Monument record 1080/1/0 - Roman fort, Harbutt's Field

<1> Thompson Watkin W. T., 1886, Roman Cheshire: A Description of Roman Remains in the County of Chester, 243-245 (Book). SCH2878.

Harbutt's Field (Watkin Fields Nos. 22 & 28). Reports in the C18 & C19 identified this field as the site of a Roman fort, situated at the junction of the Rivers Dane & Croco, but present opinion suggests the features here are purely natural. 1750 - Mr Ralph Vernon described the site as 'regular a Roman camp as any in England'. Whitaker described the site as a parallelogram of 10 acres, bounded by natural steep slopes and the rivers Dane & Croco on the north & west, a ditch on the east, and a steep ditch c.lO yds deep and 8 yds wide at the top. Archdeacon Wood similarly records the site and notes the finding of coins, pottery and other artifacts during ploughing and levelling

<2> Thompson, F.H., 1964, A History of Cheshire. Volume Two: Roman Cheshire, /91-92 1965 (Book). SCH2862.

In 1921 Professor Donald Atkinson excavated here, but failed to find any evidence of military occupation

<3> Various, Various, Oral communication to the HER, Stubbs D A 1986 (Oral Communication). SCH2330.

In the 1950s Manchester University excavated here, but no report was published.

<4> Gifford and Partners, 1993, Report on an Archaeological Evaluation Carried Out on Land Between the River Croco and King Street, Middlewich, Cheshire., R2103 (Client Report). SCH4183.

Remains of the fort were positively identified through a programme of archaeological evaluation in 1993 in advance of a residential housing development. The archaeological work comprised a desk-based survey, geophysical survey and trial-trenching.

The geophysical survey (resistivity and magnetometer) was carried out by Stratascan in August 1993 over two fields - Harbutt's Field and Field 2981 to the south. The resistivity survey was more successful revealing four sides and corners of a Roman marching fort/camp with opposed entrances in Harbutt's Field. The possible existance of a fort here had been postulated by eighteenth and nineteenth century antiquaries. There was no firm evidence for buildings or roads within the camp but the linear lower resistance features possibly represent drainage or enclosures contemporary with the camp. The magnetic survey supported the findings of the resistance survey although generally it was less productive. This may be the result of considerable depth of overburden on the site which was to a depth of approximately 700-800mm. No recogniseable features were located in the more southerly field, 2981. The geophysical report is included as Appendix E in the wider report for the programme of work SCH4183, R2103.

Following on from the geophysical survey a series of trial trenches were excavated between August and September 1993. Two of the ditches were located across the defensive perimeter of the fort and a third within its interior; these trenches confirmed the existance of the fort. It is a roughly square enclosure circa 125m (N-S) by 113m (E-W) with rounded corners and entrances in each side. Possible, entrance-works were also located by the geophysical survey inside the north gate. Trenches 1 and 2 located across the north and east defences of the fort, both confirmed that the fort was surrounded by a single rampart and ditch. Additionally, both contained indications that the camp's outer defences had been re-cut in antiquity. In trench 3 a beam slot was recorded indicating the possible existance of timber structures within the fort. None of the three trenches produced reliable dating evidence, the little pottery recovered was generally from mixed layers containing late first and second century material as well as medieval and post medieval pottery, however, the position of the fort in relation to King Street to the east and the Roman settlement to the south suggests that it may have precended both indicating a construction date in the first century AD. The lack of artefactual evidence from the three trenches dug across the fort suggest a short occupation, although the presence of potential timber structures, indicates an occupation of at least some months. Five further trenches were located in Field 2981 to the south of Harbutt's Field and the fort but of all these only one contained a single pit containing some burnt bone and this was not dateable. Trenches 6 and 7, the most southerly of the trenches, contained the most residual Roman material. This evidence would suggest that the main Roman settlement at Middlewich lay further to the south.

<5> English Heritage, Various, Schedule Entry (Scheduled Ancient Monuments Amendment), 12615 (Scheduling Record). SCH2950.

A Roman fort had been postulated to exist at Salinae since the 18th century but was only positively identified by field work during 1993. No upstanding remains survive but resistivity survey work, which measures the way in which electrical currents pass through the ground and uses this information to provide an insight into the nature of below-ground remains, has confirmed the location of the site and also that extensive remains of the whole fort survive beneath the present ground surface.

The fort is a roughly square enclosure, measuring approximately 110m by 125m with rounded corners and entrances visible in the middle of all four sides. There is evidence of a structure located within the northern gateway. The enclosure was surrounded by a single rampart and ditch. There are further features outside the rampart and ditch which may represent ancillary features such as the roads leading to the site, and a possible outer ditch on the north side of the fort. A number of internal features have also been identified.

The geophysical survey work was followed by limited excavation. This revealed that the ditch survives as a buried feature, as do postholes which relate to the construction of the rampart. Within the interior of the fort, deposits of burnt clay were found along with evidence of the slots in which timber beams used in the foundations of the Roman buildings would have been set. Pottery fragments found indicate that the fort was in use in the late first century AD.

<6> Gifford and Partners, 1996, King Street, Middlewich, Harbutts Field: Excavation of Two Sample Trenches., R2104 (Client Report). SCH4182.

