The archaeological site "Ulmetum" is positioned in the central part of Dobruja, located in the north-western edge of the village Pantelimonu de Sus (Constanta County). The site, representing a Roman-Byzantine fortification, played an important strategical role for the main road of Dobruja, protecting the transport of goods and merchants (between the Roman cities Noviodunum and Markianopolis). This control was performed with the help of the soldiers stationed in the area, up to the northern part of the provincial territory, near the Danubian borders.
Over time, especially in the late nineteenth century, several inscriptions, highly relevant to the history of local communities, were saved. Amongst these there are a funeral cippus on which it is carved the name of Caius Julius Quadratus, who was quinquennalis territorii Capidavensis and held the title of princeps loci. To this there can be added other inscriptions from Pantelimon, discovered and preserved by the great Romanian archaeologist Grigore Tocilescu who proved the existence of a city from the early Roman period (inhabited by a mixed community of cives Romans et Bessi consistentes), whose name was recorded on an votive altar dated back to the year 140, in the form of Vico Ulmeto, namely vicus Ulmetum. Another very important document confirming the existence of the fortification of Ulmetum appears in De Aedificiis written by Procopius of Caesarea: ".... Beyond it lays an ancient city called Ulmetum. Because the Sclavins barbarians made there a watching station, where they stayed for a very long time, the city was deserted and there was nothing left of it than the name. This city was rebuilt it from the ground and thus it protected the region from the Sclavins invasions". Subsequently, based on other inscriptions, there have been identified those who worked on its reconstruction, a detachment of legio palatina, called lanciarii iuniores.
The beginning of Ulmetum archaeological research works is marked by the great archaeologist and historian Vasile Parvan. During 1911-1914, he performed four successive excavation campaigns. The results of the research were published in three excavation reports containing various information regarding the construction elements, the monuments and many of the identified objects. The archaeological excavations of 1911-1914 first aimed to unveil the walls and to recover the epigraphic or sculptural documents reused as construction materials. The outer and inner precinct was unveiled, as there were uncovered the gates, towers and portions of the intra-muros area (revealing in the eastern part of the city a building with a large apse, for military use (praetorium, which was functional during the 4th and the 5th centuries BC). They were drawn a number of sections (along and across the city), tracking perhaps road network, sections that in the first place started from the southwest and northwest gates, then from the secondary gates located between the towers of the northwest site and the ones of south and east. Those sections also revealed "barbarian walls" from the various constructions which destination has not been specified. These excavations have permitted the drawing of a general map of the city, as well as dating it back to the sixth century AD, also being mentioned specific findings from fourth and fifth centuries AD, but in much smaller quantity.
The city’s map has the shape of a right triangle with a curved hypotenuse. On its corners the fortification was equipped with three round towers. It also had two gates flanked by two towers with a semi-circular side shaped like U. On the southeast side there were four rectangular towers to which were added two towers of the same shape, one on both sides of city. Some believe that the craftsmen who built the fortification came from the Thracian space, as there are many similarities of the plan with other centres from the same period. The precinct had a 2.60 m thick wall, the foundation of the wall being 20 to 60 cm wider than the superior part of the wall. The foundation had a depth of 1.85 m to 2.50 m. The outer side of the wall was plated with reused stone blocks, reaching dimensions of 2.70 m (a funerary stele). Inside the precinct the wall did not had any facing blocks. On most of the curtines were preserved the traces of the stone stairs allowing climbing the wall, stair steps with a height between 28 and 40 cm, their width ranging up to 1 m.
The north-east side of the fortification was bounded by two round towers which had a diameter of less than 10 m. In the middle the towers had one rectangle pillar of stone with mortar supporting the floors and the roof, the length of this side was of 135 m, inside in the middle there was located a rectangular tower which had a length of 4.43 m and a width of 5.30 m. The northwest side of the enclosure was almost equal with the aforementioned one, which had a common tower and on the other end there also was a circular tower. On this side there was a gate flanked by two towers. It seems that the door was locked from antiquity as it results from the wall built in the opening of the gate. Also on this side there was the great quadrilateral tower, (14.75 x 10.46), which had two pillars, built according to the opus mixtum technique. The effective area of the tower was 155 m, but over the time it suffered repeated rebuilding.
