Salamis (Cyprus)

Roman aqueduct of Salamis - CONSTANTIA Salamis - CONSTANTIA
For the photo's, see below
Home / the complete website

Teucer, the legendary founder of Salamis
The traditional founder of Salamis is Teucer, son of the king of the Greek island of Salamis and one of the heroes of the Trojan war. He was banished by his father because he did not take revenge after the death of his half-brother Ajax. It is said that he set sail with some captives from Troas and landed at Cyprus.


The ancient city of Salamis was one of the most opulent cities of Cyprus during the classical antiquity. Archaeological excavations undertaken at Salamis, show that the history of the settlement dates back to as early as the eleventh century BC. Archaeologists believe that Salamis was first established by newcomers from the nearby site of Enkomi following the earthquake of 1075 BC. Salamis emerged as an important commercial center towards the end of the 8th century BC, which was apparently due to the strategic location of the site within the wider maritime network of trade that prevailed in this period in the Mediterranean World.

The cemetery with richly furnished graves of the Archaic period unearthed at Salamis, clearly demonstrate the high level of prosperity reached at the site during this period. The city mint its own coins in the 5th century BC. The city was subsequently controlled by Persian Empire until the arrival of Alexander the Great into Asia Minor. Following the death of Alexander the Great in 323 BC Salamis came under the control of Ptolemaic dynasty until its incorporation to the Roman Empire in the 1st century BC.

Saint Barnabas
Salamis was the home of the apostle Barnabas who accompanied St. Paul when he visited the city on his first missionary journey in 45 AD (Acts 13 vs 6). In 75 he died as a Christian martyr, stoned to death by the Jewish community of Salamis, the city where he grew up.
Salamis enjoyed a prosperous life in the Roman period along with other important cities such as Alexandria, Antioch, Ephesus, and Athens, Two major earthquakes occurred in 332 and 342 AD devastating much of the city. The Byzantine Emperor Constantius II (337-361 AD) rebuilt the city and renamed it as ’Constantia‘. Because the harbour of the city was almost silted up, the commercial importance of the city gradually started to decline in the 4th century AD and onwards.
Several major architectural monuments demonstrating the importance of Salamis in the early Christianity, are also represented at the site. Following the continuous Arab attacks on Cyprus, the last inhabitants of Salamis moved to Arsinoe (present Gazimagusa / Famagusta) in 648.

The aqueduct

The springs of Kythrea
“... Today, our Sunday was spent by going on a car ride and our aim was to visit an enormous spring above a village called Kythrea, east of Nicosia. The spring is renowned as a landscape oddity throughout all Cyprus and truly it did measure up to this fact now in this the first tender spring greenery. We all travelled together and brought the children with us sitting on our knees. Kythrea is situated at the foot of the Kyrenia range in the middle of a desert-like landscape with small hills, but thanks to the plentiful springs and water, that rushes and spurts forth everywhere in the village, which is like one large overgrown oasis. You have no idea, what significance running water has in this dry country! A flowing river is hardly ever seen and a fountain, that does not evaporate at once, is a rarity.”
From: The fantastic years on Cyprus, illustrated extracts from Alfred Westerholms letters to his parents 1927 - 1931.

Note 1: Westerholm took part in the Swedish Cyprus Expedition which laid the foundation for the scientific studies of Cypriote archaeology and history and worked all over the island during the years 1927 to 1931.
Note 2: Because of deep pumping the springs are now dried up.

Of the aqueduct – as the crow flies 40 km long – only a few pillars remain. The aqueduct got its water from the famous Kephalovrysi (= main spring) of Kythrea / Degirmenlik (see separate frame). Close to Salamis, southwest of Yeni Bogazici / Ayia Gregorius and north of the cloister of St. Barnabas, a set of three pillars and two arches are still preserved but based on written sources, these must be dated in the 6th of 7th c AD ¹. A Roman predecessor is surmised – at that time Salamis must have had 50.000 – 100.000 inhabitants and at least two bathhouses which could not be fed only by local springs (if available), wells, and rainwater harvest systems plus cisterns – but any proof is missing. Just north of the road junction Ismele / Tricomo, Enkomi, and Gazimagusa / Famagusta some remains of the aqueduct pillars can be seen in the countryside. On the east side of the main road the conduit must have made a bifurcation, a south branch towards the big reservoir north of the Roman forum, and a north branch following the ancient city wall or just on top of it, towards the palaestra / baths complex.
From an inscription (SEG 23 [1968] nr 675) it is known that the aqueduct from Kythrea (in Greek: Chytroi) to Salamis was probably Neronian.

