The 2017 season took place during the second half of June and first half of July. The aim of the initial campaign was to define the surface of the complex of the late Roman villa. The research was based on two components, excavations and survey of the immediate surroundings of the site.
A geophysical survey showed that the remains of the buildings spread in the direction of NW-SE, mostly underneath the field, and that parts of the complex could extend underneath the forest to the south and beneath Mitrosrem agricultural company’s land. A geo-radar survey on a smaller surface was conveyed by our colleagues from the Centre for New Technologies and it established the existence of walls on the south-eastern side of the site, stretching towards the northwest. These findings lead to the conclusion that the villa complex extends for more than six hectares.
Unfortunately, in recent decades, digging deep ditches damaged the Roman architecture in several places. This was determined by finds of the remains of floors made of hydrostatic mortar and bare walls.
The first goal of the 2017 season was to examine the wall on the south side, visible for a length of more than 80 meters, as well as the ditch that existed next to it. It is our current assumption that this was an external wall of one of the monumental buildings that was typically part of a late ancient palatial complex. The wall is of high quality masonry with regularly shaped ashlars and bricks.
Openings on the wall were noted at equal distances. To the north of the wall there are a series of rooms and we therefore assume that these openings had the purpose of a drainage. On the north side of the wall, in the modern ditch, the remains of the juncture of two massive walls with bricks were discovered. A special kind of tile (tegula mammata), commonly used for wall heating, was discovered inside one room.
Remains of a plaster floor decorated with a floor mosaic were found in one of the northern rooms. The mosaic consists of colourful tesserae forming various geometrical forms (vines, braids, complex geometrical patterns) characteristic of late antique floor decorations. A destruction horizon was recognised in several places. Above the mosaic floor a hand, fragment of a marble sculpture that could originally have been portraying a Roman deity, was uncovered in spoils. In Trench 1, several finds indicate the deterioration period of the villa. A wall decorated with frescoes and traces of luxurious marble decoration were found above the floor. A child’s grave was discovered in the same room, which indicates that by the end of the 4th or beginning of the 5th century this space was no longer inhabited and that its ruins became a burial site.
Another interesting find was uncovered in the series of rooms, north of the massive wall - a pottery kiln with a clay floor surrounded by Roman bricks. Further excavations will show whether this part of the complex was used for industry, or if the kiln was built in the later period when the villa lost its residential character. A find of a coin of Emperor Justin II, minted in Nicomedia (Asia Minor) in 572 AD, is a significant find that shows that the site was a settled over a longer period of time.
In some areas, it was possible to examine the earliest stages of the site that suggest a smaller settlement existed there prior to the late ancient villa. The strongest indicators for this are the findings of fragments of ceramic dishes in La- Tene forms, and even younger ones from the 2nd and 3rd centuries.
During the excavations, a preliminary survey of the immediate surroundings of the Glac site was carried out. The purpose of this investigation was to potentially detect the remains of ancient communications and other complexes that could exist in the vicinity of the villa and suburbs of Sirmium. There are currently indicators that point to the possibility that the road in the direction of Sirmium-Glac-Fossae was positioned north of the villa. There could also be a route that led from Sirmium to Jarak, right next to the coast, between the villa and Sava River. Further research, planned for the season of 2018, will tackle these unresolved questions.