Investigations in the hamlet of Trebbio began in 1990, when a series of surface surveys began which revealed pottery, metal objects and faunal remains datable to between the end of the 9th and the 7th century B.C. In 2002 excavations began and the first campaign brought to light the remains of a production area characterized by several structures including 3 well preserved semi-interred kilns, two of which were for pottery production. Both were of the same typology with _praefurnio_ and round combustion chamber with plastered walls. Several pits were found which probably relate to the activities undertaken around the kilns. On the western side of the excavation an inhumation with two skeletons was found; it contained no grave goods. The _terminus post quem_ indicated by the Roman materials in the fill, the grave typology and radiometric tests place the burial in the early medieval period. In 2004 a second burial was found. This was similar to the first in orientation and the absence of grave goods but contained a single deposition and constitutes evidence for the presence of a necropolis in the area. The probable existence of a 6th-8th century B.C. necropolis at Trebbio is of particular importance, given the vicinity of this locality to the historic centre of Sansepolcro In 2004 and 2005 numerous pits used for domestic rubbish or connected to craft working activities and post holes were found. The recovery of a large quantity of pottery in 2006 documents the settlement’s trading contacts with areas such as the Padovan Plain, Umbria, _Latium_ and the Piceno area between the 9th and 6th century B.C. (a period indicated by the presence of a footed cup with two handles).

The sixth excavation campaign was undertaken during July-August 2007 on the site of Trebbio (Sansepolcro-AR), under the direction of the Dept. of History “E. Lepore” of the University of Naples “Federico II” and by the Research Unit on Prehistoric Ecology, Dept. of Environmental Sciences “G. Sarfatti” of Siena University, in collaboration with the Archaeological Superintendency of Tuscany. The investigations were also carried out with the assistance of the Archaeological Group of Sansepolcro (GRAS) and the Study Centre for the Quaternary Period (CeSQ). Students, postgraduates and researchers from the Universities of Siena, Naples and Rome, as well as school students from the province of Arezzo were also involved. The settlement of Trebbio is situated on the alluvial plane on the left bank of the Tiber, at 3 km S-SW of Sansepolcro, where the river Afra meets the Tiber (300 m a.s.l.). The latest excavation campaign looked at the area partially investigated in 2005-2006. The latter was enlarged in order to complete the excavation of a surface with a dense concentration of cobbles and pottery fragments, already denominated “cobbled surface” or “structure 2”. This led to the uncovering of the remains succeeding “structure 2” and the identification of several levels in phase with the latter. Excavation continued in the central sector of the area on the large pit denominated “structure 1” (now known as US 393), situated a few metres south of “structure 2”. The pit contained animal bones, including what is probably a dog’s skull, large quantities of baked-clay and numerous fragments of impasto pottery, both with coarse fabric and a finer yellowish-orange type. On the bottom of the pit was half of a cup with a curving profile and a horizontal handle which, together with all the other impasto material recovered in US 379, suggests a 6th century B.C. date. North of pit 393 an unusual feature was excavated, the remains of a small wooden structure, that is a row of small burnt posts placed in inside a small channel on a N-S alignment (US 402-401). The structure seemed to incorporate, or intentionally respect, a grey sandstone sub-rectangular cippus (US 412), which presented clear traces of hammer work on its surfaces. South of the stone an iron knife with curved blade was found, an artifact which could convalidate a ritual interpretation of this structure. This site is of great importance for the earliest history of the Etruscan-Italic populations. In the first place its great antiquity should be taken into consideration – dating to at least between the 8th-6th century B.C. – as should its position next to the Tiber, known as an important boundary dividing the territory of the Etruscans from that of the Italic populations of Umbrian-Sabellian descent. It was also an important crossroads and commercial centre. The settlement’s great size is an important new addition to the panorama of knowledge regarding the Italic civilizations. On the basis of surface survey it seems to occupy an area of 35-40 hectares, a dimension much larger than is the norm for Italic centres.