The southeast side of the fortification was provided with a gate flanked by two U shape towers, which seems to have been functional on the entire life of the complex. On this side there were four rectangular towers of different sizes. Also there were discovered two small exit doors that can be consider as secret doors.
Also taking advantage of Ulmetum excavation period, the archaeologist also researched the territory of the main city making some excavations on parts of the fort called Castra Aestiva. Using them, Vasile Parvan dated the fort as belonging to the early Roman period (during the reigns of the emperors Trajan and Valens), the entire site being abandoned before or during the fourth century (subsequently no longer being inhabited).
Many of the objects discovered in the city, were placed in the museum, which was built at about 120 meters from the northern side of the precinct. Unfortunately the museum was destroyed during First World War and objects were robbed. The premature death of the great Romanian archaeologist Vasile Parvan sharply diminished the interest for the archaeological antiquities of the area for a very long time. Between 1914 and 2004 there had not been carried out any systematic archaeological research at Ulmetum. However, through some of the discoveries made, the city still presented interest for the scientist of this field, such as the epigraphic discoveries and the coins saved in the inventories of cultural or archaeological institutions, as remote sensing (identifying centres for living that could belong to vicus Ulmetum or to some hypothetical villas, only epigraphically attested) or as rescue excavations (after the collapse of a shore located 300 m south from the south-west gate of the city it was unveiled a kiln for construction materials for the production of pantile and tiles that typologically were dated in the fifth and sixth centuries AD).
During 1969-2000, Adrian Radulescu, the manager of Constanta Museum of National History and Archaeology was the promoter for resuming the archaeological works at the site. But only since 2004, the archaeological research restarted in Ulmetum, aiming to clarify the stratigraphy of the fortification, the scientific exploitation of the new discoveries and preparing the site to be introduced in the tourist circuit through actions of planning, restoration and conservation of the constituent parties.
In terms of research methodology, the city was divided, in several sectors, corresponding to the cardinal points, being established in theory four equal areas (north, west, east and south). Important results were achieved mainly in two sectors. In the Norther sector there has been identified a funeral area, dated before the construction of the fortification, with tombs, one pf which had a very rich inventory of "barbaric" objects (assigned to the Santana de Mures / Cerneahov civilization). In the Eastern one there were discovered multi-level dwelling areas. So there were clearly attested archaeological evidence for an earlier occupation of the area before the restoration of the Emperor Justinian, who seemed quite intense (the 4th and the 5th centuries AD.). It also identified a religious building (basilica) with a three naves plan, placed in a sheltered area, near the small south east gate, which was probably built in the second half of the fourth century or at beginning of the 5th century AD. It was identified a narthex, probably added in a later period (maybe in the sixth century AD). The building was probably abandoned in the final years of the city’s inhabitation (late 6th century and early 7th century). With the discovery of the basilica, Ulmetum became an important centre on the relatively small list of basilicas discoveries of Scythia Minor.
Also, numerous studies have been conducted in the extramural area, concentrated in the vicinity of the precinct wall, in order to identify elements that could contribute to understating the development of the early Roman habitation to Ulmetum. There can be noted the discovery of a habitation level from the Ottoman period, near the city, in the Southern Sector (in front of the tower no. 8 and in the basilica area). In this respect, the near the tower no. 8 was discovered a hovels complex (according to the ceramic fragments and coins discovered there it could have been functional within a larger period between the sixteenth and seventeenth centuries). The presence of Ottoman relics only in the southern area of the fortification suggested that only this part was inhabited during the Ottoman period, which had a close connection with the dismantling of the fortress walls. The research conducted between 2012-2015 revealed in this extramural part of southern fortification more specific living complexes from the early Roman period (the second and third centuries AD), and late Roman period (fourth century AD), namely Roman-byzantine period (the fourth century AD). In terms of stratigraphic point of view, the discoveries can be presented as it follows: a first vegetable post ancient level of the sixth and seventh centuries, followed by a level containing the debris from the precinct of the fortification, followed by a levelling level which may correspond to the sixth century and successive levelling levels possible from the fifth century, and then a level dated back in the fourth century. Finally, the last level has at least two successive stages, specific to the 2nd and 3rd centuries.