The site

The palaestra and the baths in the northeast area are the most impressive buildings of the Salamis site. The theater south of the baths was probably built in the Augustan period. Once it had 50 tiers of seats for 15.000 people. The present version is partly a reconstruction with only 19 tears of seats. In between the palaestra and the theater are the remains of a stadium and an amphitheater. Other structures of Salamis of which often only the foundations remain, are the fish market with the ‘Roman baths’, the ‘Colonnaded street’, the granite forum, and the Roman agora with on the south side the Temple of Zeus and to the north a great water reservoir, often called the vouta.

Water works near the Baths complex

To the south of the baths are two large reservoirs, one of which (at present: partly) vaulted. The stepped south wall of the large reservoir was once part of the stadium. Its internal dimensions are 26 x 4,3 m and possibly 5 m deep (volume about 550 m3). It will have served as the main storage basin for the bath complex annex. Whether and how it was connected to the baths is unknown. The same counts for its supply, presumably by the aqueduct. The purpose of the smaller basin just south of the baths – its dimensions are 10 m long, 4,9 m wide and at least 2 m high (capacity over 100 m3 – too big to serve only as a settling basin) – in unclear.

The palaestra / baths complex had two interesting features. At the east side a small elevated water channel was built, some 0,50 m above the surface. It was interrupted by a foot passage. The solution was a set of two small basins and some subterranean pipes: a mini-siphon.
The palaestra / baths of Salamis is also known because of its marvelous semi-circular latrine, a 44 seater, at the southwest corner, just behind a fountain. From the inlet the water was led into a small channel in front of the users and afterwards used to flush the discharge towards the outlet.

Where the ‘Colonnaded street’ crosses the modern main road a set of three water basins were built of which details are unknown (to me).

Double cistern

To the southeast is the famous Roman–Byzantine double cistern, faced with (fading) paintings from the sixth century AD. It was excavated by dr. Joan du Plat Taylor in 1933 but it is unfortunately not open to the public. This structure consists of two cisterns, each 4,8 m high, one equipped with a 2,4 m high shaft, a second shaft of 7,2 m height, all three connected to each other by fairly small corridors, each over 1,2 m long. To cite Du Plat Taylor:
The cisterns consist of two large circular chambers, 16 ft. deep (see plan). The first is situated below room 1 with which it is connected by shaft 1; the second cistern is joined to the first by double passages, about 5 ft. long, one above the other). On the west side of the second cistern another series of double passages lead into the foot of shaft 2.
The floor of the cistern, like that of the passages, slopes slightly towards the foot of the shaft, where it is some 4 in. lower. The passages are 2 ft. 6 in. to 3 ft. wide, and high enough to admit of a man standing erect.
The two circular chambers taper slightly towards the roof and become nearly oval; the apertures are closed with large blocks of cut stone meeting at an angle in the centre; an inspection hole in the roof of cistern 2 has been closed by other blocks laid across it. The walls of the cisterns have been built of roughly cut stone, except for the last three feet, which seem to be hewn in the rock. The face of the walls has been covered with hard gypsum cement, 3 in. thick, which is well preserved. At three-foot intervals from the bottom, bands of cement 7 in. wide project from the side and seem to mark the water-levels. One also surrounds shaft 2 at the level of the upper passage. In both chambers, above the second band, traces remain of large holes, three on each side, which have been plastered over; these appear to have held the scaffolding beams used in the construction of the upper wall, or in the case of cistern 1 they may have held a platform from which the artist could do his paintings.

All the inscriptions and paintings are in the first cistern, with the exception of three red crosses in each upper passage, one overhead and one high up on each wall.
The researcher noted that this structure has, in the first instance, been built as rainwater cisterns, being filled through the aperture over cistern 1, round which the catchment area of concrete was situated. The date can probably be placed in the 1st century AD. After a period of disuse, the cisterns must have been reopened and (at a later date) possibly connected to the / an aqueduct. The date of the building over it, together with the paintings, is presumably the late 5th or the early 6th century AD. Procopius records the building of new aqueducts in the reign of Justinian, which may coincide with the reorganization of the town water-supply, according Du Plat Taylor. Given the inscriptions one may presuppose that the cisterns were used as a baptistery at one period.