Excavations recommenced in two sectors of area VI: 1) the zone of the “cobbled paving” UUSS 375 and 413 and of wall Us 400 (called NW sector); 2) the zone of the ditch US 393 and the channel US 402 (called SE sector). _The NW sector_ An in depth exploration of part of the area showed that US 375 constituted the uppermost level of the fill of a sub-rectangular cavity (US 420) cut into the natural, whose limits have only partly been identified. This fill was formed by the repeated dumping/setting of large quantities of pottery materials, cobbles and animal bones alternating with a matrix rich in organic components. The lowest levels were characterised by the presence of large fragments of pottery, sometimes belonging to partially reconstructable vases. Of note, a very good example of an impasto pyxis lid, with incised and cut decoration, dating, as perhaps much of the material from the fill, to the full 7th century B.C. Excavation of the foundation of wall US 400, on a N-S alignment and parallel to ditch US 420, but having no direct straigraphical link with it, showed that the foundation was constituted by a narrow trench, filled with cobbles, below which a pre-existing post-hole emerged. Moreover, it was suggested that US 400 possibly related to an analogous alignment of small cobbles, US 421, which filled a narrow cut made in US 409. The US 400 and 421 may constitute the remains of a structure that post dated ditch US 420. Excavation continued of the second “cobbled surface”, US 413. This formed a strip on an east-west alignment, constituted by a dense concentration of pottery fragments of various sizes and cobbles, mainly small, arranged on the level to form a sort of compact platform. The edges appeared to be regular, although not straight as they had been disturbed by robbing presumably relating to Roman agricultural activities. The removal of the first level, characterised by a greater density of pottery, produced fragments of biconical impasto vases and artefacts destined for domestic use (loom weights, reels, jars etc.), datable to within the 7th century B.C. The continuation of the excavation of fill (US 406) in the ditch adjacent to US 409, containing numerous dumps of baked clay fragments and pottery fragments, showed that it was cut by four large post holes linked by a channel. The latter joined at a right angle a further four post holes, much smaller in size, on a north-east/south-west alignment. The materials associated with the post holes date to the first half of the 6th century B.C. _The SE sector_ In this zone excavation was completed of the large ditch known as “structure 1” (now US 393), situated several metres south of the “cobbled zone” (structure 2). The southern half of the thick fill of the ditch was completely removed. The fill was very rich in charcoal remains, bone, fragments of baked clay and pottery. At its base numerous large fragments of pottery (cup, trumpet-shaped foot, jars) and a ring made of a strip of bronze lamina came to light, partially embedded in the clay layer below (US 416). The US 416 was gradually filled with dumps of burnt and over-fired material which may be associated with pottery production. The cut of ditch US 393 may be linked to the quarrying of the natural clay. Experiments have shown this clay, containing silt and a small percentage of sand, to be particularly adapt for pottery making. A preliminary examination of the abundant material from the ditch confirmed the dating to the late Iron Age (first half of the 6th century B.C.).

The eighth excavation campaign on the site of Trebbio (Sansepolcro - AR) took place in July-August 2009. The investigation concentrated on the area opened and partially excavated during the previous years, denominated area VI. The main excavation worked on the area of cobbled surface US 409. A second and deeper level was exposed characterised by larger cobbles and concentrations of large fragments of terracotta and faunal remains that were even denser than in level I. It is possible to interpret this as the result of levelling of the collapsed material from a number of structures, as attested by the find of a strip of cobbles on a north-south alignment on its eastern edge, which may be interpreted as the remains of a foundation. To the south-east of the cobbled area excavation of US 413 continued. This exposed a dense make up constituted mainly by cobblestones of various sizes, terracotta fragments and animal bones. It was on an east-west alignment, its width varying between 80-90 cm and 1.10m, and the preserved section reached a length of circa 6.50 m. The excavation revealed the presence of two layers of material, with large terracotta fragments often showing clear signs of exposure to the weather. The Construction method, examined by the excavation of a trench, suggests that this was a sort of modest road, perhaps a pedestrian pathway, that was periodically remade. In proximity to UUSS 409 and 413 a pit appeared which contained materials that seem to be of the same date as the US mentioned (full or early 7th century B.C.). Therefore, it is possible that the succession of actions of accumulation and removal of material constitute a single cycle of activities over a relatively short time span. The excavations demonstrated that the area of Spinellina, throughout the chronological arc of its occupation, saw intense construction activity which, in part, involved the rebuilding of the structures on the same alignment, although sometimes using different techniques. On the contrary, part of this activity involved the obliteration of pre-existing structures with the subsequent change in use of the zone involved. Such dynamics were particularly evident when the wall foundation US 400 was dismantled revealing a pre-existing construction phase below it constituted by rows of small post holes. In another zone it was possible to reconstruct a structural alignment constituted by large posts (including those uncovered in 2008), later removed, the holes from which were below the dumped remains of burnt habitations. This building activity appeared to have two phases: the first dating to between the late 8th century and the 7th century B.C., the second – in part comprising rebuilding and obliteration of pre-existing structures – dating to the 6th century B.C. Thus, at least for the period between the end of the 8th century and the mid 6th century B.C., the intense and prolonged character of activities in this area of the site of Trebbio was confirmed and to which evidence for the building of infrastructures (possible road) also corresponded.