Staring with the archaeological campaign of 2009, based on the results of the excavations of the recent years and by adopting a new strategy for the research of the field (excavation in open area, dividing the sectors in squares, performing monitoring surveys etc.), there could have been established five major phases of habitation in the fort, with clear chronological elements. Thus there can be mentioned: period A - between the second and third centuries - expressed by the emergence of a discontinuous level from the early Roman period on the site of the future Theodosian city, which complete the perieghetice research of the centre of the actual village, which allowed the identification of an intense habitation during the second and fourth centuries (along the terrace of the NE side of the valley of Ceatalorman river.); period B - during the second half of the fourth century – namely a level where had been discovered Roman and Santana de Mures / Cerneahov pottery or coins dating from the second half of the century fourth AD; period C - in the fifth century – the numismatic material found indicating a clear terminus post quem between 395-401 (according to the Virtvs exerciti coins) for the first level inside the fortification); period D - during the sixth century - after the abandon of the city (arround 470/480 - 540/550), expressed in terms of stratigraphic terms by a levelling during the restoration of the Emperor Justinian, followed by a inhabitation period datable back in the fourth century consisting of at least three sub-periods which from chronological point of view date back to the early seventh century and the period E - during the sixteenth and seventeenth centuries AD.
From the investigations carried out up to now in the Roman-Byzantine fortress area, inside the fortress, there have not been identified clues to demonstrate its inhabitation during the early medieval period. However there is evidence of the presence of the pottery characteristic for that period, and in 1912, Vasile Parvan mentioned a potters stamp among the discoveries of such elements. Based on accidental discoveries made over remote sensing , in 2010 had been opened a new research sector (extramural), located in a point approximately 300 m southwest of the city, near the creek Pantelimon, on the hill on its northern side, where were concentrated ceramic materials specific to that period. The archaeological campaigns carried out in the following years have confirmed the existence of an area with dwelling structures (where have been identified several huts). Based on the analysis of the ceramic material (baked oxidant made from a sandy mass and decorated by perforation, grey lustred decorated pottery), the inhabited area was dated back in the late eighth century - the first half of the tenth century AD.
In conclusion, the fortress was built around the end of the fourth century and early fifth century AD (according to the archaeological and numismatic evidence – minting of coins, Salvs Reipvblicae during the years 388-395 and Virtvs during the years 395-401), probably under the Emperor Theodosius I, along with many buildings with a large variety of rooms as military barracks, a basilica and probably a military headquarters. Fortified on all sides, it housed a population of farmers, who supplied grains to many centres near the limes. Among the military units stationed over time in the city can be mentioned Cuneus equitum scutariorum and Paedatura lanciarium iuniorum. With some interruptions it lasted until the late sixth century and early seventh century, at the end of the Roman-Byzantine period. It is a fortification which can be included in the small or medium categories, which gradually became a fortified settlement and which in the fifth century AD had been attacked and quasi-destroyed by successive attacks of the Huns, being partially rebuilt in the sixth century AD under the Emperor Justinian (once again included in the administrative and economic provincial system). The research conducted in the last 10 years suggests that the fortification built a fundamentis by Justinian was located almost in the same place with the one erected at the end of the fourth century or early fifth century AD.
The last restoration phase of the city (which seems to have been mainly done only in certain of its sectors) seems to have been performed in the first half of the 4th century, when due to the desperate need were reused all the available funerary monuments, fragments of shaped stones from the ancient monumental buildings, all previous secular inscriptions etc. the Slavic and Avar attacks in the late sixth century destroyed any form of continuity in terms of quasi-urban fortification, the last locally present elements actually disappearing at the beginning of the next century.