MSO6804 - Roman Lode Ironworkings (Monument)


The remains of medieval or early post-medieval ironworkings, known as Roman Lode, are visible as four shafts and a number of adits, as well as spreads of spoil and dumped material. Prehistoric and probably Roman mining also took place on the site.

Please read the Exmoor National Park Historic Environment Record .

Type and Period (4)

Protected Status

Full Description

SS 752 382 Remains of "early" iron workings are visible at Burcombe. The iron deposits on Exmoor have been worked since Roman times and a systematic attempt to exploit them was made in the third quarter of the 19th Century. [1] Centred at SS 753 382. A new survey at 1:1000 scale was undertaken of the openworks known as Roman Lode at Burcombe. These workings form part of the extensive abandoned iron workings in the Burcombe Valley. These openworks, some 660 metres in length, are abandoned and silted. Four shafts and a number of adits are traceable as well as spreads of spoil and dumped material. The survey was undertaken by the RCHME Exeter office in the autumn of 1997 and was commissioned as part of the Exmoor Ironworking Project. The full survey and report are deposited in the NMR Archive: collections number 1115992. [2-5] In 1550 Micheal Wynston was given license to dig for iron ore and may have been responsible for the "Roman mine" on Burcombe. [24] A detailed survey was undertaken by RCHME in October 1997 of the main working from SS 7481 3832 to SS 7547 3812. The quantity of waste material evident on the surface appears to be insufficient so that much must have been dumped in old workings, obscuring them. From SS 7481 3832 to SS 7492 3829 there are a number of prospecting pits on the crest of the hill slope. There are also 4 larger pits with spoil on the slope and 3 at the foot (shown on 1889 Ordnance Survey map). Centred about SS 7493 3828 are dumps of material around the stream covering the sites of pits shown on the 1889 map. Earthworks of the track, a tramway and (probably) the line of the flat rod line powering the Burcombe Valley Shaft are visible here. From SS 74963 828 to SS 7503 3825 is a series of pits on the steep west facing slope and a possible adit. These have been filled with material washed from a large pit at SS 7504 3824. This is 11 metres in diameter within a fenced area and possibly that sunk in 1910. It has a smaller pit (0.6metres in diameter) adjacent. From SS 7506 3824 to SS 7516 3819 is a sinuous and broad section of openwork with three silted pits, each with a spoil heap, and a possible blocked adit (SS 7515 3819). To the south is an area of linear and circular spoil dumps. It is not clear whether these have been reworked. From SS 7515 3819 to SS 7539 3812 is the visually most impressive stretch of the workings. To the north there is a spread of mounds and banks which may represent reworking of old spoil. Within the trench are some infilled and silted pits and possible 4 infilled shafts can be identified. Two of these lie at SS 7517 3818 and SS 7518 3816 and are c. 14 metres diameter with spoil on the north. The shaft at SS 7519 3818 is visible as a large pit (5.5 metres across) with spoil to west and north. It is approached from the south by the tramway built in 1912. From SS 7539 3812 to SS 7547 3812 the openwork is broad and shallow with spoil dumped on the north. [11] The extensive openwork known as Roman Lode is clearly visible on aerial photographs of Burcombe, Exmoor. In addition, to the north of the openwork a number of hummocks and pits can also be seen. [12, 25] Excavations at these pits as part of the Exmoor Ironworking Project in the early 21st Century established a radiocarbon date dating to the mid Bronze Age. It is possible that copper was also exploited here prior to iron working, and if that is the case, it is also conceivable that iron has been extracted here throughout the Iron Age, Roman and early medieval periods as well. Certainly the size and complexity of the workings indicate that a long period of exploitation occurred here. The form of the openwork itself suggests a medieval or early post-medieval date, and it is likely that the majority of the workings date to this period. [13] A gradiometer survey was carried out at Roman Lode in June 2002, in advance of, and to inform, excavation as part of the Exmoor Iron project. The survey distinguished a number of magnetic anomalies including one area of potential heating which was recommended for further investigation by excavation. [14] Small scale excavation took place at Roman Lode as part of the Exmoor Iron project. The primary objective was to assess the possibility that Roman Lode was the source of ore processed at the ironworking site at Sherracombe Ford. Three seasons of excavation took place, in 2002, 2003 and 2004. The 2002 excavations targeted a hollow within a roughly triangular area of irregular earthwork hollows to the eastern end of the openworks. It was thought these might represent early and small scale areas of ore sorting, potentially contemporary with Sherracombe Ford. Excavation revealed this not to be the case but revealed bell-shaped pits thought to be ore extraction pits. The 2003 excavations continued to investigate the excavated bell shaped pit as well as a neighbouring hollow, which proved to contain a second similar pit. The 2004 season targeted the magnetic anomaly previously identified by geophysical survey. The anomaly was revealed to be a burnt feature of uncertain function up to 1 metre in diameter, that probably predates the creation of the bell shaped quarry pits. Two charcoal samples from the burnt feature were radiocarbon dated to the Bronze Age. The pits themselves are suggested to be at the earlier end of a probable maximum Iron Age to medieval date range. Excavations did not demonstrate a connection