Skip to content Skip to footer

The ancient Messapian and then Roman city of Egnatia is located just north of the small town of Sevelletri (BR), right at the border between the ancient Peucetia and Messapia (today’s Salento). It is undoubtedly one of the most interesting archaeological sites of Apulia, both for its extension and its archaeological contexts. The pre-Roman city is undoubtedly connected with the Messapian culture, as demonstrated also by an important IV-III century B.C. ceramics class that was named after it. 

It is undoubtedly one of the most interesting archaeological sites of Apulia, both for its extension and its archaeological contexts.

Even though little architectural evidences have survived from the Messapian period (mainly the massive city walls), more important interventions took place in Roman times and in Late Antiquity. The Roman remains include some buildings (e.g. the civil basilica, the forum, the sacellum dedicated to Eastern deities and the amphitheatre) and some kind of urban planning consisting of irregular insulae, but especially the construction of the via Traiana that crossed the whole city and contributed to its economic development.

In the I century B.C. a port was built, probably thanks to Augustus’ friend and companion, Marcus Vipsanius Agrippa.

During the Late Antiquity, two paleo-Christian basilicas were erected; together with other elements, they are evidences of the importance of the city at the times, as it became also an episcopal see.

The archaeological and documentary testimonies attest that Egnatia was still inhabited in the Middle Ages, till it was slowly abandoned in the late-medieval age.

In modern times, the ancient site was extensively looted by grave robbers.

The first official excavations started in the XIX century and have continued until the present day with targeted research projects.

The Via Traiana that passed through the city

The city

The first mention of the ancient Egnatia in the modern ages can be found in Descrittione di tutta Italia by Leandro Alberti shortly after the middle of the XVI century.

The first map of the city was drawn up by the scholar and antiquarian Francesco Maria Pratilli who published it in his work entitled Della Via Appia in 1745.

The area was heavily plundered several times since at least the XVII century. For example, in 1846, during a serious economic crisis, the citizens of Fasano and Monopoli started looking for artefacts to sell to Neapolitan antique dealers, with the indirect complicity of the inspectors. The continuous looting was strongly condemned by the great historian and philologist Theodor Mommsen; the situation did not change and Ludovico Pepe, in his detailed monograph of 1882 dedicated to the remains of the city, still complained about the looting.

The first systematic excavations in Egnatia started right before the First World War. Quintino Quagliati, the then Superintendent, brought to light the so-called porticoed square, a paleo-Christian basilica and a portion of the Via Traiana.

Shortly before the Second World War, archaeological surveys started in the acropolis, where phases of the Iron Age were identified.

After sporadic works conducted in 1959 on the porticoed square, in 1963 it was decided to carry out a more organised excavation by identifying the temple on the acropolis and highlighting a monumental area where shrines dedicated to Eastern deities (Cybele-Magna Mater and Syria) were located. In 1967 the attention of the Superintendency focused on the walls and the city gates.

The inauguration of the Archaeological Museum (1975) coincided with the excavation works in the eastern necropolis and a general reorganisation of the excavations to open them to the general public.

However, it is due to the Project “Egnazia: dallo scavo alla valorizzazione” [“Egnazia: from excavations to development”], created by the Department of Science of Antiquities at the Università degli Studi di Bari (Raffaella Cassano), in collaboration with the Archaeological Heritage Superintendency of Apulia and the Municipality of Fasano, that the excavation campaigns conceived to answer to specific requirements – in particular for to the ancient and late-antique urban planning – were initiated.

Thanks to its position, Egnatia had always been an active hub for exchanges and trade. This particular vocation is evident due to its location facing the sea and the construction of the Via Traiana (II century A.D.) that crossed in length the whole Apulian city.

The city became a Roman municipium probably after the Social War. The most interesting element for us – as we would like to concentrate mainly on the structures of the Roman sumberged port – are the construction works undertaken in the I century B.C. most likely by Marcus Vipsanius Agrippa, Augustus’ companion and the city’s patron, as attested by an inscription. 

In all likelihood, the buildings were a reward for the old Messapian settlement for the support given to Augustus during the civil war.

Besides the port, also the amphitheatre, the public baths, the cryptoporticus, civil basilica and the porticoed square were built under Agrippa. The above mentioned square – actually pre-existent and built in the mid-Hellenistic age – seems to be in close connection with the nearby port and therefore with commerce, especially in the late-antique age. In fact, starting from the late IV century A.D. till the VII century, the area underwent renovation and modification works where several buildings were built with reused materials, located on the inside and on the outside of the re-purposed portico. The commercial and artisanal vocation of these spaces was confirmed also by the discovery of amphorae of Eastern and African production, fine red pottery, lamps, weapons, fishnet weights and needles for repairing the nets.

The city must have retained a certain importance also in the late-antique age. It is in fact mentioned in Itinerarium Burdigalense, a very important document from the IV century A.D. (around 334 A.D.) used by travellers from and to the Holy Land. The document records faithfully the route of a pilgrim from the ancient Bordeaux to these places, mentioning stationes and mutationes, together with villages and cities where Christian communities were present. And Egnatia appears there, with its name twisted in a interesting way.

The Tabula Peutingeriana also mentions the ancient Messapian centre as Gnathia.

In the VI century A.D., the city is an episcopal see with a bishop named Rufentius who took part in Pope Symmachus’ councils in 501 and 502. Some sources and archaeological data document the expansion of Christianism since the IV century A.D. when the basilica – to date, the most ancient of Apulia – was built in Egnatia.

Some economic prosperity in Late Antiquity, together with a resumption of constructing activity, is directly connected with the presence of the episcopal see. This was probably linked also to an economic crisis following the administrative reform under Constantine and a hypothetical partial destruction of the city after an earthquake in 365 A.D. From this moment on, the face of the ancient city was destined to change. Both the episcopal basilica with its baptistery and the great monument of the Imperial age located south to the forum undergo a heavy transformation, perhaps becoming a church with a longitudinal plan. Between the V and the VI century, the so-called southern basilica and a small single-nave apsed church with brick floors and marble presbyterial transennae in the north-west suburbs of the city were built.

Alberti L. 1551, Descrittione di tutta Italia, Venezia

Andreassi G., 1989, Egnazia, in Nenci G., Vallet G. (a cura di), Bibliografia Topografica della Colonizzazione Greca in Italia e nelle Isole Tirreniche, VII, Pisa-Roma, pp. 104-125.

Andreassi G., Cocchiaro A., Maruca A. 2002, Egnazia. Dalla terra al mare, Bari.

Biancofiore F. et al., 1994-1995, Egnazia (Brindisi), scavi 1966 sull’acropoli di Egnazia, in Notizie degli Scavi di Antichità XI, V-VI, pp. 165-204.

Cassano R. 2007a, Egnazia al tempo della diocesi, in Bonacasa Carra r. m., Vitale E. (a cura di), La cristianizzazione in Italia fra tardoantico ed altomedioevo: aspetti e problemi, Atti del IX Congresso Nazionale di Archeologia Cristiana (Agrigento, 20-25 novembre 2004), Palermo, pp. 1259-1282.

Cassano R. 2007b, Nuove acquisizioni sulla vicenda urbana, in Cassano R. et alii 2007, Ricerche archeologiche nella città di Egnazia. Scavi 2004-2006: relazione preliminare, “Epigrafia e territorio. Politica e società. Temi di antichità romane”,VIII, pp. 7-45.

Cassano R. 2008, Conoscere, valorizzare, comunicare la storia dell’antico scalo adriatico di Egnazia (con approfondimenti di Modugno F., Scutari M., Mastrocinque G., Cuccovillo M.), in Fioriello C.S. (a cura di), Rotte mediterranee della cultura. Turismo integrato e riuso delle architetture, Atti del Convegno Internazionale di Studi legato al Progetto Interreg SIRiAr (Fasano, 18-19 settembre 2008), Bari, pp. 71-113.

Cassano R. 2008-2009, Egnazia tardoantica: il vescovo protagonista della città, ‘’Rendiconti della Pontificia Accademia Romana di Archeologia’’, LXXXI, pp. 15-37.

Cassano, R. 2009a, La vicenda urbana di Egnazia ridisegnata dalle recenti indagini, FastiOnLine documents & research (159), pp. 1-20

Cassano R. 2009b, L’area archeologica di Egnazia, in Cassano R., Fioriello C. S. (a cura di), Percorsi di Storia. Itinerari informativi nel territorio di Fasano, Bari, pp. 35-75.

Cassano R. 2010, Egnazia tardoantica: nuove indagini e prospettive di ricerca, in Volpe G., Giuliani R. (a cura di), Paesaggi e insediamenti urbani in Italia meridionale fra Tardoantico e Altomedioevo, Atti del Secondo Seminario sul Tardoantico e l’Altomedioevo in Italia Meridionale (Foggia-Monte Sant’Angelo, 27-28 maggio 2006), Bari, pp. 91-106.

