Itinera Volume III 2023 PDF Free Download
1 / 87/87
100%
“A thousand roads
lead men forever to
Rome”
Alain de Lille, 1175
Mille·viae·
ducunt·h
omines·pe
r·saecula·
romam
TINERA
Journal of the
Roman Roads Research Association
Volume III
2023
Published by the Roman Roads Research Association
The Journal of the
Roman Roads Research Association
TINERA
VOLUME III, 2023
ISSN 2635-1579 (print)
ISSN 2635-1578 (online pdf)
Typeset in Gentium Basic and Cinzel by the Roman Roads Research Association
licensed under the SIL Open Font License, Version 1.1
Published by the Roman Roads Research Association
Ellerbeck Cottage, Ellerbeck, Northallerton, North Yorkshire, UK. DL6 2RY
A charity registered in England and Wales No. 1163854
All individual contributions to Itinera remain copyright of the author(s). The PDF
versions of any article that has been publicly released on our website or elsewhere
(but not as early release copies to our membership) may be freely copied and
distributed under a Creative Commons Attribution-NonCommercial-
NoDerivatives 4.0 International License, however you may not adapt or alter the
work, and may not reproduce individual illustrations and photographs without the
author’s prior consent, since they often contain additional copyright licensing. The
paper copy of this journal remains © RRRA and individual authors.
The Roman Roads Research Association is not responsible for any statement made,
or opinion expressed, in Itinera. Individual authors are solely responsible for the
content of their articles.
TINERA
ISBN 978-1-8383918-2-9
Volume III
© the Roman Roads Research Association 2023
- iii -
Contents
v About the Association
vi Editorial
Articles
1 - 48 DAVID RATLEDGE, The Roman Roads of Suffolk - a Lidar Reappraisal
49 - 72 DAVE ARMSTRONG, Gradients of the Roman Roads of Britannia
73 - 98 ROBERT ENTWISTLE, Radial Roads and the Origins of London
99 - 104 HOLLY DRINKWATER, RR72a: Further Investigations of the Roman Road at
Worston, Lancashire. An Update on the Archaeological Investigations
105 - 144 STEVE MITCHELL & JOHN POULTER, The Case for a Possible Roman Transport Canal
into Leicester
145 - 166 NIGEL ROTHWELL & EDWARD PEVELER, Roman Roads in the Chilterns Region:
Observations on a Lidar Based Reassessment
167 - 194 MICHELE MATTEAZZI, Road Building in Roman Times: an Insight from Northern
Italy
195 - 220 ELIZABETH LEGGE, The Malhata Fortress on the Roman-Judean Negev Frontier:
Associated with a Roman Road, the Frankincense Trail and a Princely Fugitive
221 - 224 MATT SPARKES, The Roman Road through Croydon
Roman Roads in 2022
225 - 302 Roman Roads in 2022: a Review of Recent Research and Fieldwork
303 - 304 Newly Allocated Margary Road Numbers
Reviews
305 - 306 CHRISTOPHER HADLEY, The Road, a story of Romans and Ways to the Past. (By Carlton
Reid, of Forbes online)
307 - 316 Index
The Roman Roads Research Association also wishes to acknowledge the contributions of all the
other individuals who have volunteered their time and expertise in the preparation, production
and distribution of this volume, without whom it would not have been possible:
People and Ofcers
contact Roman Roads Research Association
Editorial Committee
Rob Entwistle (Hon. Editor)
Dave Armstrong
Dr. M. C. Bishop
Paul Booth
Mike Haken
Advisory Panel
John Poulter
David Ratledge
Dr. Pete Wilson
Keith Abbott
Colin Adams
David Brear
Neil Buckley
John Byde
Tilia Cammagh
Hannah Collingridge
Ian Dean
Chester Forster
Dave Haywood
Tyrone Hopes
Ian Jardine
Geoff Lunn
James Lyall
Delwyn Matthews
Greg Morris
Paul Morris
Elliott Mouelhi
Dan Murphy
Karen Potts
Nigel Rothwell
Eric Rose
Chris Smart
Paul Smith
Matt Sparkes
Richard Whalley
Gary Whitaker
Sally Woodlock
TINERA
If you are interested in Roman roads or would like to know more contact us via our web site
https://romanroads.org/ or by email to one of the below;
Mike Haken (Chairman) mike@romanroads.org
Dave Armstrong (Secretary) dave.armstrong@romanroads.org
Rob Entwistle (Itinera Editor) itinera@romanroads.org
Rebecca L. Ellis-Haken (Finds Ofcer) reb.ellis@romanroads.org
Peter Webb (Membership Secretary) membership@romanroads.org
- iv -
- v -
About the Association
he RRRA was formed in 2015 as a registered charity to bring together disparate individuals
who were researching Roman roads, and to coordinate a nationwide programme of
consistent and high quality research, promoting the study of Roman roads and Roman heritage
throughout the former Roman province of Britannia. Over the last couple of decades, it has often
been a race against time to discover and record what we can of the 60% of the Roman road
network about which we are still uncertain, since modern agricultural methods and urban
development have been steadily removing surviving features from the landscape. Fortunately,
new technologies such as lidar and geophysical survey have helped enormously and enabled
researchers to identify the remains of hundreds of miles of previously unknown Roman roads,
along with associated Roman sites, and we continue to work to ll the many gaps. Research is
only half the story though, we also have to ensure that the results of our work are readily
available. We aim to:
1. bring together all known information on Roman roads in Britain, summarised in a freely
accessible online interactive gazetteer.
2. identify key sites where important questions remain, and organise eldwork necessary to
answer those questions. 100 Ha of geophysical survey have been completed, with a further
500 Ha already planned, and several future excavations are currently at the planning stage.
3. encourage the involvement of as many people as possible in our activities. We care
passionately about community archaeology, and will always encourage local people to get
involved in our work, without any charge (unlike some organisations, we will never do this!).
4. make resources available to researchers and other groups, organise events to keep people up
to date with research including online talks & seminars.
5. ensure that all our published work is Open Access, including our quarterly newsletter, Itinera
(following a very short initial members only embargo). Our eldwork and research reports
are, from 2023, being submitted via OASIS to the Archaeology Data Service.
Membership is open to everyone, and our four hundred and seventy or so members come from
a wide variety of backgrounds ranging from those with just a general interest in our Roman
heritage to professional archaeologists from both the public and commercial sectors, alongside
seasoned Roman roads researchers. The Romans tended to apply their technology reasonably
uniformly across the empire; this is especially so for Roman road layout and construction.
Consequently we do not just restrict our interest to Britannia and our membership now includes
international members. Joining the RRRA gives you the knowledge that your modest
subscription (just £16 a year for a single adult) is helping to support our important work. You
might even get a warm and fuzzy glow.
T
Editorial
Robert Entwistle
ith the issue of Volume III, Itinera becomes a xed feature of the
landscape in archaeological publishing. As has become our tradition,
this volume comprises a set of papers taking widely ranging approaches to
our theme of Roman roads and transport – the common factor being their
stimulating and thought-provoking nature, anchored in rigorous research.
A peer-reviewed journal such as Itinera depends heavily upon the
willingness of referees to review and comment upon proposed papers. We,
along with many in the world of Roman studies, feel keenly the loss of Paul Bidwell after a
sudden illness, a scholar of international reputation who had given generously of his
expertise to the service of our journal since its launch. We owe a debt of gratitude to the
scholars who continue to support our work.
The RRRA, the parent body of Itinera, has always seen a major part of its remit as supporting
research into understanding the Roman road network in Britain. 2023 marks the 50th
anniversary of the third and nal edition of Ivan Margary’s landmark Roman Roads in Britain.
Remarkably it remains the accepted reference on the subject, yet inevitably research has
progressed, lling in (and occasionally correcting) our knowledge of the network. Several
papers in this volume demonstrate this work. David Ratledge once again contributes an
authoritative paper on the roads of a particular region – in this case Suffolk – demonstrating
the results of his skilled use of lidar. Further papers examine roads in the Chilterns, or revisit
locations examined in earlier volumes. An associated role of the RRRA is classifying new
discoveries by tying them into Margary’s numbering system, extending the record for a new
generation. Itinera thus raties and publishes new ‘Margary numbers’ when appropriate,
extending the recognised corpus of catalogued roads. An overview of work and research
around the country (and the contribution of local societies) is maintained through our
section ‘Roman Roads in 2022’.
A glance through this volume, however, demonstrates that Itinera Vol III reaches out beyond
detection and classication into other areas of Roman road studies. Dave Armstrong
crunches the data to examine an aspect of Roman road engineering much overdue for study:
the issue of hills. Roman road builders are well known for tackling slopes where successors
feared to tread – but was there any consistent understanding of what they deemed
acceptable, or did surveyors proceed on a case-by-case basis? The paper is the rst extended
data-led analysis of what the Romans considered acceptable, leading to reections upon the
nature of anticipated trafc.
Road studies also feed into areas of academic debate. The origin of London is one such: an
issue on which there are divided views, with Londinium recognised as a town difcult to t
Journal of the Roman Roads Research Association
Web page: www.romanroads.org/itinera.html
TINERA III (2023)
- vi -
W
- vii -
into established urban categories. A paper examines how the evidence of road surveying
may contribute to understanding Roman intentions for the site of the future city.
A further paper, focused on water rather than land, argues a remarkable (and perhaps
controversial) case for a Roman transport canal serving the growing town of Ratae
(Leicester), challenging certain original ndings of Kathleen Kenyon and offering a
proposed solution to apparent insufcient water supplies for the town.
Britain being at the furthest reach of empire for the Romans, we welcome international
contributions highlighting research from other provinces. Michele Matteazzi, of the
University of Trento, provides a refreshingly different view from the one we are accustomed
to with lively insight into road studies in Italy, the imperial heartland. Occasionally, also, we
are attracted by worthy papers in which roads play an important, but peripheral role. Thus
our nal paper examines a mysterious phase in the life of Marcus Julius Agrippa, grandson
of Herod the Great, reconstructing his period of exile in the Judean desert.
This editorial cannot conclude without recording the devoted work of our editorial team,
and those who do the background work, upon all of whom Itinera depends. Chester Forster,
a valued member of the group through Volumes 1 and 2, has stepped down from the
Committee while continuing to provide our Index. Mike Bishop provides valuable expertise
both in Roman archaeology and publishing, and Paul Booth is a welcome and highly
experienced recruit. Mike Haken (our RRRA Chairman) and Dave Armstrong remain key
hands-on members of the team, with vital skills, appetite for hard work, and wide-ranging
understanding.
Now into its third volume, Itinera has become an established outlet for authoritative and
wide-ranging work in Roman road and transport studies. As the thoughts of our editorial
committee turn towards next year’s Volume IV, we welcome proposals for papers from
authors considering publication in our pages (deadline mid-November), whether they be
internationally established academics or contributors new to academia and publishing.
Potential authors should expect us to be willing to support those less experienced in
academic publishing. If your topic is important and interesting, we look forward to hearing
from you.
Robert Entwistle
Hon Editor, Itinera
itinera@romanroads.org
Editorial
- viii -
- 225 -
Journal of the Roman Roads Research Association
Web page: www.romanroads.org/itinera.html
TINERA III (2023) 225-302
This section of Itinera briey describes work on roads that has recently happened, often in
advance of formal reporting, in a similar style to Britannia reports.
This data has been assembled by RRRA members in the local areas as listed below and then
collated by Dave Armstrong. For future years please contact these with any news via one of
the Association contacts listed in the Frontmatter. Data has come from enquiries to the
county HER and to local Archaeological Societies and we appreciate the help that has been
willingly given, often while busy themselves. Archaeological activity in 2022 has still been
impacted by the pandemic, but we have been able to catch up on some unreported work
from earlier years.
Local correspondents
Neil Buckley; Cheshire, Derbyshire, Greater Manchester and Lancashire,
David Brear; Yorkshire and Lincolnshire,
Dave Haywood; Northamptonshire,
Tilia Cammagh; Hampshire,
Sally Woodlock; Dorset,
Eric Rose; Wiltshire,
Elliott Mouelhi; Somerset,
Paul Smith; Shropshire, Staffordshire, Warwickshire and Municipal W. Mids,
Ian Jardine; Herefordshire,
Geoff Lunn; Essex, Suffolk and Norfolk,
Keith Abbott Berkshire,
Delwyn Matthews; Devon and Cornwall,
Chester Forster; Cumbria,
Paul Morris; Scotland,
Dan Murphy; Glamorgan & Gwent,
Colin Adams; Dyfedd,
John Byde; Gwynedd,
Greg Morris; Clwyd & Powys,
Matt Sparkes; Surrey & Greater London,
Karen Potts; Kent,
Ian Dean; Gloucestershire west of the Severn,
Dave Armstrong; Durham, Northumberland and Tyne & Wear, as well as assisting with
Hertfordshire, Gloucestershire east of the Severn, East & West Sussex,
Worcestershire, Cambridgeshire, Nottinghamshire, Oxfordshire,
Bedfordshire, Leicestershire and Rutland
We are grateful for the assistance and contribution from Richard Whaley of the North East
Hampshire History and Archaeological Society, Field Archaeology Branch (NEHHAS).
Roman Roads in 2022
Thanks are also due to Dr. Chris Smart of Exeter University, for his advice and oversight in
the south west of England .
Summaries, on the following pages, are arranged rst by Country (ie Scotland, Wales and
England), and then in alphabetical order by county. Researchers can follow the links or
contact the appropriate HER, commercial company or Society direct.
Scotland
Angus
RR9, Stracathro fort, Angus, NO 62200 65995
Stracathro Roman Fort Field visit by Andrew Tibbs and Irene Hallyburton
An attempt was made to trace the route of the Roman road as it exits the eastern gate of the
fort at Stracathro. The work was done from the 5th to the 9th of September 2019. A modern
road follows the line of the Roman route along what may be an articial ridge, before
sweeping southeasterly towards modern buildings. A track continues on the line of the
Roman road, descending to the banks of the West Water, where it allegedly crosses the river
at a ford known as the King’s Crossing - there was no trace of a ford at this point, with
bedrock and crevices visible underwater on the far side of the river.
At the point where the track turns to follow the river bank, a series of around 12 large stones
were identied in the water. The stones appeared to have been placed randomly, although
there was some speculation that they may have formed part of an articial embankment
which had subsequently collapsed. Some of the stones appeared to have been dressed and in
one stone a Lewis Hole was visible. Another block had the remains of lead partially covering
another hole. One of the larger stones may have had a carved edge. A storm increased the
river level quite quickly by at least 30cm and no further examination of the stones was
possible. In a subsequent discussion with the landowner the following year, he thought the
stones had been placed there by a company wanting to create a salmon pool at this point in
the river, which was never completed.
Summary report in Discovery and Excavation In Scotland, Vol.20, 2019, 33
Inverclyde
RR780, Roman Road, Lurg Moor, Inverclyde, NS 28900 73486
Editor’s note:
The following account was included because of its previous inclusion in Discovery and
Excavation In Scotland Vol. 14. Readers should be aware that the editorial team are somewhat
sceptical about this entry, with the possible exception of the east west aligned segment
immediately south of Lurg Moor fortlet.
- 226 -
Roman Roads in 2022
Greenock/Kilmacolm Lurg Moor Survey, by Louie Pastore and Stephen Jennings
The area of western Scotland, known today as Inverclyde, at the point where the River Clyde
meets the Firth of Clyde is the location of the Roman fortlet of Lurg Moor. The fortlet
occupies a commanding vantage point on the southern shore of the Clyde, looking north to
Dumbarton Rock on the northern side. From here, the Roman army could monitor maritime
trafc on the Clyde and movements on land below and around.
A short stretch of road, heading south from the fortlet, has been identied and recorded.
However, the wider local network has not been conrmed. Work, undertaken by Frank
Newall in the mid 20th century, has not been formally conrmed.
- 227 -
Roman Roads in 2022
A possible section of Roman road, near Lurg Moor Roman Fortlet, Inverclyde (NS 28900
73486). Lidar image taken from National Library of Scotland https://maps.nls.uk/geo/
explore/side-by-side/
In 2013 a survey was undertaken by Louie Pastore and Stephen Jennings. This included
searches for features believed to relate to Roman roads, such as quarry pits, earthworks,
cropmarks, possible sites for Roman signal stations, stone route markers, kerb-stones, etc,
not necessarily in situ but found close to a likely line. The search area covered Lurg Moor,
Knocknairs Moor, Corlick Hill, Maukinhill Moor, Burnhead Moor, Crawberry Hill, Whitelees
Moor and Darndaff Moor, though this report covers only Lurg Moor, Corlick Hill, Maukinhill
Moor, Knocknairs Moor and Burnhead Moor.
