The walled town of Colchester is situated on a spur of land overlooking the River Colne, lying within the bank and ditch (dyke) system that defended the Iron Age settlement of Camulodunum. Over 26km of dykes crossed the gravel terrace that separates the Colne and Roman Rivers, creating a defended area of some 31km square. Current evidence suggests that prior to the foundation of the legionary fortress the site of the current town was unoccupied. The Romans placed the longitudinal axis of the settlement on an east-west ridge formed by the steep slope down to the river to the north, and a gentler slope to the south, thus providing both a commanding view and access to the Colne waterway; a site from which the conquerors could effectively police and supply their new territory.

See <64> for a summary of the current understanding of the Roman wall.

Because of the lack of local building stone, calcareous mudstone called septaria from an outcrop near Harwich was imported and used as the main building component, supplemented by courses of tile. The walls were constructed over a foundation trench 3m wide and 1.2m deep, which was filled with layers of mortar and septaria. The wall itself is of ashlar construction, meaning the inner and outer faces were constructed independently of the core. The faces were raised, presumably a few courses at a time, and then the space between the two faces was filled with layers of septaria and mortar laid alternately just as in the foundation. The faces were made of neatly coursed septaria and tile, usually four courses of tile being followed by four courses of septaria. A small offset was formed at the base of the wall by making the lowest one or two courses slightly wider than the rest (Crummy 1997).

The wall appears to have been built over a number of years between AD50/85 and AD125 - making it the first of its kind in Britain - to protect the 48 ha town, following the refounding of the colonia after the Boudican revolt. The wall originally formed a circuit of 2,800m long and 2.4m thick and stood to an unknown height of probably over 6m. It incorporated six gates and between 12 and 24 towers, with a V-shaped ditch dug along the foot of the wall to improve its defensive capabilities.

The town's main entrance from the west had been a monumental arch built in c.AD 49 (MCC718). This was incorporated into the walls with the addition of a footway and a bastion on both sides to form what is now known as the Balkerne Gate (MCC555). Most of the gate was deomlished and the gap filled in c.AD 300.

At some point between AD 150-200, the defences were again strengthened with the addition of an earthen rampart behind (i.e. on the interior of) the wall.

Around AD 275 the town ditch was widened to convert its profile from a deep V shape to a very wide truncated U shape. At Balkerne Lane, the widening had been achieved by digging out the far side of the existing ditch and piling up the excavated material much further out to form an outer bank (counter scarp). Also at this time the number of gates was reduced, the Balkerne Gate was closed and the external ditch extended to close off the road. The monumental arch and part of the rest of the gate was demolished and the rubble used to block the gap, apart from the south pedestrian walkway which seems to have been left open. Duncan's gate was treated in a similar way as the counter scarp bank added to the outside of the town ditch passed unbroken across the front of the gate.

For a time it was thought that the wall had been white washed during the Roman period. However, recent studies have shown that the leaching of lime from mortar or building stone can produce the effect of whitewash on wall surfaces.

Little is known of the wall's fate during the 5th-9th centuries. The Saxons appear to have repaired the town after the Danes were defeated in the siege of 917. The Anglo-Saxon Chronicle records 'King Edward went with the army of the West Saxon's to Colchester and repaired and restored the borough where it had been broken'. During the Medieval period the wall was valued for its protection as various Lords and Kings besieged the town from time to time. The Court Rolls of the town show that the bailiffs and commonality kept a close eye on the wall, summoning those who damaged the structure or built against it, or dug pits too near the foundations (Hull 1958, p.15). Further repairs are recorded for the reign of Richard II (1380s and 90s) and it seems likely that the bastions were added at this point and the stretch along Vineyard Street refaced.

Significant damage was inflicted during the Civil War siege of 1648, when the tops of the bastions were blown off and part of the south-west angle was demolished to allow for the mounting of a Royalist Battery. Other sections of wall were badly damaged and collapsed along the southwest corner of the town (south of St. Mary's Steps). The Civil War was the last military action seen by the walls and as they ceased to be looked on as a defensive asset less care was taken of them. Houses were allowed to be built up against the wall and passages cut through, extensive robbing occurred along exposed sections, especially at Park Folly and east of Roman Road. Some of the gates damaged during the siege limped into the 18th century, however the East Gate was pulled down in 1651 and Ryegate in 1659. Antiquarian Philip Morant records that the main medieval gate, Head gate, was gone by 1766. Furthermore, in 1795 antiquarian records note that a large section of Balkerne Hill stretch of wall fell into the road.

