Aqua Marcia

Roman aqueducts: Rome Aqua Marcia (Italy) Rome - ROMA Aqua Marcia
For the photo's, see below
Home / the complete website

Waters flowing into the city via the Aqua Appia and the Anio Vetus satisfied the needs of Rome's population for about ninety years. Or perhaps it should be said that the Romans had to be satisfied with the supply )1. The near cataclysm and associated expenses of the Second Punic War caused an understandable hiatus in building projects in Rome. When supplies became inadequate to support Rome's public fountains private users were removed from the system by cutting off their pipes. Marcus Aemilius Lepidus and Marcus Fulvius Nobilior, censors from 179 BC to 174 BC, let contracts to construct a new water supply, but Livy tells us that the project was blocked by Marcus Licinius Crassus, who would not allow the aqueduct to cross his property (See XXX 6.19). Consequently, no new water was brought into Rome for another thirty years, until the praetor Quintus Marcius Rex was charged with restoring the existing aqueducts and building a new one (Heiken, Funiciello & De Rira 2005: 145).
The only aqueduct built by a praetor, the Aqua Marcia )2 was constructed between 144 and 140 BC, one hundred and thirty years after the construction of the Vetus and became perhaps the most famous of the Roman aqueducts. It was financed with booty taken from Carthage and Corinth after 146 BC (Evans, 2000:84). Frontinus states that Q. Marcus Rex was also charged with the responsibility of repairing the Appia and Anio Vetus, which by this time where leaking badly, and many citizens where stealing water for their own use without paying taxes. The Marcia provided clean water to the city that had more than doubled in size since the previous aqueduct was built, and was continuing to expand as a result of military success against Carthage and Macedonia. In the years following the Second Punic War water was in such demand that private lines were reclaimed for public usage. Both Livy (39.44.4-5) )3 and Plutarch (Cat. Mai. 19) )4 indirectly support the notion of this water shortage, and indicate that it was a limited resource. Frontinus hints that the old aqueducts were in such bad repair that their supply was wholly inadequate.
Model of the Ponte Lupo
The Marcia's source )5 was a series of springs located on the right bank of the Upper Anio, just below Agosta on the road to Subiaco. This is in the same area where numerous spring houses gather water today for the Marcia's modern counterpart, the Acqua Marcia Pia (Aicher, 1995:36). The ancient channels are now approximately eight metres below ground, the floor of the Anio Valley having been raised by calcareous deposits and the springs themselves (Ashby, 1935:95). 'Apparently, the pools of water that seeped from the ground until the 1920s was from leaks in the ancient channel'. Several underground catchment channels and the run-off from the slopes of the Simbruini ridge may also have contributed. Frontinus describes the reservoir at the source, 'Its waters stand like a tranquil pool with a deep green colour. Tacitus ('Ann'. 14.22) states that Nero swam in the sacred pool, and shortly afterwards fell sick. From its source, the Marcia descended mostly underground along the river's right bank, until it crossed to the left bank near Vicovaro and took almost the same route to Rome that the Anio Vetus took.
The Marcia emerged from the ground to finish the last ten kilometres to Rome aboveground, near the farmhouse named Romavecchia. Incorporating both sub-channels and arches, the aqueduct entered the city through the Porta Maggiore and terminated in a large tank on the Viminal hill, located north of Diocletian's Baths. This would be under the present Ministry of Finance. Near the Porta Tiburtina, however, a branch of the Marcia, called the rivus Herculaneus, diverged from its original path only to transverse the Caelian Hill and terminate at the Aventine Hill. The Aqua Marcia was the longest aqueduct spanning 91 km and yielded and estimated 190,000 m3 per day. Eighty kilometres of the channel lay underground, 1.5 kilometres on substructures and 9.5 kilometres on arches.
Inside the City
Martial ('Epi'. 9.18) gives us some evidence that the Marcia was also delivered to the Quirinal Hill.
I possess, and pray that I may long continue to possess, under your guardianship, Caesar, a small country seat; I have also a modest dwelling in the city. But a winding machine has to draw, with laborious effort, water for my thirsting garden from a small valley; while my dry house complains that it is not refreshed even by the slightest shower, although the Marcian fount1 babbles close by. The water, which you will grant, Augustus, to my premises, will be for me as the water of Castalis or as showers from Jupiter.