between this site and Sherracombe Ford, but it was not disproved and Roman Lode remains possible source for the Iron Age to Roman bloomery. [14, 16] The charcoal sample taken from the burnt feature at Roman Lode indicates extensive late Bronze Age exploitation of oak, with less frequent use of birch, hazel and possibly heather. [10] A deep blanket peat deposit c. 100 metres to the southeast of Roman Lode was sampled in August 2004 with the aim of providing an environmental context for the site for the periods prior, during and following the iron extractive activities at Roman Lode. Specific aims included; pollen analysis to examine changes in the local environment; assessment of the deposition of atmospheric pollutants associated with iron extraction; determining the local impact of the industrial activity on the landscape, for example through peat deposition and vegetation management. Radiocarbon dating of the peat samples indicate peat initiation began during the late Iron Age (350-115 cal BC) and continued into the modern period. The pollen data provides an unbroken sequence from the Iron Age, indicating Roman Lode was situated within a predominantly open upland heath landscape with changes in vegetation between a species rich grassland landscape towards the species poor dry heath and blanket mire, similar to the landscape of today. The vegetation record shows no evidence of any human impact from the iron extraction activity at Roman Lode. [17] Two geochemical surveys were undertaken at Roman Lode. The first survey has been compared to the magnetometer survey results. The second higher resolution survey targeted the area to be excavated. The low resolution sample produced little archaeological information. The high resolution survey was successful and compared well to the excavation results, but ultimately this site proved to not be a metal working site and the geochemical techniques technique provided little new information. [19] Radiocarbon dates were obtained of 1950-1750BC. It is highly probable that Bronze Age Roman Lode was a ridge of quartz rock. [21] It is possible that the Bronze Age activity at Roman Lode was not only concerned with the extraction of metals, although the presence of copper suggests there may have been a small and now completely mined out source of copper. It is possible that the ores at Roman lode were mined for colour pigment, the symbolic meaning of colour in the Early Bronze Age is well attested, the iron ores could have produced red, brown and yellow pigments and the copper ores green and blue pigments. Another possible material being mined at Roman Lode is white quartz, which can be seen incorporated into contemporary monuments such as White Ladder stone row (MSO6810). [23] Roman Lode is recorded as having been active at some point between the years 1853 and 1859. [26] During a walkover survey by Cornwall Council, a tight cluster of 6 to 7 pits and associated spoil dumps cut into the break of the slope was noted at SS 74842 38315 at the head of the steep descent to the valley bottom. The pits are 4 to 5 metres by 3 metres, with some smaller; the downslope spoil heaps are up to 0.75 metres high. Some spoil dumps appear to partly overlie earlier pits and there are probably several phases of working. These pits are on the axis of the Roman Lode, clearly visible on the facing hillslope, and the disturbance can be seen to continue beyond the boundaries of the survey. Another pit was noted nearby measuring 2 metres across. [27] A follow up survey by the Mires Historic Environment Officer confimed that the features noted by the previous survey [27] correspond to the Roman Lode Ironworkings. [28] An example of an openwork trench or dyke following the iron lode is at SS 75 38 near Cornham Ford. It is about 600 metres long and was originally more than 15 metres deep, now about 8 metres deep after infilling. A shaft would be sunk on to the outcrop of the lode until water ingress became too great, when a new shaft would be sunk a few metres along the outcrop. This would leave a section of ground between the two that would be removed later, creating a continuous trench with sides inclined to follow the dip of the lode. It would have occurred intermittently on a small scale and may have produced as little as one tonne of iron a year. It may have been in operation between the 2nd or 3rd Century BC and the 11th Century AD, when Exmoor became a Royal Forest and casual opencast working was no longer possible. 19th and early 20th Century workings at the site are also discussed in the report; these have been recorded as child Monument records and linked separately. [29] This record was enhanced as part of the National Record of the Historic Environment to Exmoor National Park Historic Environment Record data transfer project. [30] Radiocarbon dates from samples from the site and from a peat section to the southeast submitted between 2002-6 have been published by Historic England. [32] The mine is mentioned in a publication on the industrial archaeology of Somerset. [33] A newly visible shaft was noted October 2023 at SS 75101 38221, any original capping is believed to have collapsed. It has been fenced off. [34] The heritage as the importance of it's age is high with complex prehistoric origins to 19th century activity. It was also considered highly rare being quite rare to find evidence for mining activity over such an extensive period on one site (not continuous). It was assessed as having a high historical association being associated with Knight Estate exploitation of resources. It's evidential value is high being a rare example of early mining on Exmoor with high level of archaeological evidence surviving plus association with peat deposit in wider vicinity. It was decided to add the asset to the Local Heritage List. [35]

