Excavations carried out during the restoration of the theatre, in 2002, revealed the existence of a large building aligned on an ample terrace to the west of the monumental temple-theatre complex. Excavations in the following years brought to light a large atrium house contemporary with the temple (end 2nd century BC). The building, directly linked to the area of the sanctuary and entered by an open passage in the precinct wall, has the distinctive feature of having at the back, in place of the peristyle, a great portico with two naves containing the bases of ex-voto, altars, votive objects and a dedication in Oscan. The internal nave was in part occupied by closed rooms: a _sacellum_, a thesaurus and kitchens for ritual banquets; along the walls are the remains of walled benches for the display of the votive objects. The entire _domus_ complex with the portico represents the first example of a _domus publica_ whose plan is completely preserved, along with some elements relating to the standing structures (columns with capitals). The dedication in Oscan refers to a deity who can be recognised as _Ops Consiva_, given also the presence in the sanctuary of a slave called _Opalis_. The _sacellum_ with an altar in the internal nave of the portico behind the domus, therefore, must have been dedicated to Ops Consiva, who had the _sacrarium_ in the _domus publica_, as she had at Rome in the _Regia_. Furthermore, another dedication in Oscan, discovered in the past in the area of temple B, attests the cult of Vittoria. In Ops and in Vittoria, therefore, it is possible to recognize two of the divinities which formed part of the triad to which the largest temple of Pietrabbondante was dedicated.

Work continued in the zone of the _tabernae_, where the original layout, pre-dating the construction of Temple A, began to emerge, with the presence of constructions with the characteristics of cult buildings. Some had been destroyed by a lightning bolt. A _sondage_ revealed the presence of another temple, already in use at the time of the Samnite wars, situated about 100 m east of Temple A. The excavations in the _domus_ _publica_ defined the layout of the slaves’ quarters and the building’s various phases. Phase one (120-30 B.C.: _domus_ _publica_ of the sanctuary). The building occupied a terrace of 110 m south-west of the temple and theatre. From this terrace the sacred area was accessed through a large opening in the enclosure wall that led directly to the space between the theatre and the temple. The size and architectural characteristics are those of the great Italic-Roman aristocratic houses of Republican date, with an atrium, _impluvium_, _tablinium_ and _alae_. The _domus_ stood in an open space with a double-aisled portico that had nine columns at the front where the offerings were placed. Following the Social War, the entire complex of Temple B and the theatre were no longer used as public cult structures. The complex was neither destroyed nor abandoned but preserved as part of the public patrimony, although no maintenance was carried out on the structures for at least fifty years. During this period, the buildings decayed and were damaged. In particular, most of the rear portico of the _domus_ collapsed and was never rebuilt. Phase two (30 B.C. – mid 2nd century A.D.: farm villa). In the Augustan period, the sanctuary’s properties were assigned to veterans: a _fundus_ became the property of the Socelli, who installed themselves in the ex _domus_ _publica_; in the nearby locality of Arco another _fundus_, already owned by the sanctuary, was assigned to the Munatii. In this phase, the _domus_ underwent a number of transformations, with the creation of a bath suite and a direct passageway between the space in front of the building and the service area. The link with the sanctuary buildings was maintained and maintenance work on the roof of Temple B began but was not completed. During the 1st century A.D., the temple began to fill with earth and the terracottas from the roof fell down. Phase three (mid 2nd cent. A.D. – 3rd cent. A.D.). Production activities underwent a radical transformation with the creation of kilns and metalworking structures. The demolition of Temple B began in order to recover the metals in the structure. The opening between the _domus_ terrace and the sanctuary was blocked. The platform on which the portico to the left of Temple B stood was used as a cemetery area. Phase four (first half of the 4th cent. A.D.). Decline in production activities, now limited to stock raising. There was evidence of large scale occupation in the _tabernae_ situated between the theatre and Temple A. Part of the _domus_ was in ruins and only occasionally occupied, perhaps by shepherds.