Cassano R. et al. 2004 = Cassano R., Di Grazia V., Fioriello C.S., Pedone A., Tedeschi L. 2004, Ricerche archeologiche nell’area del ‘foro’ di Egnazia. Scavi 2001-2003: relazione preliminare, in Pani M. (a cura di), Epigrafia e territorio. Politica e società. Temi di antichità romane,VII, Bari, pp. 7- 98.

Cassano R. et al. 2007 = Cassano R., Fioriello C.S., Mangiatori A., Mastrocinque G. 2007, Ricerche archeologiche nella città di Egnazia. Scavi 2004-2006: relazione preliminare, in Pani M. (a cura di), Epigrafia e territorio. Politica e società. Temi di antichità romane,VIII, Bari, pp. 9-21

Cassano R. et al. 2008 = Cassano R., Conte R., Annese C. 2008, Forme della circolazione e della produzione delle merci ad Egnatia in età tardoantica: nuove indagini e prospettive di ricerca, in Biegert S. (a cura di), The Pottery of the Via Egnatia. Cultural Exchange between East and West, Acts of the 25th Congress of the Rei Cretariae Romanae Fautores (Dürres, 24 september-1 october 2006), Bonn, pp. 417-441.

Chelotti M. 1993, Regio II. Apulia et Calabria. Gnathia, in Supplementa Italica, n.s. a cura di Panciera S., 11, Roma, pp. 11-58.

Chelotti M. 2007, Regio II. Apulia et Calabria. Gnathia, in Supplementa Italica, n. s. a cura di S. Panciera, 23, Roma, pp. 468-487.

Cinquepalmi A., Cocchiaro A. (a cura di) 2003, Egnazia: trenta secoli di storia, Bari.

Ciurletti G. (a cura di) 1991, Tabula Peutingeriana, Codex Videbonensis. Completa di tutti gli 11 segmenti nel formato originale 76X42, Trento.

Cuccovillo M. 2008, Nuove acquisizioni dal settore nord-occidentale della città, in Cassano R. 2008, Conoscere, valorizzare, comunicare la storia dell’antico scalo adriatico di Egnazia, in Fioriello C.S. (a cura di), Rotte mediterranee della cultura. Turismo integrato e riuso delle architetture. Atti del Convegno Internazionale (Fasano, 18-19 settembre 2008), Bari, pp. 71-113., pp. 101-108.

Donvito A. 1988, Egnazia, dalle origini alla riscoperta archeologica, Fasano.

Felle A. E., Nuzzo D. 1993, Testimonianze paleocristiane in Puglia: recenti studi e ritrovamenti, “Vetera Christianorum”, 30, pp. 307-353.

Fioriello C. S. 2007, Saggio IV. L’area della basilica episcopale, in Cassano R. et alii 2007, pp. 93-119.

Fioriello C. S. 2008, Merci e traffici commerciali lungo le rotte del Mediterraneo: il caso di Egnazia in età romana, in Fioriello C. S. (a cura di), Paesaggi e rotte mediterranee della cultura. Turismo sostenibile e riuso delle architetture, Atti del Convegno di Studi (Fasano, 18-19 Settembre 2008), Bari, pp. 157-185.

Guidoboni E. 1989, Catalogo: autori antichi e medievali. Repertorio, in Guidoboni E. (a cura di), I terremoti prima del Mille in Italia e nell’area mediterranea, Bologna, pp. 725-738.

Jacques F., Bousquet B. 1984, Le raz de marée du 21 juillet 365. Du cataclysme local à la catastrophe cosmique, in Melanges de l’Ecole Francaise de Rome-Antiquite 96, 1, pp. 423-461.

Kehr P.F. 1962, Italia pontificia, vol. IX, Berlino.

Lippolis E. 1994, Gnathia, in Enciclopedia dell’Arte Antica Classica e Orientale. Secondo Supplemento 1971-1994. II, Roma, pp. 818-821.

Mastrocinque G. 2014, Spazio residenziale e spazio produttivo ad Egnazia (Fasano – BR) in età tardoantica, in Pensabene P., Sfameni C. (a cura di), La villa restaurata e i nuovi studi sull’edilizia residenziale tardoantica, Atti del convegno internazionale (Piazza Armerina, 7-10 novembre 2012), Bari, pp. 415-426

Otranto G. 2010, Per una storia dell’Italia tardoantica cristiana, Bari.

Pepe L. 1882, Notizie storiche ed archeologiche dell’antica Gnathia, Ostuni.

Pratilli F.M. 1745, Della Via Appia, riconosciuta e descritta da Roma a Brindisi, Napoli.

Sardella T. 1996, Società, chiesa e stato nell’età di Teoderico: papa Simmaco e lo scisma laurenziano, Messina.

Sartin G. (trad.), Althoff G. (rev.) 2014, Itinerarium Burdigalense vel Hierosolymitanum (Itinerário de Bordeaux ou de Jerusalém): texto latino, mapas e tradução comentada, Scientia Traductionis. n. 15 .

Silvestrini M., 2005, Le città della Puglia romana. Un profilo sociale, Bari.

Volpe G. 2007, Il ruolo dei vescovi nei processi di trasformazione del paesaggio urbano e rurale, in Brogiolo G. P., Chavarria Arnau A. (a cura di), Archeologia e società tra Tardo Antico e Alto Medioevo, Atti del 12° Seminario sul Tardo Antico e l’Alto Medioevo (Padova, 29 settembre-1 ottobre 2005), Mantova, pp. 85-106.


The massive walls of the city, dating back to the Messapian age (IV century B.C.), were built as opus quadratum with a double row, a rampart and a moat.

Their original perimeter of two kilometres was doubled during the III century B.C. and the space between the two walls was filled with stones and soil. These new walls were 8 meters high and provided with a new moat that was 20 meters wide, while inside the walls there were chemins de ronde and stairways leading to the terraces.

                                            Surviving structure of the north wall, today swallowed up by the sea

The technical examination of the construction was possible thanks to the only surviving fragment, locally called “muraglione” i.e. massive wall, preserved today in water for more than 7 meters in height along the north side of the city walls. The base of this structure was carved in the foundation rock depending on the slope of the ground and laying the first two rows of ashlars, unlike the Opus isodomum. Three surviving blocks of the last course are headers with a rounded and prominent surface with respect to the external row of the wall. These blocks seem to belong to the top part of the structure.

The city walls of Egnatia must have had 4 gates. Their location has been only partially identified. During the excavation campaign of 1967, two of them were studied: the north-western one, already identified in an aerial photo in 1966, and the northern one, close to a modern road and probably hosting a tower.

 The northern section of the walls of Egnatia in a photo dated 1927 (C. Colamonico, Da Torre Pelosa a Egnazia, in Le vie d’Italia, Milan, August 1927)

The monumentality of these walls, as it was already noted, is only partially visible today along a brief stretch of the northern side. It is currently submerged, but it was described and immortalised numerous times in photos and illustrations.

                                                                    The north side of the walls today


Andreassi G. 1979, Scavi a Gravina, Salentino ed Egnazia, in Atti del Convegno Internazionale di Studi sulla Magna Grecia, 18 (1978), Napoli, pp. 437- 443.

Cocchiaro A., Dell’Aglio A. 1982, L’impianto difensivo, in AA.VV., Mare d’Egnazia, Fasano, pp. 45-56.

Diceglie S. 1981, Gnathia. Forma della città delineata mediante la prospezione archeologica, Bari.

Drago C. 1960, Gnathia, in “Enciclopedia dell’arte antica” III, Roma

Ginouves R., Martin R., Dictionnaire méthodique de l’architecture grecque et romaine, I, Paris, 1985

Lattanzi E. 1969, Il dibattito, in La Magna Grecia e Roma in età arcaica, Atti VIII Convegno di Studi sulla Magna Grecia (Taranto 1968), Napoli 1969, pp. 227-228.

Lattanzi E. 1970, Integrazioni e discussioni, in La Magna Grecia nel mondo ellenistico, Atti IX Convegno di Studi sulla Magna Grecia (Taranto 1969), Napoli 1970, pp. 268-270.

Lattanzi E. 1974, Problemi topografici ed urbanistici dell’antica Gnathia, in Cenacolo 4, pp. 9-21.

Lugli G. 1957, La tecnica edilizia romana, I, Roma

Pepe L. 1882, Notizie storiche ed archeologiche sull’antica Gnathia, Ostuni (ristampa, Fasano 1980), pp. 141-169, tav. 1.

