The castrum de Montecastresi stands above Camaiore in a strategic position controlling the territory and roads. The castrum is mentioned in two archive documents dated 1219 and 1231, but is perhaps to be linked back to the loco ubi dicitur Castro mentioned in a document of 950. The fortified settlement developed in an elongated form, on a north-east/south-west alignment, covering a surface area of circa 25,000 m2 and surrounded by a wall circa 1 km long. The summit was occupied by the keep, characterised to the east by a square tower and to the west by a subsidiary tower. Circa one hundred dwellings (some around the keep) and the church of S. Barbara were present on the site. In 2007-2008 an intensive survey and excavations were undertaken on the north-eastern part of the keep. The survey revealed the presence of a ditch, 9 x 5.4 m, to the south-west of the summit, on which stood an internal curtain wall on a north-east/south-west alignment. This wall surrounded the base of a quadrangular tower, of which only the imposing demolished remains were left. The survey also revealed an external curtain wall which completely enclosed the settlement’s upper part including the western tower. The stratigraphic excavation looked at an “L” shaped area situated along the north-west and north-eastern sides of the quadrangular tower. The internal curtain wall was revealed to be a quadrangular structure of circa 11 m built around the tower, with a double access on the north-eastern side which closed a long narrow space (8 x 1,10 m). On a north-west/south-east alignment this was interpreted as the guardroom. The stratigraphy clearly showed how the curtain wall was built at the same time as the quadrangular tower. The limestone bedrock was artificially levelled and the tower’s base constructed with limestone ashlar blocks bonded with mortar. The tower, with a quadrangular base of circa 5,40 m each side, was built in a single construction phase by specialist artisans, whilst the curtain wall was built by non-specialists, or with the waste materials from the main building. The pottery, currently being studied, and the analysis of the walls date the construction to the 12th century. The absence of housings for anchoring the wooden floors, of openings (embrasures, slits for light) and the presence of an entrance on an upper level, suggest that this was simply a watch tower. Its use would thus have been limited to the guards floor at the top, which was reached via a timber ladder, probably retractable, positioned, at least to a certain height, on the south-western side of the building. There was a cistern in the basement of the structure. The structural remains of the tower (base and rubble remains) showed signs that the building had been deliberately demolished at the beginning of the 13th century. Subsequently the area remained abandoned until the end of the 15th-beginning of the 16th century when it was converted to agricultural use through the construction of terracing for olive cultivation. The discovery of a stretch of wall abutted by the guardroom and of early medieval artefacts including a bone pendant decorated with dice eyes, indicate early medieval occupation of the site. This will be the object of the next excavation campaign.

The 2009 campaign completed the excavation of the stratigraphic sequence in area 2000 (begun in 2008) and opened two new sectors within area 3000 (area to the east of the collapsed tower), respectively denominated 3000 and 3500. Area 2000 revealed a first occupation level, datable to the 10th century, characterised by structures in perishable materials. These included traces of a hut with clay walls on a stone footing, a roof of perishable materials held up by central timber supports and a hearth. The first stone construction, of very roughly worked split stones bonded with mortar, datable to the 9th century, was built on top of the collapsed hut. Only the southern corner of this structure was preserved which belonged to a defensive structure, perhaps a tower. Occupation evidence comprised a simple beaten earth floor and a hole, perhaps a small silo. That the structure went out of use was shown by the presence of a large pit used for lime making in relation to the building of a new square tower in the 12th century (2008 excavation). In fact the pit was obliterated by waste materials from stone and metal working. The construction of the gatehouse (2008) began with the obliteration of the lime kiln. The excavation within sector 3000 revealed the presence of a post-medieval agricultural phase characterised by the presence of containing terraces created with reused materials from the tower collapse. This phase differs from that present in area 2000 which began in the late 15th-early 16th century in that it began in the 19th century. The tower collapse was found below this phase. Its orientation and stratigraphic position revealed the dynamics of the collapse itself, which occurred at the same time as the tower was deliberately blown up using mines, taking down the walls of the gatehouse with it. Over time the long exposure of the collapse and the sloping terrain caused the materials to wash away, a process only interrupted by the agricultural phase. Below the collapse was a parallel cut in the bedrock thought to relate to a passage, perhaps a footpath, the end of which was identified within sector 3500. Excavation of sector 3500 also revealed an agricultural phase constituted by 19th century terracing, overlying the washed out material from the collapse. Below the collapse was the terminal part of the pathway identified in sector 3000 and a patch of cobbled surface, which may attest the existence of an open space in this part of the site. Traces of a wall bonded with mortar pre-dated the cobbled surface. The wall was part of an earlier curtain wall of the “cassero” perhaps linked to the castle’s first stone built phase identified in area 2000. The south-western sector revealed a series of rock-cut post holes and traces of a hearth which a preliminary hypothesis suggests area evidence for an early medieval building in perishable materials.

The third excavation campaign saw the completion of excavations in sector 3500, the opening of sector 2500, by the east tower, area 5000 situated in the central part of the keep, and area 7000 in the western part of the _borgo_. Traces of early medieval occupation were attested in sector 2500 by five cuts in the bedrock and the construction phases, respectively 10th and 12th century, of the buildings pre-dating the tower, and of the curtain wall surrounding it. Sector 3500 revealed evidence of 3rd century B.C. occupation relating to the presence of a Ligurian settlement. A probable early medieval phase was attested by the presence of a wall made of perishable material, perhaps a hut, associated with a hearth. In this area, the curtain wall around the keep was seen to be constructed directly on the levelling of a rocky slope. Area 5000 comprised three pairs of rooms, arranged on two different levels and facing north. They were separated by alleyways, running parallel to the buildings, built directly on the bedrock, which from the main path on the ridge provided access to peripheral structures. The excavation of the eastern pair of rooms showed they were probably built between the 9th and 12th century and abandoned in the first half of the 13th century. The upper room was constructed by levelling the rocky profile of the hillside and partially exploiting it for the walls. The interior comprised a single space with the bedrock forming the floor surface and a hearth in the central-eastern sector, in correspondence with a vertical cut in the rock in the room’s eastern corner, interpreted as a chimney. Only the eastern half of the room on the lower level was excavated. It was on the same alignment as the upper room with which it shared its long wall. A semi-cylindrical storeroom/larder opened in this wall. The post-medieval agricultural levels overlay the collapsed walls and, partially the collapsed roofing made up of schist slates, robbed prior to its complete collapse. The walls of this room also abutted the rock, while the floor surface had been created by levelling the bedrock with dumps of the waste materials from stone working and the smoothing of the rock surface. Excavations in sector 7000 revealed the presence of two overlying, partially rock-cut rooms. The interior of the lower room was filled by the collapse of the walls and underlying roof collapse. The materials collected in the surface layers were richer both in quantity and variety with respect to what was seen in area 5000. Their dating showed a continuity of occupation well into the 13th century, suggesting greater continuity of use in the lower part of the site than on the summit.