The _domus_ Work concentrated on the rear part of the building. The large hall opposite the _tablinium_ was linked by a corridor to the _culinae_, situated in the inner aisle of the portico housing the offerings, and roofed with tiles bearing Oscan stamps. One of the tiles had a double _opaion_, indicating the need for aeration due to the use of braziers. The hall, which can be identified as the sanctuary’s _curia_, was used for _concilia_ and ritual banquets. It represents an unusual example of the sacerdotal _curiae_. A great deal of literary evidence exists for such structures in Rome, although no monumental remains are known there or in other ancient towns. The exploration of the earliest levels revealed a floor of the so-called _scutulatum_ type, relating to the building’s first phase and so datable to around 120 B.C. _Building L_ In the area south-east of Temple A the investigations aimed to define the architectural characteristics and identify the dating of the cult buildings, of which at least two were discovered. The two buildings were separated by a wide walkway. Among the materials collected from the wall and roof collapse were a very ancient type of decorated ridge tile (_kalypter_ _hegemonies_), an _aes_ _grave_ (275-270 B.C.), and a terracotta votive in the form of a leg. Other finds included acorn-shaped lead projectiles found grouped within a small space, and a bronze _paragnatide_ with an iron nail in it for fixing it to a wooden beam. This detail shows that weapons and arms taken from the enemy were also dedicated to the deity here, as widely attested in other parts of the sanctuary. Among the remains of the roof, an Oscan stamp on a tile fragment attests its use in a building dedicated to _Venus_ _Erycina_. Stamps of this type have been found elsewhere, including the area of Temple B. Therefore, the stamped imbrex, which may have had a secondary use, does not provide concrete evidence for the presence of a temple of _Venus_ _Erycina_ here, although it is a possibility. _Zone N_ The area east of Temple B was partially excavated in 1979 and then back filled. It was reopened this season and the excavations revealed a series of limestone slabs that were being worked on when activity on the building site stopped and it was abandoned. The blocks were still in a row as they had been positioned for the final chiselling and dressing. Each block was at a different stage of completion. They had a smooth surface bordered by two cyma reversa mouldings, and were destined for a building a short distance from the construction site. The type of working and the mouldings are of the type present on Temple B. The sudden interruption in the construction work must have been caused by an important event, probably the beginning of the Social War, the outcome of which meant the building was not completed.

Restoration work began on the portico of the offerings, with the consolidation of the walls and restoration of the benches abutting the walls that were used for displaying the votive offerings. The space between the rear wall of the portico and the parallel containing wall behind it were excavated. This wall was made up of large, irregular, uncut stone blocks and was badly bulging and sloped due to the movement of the terrain behind it. In antiquity, the offerings portico was filled with sculptures, dedications and other objects donated by the faithful, aligned on the floor and on two benches running along the back wall. There were nine columns in the portico’s facade, which opened onto the space in front. The missing central column had covered a well built for the foundation ritual. Its excavation remains to be completed. Inside the portico there was a room that was not freely accessible, closed by a door, perhaps the _sacrarium_ of _Ops_ _Consiva_, goddess of plenty. She is mentioned on a base with an Oscan dedication situated in the public part of the portico and in another inscription found, in the past, in the area of the temple behind the theatre. In Rome, _Ops_ _Consiva_ was believed to be an Italic deity introduced by the Sabine king Titus Tatius. The goddess had a _sacrarium_ in the (_domus_) _Regia_ in the forum that could only be entered by the _Pontifex_ _Maximus_ and the Vestals. The excavations produced more information regarding the nature of Italic religiosity and its relationship with that of Rome. In fact, inscriptions were found that mention the cult of _Honos_, the divine personification of military and civil honour, which can be added to those mentioning the cult of Victory documented by previous excavations. This group of abstract deities, Plenty, Honour and Victory provide a picture of the ideological characteristics of the state cult practiced by the Pentrian Samnites in the monumental sanctuary of Pietrabbondante. The excavation of the cult building found downhill from Temple A in 2010 also continued. At the time of its collapse, caused by a fire no earlier than the 2nd century B.C., it was decorated with weapons taken from defeated Roman and Italic enemies towards the end of the 4th century B.C. They were offered to the deity in thanks for military success in accordance with a practice that was also followed in Rome. To the north-east of Temple A, a platform of large stone blocks was identified on which a Samnite building must have stood. The area was later occupied by a medieval structure, perhaps a house that seemed to have collapsed in the 13th century. An earring with a semi-precious stone was found on the floor below the collapse. This building constitutes one of the few traces of the medieval occupation within the ancient sanctuary. In the area of Temple L, excavations were undertaken in the _pronaos_ and the passageway between the temple and the other building whose sidewall was identified.