Randino V. 2013, Nuove indagini sulla cinta muraria di Egnazia (BR) attraverso il contributo dell’aerofotogrammetria e del rilievo architettonico, in Bartoloni G., Michetti L.M. (a cura di), Mura di legno, mura di terra, mura di pietra: fortificazioni nel Mediterraneo antico, (Atti del convegno internazionale, Sapienza Università di Roma, 7-9 maggio 2012), Scienze dell’Antichità 19.2-3, pp. 184-193

Schmidt G. 1970, Atlante aerofotografico delle sedi umane in Italia II, Firenze, tav. XXXVI.

Stazio A. 1968, La documentazione archeologica in Puglia, in “Atti VII Convegno Magna Grecia”, (Taranto 1967), Napoli, pp. 281-282.

The acropolis, the heart of the city, was located in a highly strategic position, facing the sea and close to two deep natural inlets. The archaeological investigations, carried out in different years, allowed for ascertaining that there is a complex and powerful stratigraphy, starting in the proto-history with two settlement phases (Bronze and Iron ages), all related to a village protected by a dry-stone perimeter wall.

A sacred building was constructed in the central part of the acropolis in the III century B.C. It underwent modifications in the II-I century B.C. and then a further monumentalisation was attested in the mid-Imperial age. This temple was facing the sea and surrounded on the other sides by a portico.

The acropolis of Egnatia, panoramic view with the temple in the foreground and the Byzantine fortress in the background


                                                     The area of the temple with the connected buildings

Since the IV century A.D., close to the NW porticoed square (see the data card of the city) there was a functional transformation that covered the eastern side of the portico and the area included between the building and the temple. The eastern portico was modified with the construction of three spaces intended for goods storage, built with dry-stone walls made of reused materials. In Late Antiquity, this area had a commercial and productive character. In the VI century it probably hosted the residence – structured around a central atrium – of a  prominent member of the Byzantine garrison.

                               One of the towers of the Byzantine castrum subject to stratigraphic investigations

This layout seems to last till the end of the VII century A.D. when a general collapse, maybe due to the fight between Byzantines and Longobards, marks the abandonment of the artisan-productive complex. Actually, some bronze coins (folles) and ceramics from the XI century, found after the collapse, document that the site was active even after 1000 A.D.

The eastern portico, after being abandoned for a long time, seems to be back in use in the XII century, as attested by the construction of a finished floor level. This new structure is perhaps linked to the nearby fortress, built in the VI century. Since this era, the acropolis took on a clearly defensive function, especially thanks to the presence of this building with its 4 angular towers. In fact, it was an entrance facing the sea and a minor passage towards the old town. The examination of building techniques and masonry parameters presents recurring elements in other Byzantine fortresses of the Justinian age.

The north-western tower of this castrum, recently investigated, re-purposed one of the spaces outside the Imperial age portico, whose access was closed to allow for a homogenous construction of the external double-row revetment and a shoe profile, typical of contemporary Byzantine defensive structures.

The castrum has preserved also a south-western thick-walled, quadrangular tower.

An important rampart, built from scratch, was intended to be the first defence line of the fortress.

                                                                    A small oratorium inside  the fortress

According to historical data, Egnatia with its castrum should have been occupied by the Byzantines till the 80s of the VII century, following an unsuccessful expedition in Italy of the Emperor Constans II. This resulted in the Lombards’ reaction led by Romuald I, Duke of Benevento, that led to the conquest of Brindisi and Taranto.

The fortress and the small village, according to the stratigraphies, were still active in the centuries following 1000 A.D., as the toponym “Augnatium” also appears in Guidone’s Geographica at the beginning of the XII century, describing it as a fortified centre in the middle of the woods.

Some ceramics fragments from the XV century A.D. allow us to assume its activity also in that period. In fact, it was ascertained that some shepherds’ families lived in chamber tombs of the ancient city.

The area of acropolis underwent yet another reorganisation during the XVI century when, at the behest of the viceroy of Naples, Pedro Parafan de Ribera, a watchtower against pirates and Barbary corsairs was built there. The tower of Anazzo or Adanazzo became the headquarters of custom guards in the XIX century.

Considering the drawings and indications provided by Francesco Maria Pratalli, it seems that this tower, now disappeared, had a square plan and was surrounded on its sides by two rivulets of water that flew into the sea.

                                                             The Byzantine fortress seen from UAV


AA.VV. 1965, L’antica Egnazia, Fasano.

Bethmann L., Waitz G. 1878, Pauli Diaconi Historia Langobardorum, Hannover.

Bianchi V. 1994, Egnazia e la sua acropoli nell’alto medioevo, in Fasano – Rivista di cultura, Anno XV, n. 30, Luglio/Dicembre 1994, pp. 129-162.

Biancofiore F. 1965, Egnazia (Brindisi). Saggio di scavo preistorico, in Notizie degli Scavi di Antichità, XIX, pp.288 – 306.

Biancofiore F. 1969, Nuovi dati sulla storia dell’antica Egnazia, in Studi storici in onore di Gabriele Pepe, Bari.

Biancofiore F. et al. 1994-1995, Egnazia (Brindisi), scavi 1966 sull’acropoli di Egnazia, in Notizie degli Scavi di Antichità XI, V-VI, pp. 165-204

Campese M., Caggese M., Cuccovillo M. 2012, Le fortificazioni bizantine dell’acropoli di Egnazia (Fasano – BR), in Bartoloni G., Michetti L. M. (a cura di), Mura di legno, mura di terra, mura di pietra: fortificazioni nel Mediterraneo antico, Atti del Convegno (Roma 2012), Scienze dell’Antichità 19 (2013), pp. 242-250.

Cassano R., Laganara C. 2012, La linea di costa tra siponto e brindisi, porti ed approdi: l’indicatore ceramico, Atti del IX Congresso Internazionale sulla Ceramica Medievale nel Mediterraneo, pp. 110-115

Cassano R., Campese M., Cuccovillo M., 2015, L’acropoli di Egnazia al tempo dei Bizantini: dal santuario alla cittadella fortificata, in Arthur P., Imperiale M. L. (a cura di), Atti del VII Congresso Nazionale di Archeologia Medievale (Lecce, 9-12 settembre 2015), Firenze: II, pp. 377-382.

Dell’Aglio A. 1983, I segni dell’insediamento protostorico, in A.A.A.V.V., Mare d’Egnazia. Dalla preistoria ad oggi, ricerche e problemi, Fasano, pp. 29-36

De Luca C., Com’era Polignano a metà del ‘700? Primi cenni di quel che si rileva dal catasto Onciario del 1752, Polignano 2016, p. 22, fig. 29

Mastrocinque G. 2016, Archeologia globale ad Egnazia: nuove acquisizioni dalla città e dal territorio, in LAC 2014 Proceedings , Actes of 3rd International Landscapes Archaeology Conference (Rome, 17-20 septembre 2014), Session IX edited by M. Guaitoli, S. Quilici Gigli, pp. 1-12.

Pepe L. 1882, Notizie storiche e archeologiche dell’antica Gnathia, Ristampa 1980, Fasano.

Pratilli F.M. 1745, Della Via Appia, Napoli

Schnetz J., Itineraria romana, II, Lipsiae 1940, pp. 11-142

Troccoli M.L. 1975, Le torri costiere, in De Vita R. (a cura di), Castelli, torri ed opere fortificate di Puglia, Bari, p. 224A small oratorium inside the fortress


– Gnatia and Ignatiae (VII century A.D., Anonimo Ravennate, Cosmographia, V, 1) (1)

– Augnatium (early XII century A.D., Guidone (Guido da Pisa), Geographica, 27) (2)

– Torre d’Adanazzo (1744, D’Anville, Analyse géographique de l’Italie) (3).

– Egnazia (1745, Pratilli, Della via Appia) (4).

Some interesting passages written by this author:

544: “while the passengers or the militia were embarking in its port (of Egnatia), and they would have landed in the city of Durrës that was almost in front of them…”

544: “Egnatia was not far away from the sea, in a very pleasant location, spacious and plentiful of sweet and clear waters, as they can be still found nearby, and especially close to the ancient walls that surrounded it from the sea side, where there is a spring of very tasteful water called the fountain of Agnazzo by the locals, the most acclaimed spring water of this shore. The vestiges of the ancient city are clearly recognisable, the castle, the walls and the ditch, the scant remains of an old building, probably a temple, and in the places they call the park and the seat, with a vaulted underground corridor accessible through a small opening, with some half-moon windows in the middle to shed light in the baths and thermae” (5).

Pratilli published a map of the city of Egnatia which also mentions the Tower of Agnazzo. The plan is also reported by Pepe (6).

