Investigation restarted on the northern stretch of the walls, on both sides of the central gate. The work included surveying, core sampling and excavation. The results confirmed the previously identified phases: construction of the great wall at the time of Aristodemos, with a curtain of orthostats which faced both sides of an enormous embankment; the construction, in the second half of the 5th century B.C. of two foreparts to form a tenaille gate; the doubling of the walls in the Hellenistic period, with the creation of an outer curtain built in horizontal coursess which was further advanced towards the north and joined to the earlier wall with a series of transverse tie blocks; lastly, the extensive restoration of the Sullan period. The previously documented evidence was integrated by the discovery of an earlier phase of the wall east of the gate, slightly set back with respect to Aristodemus’s wall, and probably constructed over the earlier town wall. A preliminary investigation was made of the only well preserved stretch of the south walls, situated north of the Croce di Cuma. It was established that the ancient ground level outside the town walls was circa 9 m lower than the interior ground level. The existence here of a double curtain wall, certainly datable to Aristodemus’s time, was also ascertained. A forepart was added which probably served to delimit a gate, now below the modern road. The late archaic wall was then incorporated into a double curtain wall in the Hellenistic period. Following the levelling, the inner curtain of the late archaic wall partially slipped lower down. After the dismantling of the structure, undertaken as part of the restoration process, a detailed stratigraphic investigation of the _emplekton_ and the natural layers behind it was made. This produced important data regarding chronology and construction techniques.

The structures of the eastern arm of the archaic gate were completely excavated. On the basis of the materials recovered from a trench dug in the embankment of the eastern arm a more precise chronology was formulated for the late archaic phase (end of the 6th-beginning of the 5th century B.C.). Furthermore, the entire plan of the eastern forepart of the classical phase (end of the 5th century B.C.) was looked at. This structure reached a length of 12 m to form a tenaille gate with the forecourt projecting and opening towards the exterior. It was established that in the Hellenistic phase (second half of the 3rd century B.C.) the doubling of the curtain walls also involved the inner circuit (width reached circa 12.40 m). Moreover, it was ascertained that the restructuring led to the moving forward of the front of the eastern forepart towards the centre of the forecourt and the creation of a spur, probably symmetrical to another on the western side. For the Sullan period the hypothesis proposing the abutment, on the inner side, of a new curtain wall in _opus reticulatum_ (width reached circa 13 m) was confirmed. It was also possible to ascertain the existence in this phase of a gate with a double arch, perhaps subjected to new restructuring when the _Via Domitiana_ was built. It remains unclear whether the major monumentalisation of the area can be dated to the Augustan or Domitianic period. This comprised the construction of a wall on a north-south alignment, up against the northern curtain wall of the eastern forepart and paving in trachytic pavers. The attribution to the period of the Greek-Gothic war of the last restructuring of the fortifications, involving the eastern forepart, as well as the upper part of the _emplekta_ of the various phases of the curtain wall, was confirmed.

Following a campaign of core sampling, “L’Orientale” university began work in the area of the northern fortifications, opening three excavation areas to the west of the central gate. In area 1, corresponding to the central gate, the paved surface of the _via Domitiana_ and a vast square, extending in front of the aulé of the ancient gate, came to light. A Roman _mausoleum_ and a fountain served by a large underground cistern were also uncovered. The cistern had subsequently been transformed into a mausoleum. The northern head of the forepart, delimiting the aulé to the west and datable to the Hellenistic period, was also excavated. Next, the edge of the gate from the period of _Aristodemus_, was uncovered, whilst to the east of the entrance an L shaped structure was investigated. This perhaps related to an earlier gate. An enormous sewer dating to the period of _Aristodemus_ was also examined. Area 2 revealed the complete system of fortifications, heavily robbed in the pre-modern period, with a tower that projected with respect to the curtain of the Aristodemian period. The tower had a cruciform plan and was later incorporated into the construction of the Hellenistic doubling of the wall. The wall was abutted on the south side by a vast rectangular structure with a shaped base. This may have supported a stairway providing access to the walkway used by patrols inside the town. In the Graeco-Gothic period the structure was transformed into a workshop. The opening of three trenches in area 3 revealed the line of the walls and the stratigraphy north of them: a substantial layer of earth had been placed up against the Hellenistic curtain wall, before it was robbed in late antiquity. South of the walls, a stairway of Republican date, over 50 m long and abutting the inner Aristodemian wall, functioned as a buttress and provided fast access to the walkway. When the walls lost their function a road was constructed in the space occupied by the stairway which obliterated the structure itself.

A number of trenches were dug in order to uncover the access stairway to the walkway abutting the eastern face of the gate and the drainage system along the road passing through it, which later became the _via Domitiana_. As early as the Republican period an in flow pipe from this water system drained into the sewer built by _Aristodemus_. Excavations in the area between the gate and the farm situated near the racecourse aimed at tracing the outer curtain of Hellenistic date. A trench opened next to the tower was deepened down to the foundation offset of the wall, thus acquiring a bench mark that was essential for the reconstruction of the system of defences. On the town side definition of the system was provided by the discovery of the late archaic inner curtain, which was fairly well preserved. Thus, it was seen that there was a recess along this line towards the south, where a stairway providing access to the walkway was inserted. This deviation in the line of the wall was determined by the need to incorporate within the fortifications of Aristodemus the earlier defensive structures. A trench dug between the two curtain walls confirmed the proposed chronology of around 560 B.C. A trench dug near the ex-racecourse, within the embankment of _Aristodemus_, revealed an earlier defensive structure, of the type with two curtain walls built in horizontal levels, faced with structures of tufaceous chippings. The emplekton was constituted by alternating layers of tufa chips and grey silt, datable generically to around 560 B.C. Furthermore, excavation of the tower with the four-part plan, and the adjacent sector of the Hellenistic fortifications continued. For the first structure the dating of the beginning of the 4th century B.C. was confirmed, whilst in the second case a substantial deposit of articulated animal skeletons was recovered, amongst which were at least three cattle and a canid.