The trenches opened in 2010 to the east of Temple A revealed the presence of another medium sized temple, the excavation of which is ongoing (Temple L). It was built using different construction techniques: walls built in stone bonded with mortar, mud bricks, and ashlars The footings were preserved to a maximum of 70 cm in height and were formed by rows of small stones, positioned horizontally and bonded with weak mortar, on top of which stood the walls built of unbaked clay bricks. Smooth, soft limestone pillars stood at the head of the _ante_, both the exterior ones and those of the _cella_, supporting the trabeations. Only the bases with mouldings and a number of the drums survive to a height of 2.92 m. The moulding suggests this was an Ionic temple. The structure differs from the other temples at Pietrabbondante, and Samnite buildings in general, not only in construction technique but also in its architectural characteristics. It did not have a podium, an essential element for an Italic temple. The plan of the new temple has not yet been completely identified, neither have the foundation levels been investigated. However, the building can be attributed to the class of temples with _alae_ of Etruscan tradition, even though its layout is unusual. An almost square space, 14.46 m wide and 13.08 m long, it was surrounded by walls with a door at the centre of the front wall. The interior was divided into three spaces, the central one being the largest. The _cella_, also quadrangular (5.90 x 5.45 m), had sidewalls extending beyond its front wall so as to form two _ante-. To the sides of the _cella_ were two rooms without a front wall, the actual _alae_. The entire building, _cella_ and _alae_, was roofed and the floor was paved with _opus_ _signinum_. The particularity of the plan lies in the double closure of the _cella_ and of the space that includes the _cella_ and _alae_. The front wall closing this space was necessary for functional purposes. The building must have been rendered secure in some way given its function. Administrative activities involving the payment and receipt of money must have taken place in the temple: it may have been used as a treasury, housing the money from taxes and tithes, as widely documented in the Hellenistic and Roman world. A stone table was discovered at the centre of the _cella_. It rested on stone table supports decorated with lion’s paws separated by a small column with an Ionic capital. Behind this table, abutting the back wall of the _cella_, were the remains of an ark, a coffer sunken into the ground, which contained numerous bronze and silver coins, a hoard constituted towards the end of the 3rd century B.C. Inscription on the table: keís • enniis • keieís • medís • túvtíss • kamatúm ekík • úpsanúm ded{•}ed ísídum prúfatted • Ceius Ennius Cei f. medix tuticus mensam (?) hanc faciundam dedit idem probavit

The continuation of excavations in Temple L led to better definition of its shape, size, construction phases and functions. The building occupied an almost square space of c.14.30 x 15.40 m (52 x 56 Oscan feet). The supporting structures were on the whole recognisable, but it remains to be ascertained whether between the front _antae_ there were timber columns or pillars. At about one metre back from the _antae_ was the temple facade formed by a wall with a central door leading into a quadrangular interior space. At its centre was the _cella_, also closed by a wall with a door on the same axis as the first. The closure of the _pronaos_, the_ arca_ abutting the back wall of the _cella_ and the _trapeza_ (altar) with magistrate’s dedication placed in front of it indicates that the cult building also functioned as an _aerarium_. Three hundred and forty-two coins, including 140 silver examples, were recovered from inside the _arca_, the latest dating to 207 B.C. during the course of the second Punic War. Most of the coins were part of a large supply of money that came from Rome to finance the war against the Carthaginians. The excavation of the Ionic temple had already shown that in this