The areas considered here from this search form a stretch of road from the southwest of the
fortlet in a north/south orientation, before turning east to pass the fort and beyond.
The rst section of route runs from Burnhead/Maukinhill Moor, to the north of Corlick Hill,
and continues north downhill toward the Burnhead track at the foot of the hill. Newall
recorded the site at NS 2892 7350, though a more accurate GPS reading of the site is NS 28900
73486. A possible cambered section of Roman road may be the section originally reported by
Newall (Discovery & Excavation in Scotland, 1963). This section is as he described,
accompanied over rocky ground by quarry pits and cambered, 4.58–4.72m wide between hill
ridges. The route ties in with Newall’s reported road nd (NS 2892 7350) on the opposite side
of the hill.
A possible oval roundhouse is located at the top of Burnhead/Maukinhill Moor, north of
Corlick Hill. It consists of an oval hollow with a west-facing entrance anked by two large
boulders. In view of proximity to the road and prominent position, overlooking the Fortlet
- 228 -
Roman Roads in 2022
Section of Roman road, near Lurg Moor Roman Fortlet, Inverclyde (NS 29608 73397). Lidar image taken from
National Library of Scotland https://maps.nls.uk/geo/explore/side-by-side/
- 229 -
Roman Roads in 2022
A possible cambered section of Roman road route (NS 29994 73065) located on a slope on
Knocknairs Moor. This may have been the easterly route which the road took, linking Knocknairs
Moor with the fortlet road on Lurg Moor. A large rectangular block of stone was found at the foot
of the slope.
at Lurg Moor and the possible road route over Burnhead/Maukinhill Moor, it is possible that
the site may have been a Roman signal post, though this has not been conrmed. Several
quarry pits, around 1 metre in diameter, are located along the route of the possible Roman
road. Pits of this nature have been associated with road construction. These may have been
the quarry pits originally reported by Newall (Discovery & Excavation in Scotland, 1963).
Possible cup-marked rocks were recorded in three locations. The rst was recorded at NS
29064 73420, with a second 5.18 metres away at NS 29067 73418. The stones are located on
the brow of a hill, on the possible route of the Roman road and may even have sat astride the
road as kerb or marker stones. A third cup-marked boulder was subsequently recorded at NS
29070 73421. At NS 29078 73431 a possible section of cambered road was located on the brow
of a hill and is anked by quarry pits. Finally, at NS 29311 73293 Burnhead Moor/Maukinhill
Moor a possible carved block of stone was found close to the route of the Roman road on
Burnhead Moor. Part of the stone protruded above the turf and measured circa 370mm x
330mm.
The second stretch of route is a cambered section of possible Roman road (NS 29608 73397)
which winds uphill, curving behind the summit of the hill upon which the fortlet stands, and
heading east. This may have been the junction leading to Whitemoss Fort (modern
Bishopton) to the east and Outerwards fortlet to the southwest (just inland from Wemyss
Bay). A cambered section of possible Roman road was located at NS 29648 73466 and NS
29680 73445. This may have been the route uphill from the plateau where the scheduled
Roman road has been located. These cambered sections may lead to a possible junction
between the Bishopton and Largs routes on the ridgeline.
A possible section of Roman road was located close to the scheduled road leading south from
the fortlet. This section of the route appears as a cutting running along the foot of the hill.
The road leads to a stream, beyond which is a cambered uphill section of road. A cambered
section of possible Roman road was located (NS 29685 73391) which appears to align with the
uphill cambered sections, and may be the route leading towards a possible section of the
route near Auchenfoil Road.
Summary report in Discovery and Excavation In Scotland, Vol.14, 2013, 121-123
Scottish Borders
RR79a, Parish of Lyne, Scottish Borders, NT 224 416
Work by Peebleshire Archaeology Society
Denite traces of this road, which doubtless provided a link between the fort at Newstead
and the west of Scotland, probably Irvine, rst appear at the south end of Standalane Wood,
500 yards to the north of Jeddereld Farm.
The section of road begins as a series of quarry-pits which cross the plantation from east to
west (NT 235 414 - NT 234 414). Obliterated for a short distance in cultivated ground, the pits
reappear just inside the perimeter of the plantation at NT 232 413, while from NT 231 413
where it enters open moorland, the road-mound is well marked and descends into the head
of the side-valley that runs northwards from Edston Farm.
- 230 -
Roman Roads in 2022
Some 80 yards southeast of the corner of the Upper Kidston enclosures a small watercourse
has removed the road-surface and revealed much of a solid bottoming of large and carefully
laid stones. At the head of the valley it coincides with the wall of the Upper Kidston
enclosures, and then swings round the north-western bay of the valley on to a south-south-
westerly course. The road-mound is here somewhat indistinct, as a result of past cultivation,
but about 500 yards beyond the turn a much better preserved stretch begins, 250 yards in
length and evidently once provided with a drainage ditch on its upper side. This stretch is
also accompanied by a row of quarry-pits for the supply of road metal; more than twenty
pits are visible, some separate but many coalescing. Then follows a further length of 100
yards where the mound is considerably attened and spread by old cultivation; it has,
however, been boldly cambered and was at least 20 ft broad. It disappears in the cultivated
elds at about NT 219 406, and is not seen again in the lower part of the valley. This section
of the road is well engineered and graded throughout its length. After crossing the Meldon
Burn, the road makes for the fort at Lyne.
The re-identication of the road-terrace on the east ank of Edston Hill as a mineral road
makes it probable that the Roman road-terrace across the head of the Edston Burn continues
on a course westwards along the south ank of South Hill Head. The search for the roadway
on this line is hampered by later land uses but a broad roadway with massive cutting,
terracing and banking for 165 yards from NT 211 411 to the Meldon Burn at NT 209 410
- 231 -
Roman Roads in 2022
Section of Roman road, Lyne Parish, Scottish Borders (NT 224 416). Lidar image taken from National Library of
Scotland https://maps.nls.uk/geo/explore/side-by-side/
probably marks the course of the road. From the Meldon Burn a broad well-graded terrace,
largely obliterated by cultivation, rises south-westwards to NT 205 408.
Accessible via the Canmore website (accessed 05/01/2023): https://canmore.org.uk/site/
354207/margary-route
Wales
Dyfed, Pembrokeshire
Potential Roman Road crossing, Llanboidy, Carmathenshire. SN 19002 17596
From M. Davies in Archwilio, 2022
Large concreted material and large slabs lie broken within the river Nant Cwmfelin Boeth,
possibly a road of Roman/Medieval date. Hard, concreted material on a deep base layer.
Some of this has been so undermined that slabs have broken off into the river.
HER aware, PRN129290
RR69c, Lledrod, Ceredigion. SN 6522 6736 to SN 6527 6752
From Archwilio, 2019
- 232 -
Roman Roads in 2022
Lidar image of RR69c near Lledrod.
A short stretch of RR69c has been revealed on lidar continuing a straight alignment where
the overlying modern road veers away for a short length.
HER aware, PRN51971
Grove Hill Road, Pembroke, SM 99038 00822
Lidar evidence from Archwilio, 2018
Lidar and GoogleEarth shows a straight feature starts in the area known as Grove, near
Grove Way SM 98968 00778 and runs approximately north east to SM 99206 00908 where it
is interrupted by urban development. However, there is a suggestion of the Linear following
the same line to the west of the built-up area (SM 98762 00690 to SM 98184 00634). The
feature can be seen near junction of Grove Drive and Grove Hill SM 98968 00778 (Google
Earth 2006)
HER aware, PRN115635
Preseli Hills
Work by Dr. Mark Merrony
What may be a ‘new’ Roman road has been proposed across the Preseli Hills in
Pembrokeshire. While marked on the rst series OS maps as a Roman road, it was not
present in subsequent series and did not gure in the output of the Dyfed Trusts Roman road
project in the early 2000’s. The researcher spoke of nding a section of what he believed was
a perfectly preserved Roman road approximately 5m wide buried in peat with further
evidence in sunken lanes and low causeways barely discernible today but which followed
straight routes and worked round hill contours. It is unclear if this is a Roman road and
further research is needed.
As described in a Guardian article; https://www.theguardian.com/uk-news/2022/jun/05/
romans-ventured-deeper-into-wales-than-thought-road-discovery-shows?CMP=share_
btn_tw and in Dr. Merony’s magazine Antiqvvs.
Gwynedd
RR67, Pentir. SH 556 671
Work by Brython Archaeology
A pre-existing crop mark prompted an evaluation conrming the presence and condition of
the road in advance of full excavation later in 2023 for cable laying. This showed that the
road was little more than a compacted surface of clay and small stones but it did have a nice
camber. A slot was not excavated through the road, it is as yet unknown if there is a
foundation of larger stones - this is to be done when the excavation is undertaken. There was
no evidence of any ditches either side of the road.
Two other trenches were excavated across the road - one of these contained a poorly
preserved stone surface, the other was just a layer of compacted clay and gravel - it wasn't
- 233 -
Roman Roads in 2022
clear whether this alone represented the road or whether any stone foundation had been
robbed (the trench was excavated to below the level where the foundation would have
been).
RR67
A series of quarry pits have been noted 200m to the north east of the work above at SH 5889
6732, that indicate the ongoing course of the road.
The HER are aware, GAT90579.
England
Berkshire
Researching the route of RR53 to Bath. Centred on SU 395 717
Work by Keith Abbott and the Berkshire Archaeological Society
A group from the Berkshire Archaeological Society (BAS) have been researching the route
taken by the Roman Road to Bath (RR53) over is nal kilometres eastwards to its intersection
with Ermin Street (RR41). In his 2013 paper, Hugh Toller projected the route of the road to
cross the Wormstall Estate before joining Ermin Street at Benham Burslot some 1.6km
southeast of Wickham close to the B4000. However, a BAS geophysics survey of the
Wormstall Estate in April 2021 failed to nd any evidence of the Roman Road along Toller’s
projected route. Furthermore, a topographical survey of the Estate identied a swathe of
permanently boggy ground along a steep escarpment close to the B4000 which would have
presented a challenge to the construction of a roadway without considerable earthworks, of
which there was no visible evidence.
The most easterly visible earthworks of the Road to Bath are located at Radley Farm 2km
southwest of Wickham, which was extensively surveyed by BAS in February 2022 using a
gradiometer and earth resistance meter. The aim of this survey was to better understand the
fabric of the extant remains of the Roman Road and determine the geophysics signature of
the Road’s surviving remains beneath various modern agricultural landscapes and local
geology as a reference for later work.
Visible remains of the Roman Road are located near Stibbs Wood where the road takes an
oblique route down the steep escarpment from Stibbs Wood to Radley Bottom via a terrace
cut into the slope. Historic maps suggest that this route was used as the main point of entry
to Radley Farm until the early 19th century and is likely to have been much modied and
repaired since its initial construction. Well preserved remains of the Roman Road can also
be seen within Three Gate Copse 1km to the east of Stibbs Wood, where an earth resistance
survey conrmed an 8.5m wide agger with 1.5m wide ditches on either side over some 60m
within the wood. Across the pasture between these two points, historic aerial photographs
- 234 -
Roman Roads in 2022
record crop marks thought to be the Roman Road making a gentle sweep to the north, and
when surveyed with the gradiometer the remains of the Road’s parallel ditches were
revealed together with what appeared to be a small areas of surviving agger on the same
alignment between the ditches that had escaped the ravages of the plough.
With the assistance of the landowner, a previously lost transcript was recovered of an
excavation of the Roman Road at Three Gate Copse carried out in 1968 by teachers and
schoolboys from a school in Reading. The transcript described the cambered road surface in
a good state of preservation, and the section cut through the road revealed its robust
construction using compacted local int - conrming the ndings of the geophysics survey.
Unfortunately the photographs and detail measurements taken during this excavation
remain lost.
Much experience was gained during this project of the geophysics survey methods best
suited to overcome the local geology and the impact of historical agricultural practices to
detect the remains of the Roman Road. East of Three Gate Copse there are no visible remains,
but it was observed that the alignments of the road’s remains at Stibbs Wood and Three Gate
Copse differ by almost 10 degrees, which together with the previously mentioned
- 235 -
Roman Roads in 2022
Map of the probable revised route of the Roman Road to Bath (© OpenStreetMap contributors 2022)
permanently boggy ground and steep escarpment at Wormstall, suggested that the Roman
Road may have taken a more northerly route than that projected by Toller. In September
2022 BAS undertook an extensive geophysics survey of the elds between Three Gate Copse
and New Copse near Church Hill at Wickham. This survey revealed a 160m long parallel
linear anomalies to the west of New Copse that were coincident with a linear anomaly of
high earth resistance as well as a linear lidar anomaly and crop marks on satellite images.
Together these features suggest that they probably relate to the surviving remains of the
Road to Bath, however this evidence is not denitive and further work, ideally involving an
excavation across the observed anomalies, will be required to conrm this initial
interpretation. The revised route of the Roman Road to Bath should this evidence be fully
substantiated is shown below, which projects that the road’s intersection with Ermin Street
was close to the centre of modern-day Wickham village.
Plans are in place for further geophysics surveys across the remaining 500m between New
Copse and Ermin Street later in 2023. The probable routes of the road from New Copse to
Ermin Street are shown below where the lidar highlights the steep escarpment at Church
Hill and other natural obstacles such as gullies that the road would have had to navigate, of
- 236 -
Roman Roads in 2022
Possible routes taken by the Roman Road to Bath from New Copse to Ermin Street. Base lidar data is © Crown
Copyright 2020.
which it is thought ‘scenario #1’ is the most likely. The HER records Wickham to be a small
rural settlement or farmstead occupied between the 1st and 3rd centuries AD, but an
additional objective of this survey will be to assess the full extent of settlement in the area
of the roads’ intersection along both Ermin Street and the Road to Bath.
A summary report of the work so far is available on the Berkshire Archaeological Society’s
site at https://www.berksarch.co.uk/index.php/all-roads-lead-to-wickham/ It is hoped
that once this season’s geophysics surveys have been completed that a full report of the
research and eldwork undertaken will be drafted for inclusion in a future issue of Itinera.
Buckinghamshire
RR16a, RR173b?, Aylesbury, Fleet Marston SP 77710 15370.
From Edward Biddulph. Work by COPA on behalf of Fusion Joint Venture HS Enabling Works
Contractor
COPA (Cotswold Oxford Pre-Construct Archaeology) undertook a programme of
archaeological recording in 2021 for Fusion Joint Venture HS2 Enabling Works Contractor at
the site of a Roman nucleated settlement at Fleet Marston as part of the enabling works for
High Speed Two (HS2) Phase 1, as enclosed image. Areas of excavation (areas C24014 and
C24015) crossed the route of Akeman Street (RR16a, Verulamium – Alchester) and evidence
for the road, including the gravel surfaces forming the agger and a single pair of anking
ditches and its re-cuts, was recorded in several interventions. Curiously, the agger, aligned
WNW–ESE, was not of equal width along its course, ranging as it did from c10m wide to 28m,
its widest parts possibly accommodating a market or other communal or public activities.
Pottery recovered from the surfaces and associated ditches dated to the middle to late 2nd
century AD or later, but an early Roman date for the laying out of the road is indicated by
dating evidence from the road uncovered at other sites, including the neighbouring
Berryelds site, which also forms part of the Roman settlement (Biddulph et al. 2019, 159).
The excavation in Area C24016 exposed a Roman road that provisionally has been identied
as part of the Dorchester-on-Thames – Fleet Marston road (RR173b). The road was
orientated ENE–WSW and, as indicated by the results of a geophysical survey carried out
prior to excavation, met Akeman Street. The agger was up to 10m wide and comprised a
surface of limestone rubble. There was a single ditch on each side of the agger, although both
ditches had been re-cut several times. Pottery and coins from the ditches point to road use
after AD 150, although an earlier date for construction cannot be discounted.
A third Roman road was recorded in Area C24017. The agger was aligned NNE–SSW and c4.5–
6m wide and consisted of one to two layers of limestone rubble mixed with silty clay. It had
a single pair of anking ditches, which had multiple re-cuts. Pottery and coins recovered
from its surfaces suggest that the road was in use by the middle or late 2nd century AD, but
again it is possible that the road had been set out earlier. Though as well constructed as the
other two roads, this road appears to have been a minor one and connected the settlement
with elds, a farmstead or villa in the settlement’s hinterland.