During the campaign which took place in July-August 2010 work continued in area IV which had already been investigated between 2005 and 2009. Trial trenches were also dug in the locality of Olmo di Costolino. This year the excavation at Spinellina aimed to clarify the dynamics of the stratigraphic sequence, that had already been uncovered, and the relationships between the layers. The removal of the deep layers belonging to the dense band of materials denominated US 413, datable to the mid 7th century B.C., was completed. In particular, a dense sequence constituted by a succession of horizontal levels of detritus, pottery and abundant charcoal, intentionally dumped to fill a cut, or foundation trench was excavated. The latter, denominated US 530, had a sub-rectangular section and flat bottom. On average it was one metre wide and at the best preserved point was about 0.25 m deep. The interpretation of US530 is uncertain. On the basis of the data from 2010 it seems likely that these layers formed the make-up, also providing drainage, for a large structure (ascertained length about 6.50 m) which has been completely obliterated by ploughing. The robber trench US539 is also datable to the early-mid 7th century B.C. Excavation of the fill, containing abundant pottery, was completed. This trench was situated at the centre between various structures denominated US413-530 and 409-555 and partially cut them. It was also seen that the fill of US539 had subsequently been cut by another ditch, filled with alternating dumps of grey earth with abundant charcoal and concentrations of medium to large sized cobbles (US 592 and US594). The latter is probably datable to the end of the 7th-first half of the 6th century B.C. The narrow curving cut US571 can be interpreted as a robber trench filled with rubble from disused structures. The fill contained large fragments of baked clay and cobbles (US570, 583). From a stratigraphic point of view it seemed to coincide with the first phase of the cycle of activities removing and accumulating materials, which concluded with the creation of the cobbled surfaces US 375-409, investigated in 2005-2009. In the eastern part of the excavation area work continued on the negative stratigraphy and the post holes associated with timber structures, some of which later than the 7th century structures. One of these structures was constituted by rows of small post holes (US510) which formed a corner, or more probably a curve, to the west. Another post hole (US581-582 was a continuation of this structure. Other cavities relating to timber structures were found further to the south (US 573, 575). In the locality of Olmo di Costolino, situated north-east of the main concentrations of finds relating to the proto-historic settlement of Trebbio, a number of trenches were dug with the aid of a mechanical digger with the aim of defining the extension of the ancient settlement. The presence of the remains of rural structures of Roman date was ascertained (drainage channels, well, cobble-built walls), some of which overlying alluvial deposits relating to a probable pre/proto-historic expanse of water. This would seem to confirm the hypothesis that this area, although presenting traces of occupation in the same period as the occupation of Trebbio, was situated outside of the settlement and partially formed its north-eastern border.