period raiding by Hannibal’s army badly damaged the sanctuary. The sacking of Temple L can be attributed to one of these raids, if not the same one that led to the destruction of the Ionic temple. However, the temple was not destroyed, given that during the 2nd century B.C. it was embellished and maintenance work was carried out, although no structural rebuilding tool place: the interventions involved the floors, the application of red, black and white plaster with mouldings, and the substitution of the roof with tiles that were destined for other buildings, including Temple B, as attested by the tile stamps. Temple L remained in use until the Social War, although it is not known with what function. It was not destroyed during this conflict but at the end of the war was closed and not used again. There are no traces of destruction at Pietrabbondante attributable to the Social War: the sanctuary was respected, although deprived of its public cult function. The building’s end can be attributed to a lack of maintenance during the 1st century B.C. with the consequent collapse of the roof, while the _trapeza_ placed inside it by the Samnite magistrate in the 3rd century B.C. was still _in_ _situ_. The temple, certainly the earliest among those found to date at Pietrabbondante, was in use for at least three hundred years, between the 4th century and the beginning of the 1st century B.C. The remains of a small rectangular building of unknown function, dating to the 2nd century B.C., were excavated c. 15 m north of the temple. It had a low platform 12.70 x 8.50 m with three rooms measuring 2.90 x 3.50 m opening onto a portico with four columns.

La campagna 2015 ha previsto la prosecuzione di scavi in estensione presso l’ _Aerarium_, le zone N e DN, la realizzazione di 5 saggi stratigrafici presso la _domus publica_, il restauro del portico della _domus_ e degli ambienti annessi, delle sedute del teatro e del muro della cella del tempio B. Il saggio 1 nei vani 14 e 15 ha riportato in luce un pavimento in signino ascrivibile alla prima fase di frequentazione (fine I a.C.) e confermato l’aggiunta del complesso termale durante la seconda fase (I-II d.C.). Il saggio 2, presso la canalizzazione esterna, ha consentito di completare il ripristino del drenaggio grazie al ristabilimento della pendenza originaria nella zona 27 e tramite la posa di un tubo drenante al di sopra della pavimentazione del tratto di canale che segue in parallelo il muro di facciata della _domus_ (zona 24). Lo scavo ha evidenziato due quote d’uso della canalizzazione. Il saggio 3 nel vano 7 ha confermato l’esistenza del collegamento tra il canale esterno e il sistema di canalizzazioni interno alla _domus_. Il saggio 4 presso il pozzo della _stoà_ ha previsto lo smontaggio parziale della struttura, in seguito ricostruita fino al piano di appoggio della quinta base di colonna. Il riempimento era costituito da alcuni recipienti ceramici, oggetti di legno, reperti vegetali e osteologici. L’entità dei campioni chiarirà la destinazione d’uso: se funzionale all’approvvigionamento d’acqua o connessa a rituali. Il saggio 5 è stato realizzato per assicurare il deflusso delle acque e preservare le strutture del portico grazie alla posa di un tubo drenante all’interno della canalizzazione antica che percorre l’ambiente 26 F e ne fuoriesce con andamento parallelo al braccio occidentale del porticato. Lo scavo dell’ _Aerarium_ ha consentito di ultimare l’indagine dell’ala est con il raggiungimento del suolo vergine e di individuare una canalizzazione la cui funzione è da porre in relazione a specifiche pratiche rituali. Nell’ala ovest è stata completata l’asportazione degli strati di crollo recuperando parte della decorazione parietale e reperti relativi alla cerimonia di chiusura che ha interessato il complesso (fine II-inizi I secolo a.C.). Lo scavo dell’ambulacro esterno ha evidenziato la presenza del muro di contenimento che circonda la struttura a est e a nord. Nella porzione occidentale del camminamento, al di sopra degli strati di crollo, è stato individuato un cospicuo quantitativo