Baratti Gulf the undulating landscape of beautiful Tuscany Archaeology Archeodig

responsible archaeology

Poggio del Molino

Area B
Area B lies in the north-western part of the site of Poggio del Molino. It was partially excavated by the University of Florence in the 1980s, revealing what was interpreted to be bath complex, containing three rooms identified as a sudatio (sauna), a calidarium (hot bath), and a tepidarium (warm bath), flanked by drainage canal to the west, constructed at a later date.
During this season, we decided to complete the excavation, in order to better understand the precise workings and characteristics of this bath complex and its temporal placement in the chronology of the villa.
Numbers have been assigned to the rooms:

  • Number 1: the room to the North of the complex
  • Number 2: the heated room in the South-West
  • Number 3: the room in the South-East
  • Number 4: the entrance room in the West

The drainage canal: First phase (Figg. 3-5)
A drainage canal lies to the East of Area B and runs South-North (Fig. 3), reaching North Area H; it has a concrete bottom and two walls 40 cm deep made of stones bound with mortar (Fig. 5). The cover was made using both roof-tiles and stone slabs.
In excavating Room 3 of the bath, we discovered that the canal probably ran in a straight line, but was destroyed when the thermal complex was built, as the apse of the room covers its former path. This indicates that the canal is older than the bath, belonging to a previous phase of the villa (Fig. 4). We have yet to understand from where exactly the canals were draining.
A preliminary study of the relationships between the walls and of their building technique shows some of the walls of Area B are contemporary to the canal but the imposing building of the bath doesn't allow us to reconstruct both plan and uses of this phase.

The building of the bath (Figg. 1-12, 15-24)
At the end of the 2011 excavation we have been able to understand the full plan of the bath; our description will start from the entrance and follow the path the ancient bath-goers would have used. Since Rooms 1 and 3 have been almost fully excavated during the 1980s, our research focused on Rooms 2 and 4, where the original ancient stratigraphic sequence was still preserved. Inside both of these rooms the humus layer (Fig. 6) covered layers of collapse of the walls (Fig. 7) and of the roofs (Fig. 8); thin layers of soil, containing almost no materials, covered the mosaic floors (Figg. 9-10).
Room 4 (Figg. 8-16).
The entrance was through Room number 4, a corridor occupying the North-western part of the building; a marble threshold (Figg. 13-14), where the circular holes for the door jams are still visible, gave access to a rectangular room decorated by a simple black and white coloured mosaic floor (Figg. 11-12).
A second marble threshold (Figg. 15-16), located at the southern end of the room, gave access to Room 2. The shape of the upper surface of the threshold is very interesting; the part facing the corridor is higher than that facing Room 2, This shape was useful in preventing hot air from escaping Room 2 once the door jams were closed; Room 4 was thus a sort of antechamber of the bath, a space used to keep the warmer air inside the very hot Rooms 2, 3 and 1.
Room 2 (Figg. 17-20).
In 2009, a test pit in the south-west corner of the room revealed a small section of mosaic floor, made entirely of white tesserae. Further excavation revealed the room had a mosaic floor lying on a concrete preparation underneath (Figg. 17- 18); the mosaic, badly damaged by roots and human activity, it remained in situ in large pieces at the western and eastern ends of the room, allowing us to see that it had one constant pattern of decoration. Around the edges of the room is a border of plain white tiles (tesserae), which seem to be made of local marble (marble of Campiglia Marittima). The background of the inner part is of the same white stone while the design itself consists of blue lines in a criss-cross pattern; the lines are at regular distances from each other and form perfect squares, which are set diagonally to the walls of the room. At the intersections of the blue lines is a reddish purple rosette, made of broken pieces of small round stones. Room no. 2 yielded many fresco fragments belonging to the paintings on the walls. We have distinguished two different patterns, both containing criss-crossed lines to create diamonds, similar to the pattern on the mosaic floor:

  • criss-cross pattern with black lines on a white background. In the center of each diamond (the diamonds fill the entire width of the border), there is a small red mark. It is not clear whether it is simply a dot or whether it is perhaps some sort of shape. In one of the pieces the border, coloured in black and yellow, can still be seen.
  • criss-cross patten with yellow lines on a blue background. Occasionally, there are small red shapes (indistinguishable), in the centre of the diamonds. They are not found in every diamond, and we do not yet have enough information to discern the regularity of the red painted shapes. However, we have one fairly large piece of this fresco, which contains a red border in what appears to be the corner of the decorated area. Although yellow is not present in the mosaic, the red and blue used are quite similar to the colours of the stones themselves.