The Marcia supplied supplemented the Tepula and Anio )6 (2.67). On the surface this fact complicates the task of assessing the number of aqueducts in Rome. However, the supplementary volumes are so low that in this case the aqueducts can maintain their separate identities. The Rivus Herculaneus crossed the valley between the Caelian and Aventine Hills on an arcade, like the Appia. Lanciani's (1990) hypothetical reconstruction of the channel has the arches of the Marcia parallel and abutting the Servian Wall, on the basis of references to an old arcade and a wet gateway at Porta Capena by Juvenal (3.2) and Martial (3.47.1). Juvenal refers to the arch 'veteres arcus madidamque Capenam'. Martial refers to it as 'arcus stillans'. Aicher (1995:37) thinks these descriptions may refer to an even older arcade of the Appia, on the basis of evidence in Frontinus (1.5) which supports his version. By Frontinus's time a higher branch on arches delivered water to the heights of the Caelian and Aventine Hills.
Aquae Marcia, Tepula and Julia
The Marcia was appreciated by the Romans for the quality of its clear, cold water, which derived from rainwater on the slopes of the Simbruini ridge west of the valley. Here, Mt. Autore reaches a height of 1,850 metres. The rain takes several months to percolate through the porous limestone before it wells up in the valley springs. This makes the water hard, and the Marcia's channels were quickly coated with a calcareous deposit that had to be removed periodically (Aicher, 1995:37).
The Marcia underwent several restorations and additions during its lifetime. Augustus significantly increased its capacity by adding a supplemental source called the Aqua Augusta. This source, after the introduction of the Claudia, was reserved as a supplemental supply for the Marcia and occasionally for the Claudia. Evidence regarding Augustus' overhauling of the line appears as an inscription on the Porta Tiburtina, or in literary sources such as the Res Gestae. Finally in AD 212, Caracalla added another secondary channel, the Aqua Antoniniana, near Tor Fiscale, in order to supply water to his baths. Diocletian also made renovations for the same reasons as Caracalla. The result of these extensive restorations and additions was a complex distribution system that delivered water to a diverse area. The Marcia was the first aqueduct that supplied the high elevation districts of Rome. Archaeological evidence suggests that widespread distribution occurred in the area of the Porta Viminalis. The only evidence visible until quite recently, of this distribution was marked by a small circular structure outside the line of the Servian Wall (Evans, 1997:85). Its other location outside the Porta Viminalis, coupled with its small size, indicate that this was part of a secondary branch. The Marcia supplied the Palatine and by means of a siphon, the Caelian, the Aventine, the Forum Romanum, the southern Campus Martius, and the locations too high for the Vetus on the eastern hills of the city.
The Aqua Antoniniana )7 ended in the large cisterns of Caracalla's Baths. These remain on the south side of the baths (below Via Baccelli), buttressed against the hill on which the aqueduct arrived. The water was stored here for distribution from 32 chambers of approximately two stories each. Such a high capacity would have served the baths well should the water supply have been interrupted. A branch of the Marcia was also taken to the Capitoline Hill. This against the opposition of a number of politicians, who were rivals of the builders. They cited an oracle of the Sybilline books that prohibited water of Anio Valley from touching the Capitoline (Frontinus, 7). According to Livy, the Anio Vetus was also brought here.
Frontinus observes that only a small portion of the Marcia's flow was allotted to public buildings, public works and ornamental fountains. The greatest volume of water was delivered to privati (49.3%) and to public lacus (23.2%). Approximately three-fourths of the Aqua Marcia was reserved for drinking, either for private citizens or for public basins. This explains Frontinus' efforts to keep the integrity of the line, saving it for human consumption whenever possible. The only regions not supplied were the Circus Maximus and Piscina Publica.
The Marcia has a number of well-preserved 'cippi'. Ashby (1935:93) lists ten. Their inscriptions are mostly preserved in CIL 6.3156 and 6.3157. He mentions another fourteen that are joint 'cippi' for the Marcia, Tepula and Julia. These mostly in CIL 6.31561 and 6.1249. The Aqua Marcia was an ingenious, well-built and handsome engineering system. Its length set a record that would stand for centuries, and would never be broken in Rome. It supplied two and a half times as much clean water than the Appia, and more water even than the Anio (Hauck, 1988:35). It was in use until the 10th century.