Wire's diary records ' Received information that an arch had been discovered in the Town Wall a few yards east of Sheregate went and examined it when I found it formed of two Roman tiles set on the ends at each side and a double row of the same tiles formed the arch which is 3ft thick from the under to the outer crown and two feet six inches wide inside and four foot six inches high to the inner crown how far it lead under the wall could not be ascertained an oven is now made in it'.<42> Hull records that a pencil drawing is to be found in Wire's notebooks, the arch cannot now be seen.<43>

An original drawing is held by the Museum of a section of the Roman town wall at the end of St Peter's Street, near Ryegate 24, February 1908, prepared for Henry Laver.<50>

In June 1920 the steps of the South West postern in St Mary's Churchyard were partly removed for excavation and revealed the drain of an arch. Drawings of the remains were made by Mr. W.D. Clarke.<28><29>

In 1923 clearance work was undertaken on one side of the interior of Bastion 1 by P.G. Laver. A shaft was sunk to a depth of 7 ft. 3 in. below the foundations of the Roman wall.<30><31>

In 1931 Hull examined bastion V. This is an external tower, D-shaped in plan and with an open back. The bastion and wall at this location were clearly not Roman and were found to have been built on the ruined remains of the Roman wall with the line of the rebuild diverging slightly from that of the original wall.<32>

A portion of an arch of Roman tile was observed around 1933, close to Head Gate and the cellar of the Old Liberal Club.<58>

In 1934 excavation on the site of Bastion 4 of the town wall followed the demolition of some small cottages. The investigations revealed the foundations of the Bastion.<33>

In 1938 the steps through the town wall at St Mary's were 'disturbed' and Rex Hull was able to make a number of observations. (See 'Roman Colchester' for full report).<51>

In 1938 a trench was cut inwards from the broken end of the Town Wall across the line of the rampart. The excavation revealed a faced section of wall and part of an internal tower.<34>

Graffito cut deep before firing, on the end of a building-tile used in the east face of the first internal tower of the Roman city-wall north of the Balkerne Gate. Noticed in 1938 when a tree was removed, still in situ.<46><47>

Excavations for 'fox-holes' during the invasion scare of 1940 exposed the corner of an interior tower standing approximately in the centre of the curve of the NE angle.<52><53><54>

Brick repair to south side of wall excavations for 'fox-holes' during the invasion scare of 1940 exposed the corner of an interior tower standing approximately in the centre of the curve of the NE angle.<52><53><54>

A section was cut through the town wall rampart by G. Farmer and E.J. Rudsdale, whilst still at school (ECC785). The excavation revealed the inner face of the Roman town walls.<41>

In 1948 a trench was cut at right angles to the town wall at a point 70 ft. south of the centre line of Bastion 1 (ECC864). The point was selected in an attempt to avoid earlier excavations by P.G. Laver, and to investigate the possibility of an internal tower on the site. The work was undertaken by volunteers, in particular Lt.-Col. Appleby. The foundations of the Roman wall, were uncovered.<35>

In 1951 a section was dug through the town wall rampart within the grounds of the Sixth Form College (ECC933). When completely excavated the wall, with buried face in mint condition was found to be standing to a height of 13 ft. as from the Roman ground level.<40>

Hull records 'only a few feet south of East Hill, the Roman wall is pierced by an opening with a double ringed arch of tiles. It is 11ft above the present ground level (to inner crown of vault). The lower part is covered by a modern stone apron and has never been examined. The opening is 2ft wide. It is the outfall of a drain.'<59>

Hull records an excavation by P.G. Laver on Bastion 2 of the town walls but there is no other record of this event (ECC952).<48>

Hull also records the sinking of a shaft by P.G. Laver between Bastions 1 & 2 of the town wall, but there is no other record of this event (ECC953).<49>

In 1957 the demolition of some houses in Vineyard Street together with the removal of outbuildings in the late summer revealed the Roman wall where it runs parallel to Vineyard Street. Immediately to the rear of No.19 Vineyard Street an arched opening about two feet square was revealed which led to the excavations by members of CAG, who concluded it was a Roman drain.<14>