Two trenches were dug 25m away from the west edge of the fort in 1996, prior to landscaping the area. No archaeological deposits were found to a depth of 0.6m, the trenches were not dug deeper

<7> Giffords Consulting Engineers (Chester), 1997, Report on an Archaeological Watching Brief: Harbutts Field, Middlewich, R2202 (Client Report). SCH4187.

An archaeological watching brief was undertaken in 1997 at Harbutts Field, Middlewich, in order to supervise the depth of penetration of ploughing activity. Field walking after the ploughing was undertaken to recover any artefacts revealed, potentially related to the Roman fort. Romano-British artefacts were recovered including fragments of pottery, brick/tile, and clay daub from buildings. A scatter diagram of the finds was prepared which proved a general distribution with no particular pattern. The discovery of brick/tile and daub indicated the fort was not just a temporary encampment. Nevertheless, the pottery idicates a relatively short occupation, possibly no more than two decades in the second half of the first century AD.

<8> Chester Archaeological Society, Journal of the Chester Archaeological Society, Vol 83, 2009, p.37-93 (Journal/Periodical). SCH1595.

Article assessing the current state of knowledge for Roman Middlewich, including the fort, vicus and road network, based on evidence from a series of developer-funded archaeological evaluations occurring in the 1990s and 2000s, but also drawing from the results of the significant body of work undertaken previously.

The 1.4ha fort in Harbutt's Field represents the earliest structural evidence for Roman occupation at Middlewich. Given the lack of datable material from the excavations within the fort, it can only be dated tentatively by finds from the extramural settlement. The coins and samian agree in suggesting a Flavian (AD69-97) start for occupation. Minimal excavation on the site of the fort also means that very little is known of its internal layout and nature of occupation.

The existance of a contemporary road leading to the southern gate is indicated by the presence of a single, strip building, fronting onto the road, just outside the southern entrance of the fort. Additionally, the road itself was detected further to the south. No further roads entering the fort have been identified. Additionally, the nature of the crossing of the River Dane is unknown. The strip building at the southern entrance appears to have fallen out of use circa AD 130, perhaps also indicating the end of use of the fort. This is supported by the coin evidence which demonstrates a marked decline in the value of average coin loss across the whole settlement after AD 138.

Along with the building at the southern entrance, only one other building of the same phase as the fort was excavated in the area to the south, both buildings had a related plot. However, in the area to the south of the fort, the second phase of settlement in the 2nd century saw an increased number of buildings which may overlap with the life of the fort. Boundaries of fields have also been recorded aswell as a pottery kiln. Only one possible stone building has been discovered by Bestwick (site C) where a stone wall footing was found (but only referred to in the artefact records). However at the same site discoveries of box flue tile, window glass and unguent bottles possibly indicate the presence of a bath house. Possible boundary ditches delimiting the extramural settlement to the south of the fort were discovered in three of the excavated trenches. The fill of two of these contained material dating to the late 1st and early second century, the fill of the third trench has been dated to the 3rd or 4th century perhaps indicating the boundaries were used throughout the life of the settlement.

The main activity in the extramural settlement in the 1st and 2nd century appears to have been salt production. Brine wells/pits and brine hearths have been found at several sites and one, possibly two, brine tanks have also been recorded. Some of the pits seem to have gone out of use, and were being filled in with rubbish, by the early 2nd century. Pottery production is also attested in this period by the kiln and also wasters found at one site. There is also some evidence for secondary iron-working.

Evidence of burials for the settlement were recorded and include two urned cremations of the late 1st or 2nd century date, a 3rd century urned cremation discovered in a roadside ditch, and two burials found within the settlement (one an infant contained in an amphora, the other an urned cremation of 3rd century date).

In summary, the main impetus for the establishment of the Roman settlement at Middlewich was the construction of the fort in around AD 70, probably to accommodate a small auxiliary infantry unit. Whether the choice of site was influenced by the presence of brine springs is not clear. It may be that the known fort may have been preceded by some form of campaign camp but this is unclear. From the recent investigations it would appear that Middlewich was a military vicus. It could represent a 'street type' vicus where the buildings were concentrated on either side of the road leading to the main gate. Another form of vicus is the 'tangent type' where the major road by-passed the fort for topographical reasons. At Middlewich the orientation of the fort has yet to be established and it is not known whether its main gate lay to the south or to the east adjacent to the Roman road King Street. The investigations have shown that King Street was partially realigned , probably in the second quarter of the second century, by-passing it in the manner of the 'tangent type' vicus. It is possible that the settlement outwith the fort may not represent solely a vicus, but could also represent at least in part an separate military annexe to the fort, if this were the case, it is thought that for Middlewich the most likely location for an annexe would be to the east of the fort, or less likely, to the north.

The ordered arrangement of roads, property boundaries and buildings seems to imply a high degree of planning in the initial layout of the settlement and subsequent development. Remains for numerous timber structures have been found throughout the settlement, the majority of which appear to be strip buildings. No in situ stone buildings have been identified, although the presence of one is inferred through Bestwick's work. There are notably fewer buildings of third and fourth century date, together with a markedly reduced pottery assemblage. This may indicate a declining population and/or economic changes bought about by military changes. It was probably after the end of military occupation and the decline of the fort that Roman King Street was realigned to bypass the fort to the east rather than approach its southern gate. In the mid second century the agger and ditch of the former southern approach road were overlain by a metalled surface atrributed to a re-organisation of land holdings.