1) Accounts of the censorship of Cato the Elder (184 A.D.) include notices that the censors reclaimed public water flowing onto private property. Evans (2000:83) sees this as an indication of an attempt to make the best use of a limited resource.
2) According to Pliny (Nat. Hist. 31.41), the Marcia was originally named the Aufeia. There is no other evidence for this.
3) To quote Livy: 'The censors cut off all public water that had been piped into a private building or into private land, after giving thirty days notice'.
)4 To quote Plutarch: 'He cut off the pipes by which people were in the habit of diverting some of the public water supply into their houses and gardens...'
)5 Pliny calls the spring 'Pitonia' (Nat. Hist. 31.41)
)6 Presumably Anio Vetus
)7 The Baths of Caracalla would have required a copious supply of water. Grant (1968:101) estimates that it could accommodate 1600 bathers at one time. The baths which Diocletian and maximian built after the fire of 283 are estimated to have been twice that size; perhaps they could accommodate 3000 bathers.

From E.J. Dembskey: The aqueducts of Ancient Rome" (master thesis 2009)

Porta Tiburtina

The Construction of the Aurelian Wall began in AD 271 under the emperor Aurelian; it was completed in AD 279 during the reign of Probius. The emperor Maxentius (AD 306 - 312) commissioned its reinforcement, whilst during the reign of Arcadius and Honorius (AD 401 - 402) the height of the Wall was increased and the towers were strengthened. The towers were originally built with a square plan and were spaced every 100 feet (ca. 29,60 m) along the perimeter of the wall: both the Wall and the towers were built in opus latericium (bricks). The gates, defended by semi-circular towers, consisted of two arches, or just one, as was the case with the Porta Tiburtina.
This gate takes its name from the Via Tiburtina. It encapsulates a pre-existing monumental arch composed of travertine stone that vwas erected during the time of Augustus (5 BC) in order to permit the flow of water in three superimposed aqueducts (Aquae Marcia, Iulia and Tepula). The inscriptions which are preserved indicate repairs made to the Aqua Iulia by Augustus in 5 BC, Aqua Marcia by Titus in AD 79 and Aqua Tepula by Caracalla in AD 212.
During the Middle Ages the gate was also known as Porta S. Lorenzo due to the presence of the Basilica of St. Lawrence beyond the Walls.

During the reign of Arcadius and Honorius, another archway was constructed on the outer face, as attested by an inscription located below five arched windows. At the same time a double gate was installed, creating an internal court for guards; this was demolished in 1869 under Pius IX. Outside the square towers are features added by Cardinal Alessandro Farnese in 1586.

Text from an information panel near the Porta Tiburtina, Rome

HOME More literature on more aqueducts Last modified: January, 2010 - Wilke D. Schram (

Ponte San Pietro

View on the channel

Archway San Pietro

Details of archway

West side Ponte Lupo

Partly blocked archway

Lowered archway

Body of Ponte Lupo


Marcia Ponte Lupo

Marcia meets Anio Vetus

Marcia bridge

Opus Quadratum

Aqua Marcia's channel

Gloves for better joints

Cistern and Felic


Marcia-Aqua Antoniniana

Piazza Galeria


Arch of Drusus

Crossing the Via Appia

Viale Guido Baccelli

The last scant remains

Cisterns of the Baths

Thermae Antoninianae

at Porta Maggiore

Acqua Felice

Maggiore and Tiburtina

Piazza di Porta S. Lorenzo

A branch line?

Ad Spem Veterem

Tiburtina in
mediaeval times

Three inscriptions

Porta Tiburtina

Three inscriptions

View from outside the city

Three channels

Acqua Felice

Denarius 1

Denarius 2