di ossa animali e ceramica. In alcuni punti è stato possibile isolare dei nuclei di oggetti interpretabili come resti di rituali espletati al di sopra delle rovine (I a.C./I d.C.). Nella zona DN l’allargamento verso nord ha chiarito l’andamento delle strutture intercettate nel 2014. Il rinvenimento di cinturoni bronzei, di un puntale di spada e di materiali di pregio ha consentito di avvalorare l’ipotesi che riconosce nel complesso un _heroon_. L’attività di scavo nella zona N si è concentrata nell’angolo NE dove è stata individuata una struttura con due diversi livelli pavimentali sui quali si imposta un pozzo, chiuso da una lastra calcarea con foro centrale. Si è potuta inoltre accertare la presenza di un piano di cantiere che risulta attualmente inclinato per i dissesti geologici che ha subito l’area. Didascalie 1. Planimetria generale 2. Sandalo in legno rinvenuto all’interno del pozzo della Stoà 3. Veduta della Stoà dopo il restauro 4. Zona DN. Veduta generale dell’area di scavo

La campagna 2016 ha previsto la prosecuzione di scavi in estensione presso l’Aerarium e la cd. Piccola Stoà, le zone N e DN e la realizzazione di un saggio presso il Santuario Orientale. Nella zona DN lo scavo ha interessato il margine sud-orientale e ha permesso di intercettare tre setti murari, confermando lo sviluppo del complesso verso la terrazza SO. Nella zona N è preseguita l’indagine nell’ambiente SE, concentrandosi sullo scavo del pozzo al suo interno il cui riempimento ha rivelato la presenza di pietrame sistemato per dismettere la struttura ovvero per impedirne il crollo. Si segnala la presenza di uno scalpello in ferro da mettere in relazione al cantiere di lavorazione delle lastre. Le indagini presso il Santuario Orientale si sono concentrate presso l’ambiente C della cd. Piccola Stoà. Lo scavo del vano ha previsto la rimozione delle stratigrafie di crollo e l’individuazione di due livelli di preparazione pavimentale e di lacerti di pavimento relativi all’ultima fase di frequentazione. Al di sotto della preparazione è stato intercettato uno strato in pietrame sistemato in modo regolare e interpretabile come resti di crollo relativo alla prima fase di vita. Al di sotto di tale strato sono visibili frammenti di pavimentazione che continuano al di sotto del muro ovest, verso il vano B, lasciando ipotizzare una diversa partizione dell’edificio durante la prima fase di frequentazione. Le indagini presso l’Aerarium si sono concentrate nella porzione orientale dove erano già state individuate stratigrafie relative alla fase di IV-III secolo a.C. È stato messo in luce uno strato argilloso corrispondente al banco naturale sul quale poggiano le murature e al di sopra del quale è visibile uno strato compatto con pietrame e tracce di calce relativi a un piano di cantiere. I muri, con tessitura molto irregolare, dovevano ricreare una sorta di recinto riempito con materiali pertinenti ai crolli di una struttura precedente e sfuggiti alla distruzione. Il riempimento risulta contemporaneo all’edificazione dell’Aerarium come conferma una moneta databile agli ultimi decenni del III secolo a.C. L’ipotesi che le murature non corrispondano ad un edificio sembra essere confermata dall’assenza di un piano pavimentale e dal fatto che i materiali siano stati posizionati in modo ordinato all’interno del riempimento. L’ultima area oggetto di indagini è situata a poca distanza dalla cd. Piccola Stoà all’interno della quale un sondaggio ha consentito di intercettare un sacello pertinente ad una struttura più ampia. La stanza presenta al centro, impostato sul muro di fondo, il basamento di una statua di culto, al di sopra e ai lati del quale sono stati rinvenuti frammenti architettonici in tufo. Di fronte al basamento sono presenti i resti di un piccolo altare in pietra calcarea. L’ambiente era occupato da uno strato di pietrame frammisto a tegole in cui sono state rinvenute 149 monete in bronzo e 1 in argento, lucerne, lamine in piombo e oro, frammenti di una statuetta in marmo, ossa animali, ampolle in vetro. L’eccezionalità della scoperta è rappresentata dai materiali ascrivibili a un rituale pagano di abbandono dell’area e dalla moneta del 407-408 d.C. che rappresenta la testimonianza più tarda di frequentazione del sito. Didascalie 1. Planimetria generale 2. Piccola Stoà. Veduta dell’ambiente C in corso di scavo 3. Sacello del Santuario Orientale, veduta dell’area di scavo 4. Moneta di Onorio (407-408 d.C.) rinvenuta nel sacello