The plan of Egnatia by Pratilli

– Anazzo or Torre d’Anazzo (1751, Diderot – D’Alembert, Encyclopedie) (7)

Abbé de Saint-Non (1781)

Jean-Claude Richard de Saint-Non (1727-1791), better known as Abbé de Saint-Non, was a French humanist, archaeologist, and traveller. During his life, he made a long journey to Italy and was fascinated by the South, to which he dedicated an encyclopaedic work, published over a period of five years. Here his attention focuses on the archaeological area of Egnatia:

“Le lendemain , nous rencontrâmes à sept milles [da Monopoli], les ruines de l’antique Égnatia, qui font voir encore l’étendue de cette ville. Elle était considérable, et arrivait jusque sur les bords de la mer. On aperçoit encore quelques vestiges, qui pourraient être ceux d’un mòle. Il est vraisemblable que la construction de ce mòle n’était pas antique, mais qu’il avait été élevé sur les bords de la mer, des débris de l’ancienne ville, dont les murailles sont encore, en quelques endroits, de cinq pieds d’élévation, et en très grosses pierres posées à sec. Nous distinguâmes même au milieu de ces débris, et malgré le blé qui y était semé, les traces interrompues des rues et quelques angles de maisons.” (ABBÈ DE SAINT-NON 1781-86, ed. 1829, vol. III, p. 4).

– The first mention about the harbour facilities can be found in the description by Emanuele Mola (8) who in 1796 seems to have identified them in a fragment of his description of the city.

– In 1799 Gian Luigi Tanzi (1722-1804), the war commissioner of the Department of Bari, in a letter dated Pluviôse 28th (February 16th), i.e. the first year of the Neapolitan Republic, mentions a rebellion of the population of Polignano and Fasano. The citiziens also plundered the Tower of Anazzo (9).

– The ruins were attributed to the ancient Egnatia by the abbot Romanelli (1818) (10) that seems to draw much of its inspiration from Pratilli.

– Anazzo or Torre di Anazzo (1826, Nuovo Dizionario geografico Universale) (11).

– Salvatore de Renzi in 1826 describes the miasmas rising from the swamps of Egnatia (12).

– Giuseppe Castaldi recalls the source of Anazzo in 1842 (13).

– Nicola Corcia in 1847 mentions the presence of the port (14), recalling what another erudite traveller told him (15).

– The French archaeologist Lenormant in 1881 visits Egnatia and writes an interesting description of its port (16).

– Ludovico Pepe dedicates a whole book to Egnatia in 1882 (17). It includes also a general plan of the city. In Chapter IX, entitled “Rovine” [“Ruins”], he also writes about the port and discusses the previous hypotheses proposed by scholars (18).

Pepe quotes an entire passage from Mola (see note 8), but admits that he is not able to clearly see any tomb in the waters of the harbour. Actually, the whole stretch of sea occupied by the Roman port preserves the remains of tombs from the Messapic age, at a maximum depth of 3 m.


1 La Cosmografia Ravennate is a list of places and cities in the VII century A.D. that presents the world known at the time. It takes its name from the Italian city of Ravenna where the text was written by an anonymous author. It is composed of a sequence of toponyms going from India up to Ireland. The text suggests that the author had often used maps as sources. But even though it looks as a map, it is actually a methodical enumeration of places taken from a map. The critical edition is PINDER, PARTHEY 1860, IV, 31, p. 261 and V, 1, p. 329.  A more recent edition is SCHNETZ, 1942 (reprint 1990).

(2) The Geographica by Guido of Pisa is articulated in four books. The first one – consisting of fragments taken from Cosmographia by the Anonimo Ravennate, from Historia Longobardorum by Paolo Diacono and from Collectanea by Solinus – describes the territories of the Roman Empire. The second volume describes briefly the ancient Roman society, following Etymologiae by Isidore of Seville. The third one focuses on geography, following once again the Anonimo Ravennate. The fourth one describes the Trojan War, according to De Excidio Troiae Historia by Dares Phrygius and the heroic deeds of Alexander the Great by Pseudo-Callisthenes. The edition is always PINDER, PARTHEY 1860, 27, p. 467 (see previous note).

3 D’ANVILLE 1744, p. 224. Jean-Baptiste Bourguignon d’Anville was born in Paris in 1697. He was an important geographer and all his maps were much appreciated by both colleagues and sailors. In 1744 he published his map of Italy and explained how it was composed in a geographical analysis. He reduced by several square leagues the extension attributed to the country and when Benedict XIV ordered verifications to be conducted throughout triangulations, Anville concluded that these measures confirmed what he had discovered.

4 PRATILLI 1745, pp. 544-546. Francesco Maria Pratilli (1689 – 1763) was an Italian clergyman, scholar and antiquarian, member of the Accademia Ercolanese. He was often criticised by his colleagues of the time, including Mommsen, for quoting apocryphal and false sources. However, in the XX century, some Latin epigraphic inscriptions reported by the clergyman from Capua were recognised as authentic (see e.g. PALMIERI 1982, pp. 417-431 and SOLIN 1998, note 120, p. 93).

5 The presence of this fountain (toponyms Torre di Anazzo and Fontana di Agnazzo) is also mentioned in DI MEO 1819, p. 334. Also in CORCIA 1847, vol. III, p. 489 there is a mention of the fountain of Agnazzo, interpreting the Horatian quote “lymphis iratis extructa” as a foundation close to abundant springs. Other scholars, including Pepe (PEPE 1882, p. 157 and ff.), provide an opposite interpretation of Horace’s quote, as if the city had been built defying “the wrath of the waters” and with “contrary torrents”. PEPE 1882, p. 159 (see the data card on this author below) was always very critical towards Pratilli and his reliability as a scholar.

6 PEPE L. 1882, Tab. 1

7 DIDEROT, D’ALEMBERT 1751, Tome I, p. 438: “ANAZZO ou TORRE-D’ANAZZO, (Géog. mod.) ville de la province de Bari au royaume de Naples. On croit que c’est l’ancienne Egnatia ou Gnatia. Quelques Modernes la nomment Gnazzi ou Nazzi”.

8 MOLA 1796, pp. 11-12: “The greatest change in that shore, however, seems to be visible in the port of the ancient city, which lays today beneath the military tower called Anazzo, and which still retains the shape of its ancient state. I then admired, at the bottom of the water that fills it, numerous wide and square tombs, almost all without their lids; being favoured by the calm and by a beautiful spring morning, clearly offered themselves to my curious eyes. Then the upper and contiguous beach was likewise scattered with a great number of these very ancient tombs hollowed out of the stone, and likewise deprived of their lids. We could see all of them without any order, in all directions, in all sizes, and shapes… This was clearly the burial ground of Egnatia, which in the early times must have extended considerably towards the sea in a stretch of land lately took on by its waves, that would have formed the port in the lower areas in a more recent era than the digging of the burial ground itself, which I attributed to the most remote age…. Since it cannot be understood otherwise, as burials are found in the port and also on the shore, exposed to the fury of waves, winds and storms”. Pepe (PEPE, 1882, p. 153-155) makes some observations on this text which, according to him, is not entirely understandable (see below).

9 Quoted in DE LUCA 2016, p. 22, fig. 29 : “… the natives of Polignano and Fasano plundered cannons, rifles, gunpowder, bullets and other ammunition from the maritime towers of San Vito, Rapagnona and Anazzo… and despite the order issued to the Universitas to give them back immediately, it was not possible even with the assistance provided by the lieutenant Francesco Anzalone, commander of the detachment of Monopoli”. Unfortunately, De Luca does not mention the source.

10 ROMANELLI 1818, p. 146: “The ruins of this ancient city are still visible, and its name remains today in a fountain and a maritime tower, so we can’t doubt that this was the site where it was founded. It rose close to the seashore, surrounded by a very strong wall with a profound ditch that served as its defence, and in the middle it had a tower or a castle that dominated the port nearby… But the most pleasant element of these ruins is the fountain that we have already mentioned, named after the source of Anazzo which with its clear and fresh waters offers the most desired comfort to a traveller in need of good spirit in a parched land”.

11 CAVAGNA SANGIULIANI DI GUALDANA A., 1826, vol. I, pp. 514-515: “…city of the kingdom of Naples, in the province of Bari. Believed to be the anc. Egnazia or Gnatia, a destroyed city of Apulia…”

12 DE RENZI 1826, pp. 177-178

13 CASTALDI 1842, p. 55: “…though the most useful one remains the fountain of crystalline waters called Anazzo, very beneficial in this hot land”.

14 CORCIA 1847, p. 490: “Later at noon I saw the small port, also carved masterfully in the rock, in which a rivulet was bringing its scarce tribute”.

