The so-called ‘Great Baths’ were built in the 4th cent. A.D. and had an area of approximately 25,000 square meters. The archaeological excavations had so far brought to light three very large halls with refined geometric and figurative mosaic pavements or inlay-work floors, six pools for cold baths, the central part of the natatio and some sections of the heated halls on the western side. One epigraph found in the 1980s tells us their ancient name: “Thermae felices Constantinianae”. At the end of the 5th cent. the Great Baths had lost their function and were inhabited by small family groups who buried their dead outside the perimeter of the building. Towards the end of the 7th century, their ruins were abandoned, probably because the vaults had begun to collapse. From the 13th century all the walls were dismantled to recover stone blocks and bricks and the whole area of the complex was reclaimed and given over to farming. During the 20th cent. the Great Baths were investigated several times: by G.B. Brusin (1922-23), L. Bertacchi (1961) and P. Lopreato (1980-1987). In 2002 (until 2014) the University of Udine has resumed excavations on the basis of a joint research project with the Archaeological Superintendency of Friuli Venezia Giulia. The 2016 archaeological campaign (under concession to the excavations) was focused in the northeastern area of the Great Baths, where we found a complicated stratigraphic sequence with three main phases. 1. The oldest evidence is a large foundation of bricks and lime mortar, covered with ‘cocciopesto’. It was probably a monumental complex with basins for water (one of which is circular) and with a channel for adduction and drainage of water. Its limits have not yet been fully brought to light and it could also be an earlier building, possibly connected to the nearby theatre and to one of the gates in the Republican walls. 2. This structure (maybe a monumental fountain) was partially dismantled and partly reused as foundation of the mosaic in large tesserae which covered the pavement of a room on the north side of the Great Baths. 3. In the 5th century this hall was divided into two smaller rooms with tessellated pavements. One of the mosaics had large trapezoids probably surrounding a central octagon; the only preserved scene shows a Nereid riding a Triton, nowadays exhibited in the Archaeological Museum of Aquileia. Some thin and dark layers on the surface of the latest mosaics testify to a reoccupation in the Early Middle Ages. An intensive and systematic despoliation has removed all the walls and most of the other structures, saving only few mosaic fragments and the deepest foundations and making difficult to reconstruct the different phases.

The 2017 campaign concentrated on two crucial areas of the large bath building, with important results for the progress of this research. 1. Excavation in the north-eastern area (Sector C) clarified the chronology relating to the platform built of bricks, concrete and _opus_ _signinum_, interpreted as a structure with pools and fountains (structure ‘S20’), which was demolished and obliterated by two large-scale successive restructuring interventions and by the robbing of the entire bath complex (see 2016). The excavation of the robber/foundation trench separating to the west S20 from the North Hall, showed that the first was perfectly inserted into the framework formed by the walls of the Constantinian baths and, therefore, belonged to the first phase of the building (first half of the 4th century A.D.). At the end of the 4th century A.D., the Constantinian structure was partially dismantled and partly reused as the make-up for the floor in large tesserae in one of the rooms on the north side of the baths (room A19, phase II). In the 5th century A.D. (phase III), this room was divided into two rooms with fine mosaic floors, one to the east that was figured and octagonal (A17), and a smaller rectangular mosaic to the west (A18). The fill of the robber trench contained numerous small marble slabs cut from the larger polychrome pieces of an _opus_ _sectile_ revetment, and numerous green and blue glass tesserae, and some with gold leaf, which may have decorated niches and pools with fountains relating to the first phase structure. 2. The excavations in the south-western area of the building (sector F) were conditioned by an intervention to protect and conserve a badly damaged mosaic in room A10. Three deep _sondages_ were opened in the robber-foundation trench separating A10 from the south wing. The well-preserved foundations of the N-S wall and the walls at a right angle to it were reached. The N-S wall was built of _sesquipedales_ and was 1.25 m wide (while the E-W walls varied between 0.80 and 0.90 m). It was constructed on a large platform of bricks (at 0.87 m a.s.l.). The foundations of the halls, rooms and other structures of the baths were built within the framework created by the walls. The hydrological characteristics of the ground (high level of the water table) suggest the foundations rested on piles driven deeply into the ground. The fill of the N-S robber trench contained a large fragment from a heroic nude male statue, which was badly damaged by the stone quarrying. Various fragments and an almost intact, although heavily burnt, example show that granite columns stood in this part of the building.