During July, excavations were undertaken on the Iron Age site of Trebbio di Sansepolcro, with the aim of collecting data on the topography of the settlement’s western sector. With the help of a bulldozer, 10 areas were opened (trenches and test pits) in the locality of Commenda (in a field in lot 77 of the cadastral map). The trenches, circa 1.60 m wide (some were widened), revealed archaeological evidence at an average depth of 40-50 cm below ground level. In some cases it was necessary to go down to a greater depth before the remains appeared. The excavations showed that most of the proto-historic/archaic remains were concentrated in the central and south-eastern parts of the lot. Trenches 1 and 2 in the central-eastern sector of the field produced the highest concentration of archaeological remains. Several small cavities filled with dark charcoally soil were exposed in the south-eastern part of trench 1. In US 1 this soil alternated with dumps of large fragments of baked clay, charcoal and pottery sherds. Nearby there was a small hole (US 2) filled by an ashy layer containing occasional pottery fragments, among which a sherd from an orientalizzing pyxis. In trench 2 the archaeological remains, situated further towards the north, were more scattered. Towards the centre of the field, an extension to the trench identified a wide strip of terrain showing evidence of occupation, US 10, rich in charcoal and heavily disturbed by recent agricultural activities. A fragment of a thin-walled black bucchero cup was collected from its surface (end of the 7th- beginning of the 6th century B.C.). The archaeological finds dating to the proto-historic/archaic period thinned out until completely disappearing in the trenches dug in the north-eastern and north-western zones of the field. In trenches 4 and 6, at the furthest southern edge of the field, the presence of very extensive sandy alluvial deposits was noted, which presumably corresponded to the ancient bed of the river Tiber. These deposits were cut by a large channel or hollow, whose upper fill, US 14, dated to the Roman period. The lowest, blackish clayey levels (US 17), reached in a trench with a maximum depth of 2.70 m, can be dated to the period of the settlement’s greatest development (7th-6th century B.C.). These deep levels filled a large cut in the natural, of presumably trapezoidal shape, of which an edge was identified in plan and in the exposed section. The structure may be interpreted as a large channel or feature forming a boundary to the settlement (ditch?), whose topographical and functional characteristics have yet to be defined. The excavations documented that the western limits of the settlement of Trebbio coincide with the central zone of the field, to the west of which evidence of the settlement rapidly thins out, then disappears completely. This seems to confirm considerations made on the basis of the results from field survey and previous excavations. Furthermore, a structure presumed to be a settlement boundary was also identified, and will be investigated in coming campaigns.

During the 2012 excavations at Trebbio, work was concentrated in the locality of Commenda with the aim of clarifying the characteristics and chronology of the boundary and/or channel structure identified in 2011. Two trenches were opened with a mechanical digger, in one of which the line of the channel was picked up again. The trench was over 12 m long and almost 4 m deep and exposed the channel’s complete width. Based on what was visible along the sections and on the bottom of the trench it was seen that: - The channel was dug into the sterile alluvial deposits characteristic of the Tiber plain, which are constituted by alternating layers of more or less coarse sands and altered silts; - It presented a residual depth of about 4 m, was on a north-east/south-west alignment and was 3 m wide at the base and about 12 m wide at the top. The fill was of particular interest, constituted by blackish silt containing scarce fragmentary materials, except for the first 40 cm of plough soil. The lower part of the fill contained fragments of Iron Age pottery that was in phase with the settlement of Trebbio, while the upper part produced fragments of tegulae and imbrices and a Roman coin. This data indicates that, if as seems likely the channel was created during the Iron Age, it had still not been completely obliterated in the Roman period. The nature and function of this imposing structure are still to be identified. Above all, its relationship with the settlement of Trebbio remains to be determined and therefore whether or not this was a defensive ditch running around it.

As part of the project to identify the limits of the Iron Age settlement at Trebbio, a trench was opened in the area of lot n. 142, map 62. The trench, 20 m long and 1.60 m wide, was placed in the southern part of the lot, starting from the border with lot n. 30 and reached an average depth of 100 cm below present ground level. In the north-western section, a patch of an occupation level was exposed at a depth of only 38 cm, just below the plough soil. This lens of material, probably Iron Age, contained earth mixed with charcoal and occasional minute pottery fragments and in section was a maximum of 15 cm wide. The dark colouring terminated abruptly, giving way to a yellowish, silty, sterile soil. This sterile layer continued in the trench for a further 18 cm. In the final two metres was found a level with brick/tile fragments appeared in section at a depth of 35 cm, practically in contact with the plough soil. Among these fragments was part of a hemispherical cup with a cordon, Goudineau type 38/form 34 in the Conspectus (fig. 4). In order to avoid further damage to the Roman level excavation was halted.