- 237 -
Roman Roads in 2022
- 238 -
Roman Roads in 2022
Plan of excavations at HS2 Fleet Marston, showing the course of the three roads recorded
Bibliography
Biddulph, E., Brady, K., Simmonds, A., and Forman, S. 2019: Berryelds. Iron Age settlement and a Roman
bridge, eld system and settlement along Akeman Street near Fleet Marston, Buckinghamshire, Oxford
Archaeology Monograph No. 30, Oxford.
Cambridgeshire
Cambridgeshire. A14 Cambridge to Huntingdon Improvement Scheme TL 1925 7373 to
TL 4095 6178
From Ruth Beckley, Cambridgeshire HER
The A14 Cambridge to Huntingdon Improvement Scheme involved an extensive programme
of archaeological eldwork culminating in a series of Targeted Excavation Areas (TEAs)
between 2016 and 2022 along the 14 mile route. The Scheme was led by MOLA-Headland
Infrastructure (MHI) but included several other archaeological units. Extensive Roman
remains of settlement, cemeteries and industry were identied at 24 different locations.
Although parts of the Scheme followed the route of the Via Devana (RR24) no clear evidence
of this road was identied, however, evidence of RR22 from Braughing to Godmanchester
- 239 -
Roman Roads in 2022
Map showing the location of Roman related discoveries during the A14 development
was identied at TEA 21 near Offord Hill Farm (TL 2357 6815). At this location, evidence of a
16m wide road with parallel roadside ditches were identied. Roman pottery dating to the
rst and fourth centuries were recovered from the roadside ditches. Elsewhere, trackways
and roads have been identied as part of several settlements at Brampton, Buckden and
Hilton. Interpretation of the archaeological evidence is still ongoing.
Smith, A. & Bowsher, D., 2019. A14 Cambridge to Huntingdon, Cambridgeshire Archaeological
investigations Volume 1: post-excavation assessment. A digital copy is available at; https://
archaeologydataservice.ac.uk/archives/view/a14_he_2020/
Steeple Morden, South Cambridgeshire, centred on TL 3067 3942
Work by Oxford Archaeology East
Hollow way 269 was the stratigraphically earlier feature and consisted of six different
segments. Segment 1 measured 65m in length before a break was recorded. This segment
fairly consistently measured 4-5m in width. For 58m of its length a layer of dense loose chalk
1.8m in width was also recorded to the south. Finds from the backll of the feature consisted
of a sherd of Roman pottery and two tile fragments from the surface. Segment 2 represented
a gap of approximately 15m. Segment 3 continues for 58.03m on the same alignment as
segment 1. It measures 6.68m at its broadest point, datable nds were again very limited to
a single sherd of Roman pottery. A similar chalk band to the south was also recorded.
Segment 4a represents a slight kink northwards at the northeast end of segment 3. Segment
4 measures 152m in length, like earlier segments it is bordered to the south by a chalk layer
but is much wider, varying from 4.18m to 10.4m. Finds were more frequent from this section
and includes pottery of a 2nd to 4th century date. Segment 5 measures 99m in length and
2.08m-5.6m in width. At one point it merges with hollow way 270. Segment 6 measures 181m
in length and is characterised by the merging of the two hollow ways, it includes a stretch
of metalled surfacing or a causeway where it crosses the palaeochannels. Wheel ruts were
encountered along a 68m stretch of the causeway. Datable nds included a small number of
Roman potery and a medieval key.
Hollow way 270 consisted of ve segments, it appeared to be more narrow and more shallow
than the earlier feature. Segment A measured 93m in length at the southwest edge, segment
B was a 29m break. Segment C represented an 165m section that initially curved away from
hollow way 269 before coming back. This section varied in width from 2-2.5m for the most
part. Segment D is the junction between the two tracks. Segment E represented the nal
section of the hollow way and measured 181m in length.
HER aware MCB30919; Unpublished report: Atkins, R. & Graham, S. 2013. Excavation of
"Avenell" Way: A Roman Track-Way at Station Quarry, Steeple Morden. Oxford Archaeology East
Report 1415 ID: SCB22550, Location: OA Library
RR23a, Wilberforce Road, Cambridge, TL 4352 5842
Work by Cambridge Archaeological Unit
Evaluation was carried out on Wilberforce Road in 2020 in advance of proposals to develop
the site. The evaluation comprised seven trenches with a further ve trenches investigated
- 240 -
Roman Roads in 2022
in January 2021. Evidence of the Roman road was found along the projected path of the
Roman road (MCB28262) within trenches 6 and 7 and partially within trench 9. The evidence
consisted of an 8m wide agger on a northeast-southwest alignment. The foundation layer
comprised yellowish brown silty sand between 0.05m and 0.2m thick with a 0.1m thick
gravel layer above it. Two parallel ditches were identied to the north and a single ditch to
the south.
HER aware, MCB 30546, Unpublished report: Brittain, M. 2021. Land south of Wilberforce Road,
Cambridge an archaeological evaluation. Cambridge Archaeological Unit Report 1461, ID:
SCB67955, Location: ADS Library
Possible Roman road RR23a, Lucy Cavendish College, Cambridge. TL 4420 5909
Work by PreConstruct Archaeology
Archaeological evaluation was undertaken in advance of the construction of student
accommodation. The evaluation consisted of two trenches, one of which identied Roman
ditches interpreted as part of a road, possibly that from Arrington to Cambridge. The
evaluation identied only Roman pottery and animal bone from trench 1, however, trench
2 contained two parallel intercutting Roman ditches on a northeast to southwest alignment
and thought to represent a former road.
Subsequent excavation carried out in 2021 revealed ve phases of Roman activity based on
pottery dating. Early Roman phase 1 (c.70-100AD); Evidence from this phases consisted of a
ditch measuring 16m in length on the west side of the site with a large quantity of Roman
pottery, a quern stone fragment recovered from its ll.
HER aware, MCB31571, unpublished report: Pullen, A. & Woolhouse, T. 2021. Land at Lucy
Cavendish College, Cambridge: Post excavation assessment and updated project design. PreConstruct
Archaeology Report R14780, ID: SCB74478, Location: ADS Holding
Roman settlement remains, 23-33 Wimblington Road and Job's Lane, March. Centred
on TL 4151 9495
Work by PreConstruct Archaeology
A Roman period settlement with enclosures and trackways was discovered. The earliest
period of settlement activity dates to the early Roman period from the mid 1st to early 2nd
century AD. Activity at the site at this point is characterised by eld systems similar to those
at Wimblington Road with a series of ditches recorded across the site forming two regular
enclosures with two associated trackways. Trackway 1 was formed from two parallel ditches
south of enclosure 1 while trackway 2 was aligned perpendicular to trackway 1 leading away
from the enclosure area. A possible beamslot structure and four more ephemeral structures
are also associated with this phase. The beamslot structure was located in the southwest are
of the site and consisted of a main ditch on a north-northwest to south-southeast alignment
with two perpendicular ditches coming off it and a number of postholes interpreted as
supporting a structure. Four additional postbuilt structures, were identied, two were
rectilinear and located in the southern part of the site and two were possibly roundhouses,
however, they were partially obscured by later features. They were located in the northern
- 241 -
Roman Roads in 2022
part of the site. Ten additional features were also attributed to this phase. The nds
assemblage was small and mainly comprised pottery. Finds recovered from the trackway
ditches included pottery, a Millstone Grit quern and some lead and copper alloy.
The fourth phase of activity dates to the early to middle Roman period (c.100-200 AD) and is
characterised by the establishment of the small settlement. The enclosure systems became
more dened and larger while additional settlement activity included the addition of ovens
and several large pits. At this stage, seven enclosures could be identied within the
excavation, most produced moderate to large nds assemblages. A series of boundary
ditches forming elds or activity areas were also identied. To the northwest of the site, a
post-built structure was present comprising 6 post holes forming a rectilinear feature. Two
ovens were recorded close to the structure, their size suggests a domestic function and were
interpreted as bread ovens. A number of waste pits were also recorded. Two further ovens
were identied in the western part of the site and are interpreted as possibly domestic or
industrial. Other features associated with this phase included two waterholes, one of which
may have had a post built superstructure, three animal burials and a large number of pits
primarily containing Roman pottery.
HER aware MCB30551; Unpublished report: Jones, M. 2018. Land at Jobs Lane, March,
Cambridgeshire: an archaeological excavation. PreConstruct Archaeology Report R13271, ID:
SCB63376, Location: ADS Library
Roman local road? Howson's Lodge, Old Weston. TL 1025 7827
Work by Oxford Archaeology East
Archaeological evaluation and excavation carried out on ve discrete areas along the
Winwick to Weston pipeline in 2007 revealed archaeological evidence in all ve areas. Area
A was located to the east of Howson's Lodge and contained evidence of a Roman road with
associated drainage ditches. The road was aligned east-west and measured 8m wide. The
base layer comprised a poorly preserved gravel layer with three patches of metalled gravel
surface overlying it. It was interpreted as a local or minor road. The two shallow drainage
ditches were encountered on the northern side of the road only and contained no datable
nds.
HER aware, MCB31207, Unpublished report: Henley, S. 2010. Prehistoric-post medieval
settlement on land between Winwick and Old Weston Excavation Report. Oxford
Archaeology East Report 1131, ID: SCB50504, Location: OA Library
Roman, Saxon and WWII remains, Pathnder House, Huntingdon. TL 24 71
References to an excavation at Pathnder House in the 1970s on demolition of the current
buildings. Although the excavations have never been published and no record of nds have
been provided, it is believed the excavations revealed the presence of a Roman road and
remains of Saxon buildings. It was also reported that the foundations of the former World
War II Nissen huts and buildings were broken up and used as rubble for new construction
HER aware, MCB31500
- 242 -
Roman Roads in 2022
Probable Roman buildings and roads (Ermine Street RR2c), Boat Field, Durobrivae,
Chesterton, Huntingdonshire, TL 1250 9645
Work by S.G. Upex
Watching brief carried out along the fence line in Boat Field southeast of Durobrivae revealed
limited archaeological evidence. Although features were observed they were not excavated.
In the northwest half of the site, a single trench was excavated over a 180m stretch,
observing the following features; a pit 27m from the start, a pit 63.5m from the start, a 5m
wide metalled surface bordered by a ditch on the north side and interpreted as a road 93m
from the start, a oor surface bordered to the north and south by walls at 120m and 126m
from the start, a ditch like feature partially cutting the second wall and in total the walls and
oor surface are interpreted as the remains of a single building. Also recorded was an area
of loose limestone interpreted as a passage between two buildings at 127.8m-129.2m. A
second building consisted of a ditch or possible robber trench, four walls, three oor
surfaces and two possible pits between 129.3m and 156.5m. Finally, a metalled surface
measuring some 8m wide was interpreted as evidence for Ermine Street.
HER aware, MCB31714, Unpublished report: Upex, S G. 2014. A report on archaeological
observations, investigations and recording at the A1 lay-by and the area to the east (Boat Field) of the
Roman town of Durobrivae, Chesterton. ID: SCB46605, Location: Librarylink
RR22, The Maltings, Godmanchester, Huntingdonshire, TL 245 703
Excavations carried out in advance of residential development identied several phases of
activity. The earliest feature consisted of a small roundhouse of Iron Age date,
approximately 3m in diameter.
Evidence of a Roman road recorded on the expected alignment of the Roman road RR22. The
road consisted of two surfaces of gravel metalling, cut by a 2nd century pit. Three
intercutting roadside ditches were recorded on the north side containing Roman material.
The full width of the road could not be identied however several small single unit timber
structures dating to the 1st-2nd century along with small granaries, a smithing furnace and
rubbish pits were recorded on both sides of the road. Other features included evidence of
lazy beds and two major boundary lines comprising ditches and fence lines.
HER aware, MCB31836, Monograph: Lyons, A. 2019. Rectory Farm, Godmanchester,
Cambridgeshire: Excavations 1988–95, Neolithic monument to Roman villa farm EAA 170. ID:
SCB68805, Location: Not in HER
Ermine Street RR2c and structural remains, Pinfold Lane, Godmanchester. TL 2466
7040
Excavations at 8-10 Pinfold Lane in 1971 revealed the presence of metalled surfaces along
the path of Ermine Street along with structural remains of several buildings to the east of
the road. The buildings had cob walls and clay oors, evidence of a catastrophic re that
destroyed the buildings occurred in the 2nd century.
- 243 -
Roman Roads in 2022
HER aware, Monograph: Lyons, A. 2019. Rectory Farm, Godmanchester, Cambridgeshire:
Excavations 1988–95, Neolithic monument to Roman villa farm EAA 170., site 5. ID: SCB68805,
Location: Free Download available https://eaareports.org.uk/publication/report170/
Ermine Street RR2c and roadside buildings, Piper's Lane, Godmanchester. TL 2471
7030
Excavations in the garden of No 10 Piper's Lane following the discovery of Roman features
by the owner. Evidence included six sequences of metalled road associated with the line of
Ermine Street dating from the 1st to 4th century. Foundation trenches for roadside fencing
and a large boundary ditch were also recorded. These features were subsequently cut by a
timber building along with several other minor roadside buildings dating to the 2nd-3rd
century. Structures were of cob wall construction. A possible temple complex to the west
was identied, of a probable 2nd-3rd century date. A metalled entrance drive leading to a
gateway, later superseded by a 4th century open fronted building. An intact horse burial was
recorded.
HER aware, MCB31839, Monograph: Lyons, A. 2019. Rectory Farm, Godmanchester,
Cambridgeshire: Excavations 1988–95, Neolithic monument to Roman villa farm EAA 170. , site 8. ID:
SCB68805, Location: Free Download available https://eaareports.org.uk/publication/
report170/
Unrecorded Roman road, Gravel Hill Farm, Cambridge. TL 4313 5998
Work by the Cambridge Archaeological Unit
Excavations in 2018-19 included a small site to the east of the University Agronomy Centre.
This site was investigated in order to identify a possible continuation of a road seen in Site
II (MCB18267). The evidence from Site I consisted of a 25m length of ditch interpreted as a
roadside ditch based on its position on the same alignment as the road from Site II. Finds
consisted of 13 ints and 6 sherds of prehistoric pottery.
HER aware, MCB31853, Unpublished report: Brittain, M. & Evans, C. 2019. North West
Cambridge The War Field Villa (site VIII) and other Phase 2 investigations (Sites I, VI & X) Appendices.
Cambridge Archaeological Unit Report 1435, ID: SCB74373, Location: digital backlog
Cheshire
Middlewich: Archaeological excavation at land to the rear of King Street (RR70a)
centred on SJ 7042 6647, September and October 2021
Roman Sites identied through archaeological eldwork instigated by the Cheshire Archaeological
Planning Advisory Service (CAPAS)
In Spring 2021 a planning application was submitted for a new wastewater pumping station
at land off King Street, Middlewich. A programme of archaeological mitigation was secured
by planning condition, to comprise an open-area excavation (c 0.4ha), in order to identify
and record the archaeological deposits known from earlier work to be present (Hayes 2006).
- 244 -
Roman Roads in 2022
The excavation work began in the summer of 2021 and was carried out by Oxford
Archaeology North.
The site lay within the extensive area of Roman settlement that developed to the south of
the Scheduled Roman fort at Harbutt’s Field (SM1008460). A signicant proportion of this
settlement has been excavated over the last twenty-ve years in advance of development
and has produced evidence of industrial activity, including salt production, as well as roads
and domestic activity. Across the settlement, many of the deeper features have been
characterised by well-preserved waterlogged remains, including quantities of worked wood
(Garner & Reid 2013).
Previous archaeological evaluation work undertaken in 2014 across a wider area than the
present site (Strafford 2014) identied signicant remains over much of the land examined.
These included ditches, pits, building foundations and other deposits similar to those
excavated during earlier excavations in Middlewich. As expected, waterlogging and organic
preservation was evident in many of these features.
The site investigated in 2021 was occupied by parts of three Roman settlement plots, aligned
perpendicular to King Street, the main Roman street through the settlement. Within these
plots, several salt-production features were found, including two brine wells lined with oak
timbers (Illus V.1.1), most of which were reused from earlier structures (as evidenced by
redundant joints and peg holes). There were also several large pits, some lined with wood
and/or clay, which probably served as brine settling tanks. No certain evidence for buildings
was found, probably because of the location of the site away from the King Street frontage,
where any substantial buildings are most likely to have been situated. Dating evidence
suggests a limited occupation from the AD 70s, with an increase of activity around AD 80/90
and continued occupation to c. AD 150, after which the site was largely abandoned. There
was limited evidence of post-Roman activity, and the site seems to have remained as open
ground up to the present day.