The greatest quantity of fresco pieces found in Room no. 2 was by far that of solid purple or maroon. The hue of these frescos is very close to the colour of the stones of the rosettes. In total, all of the pieces of the room seem to be linked in a single colour and design scheme, which may have recurred in different places throughout the room. As with many Roman fresco schemes, it is probable that there were different large blocks of colour and pattens throughout the room, which explains the diversity of fresco colours found in the room.
Toward the south side of the room there was a large hole, perfectly circular in shape, cut inside the concrete preparation for the mosaic (Fig. 19). The most likely hypothesis for the cut seems to be that some circular object, such as a basin or small pool, was affixed to the floor in this area. When the owners left the villa, they quite possibly tore this object out of the floor to take it with them, destroying the floor and mosaic in this area.
The presence of virgin rock right below the layer that filled the hole overturned the hypothesis of the 1980s excavation that this room was a calidarium, because the virgin rock is too close to the level of the floor, the floor was clearly not hollow. Rather, it seems likely that Room 2 was a changing room; since it contains the only entrances to Room 2 and 3 and its marble threshold is thought to separate the core of the bath from Room 4, that was just an entrance.
Along the west side of Room 2 and the adjacent Room 4 is a stripe of white concrete, which replaces part of the mosaic floors (Fig. 11, 20); the most likely possibility, at this stage of the study, is that the mosaic was removed to create a trough for a lead pipe for water, leading from the big cistern South of the thermal Area B as far as the bath of Area H.
Room 3 (Figg. 3-4, 18, 21).
A small door in the Eastern wall of Room 2 gave access to Room 3, mostly excavated during the 1980s, leaving only one layer of abandonment covering the structures; we fully excavated this last layer, uncovering the rest of the so-called hypocaustum, the bath heating system. A square opening inside the apse that closes the Eastern side of the room allowed hot air to enter the hollow space underneath the floor (Fig. 4); regular rows of small pillars made of square bricks (pilae; Fig. 21), used to hold the cocciopesto floor of the heated room have been uncovered, at the same hight of the opening (Figg. 3, 18). The excavation also revealed that the walls were completely solid, meaning that the hot air was not funnelled into another room after heating this room, as had been assumed in the 1980s. The layer that covered these structures yielded many fragments of tubuli (hollow bricks used to pump steam into the room from the hollow floor), originally affixed to the curved Eastern wall. Since the room was heated both by the hot air underneath the floor and by airflow through the tubuli inside the Eastern wall, we have confirmed the 1980s hypothesis that this room was probably used as a sudatio (sauna).
Room 1 (Figg. 22-23).
During the 2011 campaign this room, already fully excavated in the 80s, has was simply cleared of vegetation, soil and the tarpaulins the 1980's team put on structures and mosaic to protect them. A door frame, whose threshold is now lacking, gives access from Room no. 2 to Room no. 1; the excavation of the University of Florence revealed a mosaic floor, decorated with a geometric pattern. The Eastern part of the mosaic, made up of tiles (tesserae) of various colours, is devoid of decoration. The Western sector is instead characterised by the presence of a rectangular space, within which there is a roar (sic); inside, finally, there is a circle with decorative plants. This room was thought to be a tepidarium in the 1980s, but, like in Room no. 2, we have been able to prove it had no hypocaust heating system; thus, we must think about a different use. Since the Eastern sector of the mosaic has no decoration, furniture or beds may have been housed here, perhaps related to the massage or other spa uses.

The abandonment of the bath and the re-use of the drainage canal (Figg. 24)
Probably after the crisis of the late third century A. D., the bath is abandoned; the channel along the Eastern side of the building is reactivated thanks to the construction of a new section that is joined to the existing one North of the apse of Room no. 3 (Fig. 24). The bottom of the new canal, whose origin is unknown, is the virgin rock in what it was dug, while the side walls are built redeploying stones of various sizes taken from other structures. The construction of the new portion of the canal obviously implies that the pruefurnium (boiler room) of Room no. 3 is no longer in use and, presumably, the entire thermal system is no longer used as such. It is likely that during this phase is when the circular basin inside Room no. 2 (Fig. 19) is removed and in Room no. 4 a hole in the floor could have been made to remove the lead pipe that passed under the floors of this and of Room no. 2 (Fig. 10).

Area H
Investigations in Area H, started during the 2010 season, led to a considerable extension of the excavation, South, East and West; the main purposes of the 2011 campaign were:

  • continuation of the excavation of the north-east of the Area H, where two tanks for the producing of fish-sausces had been detected and partially excavated in 2010
  • continuation of the excavation of the praefurnium of Area H's bath