15 HUGHES 1820, vol, 2, p. 360-361: “Farther to the south appears another small port, like the former, cut chiefly by art in the rock; into this a rivulet which ran through the city, pours its scanty tribute during the hot months of the year and a violent torrent in the rainy season”

16 LENORMANT 1881-1882, p. 42: “Le plan de la ville est un carré long dont un de grands côtés s’appuie à la mer. C’est auprès du rivage, sur une petite colline d’une faible saillie, placée à égale distance des deux petits côtés du rectangle et par conséquent au milieu de la ville, qu’était bâtie l’acropole, dont les murailles sont aussi bien conservées et aussi nettement caractérisées que celles de l’enceinte extérieure de la cité. Cette forteresse commandait et protégeait deux petits bassins carrés, en partie creusés ou régularisés de main d’homme, entre lesquels elle était placée, l’un à nord et l’autre au sud de la colline. On y distingue encore sous les eaux les divisions, partiellement conservées, des cales de galères.”

17 PEPE 1882

18 PEPE 1882, pp. 152 ff.



Abbé de Saint-Non J. B. C. R. ed. 1829, Voyage pittoresque ou Description des Royaumes de Naples et de Sicile, Paris 1781-86.

Castaldi G. 1842, La Magna Grecia brevemente descritta da Giuseppe Castaldi, Napoli.

Cavagna Sangiuliani Di Gualdana A. 1826, Nuovo Dizionario geografico Universale, Statistico – Storico – Commerciale…, Venezia.

Corcia N. 1847, Storia delle Due Sicilie, dall’antichità più remota al 1789, tomo III, Napoli.

D’Anville J.B.B. 1744, Analyse géographique de l’Italie, Paris.

De Luca C. 2016, Com’era Polignano a metà del ‘700? Primi cenni di quel che si rileva dal catasto Onciario del 1752, Polignano.

De Renzi S. 1826, Miasmi paludosi e luoghi del Regno di Napoli dove si sviluppano, Napoli.

Diderot M., D’Alembert M. 1751, Encyclopedie, 1re éd., Paris.

Di Meo A. 1819, Annali critico-diplomatici del Regno di Napoli della mezzana età, Tomo XII, Napoli.

Hughes T.S. 1820, Travels in Sicily, Greece and Albania, London.

Lenormant F. 1881-1882, Notes archéologiques sur la terre d’Otranto, in Gazette archéologique, 7, pp. 40-53, Paris,

Mola E. 1796, Memorie per servire alla storia letteraria e civile, Venezia.

Palmieri R. 1982, Su alcune iscrizioni pratilliane, in Misc. Greca e Romana, 8, Studi Istituto Italiano per la Storia Antica, 33, pp. 417-431.

Pepe L. 1882, Notizie storiche ed archeologiche dell’antica Gnathia, Ostuni.

Pinder M., Parthey G. (eds) 1860, Ravennatis Anonymi Cosmographia et Guidonis Geographica ex libris manuscriptis, Berlin.

Pratilli F.M. 1745, Della Via Appia. Riconosciuta e descritta da Roma a Brindisi, libri IV, Naples.

Romanelli D. 1818, Topografia istorica del Regno di Napoli, Naples.

Schnetz, J. (eds) 1942, Itineraria Romana, vol. II: Ravennatis Anonymi Cosmographia et Guidonis Geographica, (ristampa 1990), Stuttgart.

Solin H. 1998, Corpus inscriptionum Latinarum, vol. X. Passato, presente e futuro, in SOLIN H. (eds) Epigrafi e Studi epigrafici in Finlandia, (Acta Instituti Romani Finlandiae 19), Rome.

The original form of the toponym Γνατία is attested by inscriptions and seems to be of Rhodian origin. It is naturally transmitted into Latin under the form of Gnatia, recalled by Horace in the I century B.C. and mentioned in Pomponius Mela  and in Naturalis Historia by Pliny the Elder.

In Greek, the geographers Strabo and Ptolemy refer to it as Έγνατὶα in the I and II century A.D., respectively.

In the Imperatoris Antonini Augusti Itinerarium (IV-V century A.D.), it was mentioned both as Egnatiae and Gnatiae, while in Tabula Peutingeriana the name is corrupted into Gnatie.

The Tabula Peutingeriana

In the early medieval period (VIIcentury A.D.), the anonymous author from Ravenna writes about Gnatia and Ignatiae. However, after 1000 A.D., the toponym changed substantially. Guidone in his work Geographica (XII century A.D.) mentions Augnatium, while in the nautical maps and portolan charts from 1300 onwards it is located precisely as Annaso or Anazzo. Finally, in some modern and contemporary maps, the Tower of Adanazzo is frequently mentioned.

Evolution of the toponym according to the sources:

– Gnatia (I century B.C., Horace, Sermones, I, 5, 97) (1)

– Έγνατὶα (I century B.C. – I century A.D., Strabo, Geographia, VI, 3, 8,) (2)

– Gnatia (I century A.D., Pomponius Mela, Corographia, II, 66,4) (3)

– Gnatia (I century A.D., Pliny, Naturalis Historia, II, 240 and III, 102) (4)

– Έγνατὶα (II century A.D., Ptolemy, Geographia, III, 1, 15)

– Egnatiae and Gnatiae (IV-V century A.D., Imperatoris Antonini Augusti Itinerarium, 117,4 and 315,4) (5)

– Gnatie (IV century A.D., Tabula Peutingeriana, VI, 5)

– Gnatia and Ignatiae (VII century A.D., Anonimo Ravennate, Cosmographia, V, 1) (6)

– Augnatium (early XII century A.D., Guidone, Geographica, 27) (7)

– Anazzo or Annaso with other variations in portolan charts starting from the XIV century

According to Paul Collart (COLLART P., Une réféction de la ‘Via Egnatia’ sous Trajan, in Bulletin de Correspondance Hellénique, 59 (1935), pp. 397-400, n. 1-4), the name of the Apulian city does not derive from the Latin nomen Egnatius. Its original form, Γνατία, attested by the inscriptions (IG, XIV, 685), is of Rhodian origin, transmitted into Latin under the form Gnatia. The existence of a road named Egnatia might have favoured the passage from the toponym Γνατία to more familiar forms in Latin, like those mentioned in Strabo, Geogr., VI, 3,8, C 283 and Ptolemy, III, 1, 15 (Egnatia), in Itinerarium Antonini 117,4 and 315,4 (Gnatiae) or in Tabula Peutingeriana, VI, 5, (Gnatie).

However, Collart reiterates that when the Roman road was built, the city had existed already for centuries. The Via Traiana that crossed the city is two centuries older than the Via Egnatia. The role of Egnatia with respect to this route is mentioned as of a lesser importance in the sources. Collart observes that Bari and Brindisi had rather a major importance, as the Via Traiana reaches the sea and Brindisi was attested as its terminal point, both in epigraphic and historical-literary sources. The only mention of Gnatia as a port concerns coastal navigation. Collart concludes that there seems to be no evident reason why the Via Egnatia, actually a prolongation of the Appian Way beyond the sea, would have derived its name from this city.


1 : “dein Gnatia Lymphis iratis exstructa dedit risusque iocosque, dum flamma sine tura liquescere limine sacro persuadere cupit.

2Strabo. ed. A. Meineke, Geographica, Leipzig. 1877. : “παραπλέοντι δ᾽ ἐκ τοῦ Βρεντεσίου τὴν Ἀδριατικὴν παραλίαν πόλις ἐστὶν ἡ Ἐγνατία, οὖσα κοινὴ καταγωγὴ πλέοντί τε καὶ πεζεύοντι εἰς Βάριον: ὁ δὲ πλοῦς νότῳ.”

3 : “Sinus est continuo Apulo litore incinctus nomine Urias, modicus spatio pleraque asper accessu, extra Sipontum aut ut Grai dixere Sipuntem, et flumen quod Canusium adtingens Aufidum adpellant, post Barium et Gnatia et Ennio cive nobiles Rudiae, et iam in Calabria Brundisium, Valetium, Lupiae, Hydrus mons, tum Sallentini campi et Sallentina litora et urbs Graia Callipolis”.

4 : “….in Sallentino oppido Gnatia inposito ligno in saxum quoddam ibi sacrum protinus fiammam existere, in Laciniae Iunonis ara sub diu sita cinerem immobile esse perflantibus undique procellis;” (II, 240). : “Poediculorum oppida Rudiae, Gnatia, Barium, amnes Iapyx a Daedali filio rege, a quo et Iapygia Amita, Pactius, Aufidus ex Hirpinis montibus 5 Canusium praefluens.” (III, 102).

5CUNTZ O. (eds), Itineraria romana. Itineraria Antonini Augusti et Burdigalense, 1929, reprint Stutgardiae 1990. The title of Itinerarium Antonini attributes the authorship of this collection of itineraries to the emperor Antoninus Augustus, usually identified with Antoninus Caracalla (211-217 AD) or with one of the Antonines; but, in reality, it is the work of one or more editors – anonymous to us – active in the late Imperial age (IV-V century A.D.), who compiled sources of different chronology and nature.

