K. Palágyi Sylvia szerk.: Balácai Közlemények 2005/9. (Veszprém, 2005)

T. LÁNG, ORSOLYA: Control Excavations in the Territory of the Civil Town of Aquincum: the so-called „Testvérhegy-villa"

contact with water (basin? cistern?). Huge thresholds stones and roof debris were also documented here. Summing up the results of the test excavations, our research revealed „new" walls and an internal roadway but missed two wall-segments, that S. Garády once found. Althoguh the resaerch is not yet complete and the find-material has not yet been studied in detail, these preliminary results may already alter the ground-plan of the complex and raise the question of whether we should deal with not three separate, but one large building-komplex. This last hypothosis is further demonstrated by the results of the geophysical survey carried out over the slope, which in addition to other wall-segments reveal the presence of an exedra as well (Fig. 9).

9 The Entrance-Unit The better part of the modem housing-estate was intended to occupy the flat area at the foot of the Testvérhegy. Thus, more than 8000 m

2 area needed to be excavated to investigate any possible Roman remains which came to light there. In addition to the road segments (see below) the so-called entrance-unit came to light in one of the trenches. This unit consisted of fencing walls, a gate, an approach road and an out building (Fig. 10). All these features were probably part of the building complex on the slope and thus, form a complete settlement-unit. As shown by the layers and the geology in this last trench this area sloped gradually towards the east. Remains were better preserved in the western part of the trench, while the eastern segment was badly damaged by modem levelling activity. In addition, eroded earth often came down from the slope, requiring continuous renovation of buildings, roads etc. in this area. The east-west oriented stone out-building came to light in the northwestern part of the plot. The construction must have extended further to the west, but as it was beyond the area of the future house, it was not possible to follow it. The building had at least 6 rooms and we could distinguish four periods, separated by thick erosion layers. 1. The earliest phase was represented by the impressions of a wooden construction under the eastermost rooms of the later stone building (Fig. 11). Although a coherent plan could not be observed, these impressions suggest that this may have been some kind of large timber building (horreuml). Fragments of Roman coarse ware were found in the impression which cannot be precisely dated yet. Although the date of the construction is yet unknown, it seems certain, that the building was intentionally abandonned since there were no traces of burning or destruction. Rooms from the following phase were constructed directly above bove the older wooden building. 2. The next phase was represented by a large, open-air courtyard, which was at least 15 meters long (Fig. 12). This was shown by the fact that its walls continued to the west in the nearby ditches created for modem canalisation. The floor of the court was covered with large limestone slabs. Below this floor, an east-west oriented channel