Area H: North-Eastem sector
The fish factory (Figg. 1-10)
During the 2010 campaign two twin square tanks (2.7x2.7m each) have been identified near the North-Eastern edge of the hill of Poggio del Molino (Figg. 7-8). The tank N. 1, the only one to have been fully excavated, revealed to have the same structural characteristics of basins used for the production of fish sauces currently known in the Roman world.
During the 2011 campaign the excavation was extended to the second tank, while to the east of these, four other tanks have been identified, all belonging to the same building phase.
A destruction layer of roof-tile (Fig. 1) indicates that tank no. 2, as well as tank no. 1, had a cover; the digging of the inner part of the tank revealed a series of layers containing a large quantity of pottery fragments, iron slags and building materials (Figg. 2-5), related to the use of the tanks as dumps after the end of the production of fish-sauce. They cover the bottom of the tank, made from waterproof mortar (Fig. 6).
In the East, adjacent to tanks nos. 1 & 2 is a third basin (no. 3; Fig. 9), built of the same construction technique but of different size (4x2x7m); further East we have been able to identify two more square tanks (nos. 4-5), similar to nos. 1-2. Since all of these basins share the same walls, we can assume they belong to the same factory; a sixth one, located East of nos. 4-5, was added later to the others, perhaps in order to increase the production (Fig. 10).
All the tanks we found are surrounded by a large wall (80cm) which was instrumental in fish processing, as it prevented the dispersal of the heat necessary for the fermentation of sauces in tubs.
The data collected so far indicate the presence of a fish factory capable of producing a significant amount of fish-sauces and active between the late Republican and early Imperial period; this activity is certain to have played a major role in the economy of the settlement of Poggio del Molino and confirms the existence in the territory of Populonia of a developed fish industry similar to that known in other regions of the Roman Empire, as the Iberian peninsula and Morocco.

The drainage canal (Figg. 11-12)
A major change occurs when a drainage canal with South-North pattern is constructed in the Eastern sector of Area H (the Southern part of the canal is inside Area B and was discovered during the 1980's excavation); starting North of the big cistern, it was probably used to collect dirty water. The construction of the canal involves the abandonment of the tank no. 3 (Fig. 9) and, perhaps, all of the fish factory's tanks. The channel, approximately 40 cm deep, consists of a concrete flat bottom and of two walls of stone and brick bound with mortar (Fig. 11); large hewn stones, not bound with mortar, have been used to cover the canal (Fig. 12).
We don't yet know when the canal was built; we can assume that it happened before the settlement of Poggio del Molino was transformed into a villa, because the apse of Room 3 of the bath building of Area B overlays the Southern part of the canal.

The phase of the villa (Figg. 13-18)
The construction of the villa, dating between the late second and early third century. A D., has led to a general reorganisation of the entire Area H, as well as all the other parts of the site; in the Eastern sector some of the major changes are the abandonment of the canal and the use of all the fish-sauce tubs as trash-pits. A new structure, almost certainly a portico, occupies most of the area; the roof was supported by a long wall with North-South orientation, parallel to the abandoned channel, and at least by two pillars, probably of wood (Fig. 13). Their bases consisted of two large square blocks (50x55x22cm) placed at a distance of about 2.5m from the wall (Figg. 14-15); one of them, to the North, has been placed inside tank no. 2, right above a layer of materials thrown into the tank (Fig. 15).
As for the functions of the portico the best hypothesis is that it would serve for the storing of tools (such as billows, shovels, etc.) and of wood used by the slaves in charge of the pruefurnium of the Area H bath.
During a phase not yet datable, the roof of the porch collapses and is not rebuilt; the excavation identified a significant layer of roof-tile collapse in the entire area between the wall of the porch pillars(Fig. 16).
The area was subsequently rearranged; two new walls, which are also oriented North-South, occupy the central sector of the area, not far from the pillars (Fig. 17). The two walls, built directly on the ruins of the collapsed roof, are interrupted by an opening 1.5m wide, equal to 5 Roman feet. A new floor of pressed clay, now covers the entire area.
We currently don't know why the portico was no longer rebuilt and the functions of the new walls and of the floor; the excavation allowed us to identify the collapses of all the structures (Fig. 18), whose destruction was probably contemporary to the crisis of the whole villa at the end of the third century A. D.

Late imperial Age ironworking (Eastermnost part of Area H; Figg. 19-20)
The Eastern most sector of Area H showed many traces of the re-establishment of iron-working in the settlement after the crisis phase of the late third century. A. D. Next to the south edge of the large wall that surrounds the tanks for the fish processing were in fact identified a series of layers consisting only of iron slag; they covered irregularly shaped holes (Fig. 19) and flat plans of vitrified iron (Fig. 20) that can be identified with remains of furnaces. This phase, which is earlier than the abandonment of Area H, can be dated to the fourth century A. D.

Area H: the praefumium of the bath (Figg. 21-23)
During the 2010 season, we found an intact praefumium, connected by an arch to a room with suspensurae (Room 1), which is completely disparate from the bath complex in Area B. Most of the rest of the bath complex in Area H has collapsed off the cliff, and thus is unknown. In order to answer the newly created questions pertaining to the bath complexes (for example the re-use of an inscription in one of the walls), we decided to continue the excavation of the praefurnium.
The southern sector of Area H was then extended to the west and south until it joins with the peristyle (Area G) and the thermal complex of Area B; it was possible to dig the western half of the space where the slaves prepared the piles of wood used for the heating of Room 1 (excavated in 2010).
An extensive layer of the collapse (Fig. 21) covered the walls that divided the working area of the praefumium from the peristyle and the bath of Area B (Figg. 22-23); in the Western sector of this space, the structure that enclosed the area where the pile was made is still well preserved. The excavation reached the level of virgin rock, on which are clearly visible signs of metal tools used during the construction of structures of praefimaium.