In 1957 the demolition and removal of houses in Vineyard Street during the summer revealed a stretch of the Roman wall where it runs parallel to Vineyard Street. Immediately to the rear of No 19 Vineyard Street an arched opening about two feet square was revealed. The drain was partially cleared.<44>

In September 1957 Rex Hull noted that the demolition of a large building on the N side of Vineyard Street had exposed the face of the Town Wall. The site had been levelled and gravelled to make a car park. Just off this gravelling, at the NW corner of the area, an opening in the wall was apparent. Almost as low as ground level a Roman drain about 2ft wide and high, plus the height of the brick arch spanning it, debouched from the wall. It had been covered, or partly covered, by modern patchwork of brick, which was now broken, revealing the Roman work. Within lay a mass of broken bricks, mortar and dust. The underside of the arch could be seen to run back a good distance.' The arch was 123m (370ft) east from the centre line of Sheregate steps. The drain was cleared for 9ft revealing a heap of septaria; the arch continued at least another 6ft.<45>

In 1962 an excavation by B.P. Blake in the south-west corner of the town (ECC670) showed that the 17th century destruction of the town wall had left little of its bank intact.<13>

In c.1965 excavations were undertaken by members of the Colchester Archaeological Group (ECC689) under the auspices of the Colchester Excavation Committee and directed by P.R. Holbert. The object of the excavation was to recover details of the town ditch. On completion of work on the inner lip of the ditch, a small section of masonry belonging to Bastion No.3 was identified and included in the investigation. Mechanical excavation on the ditch consisted of a trench 40 feet long and 10 feet wide at a distance of 2 feet from the western wall. The base of the excavated ditch was waterlogged. Work on the bastion was concentrated on the most important features of the eastern half and the point where the western end joins the Roman wall.<27>

In 1967 a section was cut across the Roman defences during excavations at St Mary's Rectory (ECC661). The trench cut through the rampart and revealed that although the western face of the wall had been robbed out, there was no sign of other destruction or later re-build. The eastern face was found to be well preserved.<8>

In 1972 a small excavation resulted from the installation of a large sewer through St Mary's Steps. The investigation revealed a Roman drainage arch that was then used as a postern in the post-medieval period.<57>

In 1976 a watching brief was maintained during the construction of an office extension along St Peter's Street. 'During the contractor's initial clearance work the retaining wall against the Northgate Street boundary collapsed, revealing the top of an arched drain outlet in the north side of the town wall.'<6>

In 1977 a watching brief was held during the construction of an extension to existing buildings at St John's Street with new first floor shops fronting onto Eld Lane. 'The Roman wall in both plots was in a very poor state. The face has been removed and there was an indication of the offset at the foot of the face. The wall has been partly repaired/refaced in brick. The area affected had been cut about during pit digging. The base of the Roman wall was uncovered and it proved to be 0.6m below the modern ground level and resting on natural sand.'<9>

In 1979 a watching brief was held during the redevelopment of a site for small shop units in Sir Isaac's Walk. Two stanchion holes and a foundation trench were inspected. The largest of the stanchion holes provided a section across the Roman rampart behind the town wall. The outer face of wall was missing but the line indicated by some tiles set in mortar probably part of the base of the wall. This showed that the rampart was multi-phased and included at least one temporary street, and was predated by a substantial metalled street.<7>

In 1980 prior to the demolition of two houses and the redevelopment of the site for a two-storey office block, Nos. 70-74 Northgate Street were surveyed. The houses contained cellars that extended to the frontage of Northgate Street. Projecting through both of the cellar walls adjacent to the street is a length of Roman town wall that was also exposed when two trial holes were dug on the site following demolition.<55>

In 1982 a watching brief (ECC1387) was held during redevelopment of the former Cameo Cinema site. Substantial alterations were carried out internally in order to convert the shell into shops. The north wall of the cinema had been built on top of and down the face of the Roman town wall. The cinema wall was buried and left intact.<10>

In 1983 a watching brief was held during the laying of a trench along Ryegate Road (ECC1400). The trench was dug through the surface of the stump of Roman wall under the modern street. Only the southern face of the wall was detected. Part of the rampart appeared to be in situ to the south of it but elsewhere the rampart was not apparent, probably because it had been destroyed by later trenches and other features.<11>