The main reservoir

The main reservoir, also called loutron or vouta, adjoining the agora, or Roman forum, consists of a large rectangle which have had a vaulted roof supported by 36 square pillars. It appears that it was supplied with water from the Kythrea aqueduct. The walls are 2,0 – 2,5 m wide, the pillars 1,1 x 1,1 m, these and the walls are made of sandstone blocks and pointed with mortar. Remains of a watertight lining are missing. The dimensions of the basin, partly below and partly above ground level, are 52,5 x 15,3 m and about 5 m deep (its capacity some 4.000 m3). The soil has been cleaned from half of the basin. The structure parallel to the cistern and a few meters to the south is part of the city wall of Constantia / Salamis within which must have run a branch pipeline or channel from the aqueduct to fill the reservoir.

Almost above the east wall of the reservoir three major apertures will have served as water inlets. Both a hole in the south wall, close to the bottom, and a set of stairs in the north wall will have facilitated the cleaning of the basin.
The reservoir could have been built in combination with the aqueduct, so possibly in the 6th century AD ¹.

Cyprus and the Tabula Peutingeriana

The Tabula Peutingeriana (Peutinger map) is an itinerarium depicting the road network in the Roman Empire. The original map of which this is a unique copy was last revised in the fourth or early fifth century. It measures about 0,35 m in height and 7 m in length. Dr. W. Bruijnesteijn van Coppenraet made this outline (from: De Romeinse reisgidsen, 2006). Distances in Roman miles (1482 m). Vignettes are Mansiones (official stopping place along a Roman road with (or without) some facilities).

On map In Greek In Turkish On map In Greek In Turkish
Soloi NW of Lefka NW of Lefke Tremithous NW of Tremetousia NW of Erdemli
Lapethos Lapithos Lapta Kition Larnaca  
Kerynia Kyrenia Girne Amathos E of Lemesos  
Chythroi Kythrea Degirmenlik Kourion SW of Episkopi  
Salamis N of Famagusta N of Gazimagusa Palai Pafos Palia Pafos  
Tamassos SE of Politiko   Nea Pafos Kato Pafos  

Ancient descriptions of a second aqueduct

From: G. Jeffery, A description of the historic monuments of Cyprus (1918)

The most remarkable surviving monument of the Byzantine epoch is the great aqueduct constructed at some period when Constantia was in a flourishing condition. The history of this colossal work about 35 miles in length is completely unknown, the very imperfect inscription on one of the arches near the village of Ay. Serghis conveying but an indefinite idea as to its date.
There is a reference in Procopius (Pal. Pil. Text. Soc. translation, p. 151) to an aqueduct of ’St. Conon’ in Cyprus as one of the works of Justinian, but although not improbable the statements of Procopius must be received with caution.

The Kythrea-Salamis aqueduct, although one of the most remarkable monuments in the island, is not mentioned by any author or traveler until somewhat late in the middle ages. There is no evidence about the date of its disuse or destruction, which must however have been previous to the foundation of Famagusta in the XIIth century ; otherwise it would probably have been made use of to supply the new city. The first historical reference to the aqueduct is by Nicholas Martoni, an Italian traveler (a native of Bologna), with a taste for antiquities:
”In the middle (of the city built by the Emperor Constantius, father of the blessed Catherina, i.e., Salamis), where the castle stood, is a certain ancient cistern, no bigger one I think, is found in the world, with a vault raised on thirty–six columns, and with apertures above whence the water was drawn. Into this tank water flowed continuously from a certain mountain, along a conduit built with pillars and arches, just as at Scolo, an appurtenance of the Castle of Tragetto or Garigliano.” N. Martoni, Peregrinatio, 1394.
Martoni might have instanced other examples in Italy of similar aqueducts such as the curious one on pointed arches at Sulmona in the Abruzzi.

Pococke (1738) noted many arches of the Kythrea–Salamis aqueduct in a state of preservation, and on one of them he observed a Greek inscription referring to some Bishop who may have been the original builder of this colossal work.