Sources/Archives (35)

  • <1> Monograph: Buchanan, C.A.and Buchanan, R.A.. 1980. The Batsford Guide to the Industrial Archaeology of Central Southern England. Batsford. 174.
  • <2> Technical drawing: Fletcher, M., Probert, S. and Newman, P.. 1997. Roman Lode/ink survey . 1:500. General: Permatrace. Pen and Ink.
  • <3> Collection: Fletcher, M., Probert, S. and Newman, P.. 1997. RCHME: Roman Lode, Exmoor Forest.
  • <4> Report: Fletcher, M., Probert, S. and Newman, P.. 1997. Roman Lode . RCHME Swindon. General: Paper. Typescript.
  • <5> Unpublished document: Fletcher, M.J.. Field Investigators Comments. RCHME Field Investigation, 1 December 1997.
  • <6> Report: Juleff, G.. 1997. Earlier Iron-Working on Exmoor: Preliminary Survey. 30, table 1..
  • <7> Report: Juleff, G.. 1999. Greater Exmoor Early Iron-Working Project: Summary of Activities 1997/98 and 1998/99. 3.
  • <8> Report: Juleff, G., Rippon, S. + Wilson-North, R.. 2000. Greater Exmoor Early Iron-Working Project: Project Outline. 8.
  • <9> Report: Juleff, G., Rippon, S. + Wilson-North, R.. 2001. Exmoor Iron: An Exploration of the Impact of Past Iron Production on the Environmental and Cultural Landscapes of Greater Exmoor: Project Design. 12.
  • <10> Report: Gale, R.. 2005. Exmoor Iron Project - Brief Summary of Results from Charcoal Analysis.
  • <11> Report: Fletcher, M. 1997. Roman Lode, Burcombe, Simonsbath, Exmoor. Unpublished RCHME report. Unpublished Report.
  • <12> Aerial photograph: Various. Various. Vertical Aerial Photograph. NMR OS/96507 42-43 (30 March 1996).
  • <13> Verbal communication: Various. 1993-. Exmoor National Park Historic Environment Team staff comments. L.Bray, 1 December 2008, Exmoor National Park Authority.
  • <14> Report: Dean, R.. 2003. A Gradiometer Survey at Roman Lode, Burcombe, Exmoor, Somerset.
  • <15> Report: Bray, L.. 2002. Roman Lode Excavation 2002.
  • <16> Report: Bray, L.. 2005. Roman Lode Stratigraphic Report 2002-2004.
  • <17> Report: Fyfe, R.. 2009. Roman Lode, Burcombe, Exmoor, North Devon: Pollen analysis of blanket peat deposits.
  • <18> Monograph: Bray, L.S.. 2006. The Archaeology of Iron Production: Romano-British Evidence from the Exmoor Region. 167-170.
  • <19> Monograph: Carey, C.. 2005. Geochemical Survey and Metal Working on Archaeological Sites. 107-112.
  • <20> Report: Dean, R.. 2004. Exmoor Iron Methodology for the Magnetic Mapping of Iron-Working Sites.
  • <21> Article in serial: Wilson-North, R.. 2005. Not so Roman Lode. Exmoor Park Life. 17.
  • <22> Article in serial: Charters, A.. 2002. Major Excavations on Exmoor Proves Romans Were Here. North Devon Journal.
