ESSEX

(1) Ardleigh, Crown Quarry, Old Ipswich Road (TM 02574 29773): an archaeological evaluation in an area of known Iron Age and early Roman occupation containing, among other remains, a D-shaped enclosure and an annexe likely used for industrial purposes. This evaluation has uncovered a similar concentration of features dating to the late Iron Age–early Roman transition, but has also revealed evidence of later occupation during the second and third centuries, including a possible well and a possible trackway. Fragments of worked stone which were recovered suggest the presence of a high-status Roman building in the vicinity. A post-medieval or modern ditch which likely represented the remains of a former field boundary was also recorded.Footnote ⁹⁷

(2) Colchester, a) 159 Maldon Road (TL 98553 24338): archaeological monitoring was carried out in an area of known archaeological remains, south of the Roman road from Colchester to Gosbecks and in close proximity to the sites of a number of Roman burials. The partial remains of two straight-sided Roman features were revealed, one of which contained a flanged bowl and folded beaker. Although no human remains were recovered, it is highly likely that both features are graves of the Roman period.Footnote ⁹⁸

b) Hedge Drive (TL 97538 23073): an archaeological evaluation on the projected route of the Colchester to Gosbecks Roman road was targeted to locate any surviving remains associated with the road. The evaluation successfully located the remains of the Roman road, which consisted of four ditches, set out as two pairs, defining narrow footways on either side of a central carriageway. The footways were 2.1 m and 2.6 m wide, and the central carriageway c. 6.2 m across. The roadside ditches are well defined at 1.05–1.45 m wide and 0.28–0.44 m deep. The road surface had not survived.Footnote ⁹⁹

(3) Elmstead, Land north of Clacton Road (TM 07300 23885): an archaeological evaluation in advance of development was undertaken on an area of cropmarks. The most significant discoveries were two features containing considerable quantities of early Roman pottery wasters and kiln debris. This would indicate the likely presence of pottery kilns on the development site, and therefore the presence of an early Roman pottery industry at Elmstead. Ditches also crossed the site, likely forming field boundaries or enclosures. It is highly likely that both the pottery industry and field boundaries are associated with archaeological remains to the south of the Clacton Road. Interestingly, though, aside from the pottery wasters and kiln debris, very few Roman finds were recovered from the evaluation, suggesting that the site is located away from the main focus of settlement. Also identified were the roadside ditches of a trackway or droveway, recorded initially as a cropmark and first investigated during excavations south of Clacton Road, where it led to the larger of the two late first- to third-century enclosures.Footnote ¹⁰⁰

(4) Epping Forest District, Hobbs Cross to Harlow Roman Road: Harlow was an important Roman site and temple located at a crossing of the River Stort, a tributary of the River Thames.Footnote ¹⁰¹ A connecting road from Margary 30 was therefore to be expected. A section of such a road was suggested by KembleFootnote ¹⁰² and with the release of full LiDAR data for this area there is now visible evidence for most of the route, although across Harlow we are reliant on interpolation. The route (fig. 25) is based on two main alignments (possibly a third for Harlow) with the major turn and setting-out point on the high ground just north of Thornwood Common (TL 46288 05720).

FIG. 25. LiDAR Image showing the route of the road from Hobbs Cross to Harlow. (Base LiDAR data © Crown Copyright. Inset Location Plan – base mapping Ordnance Survey Opendata)

At its southern end the road branches off the Great Dunmow to London Roman Road (Margary 30) at Hobbs Cross (TQ 47600 99094) and there are initially a few minor deviations to negotiate a couple of valleys, but beyond these the straight alignment to Thornwood begins (TL 47262 00802).

There are more than enough clues to confirm this first alignment (TL 47033 02080), although surprising in Epping (Lower) Forest, where it would be expected for the agger to have survived, there is only the faintest trace in the Digital Terrain Model data (TL 46823 03150). However, beyond Epping Forest there is a continuing string of clues (TL 46758 03404) establishing the first main alignment. This stretch to Thornham Common was the section suspected as Roman by Kemble based on aerial photos and is therefore now confirmed by LiDAR (fig. 26).

FIG. 26. Oblique 3D LiDAR Image looking towards Epping and Harlow from above Hobbs Cross. (Base LiDAR data © Crown Copyright)

The route across Latton Common is clear in LiDAR imagery but not so obvious on the ground. It is, however, the last secure evidence heading north (TL 46685 081549). The route onwards becomes uncertain due to modern development of Harlow with possibly just a single indication at (TL 46927 09086). The old road through ‘Old’ Harlow (London Road) is straight and looks very tempting, as are the local names of Potter Street and Latton Street. However, these are all a little to the east of the Latton Common alignment and would require a deviation to reach them. Without any visible evidence, then, it is prudent to regard the length across Harlow as still to be proven.Footnote ¹⁰³

