.

This valley is admirably calculated for the sources of aqueducts; the ground is full of springs of beautiful water, and the watery meadows have only to be dammed up a little to form lakes or reservoirs. Probably the water from these springs, which now runs into the river, was entirely intercepted and carried into Rome in the aqueducts. The Anio is in general such an abundant stream, that these additional springs would scarcely be missed, although in time of severe drought the water did sometimes run short[151].

The piscinæ, mentioned by Frontinus, were large subterranean reservoirs and filtering-places, at seven miles from the old City or from the inner gates, and six from the outer gates: the portion of Rome outside of the walls of the city was considered as the suburbs only, until the time of Aurelian, when the City was made to extend to the outer wall, then newly raised and fortified, but this was long after the aqueducts were made. The present appearance of these two piscinæ of the Claudia and the Anio Novus is merely that of earthen mounds or tumuli. They are situated about midway between the old Via Latina, which now in this part is made into a carriage-road to Frascati, and the Via Appia Nova. The distance from one road to the other is about a mile, and the piscinæ are about half-a-mile from each. The stream, which was originally a branch of the Almo, and now conveys the water of the Marrana and the Aqua Crabra united, runs near them, and received the surplus water from all the aqueducts on this line. It is divided into two parts near the piscinæ, one branch going through the valley called the Caffarella, and falling into the Tiber near the church of S. Paul’s outside of the walls; the other keeps near the arcades and reservoirs, and coming through Rome, passes on the south side of the Cœlian. This branch had come from the piscinæ nearly parallel to the aqueducts; it winds about a good deal, but is never distant from the line of the latter, now on one side and then on the other. A further account of this will be found under the head of the Marrana, in the second part of this chapter.