La campagna 2017 ha previsto la prosecuzione di scavi in estensione presso l’Aerarium e i Sacelli del Santuario Orientale e presso la zona N. Le indagini presso l’Aerarium hanno permesso di confermare l’esistenza di una fase di frequentazione di IV secolo a.C., precedente la costruzione del tempio (ultimi decenni III secolo a.C.). Il settore centrale è caratterizzato dalla presenza di un interro costituito da materiali eterogenei e da armi sistemate in aree circoscritte. L’interro può essere considerato un’azione volontaria svoltasi al termine di un pasto rituale, come testimoniato dalla presenza di numerosi ossi animali. A seguito di tali pratiche sarebbe avvenuto il seppellimento dei materiali, verosimilmente recuperati da edifici distrutti al passaggio delle truppe annibaliche. Il riempimento oblitera un piano di cantiere relativo alla costruzione dell’Aerarium di cui è stato possibile individuare livelli di argilla frammista a calce. Nel settore meridionale è stato possibile riconoscere la presenza di una struttura che si estende oltre i limiti di scavo, obliterata a seguito della costruzione del tempio. Nelle stratigrafie di interro si riconoscono tracce di un rituale di chiusura: su un piano di scaglie in pietrame furono sistemati ex voto e resti di un pasto, comprovato dalla presenza di coltelli, spiedi e ossi animali. Il ritrovamento più importante si riferisce ai resti di 5 supporti in bronzo conformati a zampa equina che sorreggevano un piano circolare in bronzo, di cui sono stati recuperati 11 frammenti. Le zampe e i frammenti del ripiano risultano rotti in antico e sistemati in terra insieme agli ossi e agli altri oggetti. All’interno dell’Aerarium sono stati eseguiti due sondaggi in corrispondenza dei muri d’anta interni. In entrambi i casi sono stati messi in luce tratti di murature pertinenti ad una struttura a pianta quadrangolare (5 x 5 m). All’interno del saggio A è stata intercettata una fossa dalla quale provengono resti della pavimentazione e degli intonaci pertinenti al sacello più antico e 54 monete in bronzo e argento, che datano il riempimento alla fine del III secolo a.C. Due oboli in argento rinvenuti in corrispondenza delle murature del sacello, confermano la frequentazione dell’area a partire dal IV secolo a.C. L’indagine archeologica estensiva ha riguardato l’area pertinente il Sacello rinvenuto nel 2016 dove è stata messa in luce una struttura molto estesa e con diverse fasi di frequentazione che si protrassero fino al V secolo d.C. L’edificio si compone di tre ambienti contigui e delle medesime dimensioni, circondati esternamente da un muro che delimita un camminamento sui lati N, E e O. È stato possibile appurare come una parte consistente della struttura fosse realizzata in tufo grigio di provenienza campana nella prima fase di frequentazione. È stata ultimata l’indagine nei sacelli B e C. In quest’ultimo è stato possibile individuare numerose analogie con il vano C della cd. Piccola Stoà sia cronologiche sia nella resa del pavimento di ultima fase che presenta un ampio risparmio di forma rettangolare situato nella porzione centrale dell’ambiente, probabilmente occupato da un basamento di statua. Nella zona N si è continuato ad indagare l’ambiente posto nell’angolo NE ampliando il saggio verso SE. È emerso il piano di calpestio realizzato con tessellato laterizio. Didascalie 1. Planimetria generale 2. Supporti e ripiano in bronzo rinvenuti nelle strutture ad est dell’Aerarium 3. Spallaccio in bronzo rinvenuto presso le strutture ad est dell’Aerarium 4. Veduta aerea dei Sacelli del Santuario Orientale in corso di scavo