The artefacts recovered during the excavations included well-preserved pottery from the
brine wells and pits, along with an entire vessel recovered from the base of one of the
probable settling tanks; this was probably deliberately and carefully deposited when the
tank went out of use. The pottery comprised locally produced and imported wares, including
signicant amounts of samian ware. The latter phenomenon has been noted on other
Roman salt production sites in Cheshire and is likely to represent evidence of military
inuence (Garner & Reid 2013, 64). Few personal objects were found, reecting the
industrial character of the site, although two copper alloy brooches and a small gold
pendant that would have been threaded onto a necklace were recovered from a settling tank
in the southern plot.
The on-site works are now concluded, and the project has moved into post-excavation
assessment phase, which is likely to result in a recommendation for publication in an
appropriate journal (Oxford Archaeology North forthcoming). The excavations of 2021 are
likely to represent one of the last opportunities for extensive excavation in Middlewich,
which will make any report all the more signicant.
Garner, D & Reid, M 2013 Roman Middlewich: re-assessing its form, function, and chronology. Journal of the
Chester Archaeological Society, New Series, 83, 37–93
- 245 -
Roman Roads in 2022
Hayes, L 2006 Roman Middlewich Community Dig: post-excavation assessment report. (Unpublished client
report 12588.RO3). Chester: Gifford. CHER ECH4173
Oxford Archaeology North forthcoming. Excavations off King Street, Middlewich Strafford, L 2014 Land at
King Street, Middlewich, Cheshire: archaeological evaluation. (Unpublished client report 2014/92). Sale:
Archaeological Research Services. CHER ECH5947
(CHER ECH numbers refer to the Event Records and CHER numbers relate to Monument
Records in the Cheshire Historic Environment Record.) that can be consulted at www.
cheshirearchaeology.org.uk/
An annual summary of archaeological eldwork instigated by the Cheshire Archaeology
Planning Advisory Service (CAPAS) has been reported in the Journal of the Chester
Archaeology Society under the heading “Cheshire Past” since 2013 (Volume 84).
For more information about how to access the Journal please see the Chester Archaeological
Society Website (https://chesterarchaeolsoc.org.uk/).
The most recent annual summary appears in Volume 92 for 2022.
Cornwall
Lostwithiel to Redgate
Identied independently by both Steve Kaye and Mike Haken
Earthwork remains of a probable Roman road can be seen on lidar from a point about a mile
north east of Lostwithiel, showing typically as a series of linear pits either side of an
elevation (agger).. Presumed to cross the R. Fowey either below Restormel Roman fort, or
further downstream at Lostwithiel, it can be traced with condence from about SX 120 614
past West and East Taphouse, Doublebois, and Redgate as far as SX 238 689, northeast of the
King Doniert’s Stone. It utilises high ground, avoiding local streams and rivers and whilst
seems to have been designed as a route north east from Restormel, its destination is unclear.
Cornwall/Isles of Scilly HER: MC066590 Source: personal communications/emails from
public 2020). Also reported in RRRA Newsletter 14, 2020, 3-4
Latchley to Callington
Work by Dr. Chris Smart/ ‘ Understanding Landscapes’ project
Lidar analysis suggests a probable Roman road from SX 412 736, next to the River Tamar
(where it presumably crossed) southwest across Hingston Down, via Cox Park and
Sevenstones, to SX 371 700, on the eastern side of Callington.
Launceston to Camelford
Work by Dr. Chris Smart/ ‘Understanding Landscapes’ project
A crossing of the River Tamar in the vicinity of Launceston has long been suspected, though
no evidence for it, or of a Roman site in the environs of the town are yet to be found. Lidar
analysis suggests a probable Roman road running from Higher Truscott, 3km (1.9 mls.) west
- 246 -
Roman Roads in 2022
of Launceston (SX 296 860) around the north side of Bodmin Moor, via Tresmeer, Halworthy,
and Davidstow, as far as SX 092 806, south west of Camelford. To the east, this would likely
have connected with that described in relation to Margary’s RR492a.
In addition an extant linear feature identied on lidar later interpreted as possibly the same
Roman road (SX 1147 8455) from Cornwall/Isles of Scilly HER: MC064613 Source: lidar sent
by email from R. Smith dated 18/07/2019)
Nanstallon to Ruthvoes
Work by Dr. Chris Smart/ ‘Understanding Landscapes’ project
Lidar analysis suggests a probable Roman road from Penaglion Downs, north of Nanstallon,
SX 034 687, via the Roman fort at Tregear Farm, to SW 928 609 a short distance northeast of
Ruthvoes Farm, Indian Queens. This route takes a level course around the lower slopes
beneath the hillfort of Castle-an-Dinas.
Callington –Possible Roman road
Earthwork remains of an unidentied linear feature visible on lidar, possibly a Roman road.
Cornwall/Isles of Scilly HER: MC067380. Source: Member of the public 23/12/2020 (Personal
communication).
Calstock Roman fort, South East Cornwall.
Work by Dr. Chris Smart/ ‘Understanding Landscapes’ project
First discovered in 2007 (by Exeter University), there have been a number of community-
based excavations carried out at this site (SX 4362 6918), the last being undertaken in the
summer of 2021 as part of the ‘Understanding Landscapes’ project. Geophysical surveys to
the east of the fort revealed a number of linear features, one of which appears to be a
trackway leading into the fort. The fort would have in all probability have connected to the
road already identied from Tavistock to Brentor (and on to Okehampton) and with there
being a likely crossing of the nearby Tamar, possibly further up river at Latchley (see also
the Latchley to Callington report above). Efforts are currently continuing to further identify
these routes.
Cornwall/Isles of Scilly HER: 176203. Source: Smart, C & Rance, H, 2012, Archaeological
Investigation to the East of Calstock Roman Fort, Church Hill, Calstock, Cornwall, October 2011 -
Assessment Report and Design for Analysis, Archive, and Public Dissemination (Cornwall Event
Report).
Cumbria
RR86d, The Military Way, Beaumont, Park Farm NY 3525 5881.
The construction of a slurry lagoon very close to a known Roman road, the Military Way, did
not reveal any features which could be interpreted as connected to a road.
- 247 -
Roman Roads in 2022
RR750? Seaton, Workington near Calva Park NY 011 296
A recent planning application to Allerdale District Council to build 180 houses on a eld at
Seaton near Workington contained an Archaeological report by Wardell Armstrong of
Carlisle and in that report is a plan of a geophysical survey of the eld. The report mentions
two features of prehistoric interest, and two dark parallel lines trending in a NNE/SSW
direction.
The parallel lines have been interpreted as the ditches of a Roman road and lie some 600
metres east of the known fort of Burrow Walls. The signicance of this road is that it lies
some four miles south of the last known frontier defences of milefortlets and towers which
were a defensive extension of Hadrian’s Wall. Archaeologists have been searching for a
southward continuation of the defences for over 140 years but mainly along the coast
margins which have been heavily industrialised. This road may indicate that defences
followed higher ground to avoid the marshy coastal area and by implication, the milefortlets
and towers (terms used to distinguish them from Hadrian’s Wall milecastles and turrets)
may also have been placed in more elevated positions.
The most southerly identied tower on the extension of the Hadrianic defences is Tower 26b
at Risehow according to Richard Bellhouse in Roman Sites on the Cumberland Coast. 1989
The planning application can be viewed at https://allerdalebc.force.com/pr/s/planning-
application/a3X3X00000IrZdcUAF/ful20220285?tabset-e3f5c=2
N.B. Allerdale District Council ceased to exist on 31 March this year and planning matters
will be determined by the new Cumberland Council-- a single tier authority. At the time of
writing it was not possible to ascertain if the application would still be viewable on the
Allerdale Planning portal but the application number is FUL/2022/0285.
- 248 -
Roman Roads in 2022
Lidar view back from Meg’s cairn to Kirkby Thore fort © David Ratledge . Base lidar data is © Crown Copyright 2020.
Might Meg’s Cairn be Roman?: Maiden Way (RR84) to Meg’s Cairn
By John Poulter and David Ratledge
The setting out of the Maiden Way, RR84, following a long distance alignment appearing to
use Megs Cairn as a guiding marker, and then skirt the constructed road around it, was
considered in the RRRA Winter 2022 newsletter. This is available at https://romanroads.
org/Newsletters/Members/Newsletter-24%20Christmas%2022.pdf
The Roman Roads of Cumbria – Some Recent Developments, Part 1, RR82
By David Ratledge
Continuing his prodigious lidar output, an update of the roads in Cumbria was given in the
RRRA Spring 2022 newsletter. This covered the Stainmore road RR82 and is available at
https://romanroads.org/Newsletters/Members/Newsletter_22%20Spring_2022.pdf
Cumbria Update, Part 2, RR75, RR755(x), RR754, RR740 and a possible route across
Morcambe Bay.
By David Ratledge
Further to Part 1, a follow up article in the Autumn 2022 RRRA newsletter covered more of
the Cumbria roads. This is available at https://romanroads.org/Newsletters/Members/
Newsletter%2023%20Summer%202022%20(1).pdf
RR75, A595 road improvement scheme, Bothel and Threapland. NY 1656 3765
A desk based assessment of the route proved that the Carlisle to Papcastle road RR75
continued to be used during the medieval period and was adopted as a turnpike road in the
late 18 century. This later became the A595 trunk road, although this section is now de-
trunked. It was felt that most features associated with the present line of the road were
medieval or later and more work was recommended.
Derbyshire
Walton/Drakelow Roman fort investigation. Centred on SK 223 194
Work by Transforming the Trent Valley Scheme, led by Dr Mark Knight
This potential fort site had been observed in aerial photographs taken by Jim Pickering in
1983 and indicated a corner of a double ditched feature that was interpreted as a probable
Roman fort or marching camp site. This appears in the Derbyshire Heritage Environment
Record as MDR2514.
The location of the return of the south easterly corner was known from the previous work
and so it was hoped to locate a second return to be able to assess the size of the fort. The
results of the survey indicated that we had indeed located the double ditched feature
extending west into and through the next eld, and had also found the westerly return and
- 249 -
Roman Roads in 2022
limit of the feature before it disappeared into the river. It had been observed from the aerial
photograph that the alignment didn’t seem to be on the course of the river, and the survey
conrmed this. The alignment seems to be with the Holme Pierrepoint gravel terrace, now
also partially eroded by the river Trent. It had also been observed that the alignment with
the proposed route of the so called Via Devana was askew, perhaps indicating that this road
was later than the fort. The fort also aligns with the paleochannel of the river Trent some
40m west of the current course of the river. The river seems therefore to have changed
course in the intervening centuries and eroded the north west corner of the feature.
These ditches correspond to the ‘playing card’ layout of Roman forts elsewhere in Britain
and across the Roman empire, this one measuring approximately 200 x 80m. Whilst it is
anticipated that this is a Roman fort guarding the shallows of a river crossing, presumably
on a pre-existing Iron Age or earlier route, we should be mindful that other interpretations
are possible. Corroborating evidence was slim, but some Roman period pottery sherds were
collected over the course of the survey.
Note: See also Staffordshire for the course of this road surveyed on the west bank of the
Trent (RR573b(x)) and Leicestershire for its course towards Leicester (RR573b(x)).
Devon
New evidence is helping to not only conrm (or dismiss) the true or suspected alignments
of various sections of Roman road (e.g. as identied by Margary) but to also identify
previously unknown sections and branches of road too. Some of this information has been
presented to RRRA members by Dr. Chris Smart in his online presentation given on 5 January
2023 (Beyond Isca; new evidence for Roman expansion in South West Britain). This is available to
watch on catch up and is recommended for further understanding. https://www.youtube.
com/watch?v=9ZW6qW4VD5o&list=PLqRwv1RsLjSctOlcQ9Voki6rglg4j55Kj&index=2. Brief
summaries of various latest surveys and reports are listed below and are intended to update
the information for 2020 (see Itinera Vol 1, 332-3).
RR492a Exeter to North Tawton and west to Okehampton
Work by Salvatore, Toller, Kaye and Stocker
Following the important new work carried out by Salvatore et. al. and already previously
referred to in Roman Roads in 2019, evidence has been provided detailing much of the
course of the Roman road from the major Roman fort at North Tawton running towards the
main legionary base at Exeter. This (combined with further re-assessment of earlier
excavations) has also helped to further develop a greater understanding of the morphology
of this road.
As a result of these investigations this primary Roman road may now be traced from North
Tawton for some 23km (14+ miles) to the Creedy valley, near Crediton, and then ESE to a
point approx. 5km (3+ miles) from the fortress at Exeter itself. Steve Kaye has further
commented that the Roman road sections identied appear to have been routed to avoid
difcult ground and steep gradients, further suggesting that it was initially designed to be
- 250 -
Roman Roads in 2022
used by supply-carts serving the North Tawton military complex, as well as possibly the fort
at Okehampton and others elsewhere.
As subsequently noted by Steve Kaye, possibilities for further accurately identifying the
remaining sections southwards towards Exeter itself may now prove difcult given the re-
working of the landscape since the Roman period (and which was why his own and other
related surveys were conducted in ‘reverse direction’ so to speak from North Tawton).
Further recent work has however been undertaken to help ‘ll in the gap’ from Gunstone
Cross, near Crediton, to Half Moon, Newton St. Cyres, just to the north west of Exeter:
RR492a Roman road from Gunstone Cross, Crediton to Half Moon, near Newton St.
Cyres: appraisal of a Roman road originally proposed by Toller (2014).
Work by Steve Kaye
As reported by Steve Kaye, since Toller’s original lidar interpretation, the Environment
Agency has released the much improved National Lidar Programme datasets, including the
2021 1.0m grid sets for the OS quadrants SX79E, SX89NE and SX89NE which cover the area
examined. In addition the availability of further Google Earth historical aerial and satellite
imagery datasets have allowed a reappraisal of the original proposed route by Toller. This
latest work has now conrmed the course of the predicted Roman road as originally
proposed. From Gunstone Cross the road ran broadly south-east to Uton and on to Chaldon
Brake before turning eastwards to descend to the Shuttern Brook, south of West Holme.
From here the road continued broadly eastwards to Ridge House and then turned north-east
to Half Moon.
The road is the eastward continuation of the road from the Roman fort at North Tawton to
a crossing over the River Yeo at Keymelford.
There are no clear indications of the road across the valley of the River Yeo but this is not
considered surprising given the changes in the course of the river over the succeeding
centuries. It is considered likely that the route continued in straight sections towards the
extant road from Uton where it crosses the railway and that this road overlays the Roman
road. From Uton crop marks visible on aerial photos further show the road continuing
south-east to Chaldon Brake, followed in part by the modern road to the east of Broomeld.
The course of the road is followed, in part, by eld boundaries in its latter sections towards
Half Moon. The predicted section of road has now been upgraded to ‘Probable’.
Devon HER MDV132948 Sources: ‘Reappraisal of proposed Roman Road from Gunstone Cross to
Half Moon (Newton St. Cyres), Devon’ , Kaye, S ,Pamment Salvatore, J, Stocker, S, & Toller, H ,
unpublished report, 2022.
Roman road west of Wallabrook Farm, Lamerton, Nr. Tavistock, West Devon
Work by Steve Kaye
The anking ditches of a Roman road are visible on the 2016 aerial photo and lidar imagery
as crop marks and earthworks running north-south across the elds to the west and north
of Wallabrook Farm, south of Brent Tor, centred at SX 469 788 (Map Sheet SX47NE).
- 251 -
Roman Roads in 2022
The anking ditches of a Roman road show well on the aerial photo and are about 10.7m
apart which appears to t well with the average width of the road of 10m-12m.
Devon HER: MDV133425. Sources: Kaye, S., 2020, Roman road Tavistock to Brent Tor
Roman road, Tavistock to Brent Tor
Work by Steve Kaye
Linked to the description above, a probable Roman road is visible on aerial photos and on
lidar imagery running from Tavistock northwards to Brent Tor (general grid ref. SX 470 803)
and then possibly onwards to the fort at Okehampton. Another road may have run
westwards from Brent Tor towards a crossing of the Tamar. Kaye considers it is possible,
therefore, that there was a Roman military presence at Brent Tor.
Devon HER: MDV 133425, see Heritage Gateway ID: SDV365044 Source: S.Kaye (2020)
Further Notes:
� Dr. Chris Smart has separately been investigating this route, including a possible branch
to the Tamar, plus continuations north of Brentor to Sourton (see also the RR492a Fowey
Cross report below)
� The Devon and Dartmoor HER presently shows a conjectural route of the Roman road
(RR492a) running westwards from Okehampton towards the Tamar (MDV18485). The
line of the road is particularly clear across Sourton Down (SX 544 917) where earthworks
survive and where stretches of metalled road have been found during excavations.