6The critical edition consulted is PINDER M., PARTHEY G. (eds.), Ravennatis Anonymi Cosmographia et Guidonis Geographica ex libris manuscriptis, Berolini 1860, IV, 31, p. 261 e V, 1, p. 329. The most recent edition is SCHNETZ, J. (eds), Itineraria Romana, vol. II: Ravennatis Anonymi Cosmographia et Guidonis Geographica, 1942 (reprint 1990), B. G. Teubner, Stuttgart.

7GUIDONE, Liber Guidonis de variis historiis, 27: «De hinc in litore civitas Dirium, quae nunc Monopoli, est, habens in silvis oppida quaedam Augnatium» (critical ed.: SCHNETZ J., Itineraria romana, II, Lipsiae 1940, pp. 11-142). See also PINDER, PARTHEY 1860, 27, cit.,  p. 467

The port

On the coast in front of the ancient acropolis, there are important archaeological presences towards the sea, among which some rectangular recesses carved in the rock that can be attributed to tombs, and massive cementitious structures underwater, to be interpreted as Roman harbour moorings.

The study of underwater structures confirmed that these elements were constructed directly in the water. The construction techniques indicate the application of methodologies described in the most famous text about Roman architecture, De Architectura by Vitruvius.

In Egnatia, there were identified two different methods of underwater construction explained by Vitruvius: opus pilarum and continuous footings. In the former, large pillars or pilae are built and placed separately with a cement casting in watertight cofferdams; in the latter, the continuous structure is made with cement casting in a submerged cofferdam (for further information regarding these construction techniques, please read the data card about the harbour).

All the pillars of the northern “arm” were built in water in dry-stone, with an exposed curtain wall in opus reticulatum. Unfortunately, only a prolongation close to the coastal reef and the two outermost pilae remained of this side.

On the contrary, the southern “pier” was entirely built with inundated cofferdams with a continuous foundation. In this area, some traces and holes of the wooden frameworks are also visible.

The first survey of the Roman port, completed by Di Ceglie in the 70s
Documentation of a pier from the north side in the framework of MUSAS project. The preservation of three courses of opus reticulatum at the base of the structure (photo: Carabinieri Scuba Divers, Unit of Pescara).


“… as one sails from Brentesium along the Adriatic seaboard, one comes to the city of Egnatia, which is the common stopping-place for people who are travelling either by sea or land to Barium…” 1. With his concise description, Strabo captures perfectly the salient features of the ancient city of Egnatia, making an important reference also to the presence of a port that, in his times, was recently constructed or expanded.

On the coast in front of the ancient acropolis, there are important archaeological presences towards the sea, among which some rectangular recesses carved in the rock that can be attributed to tombs, and massive cementitious structures underwater, to be interpreted as Roman harbour moorings (2). These were supposed to enlarge and improve the previous mooring that made the most of the favourable coastal morphology.


The facilities of the Roman port according to recent surveys

The analysis of these remains starts with the most advanced prolongation of the northern cove (take a virtual dive!). A rectangular bank made of carved rocks is still visible, whose genesis may be assumed in the northern side of the complex. Approximately 100 meters from it, in direction 45° N, there are two enormous submerged parallelepiped blocks in opus coementicium ,certainly with a staggered orientation with respect to the items near the land and placed approximately 3 meters from one another.

     One of the pilae from the north side of the Roman port. An opus reticulatum in negative (photo: Carabinieri Scuba Divers,          Unit of Pescara).

Those two plinths originally had curtain walls in opus reticulatum and ammorsature d’angolo/ angle restraining in opus vittatum, today mostly visible only in negative. This technique, widespread between the late Republican age and the early Imperal age in the Tyrrhenian area, allowed for dating these structures between the I century B.C. and the I century A.D.

Along this ideal line of conjunction that unites the aforementioned structures, there are several traces of collapse of various dimensions, always pertaining to opus caementicium with traces of opus reticulatum on the mortar. As some of them preserve the grade plane with regard to the rocky seabed, it can be assumed that the two surviving plinths were not the only ones along the same axis. All these elements might be attributed to opus pilarum, built with watertight cofferdams, as attested by the presence of the curtain wall in opus reticulatum. However, the construction of a continuous pier can be doubted if we assume that it was rather the construction of individual plinths whose real functions are not known today.

With respect to remains in the north, the southernmost part, which was likely to be a pier, is more recognisable. It is a construction with SSO-NNE orientation. It was built in opus caementicium with castings on several superimposed layers, whose base rested directly on the rock at the limits of the underwater area of the ancient port. The total currently visible length of this pier, divided now in three sections of various dimensions, is 23 meters. In every section there are visible signs left by the wooden framework with which they were built. The disappearance of vertical pillars (destinae) and horizontal connecting and reinforcement beams (catenae) made visible the insertion holes of these pillars and the parallel grooves that mark the location of the beams.

                 Remains of the 10m x 10 m structure at the landward end of the northern pier and the unloading canal.

What is left of the structure above the sea level are two ashlars in local Carparo stone and some lead elements that originally might have been fixing cramp-irons.

The south pier, unlike the north one, was a continuous structure in segments progressively placed near each other, built in a submerged cofferdam with a skeleton of circular-section destinae (diameter of 30 cm), to which catenae with a probably square section (20-30 cm per side) were repeated regularly. Some external pillars (stipites) surrounded the aforementioned structure and were connected to the seabed by iron reinforcements that were penetrating the calcarenite bank. Some traces of these particular elements were found as well.

Around the landward end of the north pier, close to the innermost basin, there are traces on the rock surface that were interpreted as a foundation of an approximately square-plan, 10 m long building, probably related to maritime activities.

Further remains of masonry attributable to harbour infrastructures must have been located by the current little beach on the north (ArcheoLido) where irregular blocks, rocks, and mortar were found. These remains belong to the Imperial age. Both the building and the structures seem to be linked with the north-east/south-west canal, parallel and close to the edge of the basin and flowing into it.

Some numbers: the water surface between two hypothetical piers described above is approximately 16,000 sqm. The theoretical entrance between the two piers in ancient times can be calculated for a width of 40 m, with a maximum depth in the current area of 6.5 m.

A bronze statue in Hellenistic style of a female was discovered in this port basin in 1969. Due to its poor conservation state, it was dated back approximately between the Hellenistic and Late-Antiquity ages (3).  This small figure could have been a counterweight for scales as it was filled with lead. A further confirmation of its use are traces of an applique on its head.

A small number of pottery materials found underwater indicate some activity between the Hellenistic age and Late-Antiquity.

In this area, at the centre of the southern cove, there are a number of limestone blocks, aligned and built on two courses with dry-stone. These are still visible and occupy an area of  3.65 x 1.64 meters. The most likely interpretation is that they can be attributed to a wall fragment in opus quadratum rather than to the load of a vessel carrying carved stone from the quarry of origin.

As far as the chronology of underwater structures of Egnatia is concerned, a series of comparisons suggests a rigorous compliance to the Vitruvian construction rules and a dating between the end of the Republic and the beginning of the Imperial age. Some scholars do not exclude the possibility that the port of Egnatia was a creation by M. Vipsanio Agrippa, patronus of the municipium, as it is attested by an inscription of which only the text survived to our days and whose terminus ante quem is 38 B.C. (CIL, IX, 262). In fact, everything makes sense if we consider the fundamental strategic position of the Salento coast during the war between Octavian and Mark Anthony, as well as Agrippa’s role, as the commander of Octavian’s fleet (4).

A recent cleaning intervention and underwater excavation on an edge of one of the northern pilae revealed three intact courses of opus reticulatum which constituted the whole revetment of the works. The opus vittatum was also spot at the corners.


1 Στράβων, Γεωγραφικά, VI, 3, 8, 282-283; Strabo., Geogr., VI, 3, 8, 282-283 (ed. MEINEKE A., Geographica, Leipzig, Teubner, 1877): “παραπλέοντι δ᾽ ἐκ τοῦ Βρεντεσίου τὴν Ἀδριατικὴν παραλίαν πόλις ἐστὶν ἡ Ἐγνατία, οὖσα κοινὴ καταγωγὴ πλέοντί τε καὶ πεζεύοντι εἰς Βάριον: ὁ δὲ πλοῦς νότῳ.”