(5) Felsted, a) Watch House Green (TL 69198 21101): investigations identified three phases of activity. The earliest occurred during the late Iron Age to early Roman period, which was represented by one ditch (F18) and three gullies (F14, F19 and F21). Probable field boundary ditch F18 produced a large assemblage of late Iron Age and early Roman pottery. Ceramic evidence suggests that there was then little activity on the site until the late second century, with the main phase of occupation probably focused on the third century and likely continuing into the fourth century, as a coin dated to a.d. 330–335 was recovered from the upper fill of one of the ditches during the evaluation. Pottery sherds deriving primarily from jars, bowls and dishes, along with amphorae and pieces of storage jar, show that the ceramic assemblage from this phase was dominated by cooking vessels, with table wares for presentation and drinking quite rare. Although small in size and quantity, fragments of Roman brick, tegulae and flue-tile from ditches and the area of metalling suggest a building of some significance must have existed in the surrounding area. The single piece of vessel glass and a key also allude to the presence and use of some higher-status material. Although the ground conditions did not allow for the survival of much animal bone, tooth fragments from both cow and horse were recovered from features, and the recovery of fragments of quernstone show that grain was being processed on or close to the site. Overall, the evidence would suggest an agricultural site located close to a Roman farmstead or small villa. Located 1.8 km south of the main Roman road running from Colchester to Braughing (Stane Street), previous investigations in the area have revealed a number of farmsteads and villas spread out on both sides of this road. Evidence from the archaeological excavation of this development site therefore adds to this growing picture of settlement and land-use between the Roman towns of Great Dunmow to the west and Braintree to the east.Footnote ¹⁰⁴

b) Land off Station Road (TL 6713 2051): excavation identified early Roman features including six c. first- to second-century a.d. cremation burials that may have been associated with a contemporary villa site 750 m to the north. These burials were dispersed across the two excavation areas, some of them located within a possible agricultural enclosure. The overall impression is of an area of Roman fields and agricultural enclosures that were also used for burying at least some of this community's dead. One of the cremations included a concentration of hobnails from a pair of shoes; another contained a copper-alloy stud.Footnote ¹⁰⁵

NORFOLK

(1) Snettisham, Land at Poppyfields Drive (TF 6788 3383): excavation confirmed the presence of a dense network of ditches and enclosures likely forming agricultural plots near a Roman settlement. These were dated to the first to fourth centuries, with a peak of activity in the mid-second to mid-third century a.d. The site lies immediately north of a complex network of cropmarks which have been identified as Roman activity and just to the west of a Romano-British industrial site which included a Roman settlement, jeweller's hoard and upstanding pottery kilns.Footnote ¹⁰⁶ The main groups of features which have been identified appear to reflect different functional areas within what was the core area of a small farmstead, related to the processing of agricultural produce (cereals and animals) as well as small-scale craft activities.

A large pottery assemblage along with small quantities of other finds including animal bone, charcoal and cereal grain indicate a rural scatter in the vicinity of a large Roman settlement. The pottery assemblage included wasters and seconds, which were likely produced in the kilns at the neighbouring Strickland Avenue site,Footnote ¹⁰⁷ as well as imported Gaulish samian ware including some with potters’ stamps. Other interesting finds included a green-glazed Roman sherd and a possible pipeclay figurine.

The remains identified on site are of local significance and make a valuable contribution to our understanding of Roman patterns of settlement and land use in the region, as well as providing a few finds of interest.Footnote ¹⁰⁸

SUFFOLK

(1) Bacton, Land north of Church Road (TM 04 67): following trial trench evaluation, three areas of the site were targeted for excavation, revealing multi-period archaeological remains. The sequence of activity began in the Middle Iron Age with part of a farmstead comprising two roundhouses and associated pits. In the early Roman period, an oven or kiln and two timber buildings, one a small structure with a rammed chalk floor, and the second a larger, barn-type building, indicate either a new farmstead or a continuation of the later Iron Age settlement. By the mid-Roman period, the farmstead appears to have gone out of use but there was evidence for continuing agricultural use of the area. A copper-alloy dress pin, similar to examples from Staunch Meadow, Brandon, hints at low-level activity here in the c. eight–ninth century a.d. By the ‘high’ medieval period (later twelfth to fourteenth centuries) there was widespread evidence for agricultural activity across the site, with indications of occupation both close to the present village, to the south, and to the west, along the Wyverstone Road. Activity continued at a lower level in the late medieval and post-medieval periods.Footnote ¹⁰⁹

(2) Elmsett, Land off Hadleigh Road (TM 0553 4635): excavation identified an early Roman (mid-first to mid-second century a.d.) farmstead represented by a complex set of ditched enclosures that showed several phases of changing organisation, albeit respecting similar alignments. The remains of a rectangular building with post-hole and slot foundations were present within the enclosure system at the southern edge of site; three ovens were recorded further to the east. A notable assemblage of pottery, animal bone and other artefacts was recovered from a ditch that formed the eastern boundary of the farmstead.Footnote ¹¹⁰

(3) Halesworth, Land at Chediston Street (TM 3814 7717): the excavation of a c. 9.1 ha site revealed evidence for sustained Roman settlement beginning during the second century a.d. when the hillside was divided up into ditched fields and enclosures for growing crops and pasturing livestock. Few finds were recovered from this initial period of Roman activity, which suggests the farm to which they belonged was located some distance away from the fields. An urned cremation burial was found dug into one of the ditches.