Oakford to Chawleigh
Work by Dr. Chris Smart/ ‘Understanding Landscapes’ project
Lidar analysis suggests a probable Roman road crossing from the Devon – Somerset border,
at SS 887 238, to Burridge Moor Cross, Chawleigh, SS 715 116. Whilst evidence is
discontinuous, what there is shows a coherant route, c.20km (12.4mls) long, via Canworthy
and East Worlington. It is possible that the route continued south to via currently-used ridge
roads to the Roman complex at Bury Barton, Lapford (SS 73325 07035, SS 73331 07226). A
junction with a south-easterly branch is suggested near Rull, at SS 768 127, with this second
road heading toward Cann’s Mill Bridge at SS 784 107.
Further Note:
The Historic England listing of the fort complex at Bury Barton states that partial
excavations undertaken within this fort indicated occupancy during the Claudio-Neronian
period.
Plympton to Clearbrook
Work by Dr. Chris Smart/ ‘Understanding Landscapes’ project
Lidar analysis shows intermittent segments of earthwork suggesting a probable Roman road
between Newnham Park, Plympton, near Plymouth (SX 548 592), northwards to Clearbrook
- 252 -
Roman Roads in 2022
and Roborough Down (at SX 515 654). Recent excavations by AC Archaeology at the Sherford
New Community, on the south side of Plympton, have revealed an undated road anked by
a drainage ditch and multiple quarry pits, and the course of this appears to be directed
towards that identied by lidar analysis. OSL dates are eagerly awaited for this excavated
road.
HER aware. Investigations at this site began in 2015. Full details will become available when
the excavation report is published. The works and preliminary ndings have been reported
on BBC TV news reports (5 February 2023) and other media sites.
RR492a. Fowley Cross, Okehampton to Tinhay
Work by Dr. Chris Smart/ ‘Understanding Landscapes’ project
Lidar analysis, and past excavations at Sourton Down (Weddell and Reed, 1997), suggests a
Roman road running from the west side of Okehampton, at SX 561 945, as far as the
conuence of the Rivers Lid and Thrushel at Tinhay. Gaps between the sections noted on
lidar, and excavated at Sourton Down, are explained by the possibility that the historic route
- 253 -
Roman Roads in 2022
Possible Roman road discovered in advance of development at Sherford, near Plympton. Image AC Archaeology
between Bridestowe, Lewdown, Portgate, and Tinhay, fossilises a Roman road as described
by Margary in his routing of RR492a.
Sources: Weddell, P.J. and Reed, S.J. 1997, ‘Excavations at Sourton Down Okehampton 1986-1991:
Roman Road, Deserted medieval Hamlet and Other landscape Features’, Proceedings of the Devon
Archaeological Society, 55, 39-147.
In a separate email response (Feb.2023) further concerning the road to the west of
Okehampton, Steve Stocker notes that he also had intended to prove the missing link
between Trehill and the Roman fort at Okehampton but that this opportunity has now
unfortunately gone following a large housing development.
Possible Roman Road north and east of Brownie Cross, near Shaugh Prior, South
Devon
Work by Dr. Chris Smart
Possibly an extension of RR495(x) described above, a linear earthwork visible in sections as
a scarp, bank and hollow, possibly anked by further hollows or pits, are identiable on
visualisations derived from lidar data captured in 2013 and 2019 to the north-east of
Brownie Cross, Shaugh Prior (SX 5449 6060). The linear elements appear to measure between
10 to 20m wide and are identiable as discontinuous earthworks for a length of up to
approximately 1700m. As for HER 128231, below, it is suggested that the line is the most
obvious for a route south of Dartmoor attempting to get north of Plymouth (C. Smart, 2020).
Devon HER: MDV128227. Sources: Interpretation: Hegarty, C., Knight, S. and Sims, R.. 2019-
2021. The South Devon Coast to Dartmoor Aerial Investigation and Mapping Survey. Area 2, Avon
Valley to Plymouth (AI&M, formerly NMP).
Historic England Research Report; Personal Comment: Smart, C.. 2020. Lidar Interpretation.
Roman road west of Hartstone Farm, Shaugh Prior, South Devon
Work by Dr. Chris Smart
Linear banks and hollows interpreted as possible evidence of a Roman road are identiable
as earthworks on visualisations derived from lidar data captured in 2013 and 2019, crossing
the slopes of a south-west facing spur to the west of Hartstone Farm, Shaugh Prior in South
Devon, and a few miles to the north of Plympton (SX 5368 6166). It has been suggested that
this route may follow the continuation of a curvilinear route previously recorded to the
south. Further assessment has been recommended. It is suggested that the line is the most
obvious for a route south of Dartmoor attempting to get north of Plymouth (C.Smart, 2020).
Devon HER: MDV128231. Sources : Interpretation: Hegarty, C., Knight, S. and Sims, R.. 2019-
2021. The South Devon Coast to Dartmoor Aerial Investigation and Mapping Survey. Area 2, Avon
Valley to Plymouth (AI&M, formerly NMP).
Historic England Research Report; Personal Comment: Smart, C.. 2020. Lidar Interpretation
- 254 -
Roman Roads in 2022
Possible Roman road, RR4f? to north of Northgate Lane, Honiton, East Devon
The recent interpretation of a long low bank visible on lidar imagery to the north-east of
White Cross (ST 1884 0079) as a possible Roman road suggests that, in turn, the aligning eld
boundaries to the north of Northgate Lane may also mark the course of the road. The
onward course of the road is faintly visible as cropmarks on earlier RAF aerial photos beyond
Cheeseway Ash. The road converges with two possible routes down towards Honiton at
Northcote Hill Farm.
Devon HER: MDV133442. Source: Interpretation: Watts, S.. 2022. Roman road to east of Honiton.
Roman Road from Exeter to Cullompton
Possible route of the Roman Road from Exeter's East Gate northwards to the Roman fort on
St Andrew's Hill in Cullompton (central grid ref. SX 971 999). Sections of the road are visible
as earthworks on lidar; other possible sections appear to be preserved in eld boundaries
and extant roads. However, some sections are conjectural. The existence of this road is
supported by the distribution of early Roman coinage and a settlement at Shortlands Lane
in Cullompton where other nds are suggestive of a roadside settlement.
A section of the road is thought to have passed through the National Trust (NT) Killerton
Estate, approximately mid-way between Exeter and Cullompton. Excavations and surveys
have been carried out at the Killerton estate in recent years (2021, 2022) into a possible fort
and other features. Exeter University and the NT are to undertake further works in summer
2023 to include investigations of the possible Roman road.
Devon HER MDV124645. Source: Horner, B., 2019, Roman Road from Exeter to Cullompton
(Interpretation). Unpublished NT/Exeter University, interim survey report, Killerton Estate,
Feb. 2023
Dorset
Historic England have largely funded Aerial Investigations and Mapping (AI&M) Projects, by
F. Fleming and C Royall, of Cornwall Archaeology Unit, Cornwall Council, assisted by C.
Pinder, Senior Archaeologist for Dorset Council. The most recent projects have been “Upper
Frome and Sydling Water, West Dorset” and the “Middle Stour”, by aerial photography and
lidar.
Historic England have made a report of this work that is available to download athttps://
h i s t o r i c e n g l a n d . o r g . u k / r e s e a r c h / r e s u l t s / r e p o r t s / 8 4 7 8 /
UpperFromeandSydlingValleysWestDorsetDorsetAerialInvestigationandMappingProject
Two Roman roads are quoted in the report. The rst RR47, between Dorchester and
Ilchester, is well known and established. The second while stated as a Roman road is not
clearly dened as such. This is RRX006, Grimstone to Exeter, the only evidence for it being
some antiquarian statements with no more recent evidence.
- 255 -
Roman Roads in 2022
Badbury to Salisbury, RR4c, centred on ST 9719 0464
Work by Dorset Middle Stour Catchment AI&M Project
A 1.7km continuous stretch of the Roman road is visible on lidar imagery near King Down
Farm and was digitally plotted during the Dorset Middle Stour Catchment AI&M Project in
2021. A nal Report from the compilers through Historic England has not yet been produced.
Badbury to Donhead and eventually Bath, RR46, centred on ST 9631 0430
Work by Dorset Middle Stour Catchment AI&M Project
This record is for the stretch within Pamphill which is intermittently visible as cropmarks
and earthworks on aerial photographs and lidar imagery and was digitally recorded during
the Dorset Middle Stour Catchment AI&M Project in 2021. A nal Report from the compilers
through Historic England has not yet been produced.
Badbury to Poole, RR4d
Work by Dorset Middle Stour Catchment AI&M Project
A section of the Roman road from Bath to Poole is visible as cropmarks and earthworks on
aerial photography and lidar imagery between Lodge Farm and High Wood. The road is
intermittently visible as a central bank with anking ditches and was digitally recorded
during the Dorset Middle Stour Catchment AI&M Project in 2021. A nal Report from the
compilers through Historic England has not yet been produced.
Frampton Roman Villa, Maiden Newton. Centred on SY 61599 95282
Work by D. Stewart, M. Russell, P. Cheetham, and H. Manley. 2022.
The site’s south east eld boundary/track is straight and runs north east to a current
footbridge, then continues straight to the known Dorchester/Ilchester (RR47) road and
stops. This may have been a connecting Roman road.
HER aware.
Ackling Dyke, RR4c, exposed. SU 0074 1442
Observation by Penny Jackson
A section of Ackling Dyke (RR4c) which has been cut across by a modern road in a such a way
which exposes the Roman construction. RR4c is such an enormous road that it is especially
clear. It's a little obscured by vegetation but the different layers can be seen very clearly.
Badbury to Dorchester, RR4e, centred on ST 9333 0141
Work by Dorset Middle Stour Catchment AI&M Project
The Roman road is visible as earthworks on aerial photographs and lidar imagery and was
digitally plotted during the Dorset Middle Stour Catchment AI&M Project in 2021.
- 256 -
Roman Roads in 2022
Durham
Hawthorn Pit Solar Farm, centred at NZ 38429 46183
Work by Pre-Construct Archaeology Limited
From a preliminary geophysics survey suggesting a high archaeological potential, a eld
evaluation was carried out on the proposed site of a solar farm. Previous work on this site by
the Northern Archaeology Group had exposed, in two sections, what they believed was a
Roman road, running approximately east/west. This was unusually narrow at
approximately four and seven feet wide respectively. Recent trial trenching expanded on
this revealing post holes, an east/west road with side ditches as well as a number of other
ditches and gullies. The geophysics had suggested there was a sub square enclosure
containing a round house. A branch from the road, revealed by surviving parallel ditches
rather than the road structure, towards the north and the cluster of features around the
junction are thought to equate to a Romano British roadside settlement with possible
livestock enclosures.
While the existence of roads in this area to the east of the known network had been
suggested by previous Northern Archaeology Group work, discovery of the slightly built
narrow road with a junction , conrms the existence of a Roman period local road and track
way beyond and between the known major routeways.
HER aware: S73414. Vance, S., 2022, Land at Hawthorn Pit Solar Farm, County Durham. PCA
Essex
Colchester to Gosbecks Archaeological Park, Essex TL969163 to TL987795, RR321
Work by Colchester Archaeological Trust (CAT)
Colchester Archaeological Trust (CAT) have recently had the rare opportunity of carrying
out a number of small excavations over the Roman road (RR321) between Colchester and
Gosbecks Archaeological Park, to the southwest of the city. The road was known to Margary
from aerial photographs, but its exact route and destination were unclear. The series of
recent excavations, which came about mainly due to small development projects (the road
is now largely built over by the ever expanding city of Colchester), has conrmed the route
in more detail, and in particular has revealed that the road was formed of a central
carriageway of some 7m in width, anked by ditches approximately 1m wide, and with outer
surfaces, assumed to be footways, of about 1m in width, again anked by outer ditches of
about 1m in width. Rather oddly the archaeologists could nd no traces of metalling of these
outer surfaces, although excavations were not conclusive on this issue. This road layout also
seems to be rare in Britain, although several instances are recorded in the area around
Colchester.
Excavations in the past have revealed a nexus of roads beneath the modern Colchester Royal
Grammar School, where routes from Gosbecks Romano British religious and administrative
centre meet roads from Braughing (Stane Street, Margary RR32) and the main easterly route
- 257 -
Roman Roads in 2022
from London (The Great Road, Margary RR3b), joining a short road to the main entrance to
Colchester walled town via the Balkerne Gate. This arrangement of roads is not well
understood, due mainly to the lack of opportunities to excavate in an urban area, but
evidence both recent and in the past has revealed that the short connecting road between
Balkerne Gate and CRGS probably went out of use sometime during the 3rd century,
following the apparent closure of the Balkerne Gate.
Recent excavations have also revealed that the main carriageway of RR321 appears to have
been widened sometime during the early 2nd century, to cover the inner ditches and
footways to a width of about 10m, as it approaches the CRGS nexus. This section of road was
also clearly metalled with packed gravel as is typical in this area and appears to have
continued in use until the 4th century. The reasons for this modication remain unclear, but
it seems that there was possibly an increase in trafc volume during the 2nd to 4th centuries.
HER Reports ECC4095, ECC4380, ECC4448, ECC 4436 Archive: Colchester Borough Council
HER Published Report: CAT reports 1587, 1228, 1167
RR33a, Borley, Essex TL851430
Work by Colchester Archaeological Group (CAG)
Ivan Margary records his route RR33a all the way from Chelmsford in Essex to Ixworth in
Suffolk where it joins RR330 and becomes known as the Peddars Way (RR33b). However, an
8 mile section of RR33a between Goseld and Long Melford was until recently unknown,
despite many attempts to trace its route over the years. Members of CAG recently spotted a
very faint cropmark of classic tramlines in a eld between Long Melford and Borley, with
an equally faint trace on lidar suggesting a buried agger. On closer inspection CAG were able
to trace the probable route of the road on lidar heading SSW towards Gestingthorpe Roman
Villa. The route crosses the Belchamp Brook where a modern road bridge still exists at
Bardeld Bridge, then follows what appears to be a terraced section along the southern
bank of the brook which rises steeply towards Smeatham Hall. The lidar trace is lost at this
point, but further south the route is picked up precisely by a hedgerow, with faint traces of
an agger to both north and south. Projecting the route a little further south, the road passes
within 200m of the villa site, and is again lost beyond that point. North of the cropmarked
eld, the road heads for the River Stour, which it must have crossed to make the approach
to Long Melford Roman settlement. A record is said to exist of a Roman bridge having
crossed the Stour just north of the modern “Rodbridge”, which would have been a perfect
location for RR33a to make the crossing, and a local newspaper report from the 1960’s
records how workmen uncovered a Roman road surface close to the crossing point on the
east bank of the river in Suffolk.
The local landowner (the son of the original discoverer of Gestingthorpe Villa) is a keen
amateur archaeologist and member of CAG, and has recently commissioned a
magnetometer survey of the area around the villa. Results are not yet available, but it is
hoped that further traces of the road might be revealed.
During investigations into the route of the road, faint traces of what appears to be a double
banked oval structure were also revealed close to the villa site, measuring approximately
- 258 -
Roman Roads in 2022
350m x 250m. The size and shape of the structure resembles a nearby monument at
Pitchbury Ramparts, which is thought to be an Iron Age bivallate enclosure.
HER Report No. N/A. Archive: N/A. Published Report: TBD
RRX017, Elmstead Market to Jaywick, Essex TM066242 to TM146146
Work by Colchester Archaeological Group (CAG), Colchester Archaeological Trust (CAT)
RRX017 is recorded by the Ordnance Survey records as connecting Elmstead Market with St.
Osyth in Essex. Colchester Archaeological Trust (CAT) were recently requested to
investigate two areas to the south East of Elmstead Market ahead of new development
- 259 -
Roman Roads in 2022
The surface of the Roman road was revealed during work to install a new bus stop in the 1960’s. From a newspaper
clipping of unknown source.
works. Traces of RRX017 and a possible trace of RRX119ax, which is suspected of connecting
Elmstead with Mistley on the south bank of the River Stour where a Roman port is suspected
of having existed. The OS have recorded that RRX115 may also have connected Elmstead
with Alresford to the south, although this has not been recently investigated.
CAG embarked on a desk based examination of route RRX017 which appeared to have been
somewhat forgotten in recent years, perhaps due to its not having been included in Ivan
Margary’s book Roman Roads in Britain. A number of cropmarks were noted which matched
those recorded by the OS surveyors during work to create the OS map of Roman Britain.