2 Various visitors and savants of the XVIII and XIX century recall and describe the port of Egnatia (see below). The first scientific investigations, mainly with aerial photos and echosounders, were carried out by Stefano Diceglie (DICEGLIE 1972; DICEGLIE 1981, tabl. II; also the more recent DICEGLIE 2002). An analysis is also present in VLORA 1975, pp. 56-61, figs. 35-39. Immersions and surveys were led in 1979 by Alice Freschi with the Società Itinera, on behalf of the Provincial Council of Brindisi (FRESCHI, ALLOA 1979-80, pp. 60-65 and p. 134; FRESCHI 1980, pp. 450-455) and, always by Freschi, in 1994 together with Cooperativa Aquarius (FRESCHI 1995, pp. 141-143). A recap of the knowledge basis at the beginning of the 1980s was compiled by ANDREASSI, SCIARRA-BARDARO 1982, pp. 107-118. More recently, Rita Auriemma has definitively ascertained its type and the building techniques (AURIEMMA 2003, pp. 77-97; AURIEMMA 2004, pp. 15-16), refuting the hypotheses that denied the maritime function of the structures (GUERRICCHIO, GUERRICCHIO, MARUCA 1996; GUERRICCHIO, GUERRICCHIO, MARUCA 1997; ANDREASSI, COCCHIARO, MARUCA 2002).

3 ANDREASSI, SCIARRA-BARDARO 1982, p. 114 , fig. 90.

4 In May 2008, the team of the project ROMACONS carried out an intervention on one of the pilae, conducting a stratigraphic, petrographic, and chemical analysis of the mortar by means of core sampling. A carbon-14 dating has also been conducted (BRANDON et al. 2014, pp. 93-94, p. 134, pp. 265-266, pp. 288-289). The weighted age of the sample collected from the internal cementitious conglomerate attested a more ancient chronology (200 to 50 BC) with respect to the original hypothesis on the basis of the revetment in opus reticulatum. The examination of the core revealed the presence of pozzolana, despite the fact that the pillar had been evidently built in a dry cooferdam.



Andreassi G., Sciarra-Bardaro B. 1982, Il porto, in AA.VV., Mare d’Egnazia, Fasano, pp. 107-118

Andreassi G., Cocchiaro A., Maruca A. (eds) 2002, Egnazia. Dalla terra al mare, Bari.

Auriemma R., 2003, Le strutture sommerse di Egnazia (BR): una rilettura, in Atti del II Convegno nazionale di archeologia subacquea (Castiglioncello, 7-9 settembre 2001), pp. 77-97

Auriemma R. 2004, Archeologia subacquea nella Puglia meridionale, in Giacobelli M. (eds), Lezioni Fabio Faccenna II. Conferenze di archeologia subacquea (III-IV ciclo), Bari, pp. 11-24

Brandon C., Hohlfelder R., Jackson, M., Oleson, J. et al. 2014, Building for Eternity – The history and Technology of Roman Concrete Engineering in the Sea, Oxford-Philadelphia

Diceglie S. 1972, Il porto di Egnazia, in Osservatorio Geofisico di Fasano (BR), Fasano

Diceglie S. 1981, Gnathia. Forma della città delineata mediante la prospezione archeologica, Bari

Diceglie S. 2002, Nel Mare di Egnazia. Telerilevamento da elicottero di ruderi sommersi in aree estese, C.L.C.A. Università di Bari,

Freschi A., Alloa C. 1979-80, Egnazia, uno studio di storia, in Sesto Continente, I, 5, pp. 60-65 e p. 134

Freschi A. 1980, Egnazia 1979. Ricerche subacquee, in L’epos greco in Occidente, Atti del XIX convegno di studi sulla Magna Grecia, (Taranto 7 – 12 October 1979), Napoli, pp. 450-455

Freschi A. 1995, Fasano (Brindisi), Egnazia, Notiziario 1994 – Taras XV, I, pp. 141-143.

Guerricchio A., Guerricchio M., Maruca A., 1996, Strutture archeologiche fisse sommerse a Vasto ed Egnazia, Arcavacata di Rende (CS).

Guerricchio A., Guerricchio M., Maruca A., 1997, Sul presunto antico porto romano di Egnazia (BR), Arcavacata di Rende (CS).

Vlora N.R. 1975, Considerazioni sulle variazioni della linea di costa tra Monopoli (Bari) ed Egnazia (Brindisi), Bari

Vitruvius and De Architectura

The primary source on the construction of ports in Roman times era is the De Architectura (V, XII) by Vitruvius. There are other ancient sources that deal with the topic in a general way, e.g. Flavius Josephus on the construction of the port of Caesarea Maritima (Antiquities of the Jews XV, 331-338; The Jewish war I, 409-413) or Procopius of Caesarea (De aedificiis 1, 11, 18- 20).

Some passages of Vitruvius are not easy to understand today. The author, in fact, takes for granted that the readers have some basic knowledge of the subject; moreover, he retrieved information directly from other sources when he was not well-acquainted with the subject. The resulting prose is often constructed in an convoluted way and the terminology, in some cases, is not easily interpretable.

After providing information about the existing relationship between the conditions of the construction sites and the projects, Vitruvius lists the most popular techniques for building structures in water. The first two techniques are directly related to casting into wooden formworks in situ, while the third describes a pre-fabricated construction made onshore and then transported to its destination.

Pozzolana: what is it?

Pozzolana (pulvis puteolanus) is the base material for creating a completely hydraulic cement, i.e. able to solidify underwater. From a mineralogical point of view, it is a loose pyroclastite containing gravel composed mainly of pumice and volcanic slag. It was called pulvis puteolanus as it was preferably extracted in the area of Phlegraean Fields.

Vitruvius himself describes it with these words: «… there is also a kind of powder which from natural causes produces astonishing results…; not only lends strength to buildings of other kinds, but even when piers of it are constructed in the sea, they set hard under water (II, VI, 1)».

Obviously, the availability of this material was a conditio sine qua non for harbour facilities. However, its absence in a given zone was often solved with imports, also by cargo ships that very often transported the material in their empty hold during the return journeys.

The three techniques described by Vitruvius

Construction in non-watertight cofferdam

«Then, in the place previously determined, a cofferdam, with its sides formed of oaken stakes with ties between them, is to be driven down into the water and firmly propped there; then, the lower surface inside, under the water, must be levelled off and dredged, working from beams laid across; and finally, concrete from the mortar trough — the stuff having been mixed as prescribed above — must be heaped up until the empty space which was within the cofferdam is filled up by the wall».

Interpreting Vitruvius, the phases of this technique could have been as follows: a basic skeleton of the cofferdam was built ashore, then dragged and finished in situ. Once it was placed, vertical poles with iron pile-shoes were wrought around its perimeter (stipites) to keep it anchored to the seabed, using a pile driver and containment boards. The wooden horizontal connecting elements (catenae), already assembled, were meant to stabilise the cofferdam and balance the push toward the outside of the freshly poured concrete. On the highest section of the catenae, the one emerging from the water, there was a plank that served as a ground base for all works, as the purgatio (cleaning) of the seabed, the casting of the cement or the scaffolding for elevated elements that were supposed to rest on the cofferdam.

       XVIII century drawing of a pile driver, perhaps very similar to the ones used in Roman times (from SILBERSCHLAG J.E.,                                                                     Abhandlung vom Wasserbau an Stroemen, Leipzig 1769)


Illustration of workers in the XVIII century with a manual pile driver (“mazzapicchio”). This tool must have been widespread also in Roman times

Obviously, this technique did not allow for constructing exposed revetments. The cementitious mortar was poured in direct contact with the boards of the cofferdam. Nonetheless, this procedure was enormously popular in the ancient world, as it was particularly versatile and adaptable in various environments that did not require casting between purpose-made formworks.

Construction in watertight cofferdam

«But in places where this powder is not found, the following method must be employed. A cofferdam with double sides, composed of charred stakes fastened together with ties, should be constructed in the appointed place, and clay in wicker baskets made of swamp rushes should be packed in among the props. After this has been well packed down and filled in as closely as possible, set up your water-screws, wheels, and drums, and let the space now bounded by the enclosure be emptied and dried. Then, dig out the bottom within the enclosure».

It is clear that, according to Vitruvius, this second technique was to be used when pozzolana was not available and another, more traditional technique had to be adopted.

In that case, the walls of the cofferdam were double, as the space between them was filled with pressed clay that was supposed to be a protection against water infiltrations.

Such construction method was well adapted to smaller structures like pilae, rather than long piers.

The two types of cofferdams: inundated in the foreground, watertight in the background (from Perrault C. 1673, Vitruve, les dix livres d’architecture [traduction intégrale de C. Perrault, revue et corrigée sur les textes latins et présentée par A. Dalmas], Paris 1965.)

Construction with prefabricated blocks

«But if by reason of currents or the assaults of the open sea the props cannot hold the cofferdam together, then, let a platform of the greatest possible strength be constructed, beginning on the ground itself or on a substructure; and let the platform be constructed with a level surface for less than half its extent, while the rest, which is close to the beach, slopes down and out. Then, on the water’s edge and at the sides of the platform, let marginal walls be constructed, about one and one half feet thick and brought up to a level with the surface above mentioned; next, let the sloping part be filled in with sand and levelled off with the marginal wall and the surface of the platform. Then, upon this level surface construct a block as large as is required, and when it is finished, leave it for not less than two months to dry. Then, cut away the marginal wall which supports the sand. Thus, the sand will be undermined by the waves, and this will cause the block to fall into the sea.  By this method, repeated as often as necessary, an advance into the water can be made.».