This season’s excavations concentrated on the north-eastern sector of the Large Baths (Sector C9), in room S20, that dates to the Constantinian phase of the complex. The organisation of the room, furnished with pools, water supply system and probably fountains, is difficult to read due to the later overlying phases and definitive robbing, which in this zone seems to date to between the 15th and 16th centuries. Room S20 comprised a rectangular platform with a circular structure at the centre (diam. 8 m), interpretable as the bottom of a large pool. Other rectangular pools on a higher level were situated to the west, towards the North Hall. The overall dimensions of the platform were c. 13 m N-S x 16 or 20 m E-W. This year the southern limit was excavated and an attempt made to identify the eastern one, which probably lies beneath the floor of the adjacent room A14. Three courses of bricks bonded with mortar on a thick foundation of large reused stone fragments (including various columns in Cipollino marble) had survived the later robbing. The floors of the pools rested on thick dumps of clay-silt mixed with fragments of various materials. The clay also formed the lining of the N-S water channel that must have been of the type built of bricks with rectangular section and a flat covering. The pipe, which probably served for both water distribution to, and drainage from the pools, was out of use and completely robbed when, in the second phase of the baths, room S20 was obliterated by the floor of large tesserae in room A19 (see 2017 entry). The fill obliterating the channel contained brick fragments from the channel, parts of lead _fistula_ and numerous small re-cut polychrome marble slabs from the _opus_ _sectile_ revetment and numerous glass tesserae in green, blue and gold leaf, that may have decorated niches and pools with fountains. Most of these finds show traces of the effects of intense heat and suggest, together with numerous charcoal fragments, some of moderate size, that room S20 was destroyed by fire. The excavation of the N-S robber/foundation trench separating S20 from the North Hall to the west also continued. The robbing extended towards both rooms. Towards the North Hall, the brick wall of the structure containing the make-up for the mosaic floor of the large hall was robbed out. On the edge of the trench and in the fill there were numerous slabs and blocks, some large, that had been removed from the walls and cornices surrounding the mosaic panels in the North Hall, and large fragments of concrete, perhaps from the bases of the pools. The removal of the disturbed layers north of the building revealed substantial architectural remains, including a granite column damaged by fire, some small patches of mosaic from the North Hall and a part of the internal facing of its north wall, with the base of the marble slabs that clad the footing of the wall. Mosaics and wall revetments were already uncovered in the 1920s directed by Paola Lopreato and then back filled, but were never documented.

The investigations conducted on the site of the Great Baths of Aquileia by the University of Udine (2002-2014 in agreement; since 2016 under concession) identified limits and dimensions of the Constantinian building, geo-referenced the previous excavations (‘900), clarified the periodization of life in this area and contributed to knowledge of the building appearance, its foundations and some of its halls. The 2019 campaign brought to light the northern limit of the thermal building, in correspondence with the large room with pools and fountains (S20), and the intersection, underlined by a protruding corner in limestone blocks, with the walls of the adjoining North Hall. The walls are completely despoiled up to considerable depth. The emptying of the plundering trenches has made it possible to read the foundation system of the rooms inside the frame formed by the brick wall structures, confirming the hypotheses advanced in recent years. Also the foundations of rooms used bricks, alternating with thick layers of filler clay with waterproofing function and completed, under the mosaic pavement of the North Hall, by thick preparations in mortar, cocciopesto, layers with ceramic fragments and fine sand. Some scraps of mortar levels outside the north of the building were also investigated; they testify that thermae were built on a light elevation. Some geognostic surveys with continuous coring allowed us to read the stratigraphy of the foundation trenches and the subsoil in the northern area of the building; a deeper coring intercepted, in the natatio, one of the wooden poles of the sub-foundations, confirming the preliminary hypotheses. The exceptional groundwater rise impeded some parts of the final documentation from being perfected and forced to postpone some consolidation works. The excavation area was however made safe, closed and covered with the greatest possible care.

The geophysical surveys carried out in 2020 on the site of the Great Constantinian Baths had the aim of obtaining a preliminary knowledge of the subsoil in areas not yet excavated, in order to orient future research. An aero-photogrammetric survey using RPA extended to the entire area of the Constantinian building was integrated with a 3D GPR survey on the two sectors to the east (Area 1) and west (Area 2) of the frigidarium (3,152 and 3,677 m2 respectively). The 3D georadar system employed was equipped with 9 shielded antennas with a frequency of 400MHz and had an exploration depth within 3 m from the ground level. The survey was carried out by the company Esplora S.r.l., a spin-off of the University of Trieste (Roberta Zambrini, Davide Martinucci, Simone Pillon, Giulia Casagrande, Arianna Mocnik). The area to the east of the frigidarium shows many anomalies that are relatively well legible and attributable to surfaces (floors or preparations) or wall-trenches (of foundation / spoliation). All the anomalies have the orientation of the structures already brought to light and therefore they most likely belong to the thermal building. The most interesting novelty concerns the organization of the part to the east of the frigidarium, where the presumed rectangular natatio is perhaps to be replaced with a large semicircular pool (diameter: about 31-32 m), as in the “Terme Erculee” in Milan and in the “Kaiserthermen” in Trier. Even the hypothesis of the two courtyards on either side of the natatio is perhaps no longer valid, at least in the first phase of the building. In area 2, west of the apse of the caldarium, the anomalies are more confused, irregularly shaped, and relatively small, probably because here the stratification is much thicker, the suspensurae are broken through by large blocks of collapsed vault and probably the foundations of the heated rooms are immersed in the groundwater.