Lidar traces were also noted which would of course not have been available to the OS
surveyors (in the 1960s we believe?).
The OS surveyors were able to trace the road as far as the St. Osyth Bypass Road to the east
of the village, but apparently no further. CAG were however able to spot a faint tramline
cropmark to the north of Jaywick at Sacketts Grove, which appears to have been unrecorded
by the HER. There is a known Roman settlement in the area of the golf course at Jaywick and
it would appear that RRX017 was making for that settlement. This cropmark has recently
been reported to the Essex HER.
It is also assumed that a branch road would have existed From RRX017 to St. Osyth where
many traces of Roman activity have been recorded, including a cropmark of a possible
fortlet or marching camp, which is still faintly visible on Google Earth imagery.
HER Report No. TBA. Archive: N/A. Published Report: CAT reports 1575, 1880 forthcoming
RR21b, Braughing to Great Chesterford. Book; The Road
By Christopher Hadley
See the Hertfordshire entry linking to the book review elsewhere in this volume.
Roman Roads around Harlow, Essex - Part 1, RR30 and RR328(x),
From David Ratledge
A lidar review laid out in the RRRA Winter 2022 newsletter starts to consider the roads
around Harlow. This is available at https://romanroads.org/Newsletters/Members/
Newsletter-24%20Christmas%2022.pdf
Gloucestershire
RR61, Mitcheldean to Gloucester
Work in 2019 by Carrozzo, F. and Garst, L.
Geophysical survey of land at Churcham suggest that anomalies with a probable
archaeological origin have been detected, in the north of the survey area, which align with
the projected route of the Mitcheldean to Gloucester Roman Road. This stretch of the road
- 260 -
Roman Roads in 2022
appears to be represented by multiple anking ditches, visible as anomalies between
Beauchamp House and Edgewood House, Churcham.
HER aware, Source Work 17350
RR541, Gloucester to Sea Mills
Work in 2019 by Thomson, F. of Headland Archaeology
Watching Brief undertaken on land at Hardwicke, Gloucestershire, during the stripping of
an easement and pipe trench excavation for the construction of a sewerage pipeline. The
investigation identied the remains of a Roman Road, likely to be the former Sea Mills to
Gloucester Road.
HER aware, Source Work 15785.
Gloucester Colonia streets
Work in 2020-21 by Sausins, D. of Cotswold Archaeology
Evaluation of land at Gloucester City Campus (former Debenhams), St Aldates Street,
Gloucester. Test pit 2 produced evidence for a cobbled Roman road surface, dating to
between the 2nd and 4th centuries AD, was identied, and corresponds with the location of
East-West Street previously identied during the 1958-1959 Bon Marche excavations.
HER aware. Quoted in part from Sourcework 17793.
Cirencester streets, NGR: 40288 20149
Work in 2021 by Reynish, S. of Cotswold Archaeology
An archaeological watching brief was undertaken at Cirencester Primary School, Victoria
Road, Cirencester at the request of Gloucestershire County Council (GCC). Roman road
surface, identied within Trench 102, is likely to be part of Roman Intramural Street D as its
projected line runs directly under the observed groundworks. Roman Intramural Street D
had been previously observed immediately north-west of the current site during
archaeological excavations in 1963 and during excavations in 2004 and 2006 at the site of the
former Angel Cinema along Lewis Lane (Holbrook and Pamment, 1998, 28 and Holbrook,
2008). The records from both of these excavations describe the road as a limestone metalled
surface. The surface is also similar to Ermin Street Roman road surfaces identied within
nearby excavations at Bingham Hall in 2002 and 9 Church Street in 2013.
HER aware, Source Work 17276.
Hampshire
RR42a, Winchester-Silchester centred on SU 60306 52550
Work by MOLA, commissioned by RPS acting on behalf of a client.
A second phase of archaeological investigation was completed at the Manydown site,
Basingstoke, Hampshire. The evaluation was carried out by MOLA, commissioned by RPS
- 261 -
Roman Roads in 2022
acting on behalf of the client. The 321ha site, centred on NGR SU 60306 52550, lies
immediately west of the Winchester-Silchester Roman Road. 1171 trenches were opened,
388 of these encountering archaeology.
Features included enclosure ditches, pits, postholes and ve Sunken-Featured-Buildings.
Dating evidence from features indicated a late Iron Age/ Roman date, whereas pottery from
the Sunken-Featured Buildings was of early-middle Saxon date. The site as a whole has been
interpreted as an agricultural settlement of intermittent activity. The development of the
Saxon settlement alongside the Roman road could indicate that the road was still in use into
the 7-8th centuries.
Evaluation report not yet in the public domain.
RR4b, Old Sarum to Silchester, centred on SU 59770 59850
Work by R. Pesnicar of Wessex Archaeology
A Ground Penetrating Radar and Gradiometer Survey were carried out for Tadley Cemetery
extension. The site, centred on NGR SU 59770 59850, covered an area of 0.5ha to the south of
the current cemetery at St. Peter’s Church. The survey was completed by Wessex
Archaeology and discovered a section of the Old Sarum-Silchester Roman road crossing the
site on a SW-NE alignment. The road’s agger is visible along almost its entire length and the
survey also identied both roadside ditches.
Plesnicar, R, 2022, Tadley Cemetery Extension, Tadley, Hampshire: Ground Penetrating Radar and
Gradiometer Survey Report, Salisbury, Wessex Archaeology.
Clausentum (Exton) Street grid.
Work by NEHHAS FAB from Richard Whaley
Clausentum is mentioned in the Antonine Itinerary VII which was the reason for looking for
a direct Roman Road Winchester - Chichester. The Iter mileages put Clausentum at Exton by
A32.
We were able to hold our Easter 2022 Dig, and the farmer kindly opened a machine trench
across what we dub the Principal E - W Street Grid of Clausentum near Exton. We call it this
because it shows up on the lidar without any special processing we use for air photos - while
Roman Street Grids usually have Principal (wide) and narrow Streets.
We found why it was clearly visible on the lidar - the chalk bedrock had been cut into, so the
material forming the Street was partly contained within the chalk bedrock. This material
was similar to that found last year for the Streets - sand coloured clay containing hard white
chalk, softer white chalk, and soft creamy chalk which forms the natural. The rst three
materials would have been brought in from elsewhere. This Principal Street probably had a
Running Surface of hard white chalk - but it was not as clear as we reported last year for the
other Streets. Its width however is not signicantly wider than the other Streets at 8m.
We also tried to nd this Street on the remains of a Causeway which carried it across the
ood plain of the River Meon over 2019 -20, but concluded erosion had removed evidence.
We found other Streets on this Causeway at a higher level. Recent post-excavation drawings
- 262 -
Roman Roads in 2022
indicated its northern edge, due to complex erosion issues worked out for a minor Street
excavated at a higher level.
On the long June Bank Holiday weekend we looked at the Principal North - South Grid. This
shows up clearly on the colour lidar print, but not on the black and white prints - though it
also shows up on air photos as a light line up to the Roman building found by the A32 - but
not beyond. It is the only N - S Grid line detected. We took a Base Line at right angles to the
eld/garden boundary along which we cut the other machine trenches for 70m west. Bruce
then cut a machine trench for us from 50W to 70W where we expected this Street to run.
This does not go across the Street at right angles, but has the advantage that we use the same
archaeological grid for all the machine trenches.
There was a plentiful amount of sand coloured clay from about 54W to 65W around 40 - 45
cms below ground surface, with hard and soft white chalk - these three materials would
have been brought in from elsewhere. There was however less white chalk than seen on
other Streets - there was probably a coating of hard white chalk as a Running surface but
again less prominent than other Streets. There was also creamy soft chalk, which is the
natural in this area - and as in the Principal E - W Street the central region of the Street the
natural had been cut into, so the material making the Street was partly contained by the
natural bedrock chalk. We thought this was why this Principal Street showed up on the lidar
without the need for special processing needed for the other Streets. The width of the
Principal N - S Street from the distribution of sand coloured clay was around 11m (correcting
for not going across it at right angles). All the E - W Streets and Roman Road Lanes have been
found to be around 8m wide - and these gures agree with the lidar.
There were differences with this Principal N - S excavation than most of the others. It was
not clear where the Street started and ended. In the others the chalk surfaces was generally
higher over the feature and tapered down on either side, or a specic revetment was found
keeping the Street material together. Though it was apparent in cases of a revetment that
the sand coloured clay material had leached out beyond it. It was generally clear where the
Street or Roman Road material ended, and the chalk bedrock began - but not so clear in this
Principal N - S excavation. There was probably a greater proportion of sand coloured clay,
and this could seep down into natural cracks in the upper chalk bedrock. A number of Box
Sections were taken down to 70cms below ground surface to trace the extent of the sand
coloured clay - but at present it is not clear whether this depth is due to seepage rather than
the Romans going this deep. However unlike most other excavations there were a number
of Roman nds - when these have been plotted into composite sections it may become
clearer how deep the Romans dug. One nd was part of a Roman necklace.
Excavation on Beacon Hill
Work by NEHHAS FAB from Richard Whaley
Over the August Bank Holiday 2022 we returned to Beacon Hill, and excavated the wide 40'
Lane, having previously excavated the two smaller lanes. It was found near the ground
surface of chalk lumps within a sandy coloured clay - which is similar material that the
Streets are made of at the Exton Clausentum site. It had been badly eroded on the sloping
- 263 -
Roman Roads in 2022
surface, but a large amount of Roman pottery was found in the soils - which are likely to be
washed down from a Roman site on the land above. The site was by the NE boundary of the
Lomer site, which we interpret as a complex of Temples. Various features make enclosures
for the Temple complex, and the NE enclosure is the 40' Roman Lane, with a bank and a
ditch. The ditch shows on air photo as continuing to the NW boundary but not beyond. The
excavation found this ditch coming through the boundary, but it was now quite small.
Excavation on the Chichester Road approaching Exton
Work by NEHHAS FAB from Richard Whaley
This report published during the year in e- News 23 found a compartment beside the Roman
Road which may have been a child's grave http://www.nehhas.org.uk/nc16.htm
RR400(x), Winchester to London
Work by NEHHAS FAB from Richard Whaley
We returned to the Winchester - London Roman Road in September 2022, where a strange
feature had been found on the edge of Chawton Park Wood. It was a 60 Roman foot wide
ditch with running lanes on both edges. But air photos suggested a light line in the middle
of this ditch - not found in excavations in the wood. An excavation in the eld in September
2022 found a large amount of crushed int, but disturbed by the plough. The 60 Roman foot
ditch had been cut some 30 cms into the clay, but at this excavation site the clay had come
back to its more usual depth below the ground surface. In the previous excavations these
running lanes were found to be quite complex features. Presumably this is a central running
lane in the middle of the ditch, which we probably did not see in the excavation in the wood
due to a tree. We will try a careful excavation in the wood when cropping arrangements
allow. Our interpretation of this feature has been of a Roman water feature, which collected
water in this ditch as it ran down to the bottom of the valley, where there is a little evidence
of a mill pond. The side running lanes could be to haul logs down the slippery clay surface.
How would a central running lane in the middle of the ditch be interpreted?
We have found other Roman water features, and on this site and on Beacon Hill we have
conducted Dew Pond experiments. On the September dig we conducted the largest Dew
Pond experiment so far, and got the largest rise in water level. These experiments started
when we found a large swimming pool size Roman water storage facility on high ground -
was it a Roman Dew Pond? This issue of e-News 25 contains a mathematical treatment of
how Dew Ponds collect dew during Summer nights. Any mathematicians are asked to send
in comments. If correct, this Roman Dew Pond could have risen by a Roman foot overnight
http://www.nehhas.org.uk/dew22.htm
RR421, Roadside settlement at Wickham, NGR 457635 111108
From Pippa Bradley, Post-Excavation Manager, Cotswold Archaeology
Excavations between 2018 and 2020 revealed part of a Roman roadside settlement. Features
of predominantly early Roman date were identied although part of a Bronze Age barrow
- 264 -
Roman Roads in 2022
and two Bronze Age cremation burials were also identied. A length of Roman road (RR421)
represented by poorly preserved int metalling in a hollow-way, with a metalled trackway
running broadly parallel to it was revealed. Roadside ditches were discontinuous and
irregular while immediately to the south a series of boundary ditches dened six enclosures.
The enclosures, and a small area on the northern side of the road, contained a large number
of wells including one with a timber lining, and pits, several waterholes and other features
were also revealed. Post-excavation analysis is currently underway.
A summary report can be obtained from Cotswold Archaeology’s site at https://
cotswoldarchaeology.co.uk/wickham-roman-roadside-settlement-and-wells/
Hertfordshire
RR21b, Braughing to Great Chesterford. Book; The Road: A Story of Romans and Ways
to the Past
By Christopher Hadley
Reviewed elsewhere in this volume, this book, while a descriptive story of the course of this
road also describes all the previous work along its length. This provides an effective and
denitive summary of the evidence for the road.
RR32, Stane Street at Bishop’s Stortford, Grange Paddocks. TL 490 219
Work by Oxford Archaeology
As well as the Roman road, press reports tell of the discovery of a roadside settlement. This
has yet to be reported. See https://www.ancient-origins.net/news-history-archaeology/
romano-british-settlement-0016751
Kent
RR1b, Watling Street at Snydale. Centred on NGR 599834 160996
Work by The Kent Archaeological Field School led by Paul Wilkinson
Previous reports of this work in Volume 2 of Itinera was brief and illustrative, a more
complete report has now been issued with the prospect of a fuller report on the Watling
Street excavations to come. Sections of the road, that were up to 14.6m wide and 1.23m
thick, revealed that it had been rebuilt at least seven times. On a clay agger approximately
50cm high, was constructed a base layer of large int nodules with further distinct layers of
sand and gravel with what seems like some added lime that had set the structure very
solidly. The initial agger, was dated to c.AD50 in the Claudian period by included pottery.
The rst road surface was constructed on this agger, 3.6m wide, dated as early in the
occupation from embedded coins. This was covered by approximately 50cm of topsoil that
sealed on the road surface quantities of domestic debris. Roman trafc had seemingly
- 265 -
Roman Roads in 2022
stopped and adjacent buildings were abandoned. The road was rebuilt in the third century
with the nal layer being just before it went out of use in the early years of the fth century.
The roadside houses and shops appear to have come to a violent end with shoes and other
items scattered across the surface of the road.
The Itinera committee are hoping that a fuller account to this work will be presented in a
future edition of Itinera .
HER aware. Summarised from the assessment report, KAFS, 2022, Archaeological Investigations
on Land at Syndale Park 1999 to 2011, Faversham, Kent. Time Team covered some of this work in
their 2003 season.
Lancashire
RR703, Ribchester to Kirkham, Lancashire. SD 478 316, Ward's House Farm Excavation,
Newton-with-Clifton, 25th June 2022
From Andy Jackson and David Ratledge
Wyre Archaeology carried out an excavation just to the east of the Springeld/Salwick
complex. This spot was believed to be the furthest west the road had been excavated. The
agger showed up sufciently well in lidar to locate the road but indicated that it was
probably not in pristine condition due to ploughing. A 6 x 1 metre trench was opened across
the top surface of the road revealing a stony surface with a distinct camber. At the end of the
dig the team dug a small pit down through the centre of the road but there was nothing
substantial revealed. No larger foundation stones were present or agger layer structure. The
construction was similar to the road excavated east of Preston at Redscar.
This extract is copied from the RRRA Autumn 2022 newsletter at; https://romanroads.org/
Newsletters/Members/Newsletter%2023%20Summer%202022%20(1).pdf
Leicestershire
RR573a(x), Roman Road from Leicester to Ryknield Street
Work by David Ratledge
Lidar has revealed what appears to be a Roman road heading north west from Leicester. The
known Roman road RR57b runs from Leicester to Watling Street near Mancetter. However
its alignment is well to the west of Leicester strongly implying that it was likely to be a
branch off an unknown Roman road heading north-west from Leicester. Lidar has now
conrmed this to be the case with this “new” road running very directly to Ryknield Street,
to the south of Burton upon Trent in Derbyshire, (see lidar image). This route could t as
part of the oft suggested Via Devana.
