Clickable image-map of geophysics results on the villa, with excavated walls etc. coloured in

Zone II A

This trench exposed the whole of the north wing of the water channel which ran round four sides of the central courtyard. During earlier examination, this stretch was found to contain vessels, set into its north and south walls, and became known as the 'Fish pond'. Approximately one third of this northern wing had been exposed in previous years, and this year the remaining portion was excavated for two main reasons. First to obtain a full plan and elevations of a structure unique in the area and very rare for central Italy as a whole (away from the coastal area only four other examples are known) and second to see whether any evidence survived of the hydraulic system which provided water for the structure and the rest of the channel around the courtyard. When fully exposed the structure measured almost fifteen meters in length, and the regular brick construction with the inserted vessels continued for just over twelve meters in the central section of this northern wing. A full photographic and graphic record was made of the walls and floor with detailed measurements of all the vessels. The southern wall was uniform in build, but on the northern wall, at the far east and west ends, the build was of irregularly set reused tile and mortar, similar to the build of the channel on the east and west sides of the courtyard. The 'Fishpond' wing was clearly a separate build, and may therefore have differed from the main build on the other three sides of the courtyard in its date of construction. There was no substantial difference in the nature of the fill removed (mainly earth, tile and pottery) in this section, and the pottery found this year confirmed the date of abandonment of the structure in the early-mid third century which had been suggested by material found in 1998 and 1999. No evidence was found for any element of the hydraulic system, which suggests that if it existed on this northern side of the water channel, then it must have been at a higher level than is now preserved. (The highest preserved portions found were at the corner of the north and east sides of the channel, and stood To a depth of over 75 cm. (to map)

Zone II B

This trench revealed a group of three rooms in the north east of the villa. The most completely excavated of these, roughly square in form, had a crude cocciopesto floor, over which was found an abandonment deposit containing a coarse pottery of the late first century AD. The construction of this room cut through an earlier surface of crude cocciopesto. The adjacent rooms lacked solid floor levels, but both produced first century AD material from their make up levels. The lifespan of the floored room and the other two rooms seems therefore to have been very short. (to map)

Zone II C

This area comprised the south west corner of the water channel, and was opened up primarily to ascertain whether an outflow was located here, but also to gain some evidence of the nature of the structure of the south side of the channel. No traces of a drainage system were found, nor were there any vessels set into the walls, suggesting that the fishpond element was confined entirely to the north side of the courtyard. Material found in this area confirmed the early - mid third century abandonment date suggested by the fill of the northern wing. (to map)

Zone II D

This trench investigated the southern part of the suite of rooms at the west of the villa, partly to gain confirmation of the apsidal nature of the south wall suggested by the geophysics, and also to see whether there were traces of habitation or other activity in the area. The south wall was indeed apsidal, and of a sub reticulate build identical to the stretch of the south wall to the east of the main staircase excavated in 1998. It seems likely that certainly this main south wall, and probably all the main perimeter walls of the villa acted as a main terracing and retaining wall for the large platform of clay and gravel fills on which the vila was constructed. The other walls located in zone II D were of a very different and much less substantial build, and the areas between them and the apsidal wall were completely sterile, mirroring the situation in much of the rest of the western suite of rooms of the villa. (to map)

Zone II E

This area in the north western part of the villa was investigated as a result of discoveries in the adjacent area in 1999. A long narrow room had been discovered with a water proof floor, with a drainage hole cut into it, partly lined with a rounded roofing tile (imbrex) leading towards zone II E. Although it was therefore assumed that the area would contain a cistern or vat of some kind, no such structure was found. Instead the area was crossed by a substantial concrete lined water channel, leading through a purposely cut hole in the wall between the two areas, from the opening of the imbrex drain and leading towards the main west wall of the villa. The fill of the drain produced a coin of Nero and pottery dating to the second half of the first century AD. This fits very well with the chronology of the adjacent room, where the floor was covered with a thick deposit of pottery and lamps dating to around the third quarter of the first century AD. (to map)

