The Salapia Exploration Project was created to study the Roman and medieval town of Salapia, the most important of the known historical settlements within the territory of Saline di Margherita di Savoia, of which little is known in the period following its re-foundation by M. Hostilius at the end of the 1st century B.C. A geophysical survey was carried out in an area west of Monte di Salpi (4 hectares) and a small zone (2 hectares) on its summit. So far, the results have revealed a large part of the town’s urban layout, with orthogonal roads and square _insulae_ (c. 40 x 40 m). In 2014, two trenches were opened in correspondence with a complex nucleus of anomalies relating to structures and roads inside the town that showed up on the geophysical survey. In trench 1, the excavations exposed the structures of what was probably a _domus_, whose original construction seems datable to between the second half of the 1st century B.C. and the 1st century A.D. During the full imperial period (2nd-3rd century A.D.), the complex was enlarged with the addition of quadrangular rooms with mosaic floors. Between the 4th century and first half of the 5th century, the building underwent further restructuring which altered the interior spaces, repaired the earlier mosaics, and reinforced the walls. The creation of a cesspit, obliterating several earlier structures, probably dates to this period. Its function was to drain water from rooms situated beyond the excavation area and consequently not investigated. From the second half of the 5th century onwards, changes occurred in the type of occupation within the probable _domus_. A number of rooms were abandoned and timber supports were inserted in other rooms as supports for roofing. The complex was abandoned during the first half of the 6th century when dwellings built in perishable materials were put in place on top of the obliteration layers and several areas were used for funerary purposes. In trench II, a section of a road was investigated, bordered by two building complexes with different functions. To the east, a craft-working installation was exposed, specializing in the treatment of textiles and hides, as attested by a series of circular and rectangular tanks. This complex may originally date to the full imperial period (2nd-3rd centuries), and gradually went out of use between the 4th and 6th centuries A.D. To the west, a room, perhaps a warehouse-workshop was excavated that produced numerous amphora fragments and a hoard of c. 400 small bronze coins. The sudden abandonment of the structure, marked by the collapse of the tiled roof, probably occurred in the second half of the 5th century. As in other parts of the town, once the craft-working complex and warehouse-workshop went out of use, the area was occupied by temporary structures, attested by huts and beaten earth surfaces.

This season’s excavations concentrated on the area between trenches I and II opened in 2014, on a north-west alignment, and a sector within trench II, immediately east of the tannery identified during the previous year. In trench I, more rooms belonging to the _domus_, partially excavated during previous seasons, were uncovered. The first phase of the _domus_ seems to date to the Augustan period. Four rooms were identified immediately south of room 6, which faced onto an open or semi-open area, interpreted as a long corridor providing access to the _domus_ itself. The corridor was paved with a mosaic of white tesserae, with a double border of black tesserae. During the 3rd century AD, the complex was enlarged and new mosaic floors were laid, such as that uncovered in room 18. The polychrome mosaic has a geometric pattern decorated with vegetal motifs. The occupation of the structures during the 4th century and the abandonment of the building in the second half of the 5th century appeared to be confirmed. All the excavated rooms showed signs of reoccupation from this period onwards, with evidence for timber structures on top of the obliteration levels and wall collapses. This reuse probably continued until the 7th century. In trench II, east of the tanning vats, the presence of a drainage channel was confirmed. Running west-east, it drained the wastewater from the tanning process and had masonry-built parapets and a tile floor. This area also presented evidence of heavy robbing and reoccupation of the spaces (originally attributed to a 2nd century A.D. phase) beginning in the mid 5th century and lasting until at least the 7th century. This involved the installation of dwellings with roofs of perishable materials and internal structures for food preparation and heating (cooking slabs and domestic hearths).

Investigations continued in the area of the Roman town of _Salapia_, on the southern shores of the ancient Lake Salpi, now the saltpans of Margherita di Savoia in Apulia. Research mainly took place in the area of the _domus_ in _insula_ XII, also excavated during the two previous campaigns. As this was the last season of a three-year concession, the aim was to define the chronology and plan of the complex as precisely as possible. The available evidence relates to only a part of a large house (450 m2 have been excavated), built between the second half of the 1st century A.D. and the early 2nd century A.D. A group of seven rooms was arranged along the south and east sides of a porticoed courtyard, or peristyle, with a with a well. The courtyard was probably a garden area. From the second half of the 2nd century A.D., the _domus_ underwent substantial restructuring. Particular care was taken in the enrichment of the decorative scheme, as attested by the use of bi-chrome and polychrome mosaics with geometric motifs for the floors of the residential or reception rooms. This season a very fine bi-chrome mosaic was found in the walkway around the peristyle. New and interesting evidence has shed light on the building’s later phases. Firstly, between the late 4th and the end of the 5th century A.D., the domestic space contracted, the unused rooms became middens, and the spaces were divided to create smaller residential units. Then, between the 6th and early 7th century, the structures gradually fell into ruin. An accumulation of rubble layers formed on which huts of perishable materials and associated structures for stabling small animals were built. Excavations also continued in the sector of _insula_ XVI opposite the tannery connected to the _domus_. Here, a shop/workshop was built on top of a pre-existing structure that was also probably commercial. The shop had spaces for the storage and sale of dry foodstuffs and for craft-working activities. The shop has two rooms, of equal shape and size, separated by a partition wall built of stones. There was only one entrance, which did not face onto the road, but towards a large open space paved with coarse _opus_ _signinum_ and extending to the south and west of the building. When, between the late 4th century and the early 5th century A.D., a fire destroyed the shop it contained six amphorae, three African (Keay XXV) and three of eastern production (two were LRA1 types). Over 200 small bronze coins were also buried in the collapse, probably the takings from sales. In the adjacent room, the roof collapse obliterated a hearth abutting a wall and, in the part of the room closest to the door, two sub-circular pits were dug into the beaten floor surface. The pit walls showed clear signs of reddening from contact with heat. In the absence of evidence for what was produced inside near the pits, it may be suggested that the activity was connected with metalworking as a large amount of iron slag was found just outside the shop/workshop, scattered on the exterior paving.

During the 2017 campaign, work took place in two distinct sectors of this settlement, which has a long and multi-layered settlement history. Trench III began the exploration of so-called _Insula_ _XIX_ of the Roman town of _Salapia_. The excavations documented evidence relating to the latest occupation phases in this part of the town (from the late 6th century A.D. onwards), when huts built of perishable materials were built on top of the layers of collapse and abandonment of the earlier Roman structures. Two distinct re-occupation phases were identified, characterised by entirely timber-built structures, with beaten earth floors and hearths. Traces indicating intense robbing of the pre-existing Roman structures were of interest: numerous robber trenches for the removal of stone and brick used in the Roman walls were found across the investigated area. Trench IV was excavated on the so-called Monte di Salpi, a manmade rise on whose summit plateau the local community, perhaps already in the 10th-11th century, reorganised its spaces and lived there until the threshold of the Modern age. In particular, the excavations uncovered a part of Salpi’s 13th century urban fabric, characterised by the presence of a probable _via_ _publica_, which from the south hill slope headed towards the interior of the settlement. The road had undergone numerous maintenance interventions and it was at least 6-7 m wide. On either side of the road, two large buildings with substantial walls were identified and partially excavated. At present, it is not possible to establish whether they were private dwellings (tower houses?), public complexes, or part of the system of the perimeter fortifications that probably surrounded the entire _civitas_, as attested by the sources. It is certain that by the mid 14th century, the buildings in the entire sector collapsed: the characteristics of the related layers and, above all, the way in which the walls themselves collapsed seem to indicate traumatic and sudden damage caused by an earthquake.