The route leaves Leicester over Bow Bridge and heads at rst along King Richard’s Road-
Gleneld Road. This initial section was believed to be part of RR57b but is now better
regarded as part of a road to Ryknield Street. It becomes extant in Western Park descending
- 266 -
Roman Roads in 2022
- 267 -
Roman Roads in 2022
The route of the newly recognised Roman road from Leicester to Ryknield Street, south of Burton upon Trent. Base mapping is © Ordnance Survey
Opendata
(SK 55724 04834) and ascending (SK55548 04854) from a small stream. It is next visible across
a golf course (SK 53585 05342). The motorway and Ratby then intervene with the main
alignment clear beyond the latter (SK 49886 06353) all the way to a reservoir (SK 47675
07659). West of the reservoir the road is less obvious but the alignment does continue (SK
47411 07815) and is visible again as a cutting at SK 45547 08957. There is a slight change of
direction just before it crosses Ellistown Terrace Road. It is then evident on the latter’s
western side (SK 43601 10149) and a nal faint agger is visible before disappearing under
Ibstock (SK 42293 10707).
Evidence reappears beyond the A447, Melbourne Road (SK 40415 11523) and again beyond
Ravenstone Road (SK 39235 12073). Ahead now is a ridge of high ground at Normanton-le-
Heath but this is negotiated via a cutting (SK 37855 12613), aptly named today “The Hollow”.
A very obvious agger is then visible in the lidar imagery (SK 37079 12916) followed further
on by a much fainter one (SK 35542 13518).
The next trace straddles the A42 (SK 34232 14153) and is on a slightly different alignment to
negotiate a steep slope. The road would have angled down the slope but the main alignment
is resumed as a terrace (SK 32548 14838) followed by prominent aggers either side of Ashby
Canal at Moira (SK 31859 15144 & SK 31452 15334). The road now enters Derbyshire and
begins its nal alignment to Ryknield Street.
- 268 -
Roman Roads in 2022
A section of the road through Moira in north west Leicestershire as excavated in the 1990s. The road was known
locally as ‘The Pink Road’ apparently due to the colour of the surface gravel, or ‘The Leicestershire Headland’ and
for many years was considered a folk tale.
The alignment recorded at Moira in Leicestershire continues and lidar imagery rst reveals
the road in Derbyshire at Mount Pleasant either side of Burton Road (SK 29151 16312 & SK
28790 16592), with the west side the more certain. There follows a short gap before the road
angles slightly to climb a ridge near The Stables (SK 27001 17328). Once the height of the
ridge is gained (SK 26789 17533) then the last alignment begins that will take the road to the
River Trent (SK 22396 19591). It passes the aptly named Hill Crest and is very obvious in
several sections (SK 25340 18227 & SK 24406 18641). This last alignment from Hill Crest is
recorded in the Derbyshire HER as “conjectural” but we can now be certain that it is indeed
part of a Roman road from Leicester to Ryknield Street, RR18c, which it probably joined
around SK 21487 20028. There are lidar indications suggesting that the road continued
beyond Ryknield Street climbing Hobhole Dingle and heading in the general direction of
Rangemore (see also the Staffordshire and Derbyshire sections). Whether or not this road
was ever perceived as part of a road to Chester, supporting the notion of a Via Devana, is
currently unclear. Work is continuing.
HER aware.
See also Staffordshire and Derbyshire for the course of this road surveyed on the west
bank of the Trent
Greater London
Below is a digest of material collated by Matt Sparkes from recent volumes of London
Archaeologist (LA), Surrey Archaeological Collections (SyAC) and Transactions of the
London and Middlesex Archaeological Society (ToLAMAS). Three reports represent negative
evidence (locations where the suspected course of a road was not found). However, in a
heavily built-up urban environment where the exact courses of Roman roads are rarely fully
understood that information was considered of sufcient value to warrant inclusion here.
City of London, 56-62 Moorgate / 41-42 London Wall, EC2M, TQ 3272 8153
Work by Shane Maher of Pre-Construct Archaeology.
Excavation revealed ‘the remains of an early 2nd century kiln/oven structure with associated [late
1st-late 2nd century] timber buildings. […] In the late 2nd century a road was established in the
north portion of the site which appears to have remained in use during the 3rd and 4th centuries along
with its associated ditches’.
Source: LA 15(S3b) p97.
City of London, 6-8 Bishopsgate, EC2, TQ 3308 8116
Work by Al Telfer & Sadie Watson of MOLA (Museum of London Archaeology).
‘Three evaluation trenches were excavated in the courtyard area of 6-8 Bishopsgate. […] In the south-
east [of the courtyard] a linear cut was observed running east-west, possibly a ditch associated with
the Roman road known to have crossed the site’. Ermine Street (RR2) runs north-south along
Bishopsgate, so if by ‘ditch associated with the Roman road’ we infer a roadside ditch (rather
- 269 -
Roman Roads in 2022
than a boundary ditch running perpendicular with RR2) then the road referred to is
presumably the assumed east-west route from Aldgate to Cripplegate fort.
Source: LA 15(S3b) p97.
Borough of Southwark, 25-29 Harper Road, SE1, TQ 3230 7950.
Work by Ireneo Grosso of Pre-Construct Archaeology.
The site is c.1.1km SSW of London Bridge and c.200m east of Stane Street.
‘The earliest Roman activity took the form of gravel quarries, dated AD 55-70, […] which were
associated with construction of a road, 40m long, on a north-east/south-west alignment. It
extended beyond the northern and eastern boundaries of the site, and its full width is unknown.
To the west of the road, a sequence of parallel roadside ditches dating from the second half of the
1st to the late 4th centuries were recorded. In the centre west of the site, and parallel to the road,
were Roman ritual shafts or wells, together with the post-holes of a possible Roman timber
building. These features had been truncated by a substantial chalk foundation, which formed
part of a rectangular or square building lying parallel to the road and extending beyond the
western limit of excavation. It is believed to have been a mausoleum. The eastern side of the
foundation had subsequently been cut back for the insertion of a plain sarcophagus […]’.
Source: LA 15(S2b) p74. (The full report of this site, Edwards 2019, can be found in ToLAMAS
70, pp81-104.)
Borough of Southwark, site bounded by Stoney Street, Bank End, Park Street, SE1, TQ
3250 8032
Work by Ken Pitt & Serena Ranieri of MOLA.
The site is c.300m southwest of London Bridge and is therefore close to the northern end of
Stane Street and within the area of Roman settlement on the south bank of the Thames. The
site was located ‘on a natural sand and gravel island dened to the south by a channel called Park
Street Creek’. The remains of a number of clay and timber buildings were found, those in the
north of the site were interpreted as domestic in nature and those to the south as
warehouses lining the creek. ‘To the south of these [warehouses] a series of gravel paths were
recorded’.
Source: LA 15(S3b), p121.
Borough of Southwark, 596-598 Old Kent Road, Livesey Place, SE15, RR1, TQ 2440 7780.
Work by David Saxby of MOLA.
‘Three evaluation trenches were excavated at the rear of the existing modern building and a
single trench was opened in a yard at the western end of Livesey Place. […] No evidence was
found of the Roman road projected to run close to the site [Watling Street] and it is likely that
later activity had removed any remains’.
Source: LA 15(S3b), p120.
- 270 -
Roman Roads in 2022
Borough of Merton: Baitul Futuh Mosque (Morden), 181 London Road, SM4 5PT, RR15,
TQ 253 680
Work by C. Sinclair of Pre-Construct Archaeology
For most of its course through South London, the line of Stane Street is clearly indicated by
the modern road network (i.e. by the A3/A24, one of the main arterial routes out of the
capital). However, for a distance of c.3km between Colliers Wood tube station and Morden
Park the modern road network diverges signicantly from this alignment and the exact line
of Stane Street is unclear. This large mosque site is located adjacent to Morden Park and
close to the assumed course of the Roman road. Unfortunately excavation found that;
‘a thick layer of made ground [overlay] heavily disturbed natural layers. […] No evidence of the
London-Chichester Roman road, or any anthropogenic activity pre-dating the 20th century, was
observed’.
Source: SyAC 104 p275.
Borough of Croydon: 32-42 Purley High Street, CR8, RR150, TQ 3138 6162
Work by Simon Pratt of Canterbury Archaeological Trust.
‘Geotechnical augering was monitored within an Archaeological Priority Area associated with a
nearby Roman road’ (RR150, London-Hassocks). Modern terracing meant that Pleistocene
deposits ‘had been truncated and overlain directly by modern construction horizons’. [Note: site is
c.100m east of the course of RR150 set out by Margary with some condence back in 1937].
Source: LA 15(S3b) p99.
Greater Manchester
The Roman Road from Manchester to Melandra Castle, RR715(x)
From Neil Buckley and David Ratledge
From lidar mapping, progress had been made in locating with high condence the long
suspected, but lost Roman road between Manchester and Melandra Castle near Glossop in
Derbyshire. In that report it was shown that an almost entirely straight alignment can be
traced between the top of Pin Mill Brow in Manchester and the Harrop Edge gap north west
of Mottram-in-Longdendale, a distance of just under 8 miles. Having passed through the
Harrop Edge gap it was evident that this road was not aligned on Melandra fort at all but on
a route into the Longdendale Valley. Since that discovery there has been a more extensive
release of lidar data by the National lidar Programme enabling the Harrop Edge/Roe Cross
area to be more effectively searched for further evidence of the road’s route towards the
Longdendale Valley, but also to reveal if possible, the course of the suspected branch road
down to Melandra Castle from Roe Cross.
- 271 -
Roman Roads in 2022
A digital copy of this work is available at https://romanroads.org/Newsletters/Members/
Newsletter_22%20Spring_2022.pdf
Manchester to Melandra – RR715(x), Anticipated Impact of A57 Improvement Scheme
Work by Neil Buckley and David Ratledge
Mottram in Longdendale centred on SK 996 958.
In the Spring 2022 edition of the RRRA Newsletter, David Ratledge and Neil Buckley
reported on the discovery of lidar traces that appeared to be the remains of the agger of
the missing section of the Manchester to Melandra Roman road RR715(x) - as above. The
suspected course of this section of the road runs from Roe Cross, to the north of Mottram
in Longdendale, through the Mottram Old Hall site and then down to the River Etherow
just below the Melandra fort site.
On its way to the river the Roman road crosses the A57, a busy road long scheduled for
improvements. It is perhaps fortunate then that the lidar evidence was found when it was
because the long planned A57 improvements were nally given consent to go ahead in
November 2022 and work was expected to start in Spring 2023. However in January 2023
the local branch of The Campaign to Protect Rural England (CPRE) launched a legal
- 272 -
Roman Roads in 2022
The possible roads through Roe Cross. Coach Road was the 'obvious' link to Melandra. Lidar 3D map by David
Ratledge. Base lidar data is © Crown Copyright 2022
challenge to the Transport Secretary’s consent thereby introducing a further delay to the
project’s start. Disappointing news if you are stuck in a trafc jam there, but a welcome
extension to the time available for the contractor’s pre-project archaeological
investigations currently being overseen by GMAAS. A copy of the Melandra road paper
published in the Spring Newsletter, was submitted to the GMHER last year in advance of
the planning decision to ensure that its presence was taken into consideration in the
approvals process. This will not protect the portion of the Roman road that will be
destroyed by the new build but it will hopefully lead to trenching across its suspected
track that will conrm the lidar indications.
The GMAAS Senior Planning Archaeologist, Ben Dyson has conrmed that the lidar
information produced by the RRRA has been passed to the contractor to ensure that it is
investigated. Ian Miller is the GMAAS Heritage Management Director and he recently e-
mailed to say,
“Please be assured that GMAAS is working quite closely with the team responsible for
delivering the A57 improvements, and we have ensured that investigating the projected line of
the Roman road is well and truly embedded in the trenching scheme that has been agreed. I
will let you know how things progress”.
If the road building excavations do conrm the lidar route for the Manchester to
Melandra road it should nally settle a question that has been unanswered since
Melandra was rst identied as a Roman fort in 1772.
- 273 -
Roman Roads in 2022
Proposed A57 improvements; compare with the previous lidar image. A large length of the probable Roman road
will be destroyed by the new road but trial trenching north and south of the A57 may conrm the line.
West Midlands
Ryknild Street RR18c, between Wall and Metchley. SP 08570 97493
From Keith Hopkinson and Paul Smith
In the RRRA Spring 2022 newsletter the well preserved remains of a segment of this road
running through Sutton Park was described. Here it was built by compacting material from
the abundantly available pebble beds that exist throughout Sutton Park. The road material
was quarried from quarry pits along the line of the road, with the pits surviving just as they
were dug by the Romans nearly 2000 years ago. The 9m wide central agger is anked by
intermittent side ditches 5m from its edge. In addition a small square earthwork alongside
the road was considered to be a possible small temporary camp for the construction gangs.
A few hundred yards to the north of the Golf Course Earthwork, a second agger diverges
from the original agger as the Longmoor Brook is approached This second agger runs
between the original agger and the outer ditch line, with the second agger slightly
overlaying the edge of the original agger. The second agger runs for approximately 100
yards from the north until the road crosses the stream. It is not visible to the south of the
stream. A quarry pit alongside this second agger seems to have been enlarged to provide
material for this rebuild. The purpose of the second agger is not known. It seems likely that
it is related to the stream crossing. Possibly the original agger was washed away by big
oods, which have occurred in the Park and the second built alongside as a replacement.
However, this does not account for why the original agger was not simply rebuilt. It also
does not explain why the second agger exists so far back up the slope away from the stream,
and not on the other side of the stream.
The article is available at https://romanroads.org/Newsletters/Members/Newsletter_
22%20Spring_2022.pdf
Northamptonshire
Blackgrounds near Chipping Warden, HS2 work. Centred on SP 513 485
Work by MoLA
Excavation in advance of the HS2 railway line has revealed, in a ‘vast wealthy Roman trading
settlement’, a Roman road. The site, which had been established as a group of roundhouses
around 400BC, included;
… discoveries showing that the settlement expanded over time, becoming more prosperous during the
Roman period, with new stone buildings and roads being built. The team uncovered a 10m (33ft) wide
Roman road, described as "exceptional in its size", running through the site. HS2 archaeologists found
evidence of workshops, kilns, several "beautifully preserved wells" and ery red-coloured earth, which
indicates activities such as bread-making and metal work occurred. Alongside the coins found, glass
vessels, highly decorative pottery, jewellery and evidence of make-up was uncovered. Traces of the mineral
galena - lead sulphide - a substance that was crushed and mixed with oil as make-up, was found on the site.
While HS2 is a controversial subject in its own right, it is undoubtedly generating a lot of
fresh archaeology. We await the nal reports with anticipation!
- 274 -
Roman Roads in 2022
The site featured on the BBC series Digging For Britain and the work hit the press as; https://
www.bbc.co.uk/news/uk-england-northamptonshire-59943179
Roman road at Priors Hall, Corby. Centred on SP 92463 91807 to SP 93342 92114
Work by Oxford Archaeology East led by Paddy Lambert
Between January and August 2021, open area excavation on Zone 3 of Priors Hall was
undertaken. Four separate areas were excavated, totalling 4.75ha. The site encompasses part
of the estate of a previously identied Romano British villa, within 100m of the current site
which is dened as an Archaeological Management Area (AMA). An Iron Age settlement was
abandoned in the rst century perhaps when the villa complex was being constructed. East
/west parallel ditches dene an approach road ‘Avenue’ to the villa with ditches 6.8 to 6.9m
apart, that was anked by a roadside rst and second century pottery kiln. This road
contained considerable surfacing of iron metal working waste.
Running north east to south west across the site were the well preserved remains of a
Roman road with a 5.35m wide surface. This was built up to 1m from the natural level with
a bottoming layer of limestone and clay, with a total width of 17.26m across the ditches. The
initial metalling was locally sourced limestone which was rutted by vehicles with a gauge of
1.6m. Later multiple resurfacing contained metal working waste but with a narrowed
surface of 3.5 to 4.5m. A large quarry 12m by 1m deep was associated with the through road
and a series of ve quarry pits and a single larger quarry anked the approach road
suggesting the source of material for the road construction.
- 275 -
Roman Roads in 2022
Image of road surface at Blackgrounds. From BBC news release.
This extensive work seems to have located either the western end of RR571, a road which
branches from Ermine Street near Ailsworth, or a previously unrecorded Roman road that
may have branched off or crossed RR57a. Also from it a local branch road serving a villa. It
was subject of an article in Current Archaeology (Issue 395, February 2023, 36-41) where a
fuller illustrated account with more descriptive detail can be read.
The Itinera Committee are hopeful that a fuller account of the roads may be presented in a
future volume of Itinera.
