During excavations at the Culver Street site between 1981-2 and 1984-5 (ECC337), a large number of Roman and later features were examined in addition to the main buildings, streets and other structures.

Evidence for the earliest military activity on the site included a group of small indentations sealed by the metalling at the junction between two streets (MCC779 and MCC781). These may have resulted from the removal of vegetation when the fortress was being laid out. Elsewhere extensive use of the military east-west streets (MCC761, MCC762, MCC763 and MCC764) prior to their metalling was uncovered. On two sites the impressions of wheel ruts were discovered and in addition, beneath MCC764 was a single hoof print from a bovine animal. Elsewhere, the earliest metalling sealed occupation debris and a hearth - see MCC2579 - (CAT BF820), which contained a military strap fitting. It was thought that the occupation and the hearth were likely to have derived from the construction phase of the barracks. Some early features occurred on Site C including an oval pit (CAT CF151) with sides and bottom stained green.

Contemporary with the barrack buildings were a number of features including: the remains of a hearth (MCC2584) to the south of the barrack block (MCC758) and a line of small pits cutting into the metalling of the street (MCC764) which may have held a structure of some kind, possibly a lean-to against the buildings veranda. Other pits were noted in the area but their relationship to the barrack was obscure.

Following the foundation of the colonia in AD 49, a new colonial street grid was laid out. The area of one of the streets (MCC733) was initially unsurfaced and instead occupation debris and silt accumulated over the area and eight pits were dug on the site of the street and southern footway. Three pits were also found cutting into the northern footway (MCC789) and the footway had a hearth (MCC2588) lying on its surface.

Following the destruction of the Boudican revolt clearance work included the removal of the debris and complete removal of the bases of internal walls, via robber trenches, and the surfaces of floors. 'This debris was presumably dumped outside the town leaving very little demolition material in situ'.

On a site formerly occupied by a large military building (MCC776), the area was left open for a period after the post-Boudican clearance when some pits and gullies were dug prior to the construction of a group of three buildings (MCC863, MCC865 and MCC866).

Following clearance work the plots which had been occupied by the southern three barrack buildings (MCC756, MCC758 and MCC759) to the west of the former via principalis (MCC760) were left open and converted into cultivation plots (MCC827 and MCC853) separated by a street (MCC764).

At the base of the cultivated soil, a few metres north of the street was a distinctive group of irregular depressions interpreted as spade marks. Close by several post holes (including GF1836-6, GF2242-4, GF2408, GF2476, GF2642 & GF2865) may have been associated with fences. Various pits were dug in the area and two groups could be distinguished, one belonging to the late 1st to early 2nd century and the others dating to the mid 2nd to 3rd century towards the end of the period of cultivation. The area was built over at the beginning of the 3rd century (see MCC902, MCC905, MCC918, MCC910 and MCC912).

An area on the south side of Insula 35 to the east of the former via principalis was also cultivated (see MCC855). Various pits and other features were dug into this area. One group was situated in the south-west corner and consisted of two north-south rows of pits/post holes. These were felt to represent either a small post-built shed or a sequence of fences or gates. Another group lay close to the southern boundary of the cultivated area and contained waste products from bronze working and possibly iron smithing. The quantity of waste material suggested that metal working either took place close to the area or that it was used as a dumping ground for the waste from that activity.

The southern edges of Insula 34 and Insula 35 were affected by the construction of the town defences. Early features included a series of wheel ruts cut into the Boudican destruction debris over the site of a building (MCC764). These were thought to be associated with the construction of the town wall (MCC859). A street (MCC1169) was built to the south of the wheel ruts which was sealed by clean sand probably derived from the digging of the walls foundation trench. Overlying this were piles of septaria chippings derived from the dressing of facing blocks and the remains of wooden mixing boards.

These deposits were then levelled and sealed to form a new street surface (MCC1167). To the north of this street several pits were dug one of which appears to have been a water tank (see MCC1020).

Following the construction of the wall but prior to the building of the rampart (MCC1170), the area between the wall and plot originally occupied by the building (MCC764) was covered by dumps of material. Contemporary with these layers were an oven (GF2139) and hearth (GF2708).

Apart from an aisled building (MCC927) and corn-drying oven, activity on the site was very limited during the 4th and 5th centuries much evidence having been removed by later activity. However pockets of deposits survived including pits, post holes, gullies and spreads of gravel and rubble, some of which may relate to former buildings which could not be identified.

The uppermost Roman layers were overlain by an accumulation of 'dark earth' which existed in substantial quantities by the time of the medieval robbing activity which took place on the site. This activity mainly belonged to the 11th and 12th centuries although limited robbing activity continued into the 13th centuries and possibly the 14th century. Much of the site was affected by pit digging during the medieval and post medieval periods. Some of these may have been associated with horticultural activities. Some of the pits (e.g. those which were square or rectangular in plan) may have been latrines associated with buildings on the southern Culver Street frontage. The largest pits formed a distinctive group because they were dated from the 15th to the 17th centuries. Elsewhere many of the pits were deep, steep-sided features thought to be cess pits of early medieval date and probably belonging to properties along the Head Street frontage. Some of the cess pits may have originated as sand extraction pits. Associated with a 13th century oven (MCC1130) was a large group of shallow pits of probable 12th to 13th century date.

In general most of the area covered by the excavation appeared to have been open until the 19th century.

Early medieval robber trenches (p.75).<1>