In 1983-1984 observations and some limited excavation took place during the construction of a new sewer through the Castle Park. This involved the laying of lengths of pipe in a trench 0.75m wide and on average 2m deep, from Cowdray Crescent through the Upper Castle Park and out through the town wall into the Lower Castle Park. Work took place between November 1983 and February 1984. Recording was mainly limited to hurried plotting and section drawing but in some cases it was possible to work ahead of contractors, in particular immediately south of the town wall and for a short stretch in the middle of the park. Information recovered from the trench included: sections through the Roman town ditch, foundation of the Roman town wall and rampart, intervallum road, series of Roman town house foundations and floors, and previously unrecorded street metalling and banks and ditches.<62>

In 1984 a watching brief was held during the construction of a rear extension in the back yard of No 42 St John's Street. Building work revealed a section of the town wall and excavation of two trenches led to the discovery of quantities of medieval and post-medieval pottery. One trench sited by the rear door was dug to a depth of 2.0m below ground level revealing black topsoil which 'was presumably the north edge of the town ditch'. A lens of burnt clay or daub was noted at 1.2m depth.<56>

In 1984 watching brief was held during the construction of a two-storey building at the rear of 37-8 St John's Street (ECC1436). The northern frontage of the new building was built over the remains of the Roman town wall. This was damaged during construction work when it was partly undermined by a trench dug too near to it. The exposed section of the wall appeared to be only core (on the south side).<12>

A watching brief was held during the excavation of three contractors’ trenches on the line of the Roman wall at the site of Rene Robert's Garage. This was undertaken ahead of the redevelopment of the site. The top of the Roman wall was observed at a depth of between 0.79 and 0.89m from the pavement surface.<16>

In 1987 three small trenches were dug to the rear of No.1 Middleborough at the request of the owner. These were excavated to ascertain the true position of the Roman town wall that was encountered in all three trenches.<60>

In 1987 during a watching brief at No.19 St Peter's Street, a section of the Roman town wall was uncovered at the rear of 'Rene Robert's Garage'. The wall had been faced with breeze blocks and concrete and used as a base for oil tanks. A modern addition to the garage on its east side had destroyed the wall. The Roman wall was located behind the original garage building. After demolition of the western building the wall core was exposed in two trenches. The works were visited on several occasions. On 7/5/1987 inspection of a machine trench exposed the side concrete wall and seemed to show that this was built on the Roman wall foundation and that the wall was 'in situ' below the footing of the modern building.<15>

Vaulted Roman drain running through town wall examined in 1988 at 11 Short Wyre Street. Unusually not found at the end of a Street, suggesting that town wall laid out before nearby street grid. The drain consisted of a single arched tile built structure mortared with opus signinum on a yellow-mortared base. The total width of the structure is estimated to be about 1.2m and it is about 1.1m high. The central drain channel was approximately 0.6m wide and just over 0.6m high with a downward gradient of about 1 in 25 north-south. The drain extended approximately 1.3m north beyond the wall face into the town area with a roughly-finished end at that point. Where the drain passed through the town wall a relieving arch of tiles had been constructed over it, creating a double arched structure, estimated to be just less than 2.0m wide.<61>

In 1988 an evaluation was undertaken at 6/7 Church Walk prior to redevelopment of the site. The trenches were located on the line of the Roman town wall. Two trenches were excavated to a maximum depth of 2m, and then probed by augur for a further metre. Neither trench produced evidence of intact masonry and it would seem that the wall (or at least a substantial part of it) had been removed at some stage.<36>

In 1988 a watching brief was held during groundwork for redevelopment at the rear of No 42? St John's Street (ECC1594). This revealed that the Roman wall had been badly damaged.<18><19>

In 1988 a watching brief was held during resurfacing work at the junction of Sir Isaac's Walk and Head Street. Several fragments of Roman wall were uncovered 0.3-0.5m below the ground level. Although these remains had been extensively damaged during the laying of modern services it was clear they belonged to the eastern side of the south-west gate of the Roman town. Part of the west face of the east side of the gate survived intact to a height of 0.7m above the original ground surface. The north-west corners and parts of the northern face of this east side were also located. The face was traced for approximately 1.3m southwards. Some Roman? gravel metalling butted on to the face of the gate. Three other fragments of wall were found which appeared to be in situ. The most important of these was a section of wall which included part of the north face. This lay 1.7m east of the north-west corner of the gate. The shop on the south side of the junction of Sir Isaac's Walk and Head Street has a cellar in which a substantial part of the Roman town wall survives.<17>