From an archaeological point of view, currently there is no evidence of the application of this method. However, Virgil in his poetry (Aen., 9, 710 et seq.) confirms that it was actually used.

Considering the importance of the Messapian and Roman city, the structures of the ancient port of Egnatia appear also in the accounts by erudite visitors and scholars between the XVIII and the XIX centuries.

In 1745, Francesco Maria Pratilli, in his text Della Via Appia, not only provided us with a first map of the city, but also mentioned the presence of a spring water close to the city walls by the sea, called “the fountain of Agnazzo” by locals.

Since Pratilli did not enjoy a good reputation as a scholar and was also known to have forged some ancient epigraphs, his work was heavily questioned by Ludovico Pepe, an Apulian archivist who in 1882 wrote the first monograph about the ancient Messapian city and did not find any evidence of the fountain. On the contrary, he highlighted the scarcity of water resources in the area. Actually, the presence of drinking water is mentioned by other, more or less contemporary sources, while recent paleo-environmental surveys confirmed the possibility of its existence. A map from the XVIII-XIX century, kept at the National Archives of Bari, reports the location of “fontane” as well. The mentions of drinking water is fundamental to confirm the thesis that the Roman port was used also in the Medieval and Modern ages, not only thanks to its fortifications that were a salient orientation point from the sea and protected it, but mainly due to the fact that it had water at its disposal in an area where water was scarce.

In the 1789s, the abbot of Saint-Non, a humanist and archaeologist, travelled to South Italy. His journey gave origin to an encyclopaedic text published over five years. When he talks about Egnatia, he confirms that “the remains of something that might have been a pier are still visible. This pier was not built in Antiquity, but on the shore, with the materials from the ancient city…”.

Emanuele Mola, Superintendent of the Antiquities of Bari, in a text dated 1796 mentions the structures of the submerged port as well preserved: ““The greatest change in that shore, however, seems to be visible in the port of the ancient city, which lays today beneath the military tower called Anazzo, and which still retains the shape of its ancient state”.

However, it was the French archaeologist François Lenormant who in 1882 in “Gazette archéologique” describes with precision the ancient port of Egnatia: “The plan of the city is an elongated square with one of the long sides that runs along the sea. Close to the shore, on a small, gently sloping hill, located at the same distance from both the short sides of the rectangle, and consequently, in the middle of the city, the acropolis was built, whose walls are also well preserved and clearly characterised as those of the external walls of the settlement. This fortress overlooked and protected two little square basins, partially dug or regularised by human hand, between which the fortress was placed, one to the north, the other one to the south of the hill. The partially preserved partitions of the galley coves are still distinguishable under water”.

The already mentioned Ludovico Pepe, in his monograph on Egnatia written in 1982, discusses the hypotheses of previous scholars about the port.


The National Museum "Giuseppe Andreassi" and the Archeological Park of Egnatia

The Archeological Museum of Egnatia, an integral part of the magnificent Park which includes the excavated remains of the ancient city, is managed by the Polo Museale (Museum Hub) of Apulia. It was dedicated to Giuseppe Andreassi, the late Superintendent who, more than anyone, encouraged the promotion of this important settlement.

The Museum was created in the 1970s, when it was decided to exhibit the artefacts that were found during the excavation campaigns since 1912.

The current fit-out describes the history and the urbanisation of the city, starting with the first evidences of the XVI century B.C. The tour is articulated in 7 sections and includes nearby sites that have contributed to the development of the city, e.g. Monopoli, Torre S. Sabina, Mesagne and Cavallino. A particular attention is paid to the Messapian, Roman, Late-Antiquity and Medieval sections.

The entrance to the National Archaeological Museum of Egnatia

Biological colonisation of the pilae

The pilae of the ancient submerged port were taken under examination. These are walls in opus reticulatum, built with squared elements of calcareous stone (cubilia) and cementitious mortars.

The degradation of the cubilia

The opus reticulatum walls present a clear difference in the state of preservation between the lower parts (well-preserved as long as they were protected by sand) and the upper parts that are heavily degraded due to their constant exposition. On the contrary, mortars are better preserved.

Scheme of the state of preservation and graphic representation of the erosion dynamics on the stone

The biological colonisation

The exposed surfaces of the pilae show a biological proliferation typical of sumberged stone substrates. There is a dense biotic coating made of many micro-organisms and organisms that sometimes cover completely the constructing material. The drawing shows a scheme of the biological growth over time, highlighting the effect caused by the covering.


The main elements of the biological coating are photosynthesising microscopical and macroscopical organisms as algae that create a thin layer with an extreme variability of shapes and colours, sometimes compacted on the stone. The chart below shows the main and the most common algal components, with species belonging to green algae (Chlorophyceae), brown algae (Phaeophyceae) and red algae (Rhodophyceae). The following table shows the main species found on the masonry structures of pilae.

Many species of red algae may either form flat calcareous layers, strongly adherent to the stone, or tree-like structures recognisable thanks to their pinkish or purplish colour.

Black and white SEM images show a calcareous skeleton that supports vegetative cells and reproductive structures.


The biological coating is also formed by numerous sessile animal forms like eggs and larvae on the surfaces. Their growth created more or less conspicuous structures that were key to the degradation process of the artefacts.

There were several groups of animals, varying in morphology, size, and colour.



This taxonomic group includes animals of a very simple structure, i.e. a kind of a sac with openings where vital functions are carried out in water. The picture shows some of the species that cover the surfaces or live inside the material.

Discover the finds in 3D

Among the artefacts of underwater origin, only the female statuette in bronze comes from the area of the Roman port of Egnatia. It is a very interesting object that quite certainly was a steelyard balance, a tool used in commercial activities.

The other artefacts presented in the Virtual Museum come from areas more or less close to Egnatia (e.g. Monopoli) and were spontaneously delivered by finders or seized by police authorities.

Stone anchors are very interesting as well, as they have always been debated among scholars, together with the so-called “doughnut-shaped” deadweights.

The discovery of a large jug called dolium along these coasts has sparked a debate on the routes of the cargo ships that transported them, adding a further piece to a puzzle that is still far from being completed.

Finally, we would like to mention an amphora of the Late Roman 1 type that confirms the commercial importante of Apulia in the Late-Antiquity and Byzantine ages.


Female statuette in bronze, discovered in one of the inlets of the port, filled with lead to be used as a counterweight. Heavily worn out (face and hands). Traces of a probable applique on its head. Perhaps it can be interpreted as part of a…


A stone anchor or a deadweight/fixed mooring, roughly shaped as a doughnut. There are two perpendicular grooves in the central hole, attributable to the wear caused by mooring ropes. Biological degradation The artefact does not present traces of biological colonisation, probably thanks to the petrographic…


Trapezoidal stone anchor with three holes. Circular upper hole with quadrangular lower holes. Traces of lithophagous organisms. Biological degradation The artefact presents clear traces of perforations caused by endolithic animal organisms. There are numerous small circular holes over a large part of the surface that…


The shape evolved considerably between the IV and the VII century A.D. (PIÉRI, 2005; REYNOLDS, 2005). Its first specimens can be traced back in the mid-III century A.D. In the first half of the V century A.D. LRA1 (Late Roman Amphora 1) was initially exported…


Biological degradation The artefact is entirely colonized by fouling organisms, consisting almost exclusively of sedentary Polychaeta belonging to various genera. There are small calcareous tubes of various dimensions, of an elongated or coiled up shape and different ornaments on the surface. There is a lesser…


A bell-shaped stone anchor with two quadrangular holes on the vertical axis. Holes made by lithophagus organisms. Biological degradation The artefact shows evident traces of perforation caused by endolithic animal organisms whose distribution seems to be different on the two flat surfaces. The whitish coloured…


Small olla (olletta) with an approximatively biconical shape with a slightly everted, thin rim and a prominently everted, tongue-shaped handle. Flat bottom. Refined, thin-walled ceramics. Biological degradation The artefact presents an heterogeneous epilithic fouling, in which there are numerous specimens of Polychaeta Serpilidae characterised by…


Stone anchor of a trapezoidal shape, with three through-holes. The upper one was used for mooring ropes, while the two lower parallel ones had a couple of wooden bills . Biological degradation The artefact presents traces of perforations caused by endolithic animals. There are numerous…


Fragment of a globular-shaped, concreted metallic container with a rounded rim (the tag wrongly indicates a shoulder with neck of an amphora). Biological degradation The artefact does not show signs of biological degradation.   References