The field ditches gradually silted up and were replaced by a single large rectangular enclosure at the top of the hill (fig. 27). Inside the enclosure was a masonry bathhouse consisting of two small rooms with flint cobble foundations, one with pilae stacks indicating the presence of a hypocaust system (fig. 28). One room was a heated caldarium and the other room a cold room or frigidarium. Just outside the bathhouse was a small rectangular tank, built of flint cobbles and lined with opus signinum. This tank was constructed to collect water from a natural springhead, as a narrow channel survived which could have delivered water to the bathhouse. A well, more than 5 m deep, was located close to the bathhouse. A large rectangular pit, lined with clay, was also found nearby and was presumably used to hold water. Discerning the precise function of these various water management features and their relationships with the bathhouse will form part of post-excavation analysis. Roman bathhouses are rare in Suffolk, thus suggesting a settlement of some wealth and status. A well-preserved pottery kiln (fig. 29) shows that the inhabitants of this farm or estate centre made some of their own pots; wasters from the final, failed firing were still present.

FIG. 27. Halesworth. Land at Chediston Street view east (drone photography: Ryan Desrosiers).

FIG. 28. Halesworth. Roman bathhouse, north to top (drone photography: Ryan Desrosiers).

FIG. 29. Halesworth. Roman kiln, view south-west (photo: Fabian Danielsson).

In the fourth century, the bathhouse and kiln fell from use and the bathhouse was demolished, with brick, roof tile and other building materials being used to backfill many of the features on the site, including the ditches, pits and the bathhouse itself. Most of the finds recovered from the site come from this period of demolition and backfilling. Further evidence for the wealth of the occupants can be seen in an array of high-status metalwork and other small items, including coins, glass beads, zoomorphic brooches, a personal grooming kit, a silver spoon and a large number of copper-alloy hair pins. There was a significant concentration of such objects in the large clay-lined water tank, indicating deliberate, ritual deposition.

One wall of the bathhouse was reused as the short end of a rectangular, post-built barn, constructed to store grain and other cereals in the late Roman period. After this the structure went out of use and there was little archaeologically visible activity on the site until the modern period.Footnote ¹¹¹

(4) Long Melford, a) Land adjacent to the Chad Brook (TL 8585 4540): excavation, following the discovery of a Roman road leaving the settlementFootnote ¹¹² to the north-west, and confirmed its crossing point on the Chad Brook. A trench across the probable road course found a wide causeway where the road crossed this low ground adjacent to the Chad Brook. Reinforced with iron smelting slag, it also included evidence of wooden inclusions together with fragments of timber from its waterlogged side ditches. Evidence suggested that the ditches could have been opened in the Iron Age but backfilled to reclaim land. Subsequent deposits from both sides of the causeway returned numerous small finds including three finger rings, two lead-alloy curse scrolls and 81 coins. The majority of the coins were of mid- to late fourth-century date, and were supplemented by late Nene Valley pottery sherds. In addition, the presence of a building was indicated to the east side by a large quantity of CBM. The road's path could then be traced for another kilometre with the aid of aerial imagery.Footnote ¹¹³

b) Land adjacent to Cricket Ground (TL 860 454): an excavation following a Community Test Pit Project undertaken in 2013 revealed an area of a gravel-topped chalk yard and a chalk floor, divided by a linear truncated robber trench together with an associated drainage gully and reclamation evidence. A further small excavation extension revealed a threshold of a wide doorway, and removal of the trench fill indicated the original building foundation probably consisted of large flint nodules. A large timber post impression was revealed within the trench terminal and was interpreted as representing a doorpost. The skull of a dog, which was deposited on the basal fill of the robber trench, was interpreted as a possible ritual deposit made before the backfilling was completed. The fill, which incorporated evidence of burning and appeared to represent a single operation, produced an Iron Age coin along with a Gallo-Belgic sherds. It was concluded that it represented a late Iron Age rectangular building of some status which was partly burnt down and demolished in the middle years of the first century a.d.Footnote ¹¹⁴

(5) Stowmarket, Land south of Union Road (TM 0291 5878): excavation, following an earlier evaluation, recorded an extensive system of ditched boundaries of late Iron Age to Roman date. Although the site did not contain direct evidence of occupation, there were clear indications of settlement nearby, including extensive quarry pits, deposits of domestic refuse, a small, ancillary agricultural structure, and several ovens. This evidence suggests the site was near to an affluent or high-status settlement on the higher ground to the north-east. The associated finds assemblage includes significant metalwork objects, such as a rare Bury E typeFootnote ¹¹⁵ silver coin with a unique obverse die, datable to c. 40–35 b.c., a complete, but fragmentary c. fourth-century Roman pewter pedestal-based bowl and a round seal box with red enamel decoration. The site is on a south-facing slope down to the Rattlesden River.Footnote ¹¹⁶