In 1989 four exploratory trenches were dug along the north side of a warehouse in Northgate Street in order to locate the Roman town wall. The wall was seen in two of the trenches.<37>

In July? 1989 an evaluation was undertaken in the garden of 89 East Hill, which revealed the Roman town wall. Missing report.<38>

In 1989 during a watching brief at the rear of No. 4 Roman Road (ECC1607) observations were made of the Roman town wall.<21>

In 1989 a watching brief was held during renovations to the ground floor structure of No. 21 Eld Lane (ECC1606). Two trenches (0.6m deep) were dug in the cellar floor and the Roman wall was exposed.<20>

In 1990, trial excavations were carried out at Queen Street Bus Station. These were limited to five small trenches. The second trench adjoined the north side of the town wall at a point 32m east of Bastion V and measured 2.9m by 1.2m. The excavation was extended to a depth of 1.4m (21.2m OD) revealing the walls north face in an excellent state of preservation, although the Roman earth rampart was in poor condition. The third trench adjoined the north side of the town wall at a point 97.5m east of Bastion V and measured 3.2m by 1m. The north face of the wall was revealed and excavation ceased at a depth of 1.4m (20.02m OD). Immediately north of the wall, modern fill was found to extend to within 50mm of the Roman rampart, which was uncovered at a depth of 1.95m (20.47m OD).<39>

In 1990 a watching brief was held during groundwork for the development of an office block on the line of the town wall (ECC1647). An exploratory trench, machine dug by the site contractor, located the inner face of the town wall.<22>

In 1991 a watching brief was held at the Roberts Pool Club on Vineyard Street (ECC1652). Part of the batter of the town wall was recorded.<23>

In 1994 a watching brief was held during rebuilding work at 22a Northgate Street (ECC2086). Removal of adjoining basement floors prior to underpinning works revealed a 7.6m length of Colchester's town wall, including the eastern part of an interval tower aligned with the north-south street between insulae 2 and 3 of the Roman town. AugEring indicated that this section of the truncated monument may survive to heights of between 0.55 and 1.2m above the wall's base offset.<24>

In 1995 a 32m stretch of town wall opposite the Castle Park bowling green was recorded during refacing work. A 1:20 drawing was produced and a set of scaled photographic prints were taken of the exposed core and tile courses.<1>

In 1997 a watching brief at 2a Northgate Street recorded the position of the south face of the town wall, projecting between 0.21 and 0.5m south into the plot occupied by No. 2a.<25>

In 1997 a watching brief at 80 Northgate Street recorded the complete profile across the surviving Roman wall, from the higher pavement level on Northgate Street to the lower level to its north.<26>

In 1997a watching brief was undertaken at St James's Churchyard on East Hill. A series of trial holes were dug in advance of the erection of railings along the east boundary of the churchyard but produced no evidence to indicate the survival of either the town wall or its associated rampart in the uppermost 700mm examined.<2>

In 1997 a watching brief was undertaken during the installation of railings along the eastern side of the churchyard (See also ECC1759). A total of twenty three 30cm x 30cm square pits were dug along the east side of the churchyard in preparation for the erection of steel railings. The depth of the pits varied between 0.35m & 0.55m. Within the limited depths examined there was no evidence to indicate the presence of masonry or rampart material associated with the town wall.<3>

In 1998 several holes for new railings were dug at the top of the steps from East Hill at the east end of St James Church, just behind the Roman wall. The holes were approximately 20cm across by 30cm deep, and revealed only post-Medieval/ modern soil accumulation to that depth.<4>

In 1998 a watching brief was undertaken at 33 Crouch Street during the refitting of 33 Crouch Street as a stretch of town wall is visible at the back (north) of the shop. The wall in the shop is exposed core with about 80cm of outer face removed. The core had been painted over, a sketch was drawn.<5>

In 2016, the upstanding section of the wall in the Old Bus Depot, on the east side of Queen Street and fronting Priory Street Car Park, was surveyed and assessed by Colchester Archaeological Trust (ECC3798). This section of wall was shown to be entirely post-medieval in date - none
of the original Roman structure survives above-ground at this location.<63>

see also: pp.61-4 CAR 6