The 2018 campaign continued the excavation of the _Aerarium_ of the East Sanctuary, in the area of the _pronaos_ and the rooms of the structure situated in the south-eastern part of the temple. The investigations on the south-western side of the _pronaos_ of the _Aerarium_ revealed a stone structure on a SW-NE alignment, interpretable as the base of a monument of which a few fragments of worked fossiliferous limestone were recovered. The base was covered by layers formed by the temple roof collapse and included a _cippus_ made of tufaceous stone. These elements indicate that the structure was built before the end of the 3rd century B.C., and was coeval with the occupation phases of the _oikos_ (4th – 3rd century B.C.) identified below the _Aerarium_. The area explored on the south-eastern side of the temple confirmed the presence of a structure formed by four rooms (building LF), which extended beyond the trench edge to the NE-SW. The excavation of room A (partially explored in 2017) revealed a floor surface made of crushed stone. Underneath this floor there was layer containing black glaze pottery, including some miniature forms, associated with numerous animal bones, and a second layer with fragments of tiles and ridge tiles. These levels covered a section of wall, aligned S-N, from an earlier structure visible to the south-west, in a section of wall already uncovered in previous seasons. The alignment and size of this second building will be clarified by the continuation of excavation towards the east. For the moment, the material finds suggest an occupation phase dating to at least the 4th century B.C. More details regarding building LF emerged from the excavation of room B. The excavated stratigraphy related to the intentional deposition of objects placed on a surface of flat tiles. Among the pottery were fragments of black glaze ware associated with cooking-ware jars and lids. Numerous iron objects were present, such as knives, skewers and two pairs of metalworker’s tongs/pliers. The tile surface was uneven and overlay a level of ash and charcoal, in turn covering a stone slab floor. The latter, very patchy, was on a gravel make-up (also found in room A). There was a hearth obliterated by a complete tile (71 x 48 cm) at the centre of the stone slab floor. This deposition of materials can be associated with the closing ritual of the room, similar to that seen in room A in 2017. The cooking and consumption of food is associated with the positioning of the ceramic objects (used for the banquet and often upside down or containing bone remains). Other objects such as miniature vases and jewellery (including a cornelian scarab, part of a seal ring), were also positioned, together with materials like the metalworking pliers, which may indicate the original use of the structure. The construction of the _Aerarium_ obliterated the entire eastern area with a fill positioned to the north-west of building LF. A preliminary study of the finds from the fill, compared with those intentionally placed inside rooms A and B, dates this intervention to the final decades of the 3rd century B.C. The new structure therefore belonged to a phase preceding the fill, in concomitance with the use of the _oikos_. The two pairs of metalworking tongs/pliers and other objects, such as a bronze support broken into pieces ready for smelting could confirm the suggestion that it was a forge. The presence of a metalworking installation close to the temple is documented by examples such as room N in building _gamma_ at the _emporion_ of Gravisca, situated north-east of the sanctuary of Aphrodite. To conclude, three occupation phases were identified in the eastern area of the _Aerarium_. The first, corresponded with the structure below building LF, contemporary with the construction of the _oikos_ (4th century B.C.); the second was the probable forge (building LF), datable to the final decades of the 4th century B.C. Lastly, the obliteration of the structure with the closing ritual occurring in the final decades of the 3rd century B.C., in conjunction with the temple’s construction after the Hannibalic destruction, which would also have involved the new building.