The explorers of 1890 (J. Hell. Studies)² found two aqueducts which crossed the plain from the west and entered the city at nearly the same point, after which the older one is lost to view. The later aqueduct on entering the city turns sharply away north – north-east occupying in all probability the line of the wall of the older town. The older aqueduct supplied the Loutron (basin), and this is proved by the ruins of a small piscina near the tomb of St. Barnabas exactly resembling the Loutron, in construction, vaulting, corbels, etc. This more ancient aqueduct was not carried to any great distance but collected the water from several surface runnels. In other words the older aqueduct and the Loutron possibly belong to the more ancient city of Salamis on the southern portion of the site, whilst the much more imposing aqueduct with pointed arches of the Byzantine period belongs to Constantia which was the city of a later period built on the northern part.

The ’Loutron’, remarkably preserved, is a nearly unique example of an ancient piscina or reservoir for the filtration and distribution of water in a city. It appears to have been a covered vaulted area with a buttressed retaining wall similar to the ’Hundred and One Columns’ of Constantinople. Its internal dimensions are 194 feet by 72 feet, divided into aisles by 36 columns.

¹ The pointed arches [of the aqueduct] are a Lusignan restoration. A number of inscrptions also survive relating to the aqueduct. Those about Kythrea indicate a Severan date. However recently an inscription came to light at Angostina on the line of the aqueduct which appears to indicate the construction of the aqueduct by Nero in the mid-first century. Certainly what must be the city terminal of any aqueduct, the 'Loutron' a large columnar masonry cistern, appears associated in design with the Temple of Zeus (of early Roman date).  From G.R.H. Wright's Ancient Building in Cyprus, pag 233 (1992)
² J.A.R. Munro, H.A. Tubbs, and W. Wroth: Excavations in Cyprus 1890, in: Journal of Hellenic Studies vol 12 (1891) pag 59 - 198

Wilke Schram

Modern - LATIN

Item Info
Length 40 km
Cross-section 0,80 m x 0,80 m
Volume 10.000 - 15.000 m3/day
Fall 0.4 - 0.6 %
Period [foundations probably
Neronian] 6th c AD
  • none

Recommended literature :
  • J. du Plat Taylor (1933): A Water Cistern with Byzantine Paintings, Salamis, Cyprus (in: The Antiquaries Journal vol 13-2 (1933) pag 97 - 108)
  • A. Baur (1989): Die Wasserversorgung der antike Stadt Salamis auf der Insel Zypern (in: SRdF vol 11 (1989) pag 203 - 218) (in German)
  • N. Hogben: Roman Cyprus (2009): Salamis ... ARA Newsletter vol 22 (2009) pag 14 ff
  • C.A. Stewart (2014): Architectural innovation in Early Byzantine Cyprus (in: Architectural history vol 57 (2014) p 1 - 29)
Recommended websites   :
  • none
How to visit                  : Salamis is situated just a few kilometers north of Gazimagusa / Famagusta (Cyprus) which can easily be reached from Lefkosia / Nicosia by motorway. The site is adequately signposted. Entrance fee, brochures and drinks can be paid in Euro's.

The easiest but not most comfortable way to visit the three pillar – two arches of the aqueduct is a ten minutes drive on the dirt road to the north just 100 m west of the St. Barbara cloister.

Not all international car rental companies let you hire a car in the south and take it to the north; the Avis branch at Larnaca airport did. Currently you can drive across the border at Bostanci, Metehan, Beyarmudu and Akyar (in Greek, Astromeritis-Zodhia, Agios Dometios, Pergamos and Strovilia). Note that the locations for border crossing are not signposted as such!
Coming from the Larnaca area I took (in 2011) the Pyla / Pergamon / Beyarmudu crossing, showed my passport and asked for a visa (a sheet of paper with stamp and signature). Be sure that no stamps are put in your passport. I bought an additional car insurance (€ 20 for three days). In total it took less than 15 minutes.

HOME More literature on more aqueducts Last modified: June, 2011 - (webmaster)

Map of Salamis

3 piers and 2 arches

South west side

Details if the specus

View from the east

Specus and nature

Person at work

Map of the major remains

Reservoir near the Baths

View into the reservoir

Lining is missing

Some opus signinum

View into the reservoir

Big and small reservoir

Detailed view

Channel heading east

Channel and path crossing


Octagonal baths pool

Semi-circular latrine

Supports and gutter

Water inlet plus basin

View from above


Triple reservoir?

Water clock?

The big reservoir

Three inlet apertures

An extra outlet

The big reservoir