  • <23> Article in serial: Bray, L. & Juleff, G.. 2007. Minerals, Metal, Colours and Landscape: Exmoor's Roman Lode in the Early Bronze Age. Cambridge Archaeological Journal. 17. 3.
  • <24> Monograph: MacDermot, E.T.. 1973. The History of the Forest of Exmoor. David and Charles Limited. Revised Edition. 223.
  • <25> Archive: 2007-2009. Exmoor National Park NMP: SS 73 NE. MD002190.
  • <26> Monograph: Siraut, M.. 2013. A Field Guide to The Royal Forest of Exmoor. Exmoor National Park Authority. 11.
  • <27> Report: Kirkham, G. and Taylor, S.. 2012. Burcombe and Wester Emmetts, Exmoor: Archaeological Survey. Cornwall County Council Historic Environment Service (Projects). p17-18.
  • <28> Report: Bray, L.S.. 2014. Archaeological walkover surveys on Burcombe: May - October, 2012. Exmoor National Park Authority. p13.
  • <29> Report: Exmoor Mines Research Group. 1995. Report on the safety condition of disused mine workings on lands owned by Exmoor National Park Department and other lands nearby. Exmoor Mines Research Group. 2, 6-11.
  • <30> Digital archive: Historic England. Various. National Record of the Historic Environment (NRHE) entry. 35006, Extant 17 November 2021.
  • <31> Report: Matthews, I.. 2008. Roman Lode, Exmoor, Devon: Tephrochronology. Scientific Dating Report.
  • <32> Monograph: Marshall, P. et al. 2020. Radiocarbon dates from samples funded by English Heritage between 2003 and 2006. Historic England. 1st Edition. N/A. p158-160.
  • <33> Monograph: Daniel, P. (Ed.). 2019. A guide to the industrial archaeology of Somerset. Association for Industrial Archaeology. 2nd Edition. p 70, W19.4.
  • <34> Unpublished document: Exmoor National Park Authority. 2023. Email regarding a newly exposed shaft in Roman Lode.
  • <35> Unpublished document: Exmoor National Park Authority. 2024. Exmoor Local Heritage List assessed by the Panel on 21 February 2024.

External Links (1)

Other Statuses/References

  • Local Heritage List Status (Listed)
  • Mires Feature Reference (Monument): EBC12(25)
  • Mires Feature Reference (Monument): EBC12(26)
  • National Monuments Record reference: SS 73 NE 10
  • National Park: Exmoor National Park
  • NRHE HOB UID (Pastscape): 35006
  • Somerset SMR PRN (Somerset): 33065



Grid reference Centred SS 275 138 (676m by 270m)
Map sheet SS21SE

Finds (2)

Related Monuments/Buildings (9)

Related Events/Activities (8)

Related Articles (1)

Record last edited

Apr 24 2024 3:43PM


Your feedback is welcome. If you can provide any new information about this record, please contact us.