Zone III

Near the remains of the mausoleum in the south western part of the field in which the villa is situated, a trench was excavated in the location of the highest concentration of material ( pottery, glass, tesserae) found during the surface survey in 1997. Excavation revealed a series of walls, one curved and clearly respecting the mausoleum, which seemed to form a funerary precinct comprising a gravel courtyard and a large solid structure which curved around the mausoleum. Next to the wall of the precinct, and on approximately the same alignment, was a square structure, behind which was a large midden of pottery, and bone waste. This waste was of a very particular sort, composed almost entirely of the ends of long bones of large animals (cow/ox??) neatly sawn off. This suggests that bone working, rather than butchery was responsible, and confirmation for this came from the square structure itself. Here, below a layer of the same cut bone, were found many hundreds of fragments of worked bone artefacts. These sometimes preserved diagnostic features such as rims, turned grooves etc. Most of the artefacts had originally been cylindrical, matching the types of bone waste found. In the mixed deposit which filled the room there were also the remains of many unguentaria (at least 20 whole and 30 fragmentary examples), drinking vessels, flagons and Black Glaze dishes. There were also a number of iron objects, including hooks, punches and blades of various sorts and three strigils. The unguentaria and Black Glaze vessels suggest dating of the later second/early first centuries BC for their deposition. The pottery from the midden behind the structure dated to the Augustan/Tiberian period, but interestingly there were several vessels in the midden including unguentaria and drinking cups which were identical to those republican vessels from inside the structure. A half as of Janus/prow type dating to the second century BC was also found in the midden, suggesting that at least some of the contents of the structure almost certainly identifiable as a mausoleum, were emptied out behind the structure before or during its reuse in the early empire. The apparent reuse of a tomb for bone working, or indeed perhaps just for dumping waste from bone working, is extremely strange, given the respect for tombs and mausolea in the Republic and early-mid Empire. (to map)

Zone IV

Situated in the south east corner of the villa field, to the west of Santa Maria, this trench investigated part of the large dark anomaly which was visible in the geophysics results of 1997. Initial cleaning revealed the reason for the anomaly, since much of the area was covered in a mass of rubble and tile. Numerous walls traversed the area, and the nature of the material found including flue tiles, marble veneer and tesserae, immediately suggested the presence of bath buildings. This was confirmed when excavations were conducted in a small rectangular structure in the west of the trench. The structure (almost certainly a bath or pool) was orientated north south. The walls were brick faced and lined with flue tiles covered by plaster/mortar, upon which the remains of a Proconnesian marble veneer were visible. Several phases were represented s shown by the presence at one of the short ends of the structure of at least two phases of mosaic, each subsequently cut or removed by later construction. The main floor of the structure as found was not in mosaic, but the top of the thick make make up levels bore the imprint of flat slabs, presumably of marble. The whole of the floor had collapsed into the hypocaust system beneath - several chunks of flooring, all at different heights were found. Two pilae (floor supports) were constructed of brick and stone, and these rested on a floor of large rectangular tiles. Over much of this floor and against the pilae was a deep deposit of ash, over which was a layer of greenish clay, almost certainly waterborne, and dating to a period when bathing in the structure had ceased but when water was still flowing into the area. To the east of the bath was an area which contained a high concentration of dark, ashy soil, and it is likely that this was the area of the praefurnium. The proximity of this area and the presence of flue tiles and hypocaust suggest that the excavated bath formed part of the Caldarium, the hot room of the complex. The other rooms presumably extended west and north of the main axis of the dark anomaly found during geophysical survey. (to map)

The history of the baths is difficult to reconstruct with certainty as only a small area was excavated, and this area was very disturbed. Very obvious robbing had occurred along the wall at the south end of the bath, the thickness of which suggests it may have been one of the main walls of the complex, where veneer and flooring has been stripped out. The main rubble fill of the structure continued sherds of amphorae and domestic pottery dating to the second or third centuries, while a coin of Commodus (AD 180-192) came from the very bottom of the robbing layers against the south wall. Interestingly the uppermost part of the rubble fill produced a sherd of pottery dated by Helen Patterson to the time of Charlemagne (late 8th century). Evidence so far therefore suggests that the baths ceased to function as such in the third century, and were thereafter used as a quarry for building materials. (to map)