Reference review
Work by Dave Haywood
Not all discoveries need to come from new work or excavation. A review of older documents
for Northamptonshire has revealed a number of potential roads not currently covered by
Margary. These include what was described as a ‘lost road’ that appears to have headed
north eastwards from the ‘dog's leg’ in RR160 at Grid Reference SP 67896 38385 towards a
junction with RR172b. Similarly a review of the papers of the Wolverton and District
Archaeological Society have highlighted two potential Roman short routes with no RR
- 276 -
Roman Roads in 2022
View along the length of the Roman road at Priors Hall. Image from the Northamptonshire Telegraph.
number, one from Deanshanger to the north east and one from Cosgrove to the north west.
Both of these routes emanate from RR171 and head towards the area of the RB villas at
Bradlem and Wakeeld Lawn. There is also an unnumbered route heading to the north west
from RR172b in the Salcey area, facing directly towards Hunsbury and Upton. Also the
signicance of Haversham at the crossroads of the River Ouse / and RR175, RR172b & RR171
(at least) is perhaps something that warrants further consideration. This location has been
identied as a late IA, early RB port on the Ouse, perhaps being the upper limit of navigation
on that waterway. The routes emanating from it could have served a number of signicant
locations including Magiovinium and possibly Upton / Hunsbury. Further work is needed to
conrm, or not, whether or not these routes were in use during the Roman period, and
whether or not they were surveyed and engineered Roman roads, or poorer quality local
roads.
Northumberland
RR88, Sharperton Edge, centred on NT 97550 04950
Work by the Holystone History and Archaeology Group guided by Richard Carlton of The
Archaeological Practice
While the course of the road past Holystone has been well established, the onwards route
over the River Coquet and past Sharperton remains to be conrmed. In 2022, circumstance
afforded the opportunity to conrm its route east of Sharperton by examining sections of
the road north-west and east of Sharperton Edge Farm, where the landowner had been
observing the road for many years in The Ladyship and East Moor elds, respectively. In
addition to doubts over the course of the Roman road, another issue thrown up by the 2018
work on Holystone Common concerned the lack of known Roman stations and the
possibility that such a site lay concealed by an enclosed coppice known as The Ladyship on
Sharperton Edge Farm.
The excavations north-east of Sharperton Edge (centred on NGR: NT 97550 04950) were
carried out where it was postulated, on the grounds of earthwork and limited cropmark
evidence, that the course of the road runs alongside rather than underneath - as projected
on the current Ordnance Survey map - the northern edge of a modern trackway. The
trackway extends eastwards from a C-road bordering the farmstead through a eld of
improved rough pasture known as Sharperton Edge East Field, entering the latter through a
long-established eld gate on the south-west side and exiting at a drystone wall boundary
with Burradon grounds to the north-east. The trackway bends sharply to the north, 30m or
so from the drystone eastern boundary wall, but the course of the road remains visible as a
cropmark as far as the next eld boundary, beyond which it crosses a small stream and
heads for the north side of a small coppice known as Greenlish Plantation.
Within the area of investigation, Sharperton Edge East eld, several areas were available for
investigation on the north side of the modern trackway, while other areas were rendered
impossible to investigate due to the presence of strips of woodland, including modern
conifer plantations, which are likely to have damaged or destroyed those parts of the road.
The area chosen for investigation was at the east end of the eld both east and west of the
point at which the modern trackway turns northwards, close to the eastern boundary. In
- 277 -
Roman Roads in 2022
this area the modern trackway exists as a slightly incised holloway occupying the south part
of a wide terrace which also includes the Roman road, and seems to have been cut into the
prevailing slope which descends from the south. Thus, the south side of the modern
trackway is dened by a c.
0.70m high, 60-degree step
onto the trackway platform
which is deeply rutted in
places and about 6m wide. The
trackway is bordered on its
north side by a ditch some 1 -
1.5m wide, beyond which is a
bank some 1.5 – 2.3m wide.
The bank contains a number
of mature trees, principally of
hawthorn and ash, the roots of
which are degrading it. Where
it turns to the north near the
east end of the eld, the width
of the road narrows and its
outer cut bank disappears,
while the inner bank loses its
ditch. East of its turn to the
north traces of a ditch on the
same line as the east-west
ditch to the west suggest that
this feature, at least, may once
have continued eastwards and
may, therefore, be an earlier
feature. North of the bank the
terrace containing the
modern trackway, ditch and
bank extends a further 6-7m
before dipping to the north,
resuming the prole of the
natural slope also seen on the
south side of the modern
trackway.
As noted above, the question immediately arises as to why the modern trackway, which
appears to be of relatively modern, perhaps Enclosure-period date, is built to the south of
the Roman road rather than upon it. There is no obvious answer to this, but it is possible that
the modern trackway was deliberately positioned to avoid damaging the Roman road, its
course having been identied by the early 19th century by local antiquarians, or positioned
there because the Roman road was seen to be in poor condition, slumping downhill to the
north and not considered to offer a sound base upon which to build a modern road.
Alternatively, the south side of the Roman road may once have marked a eld or property
boundary, dened by the existing bank and ditch described above, which constrained the
positioning of the trackway.
- 278 -
Roman Roads in 2022
Image of the road excavated at Sharperton, courtesy of Des Kelly. The
centre rib is very apparent.
Excavation through the course of the modern trackway was impossible as it remains a
designated through-route for wheeled vehicles, but a trench on its projected route east of
its turn to the north was proposed in order to ensure, as seems almost certain, that the
Roman road does not lie underneath. On nding the Roman road in its expected position to
the north this was not completed, but there remains some merit in opening such a trench,
both to prove that road remains do not lie on this projection and to detect any signs of a
putative road-side ditch related to the northern projection.
Sections of the Roman road were examined in three trenches named Trench 1 a, b & c (see
Illustration above). Trench 1a, 12 m long and 1.5 m wide, was 10.5 m from the East Field
eastern boundary wall; Trench 1b, 8.15 m long and 1.10 m wide, was a further 38 m to the
west, thus 50 m from the eastern eld boundary; Trench 1c, 9.5 m long by 1.5 m wide, was a
further 61.65 m west, its west side some 43.7 m east of the fenced boundary to a small,
modern conifer plantation.
From the three opened trenches a comparable state of preservation and construction of the
road structure was revealed. Road width was found to be 6.32m, 6.79m, and 5.25m (perhaps
truncated) wide with a large stone bottoming, cobble intermediary layer with some
evidence of a small stone metalling between the larger stones. There was some evidence of
kerbs where plough damage or robbing had not removed them. Some large stones set on
edge in the centre suggested the possibility of a centre spine that this road is noted for
further to the west. Reveting on the downhill side of the road, with some slumping,
suggested that the hillside had been ‘cut and lled’ with the uphill side excavated to make
and lled in the downhill side to give a level terrace for the agger. An impression of a
possible ditch was noted on the north side of the road. One interesting nd that may relate
to construction or a repair at the side of the metalled surface was;
- 279 -
Roman Roads in 2022
Trench plan of the Sharperton excavation, the location of which show how the road is offset from the OS
mapped route. Courtesy Richard Carlton and the Holystone History and Archaeology Group
‘… a mass of small stones, consistently 0.07-12m max. diam., also embedded in topsoil. This is of
distinctly different character to the make-up of the compacted road surface with its base layer of
large, at stones and may also be the remains of a dump of stone placed in preparation for
construction, but never used (in much the same way that bands of stone can sometimes be seen
alongside stone walls, where they were abandoned by wallers having been found unt or
unnecessary for their intended purpose).’
Overall, the course of RR88 has been proven at this point but it is not under the current
trackway as mapped by MacLauchlan and the OS but in the bank just to the north of it. Aerial
photographs show the course continuing to the east through the adjacent eld on the same
alignment.
HER aware.
RR87, the Devil’s Causeway and RR88, at Low Learchild
Work by the Roman Roads Research Association assisted by members of the Holystone History and
Archaeology Group
It has long been apparent to antiquarians that a Roman road known as the Devil’s Causeway
ran from north of Corbridge towards Berwick on Tweed, (e.g. Horsley 1732, p144 & map after
p157), which has been more recently numbered RR87 by Margary. The rst scientic step in
recording the course of the road was made through the 1850s, published in 1864, by Henry
MacLauchlan at the request of the Duke of Northumberland, with a full survey of the course
of the road accompanied by a commentary and additional notes. This also covered a survey
of RR88 from High Rochester to what we now know as the fort at Low Learchild although
- 280 -
Roman Roads in 2022
Lidar image of the Low Learchild fort site with the geophysics interpreted onto the right hand image. Base lidar
data © Crown Copyright 2020.
MacLauchlan was unaware of its presence. Cropmarks suggesting the ditches of a Roman
fort were noted on aerial photographs taken by Dr. St. Joseph in the 1940s. Trial trenching
was carried out at the fort in 1945 by Sir Walter Aitchison and Sir Ian Richmond in 1956 but
unfortunately both died before publishing their work. The road layout as shown by OS
mapping, based on MacLauchlan’s survey, has the junction of the two roads just to the west
of the Low Learchild site. If this was/is as shown it would suggest that RR88 from the west
was primary as RR87 The Devil’s Causeway appears to change alignment at the junction.
To gain a better understanding of the fort layout, RRRA applied for a Section 42 Licence from
Historic England to undertake a gradiometer survey of the fort site utilising our ve probe
array and further to survey the road junction to the west.
It had previously been thought that there had been two forts on the site, a smaller auxiliary
fort which was later surrounded by a much larger fort, both aligned north - south. The
survey provided greater understanding of the site, demonstrating that the smaller fort was
larger than previously thought (2.14 Ha), and was actually aligned east west. It also had an
annexe attached to its northern defences, all enclosed within the larger, and possibly later
north-south aligned fort (3.8 Ha). Unfortunately, the survey revealed very little of the
internal arrangement of either fort.
The survey showed no trace of RR88, although one large negative anomaly could potentially
represent the metalling of the Devil's Causeway on a line supported by evidence from lidar.
- 281 -
Roman Roads in 2022
Lidar image of the junction of RR87, the Devil’s Causeway with RR88 from High Rochester to the west. Curved
corners of Low Learchild fort are visible in the top right hand corner. The OS/MacLauchlan mapped lines are in
red but the suggested alignments from lidar are in the white dotted corridors. Image from Mike Haken. Base lidar
data © Crown Copyright 2020.
Further analysis of lidar to determine the course of RR88 suggested that the junction may
have been 135m further north than previously thought, beyond the extent of the survey.
Some uncertainty remains, however, since an extrapolation of the line of the Devil's
Causeway from Wright's 1930s excavations 1.3km further north, led him to conclude that
the Devil's Causeway was to the west of the Ordnance Survey's line, not to the east as we are
suggesting.
HER aware. This work has been issued as a RRRA report available on our site at https://
romanroads.org/reports.html.
RR8f, Dere Sreet and RR88, through Redesdale in the Cheviots
Work by Revitalising Redesdale Landscape Partnership, Redesdale lidar Landscapes lead by Paul
Frodsham
This ambitious, very extensive, Lidar Landscapes survey of 333 square kilometres, designed
to enable a team of volunteers to study all available lidar coverage within Redesdale, was
undertaken during 2018 and 2019 as part of the NLHF-funded Revitalising Redesdale
Landscape Partnership. The survey work was undertaken by a dedicated and enthusiastic
volunteer team, many of whom had not previously worked with lidar. These were overseen
- 282 -
Roman Roads in 2022
Interpretation of the RR87 and RR88 road junction. Perhaps to the north of where mapped and outside of the area
covered by the geophysics survey.
and guided by experienced archaeologists who could also help with interpretation of the
many features of many eras revealed in the survey recording a diverse range of sites, the
majority of which were previously unrecorded. 1348 new entries have now been made on
the HER. In addition the survey revealed a wealth of new detail for many known or suspected
archaeological sites.
As well as noting the courses of Dere Street (RR8f) and RR88 that branches from Dere Street
at High Rochester (Bremenium) towards Low learchild in the east of Northumberland;
‘…. Redesdale has an extraordinary concentration of Roman ‘temporary camps’, thought to have been
used by troops on the march. The project has located a probable 'new' one at Whiteld, immediately
west of Dere Street at a point at which the road changes direction (centred on NY 902 835). It almost
seems as though the camp may have existed prior to the road, which was aligned upon its east side.
There may even be two camps here, one inside the other; if so then presumably a smaller one was built
later on the site of the original, perhaps following (or during) the construction of Dere Street. The site
is located within an area that has been much disturbed by medieval and/or post-medieval ploughing,
which probably accounts for it not having been noted previously. The lidar certainly suggests a camp
here, but the site should be regarded as uncertain prior to being checked on the ground.
Another possible camp has been noted at High Rochester, immediately south of the A68, where what
look the southern ramparts of a camp, with characteristic curved corners, appear to survive as a low
earthwork. These may, however, be no more than a later eld boundary though it is of course possible
that such a eld boundary could overlie the ramparts of a camp. The site lies very close to the Rede,
and may have been linked to transport along the river, conceivably during construction work at
Bremenium. It has been suggested that such camps along the line of Hadrian's Wall could have
functioned as construction camps while the Wall was under construction; the same may be true of the
known camps at Birdhope, as well as this possible newly discovered one.
Arguably even more signicant than either of the above sites is the recognition that the camp at West
Woodburn (centred on NY 895 875) is about twice as large as previously realised. Although previously
surveyed and published (RCHME 1993, 133-4), the previous surveyors did not have access to any lidar
imagery and they mistook a recent eld drain as the camp's western rampart. On lidar imagery it is
immediately clear that the camp is much larger than shown on the RCHME plan; its southern rampart
and south-west angle can be clearly seen as low earthworks on the lidar imagery. This camp can now
be seen as one of the largest in Northumberland, indeed in the whole of Britain.’
The HER are aware of this work. A digital copy of the full report, from which the italicised
extract above is taken, is available at https://www.revitalisingredesdale.org.uk/projects/
lost-redesdale-revealing-the-hidden-landscape/redesdale-lidar-landscapes/
RR840(x), The Possible Roman Road at Kirkhaugh, South Tyne Valley, Whitley Castle
to Corbridge
Work by Altogether Archaeology, Dr. Martin Green
Part of the course of the road was located by members of Altogether Archaeology, and an
excavation in 2016 of two sections south of Hexham conrmed that it was probably a Roman
- 283 -
Roman Roads in 2022
road (as reported in Itinera Volume 1, p352, https://romanroads.org/Itinera/Vol1_2021/
Roman-Roads-2020_Itinera2021.pdf). The road had been independently noted by members
of the RRRA (Roman Roads Research Association), who identied more sections of it further
to the west. These included a section near Catton where the road descended to the East Allen
and a stretch along the hillside through Ayle and Kirkhaugh, parallel to the modern minor
road . The Roman road, it is suggested, then descended to cross the South Tyne, where an
undated stone bridge abutment survives, before climbing to reach Whitley Castle.
At Kirkhaugh, lidar shows a straight ridge running for 400m in a west-north-west direction
across elds about 70m to the south of a minor road leading to Ayle. The ridge is about 6m
wide, with (in places) possible road-side ditches, giving a total width of the feature of up to
9m. This is a possible size and conguration for a Roman road, but further assessment was
only possible by excavation.
The excavation revealed a surprisingly complex picture, but there were no structures found
that can be identied as a likely Roman road. There was no sign of an agger in the sections.
Nor were there any ditches (though many Roman roads don’t have ditches on both sides,
and if present, ditches may be a substantial distance from the road). The ridge seen on lidar,
on excavation corresponded to an irregular narrow band of stones seen in the three
trenches crossing the ridge: 1, 2, and 3. Possibly these were deposited as clearance stones at
a eld margin when the land either side was being ploughed. To the south of the ridge there
was a spread of stones in Trench 1 that may have been laid to form a rough track, but this
did not continue in the trenches further west. On its south side was a patch of smaller
limestone stones 102, presumably debris from quarrying for lime-burning. The underlying
dark, coal-rich deposit 105, may also have been debris associated with lime burning.
The spread of stones south of the ridge was also present further west in Trench 3, but very
different in nature, in that it now contained some heat reddened stones and pieces of coal
and slag. Again, the most likely source for this was waste from lime-burning. It may have
been deliberately deposited to form a rough track along the edge of the eld between the
eld boundary and the cultivation ridges. However, no evidence of a track was seen in
Trench 2, so either the track had curved further south, or no metalling of its surface was
needed there. Of course, the waste may just have been dumped at the eld edge to dispose
of it on the unploughed headland.
Of course, failure to nd a “classic” Roman road structure (such as excavated south of
Hexham: 6m wide with kerbstones) doesn’t disprove the hypothesis that a Roman road ran
along the terrace. It may have followed a different course (perhaps bending to give the best
approach to a descent into the valley). Or the extensive agricultural, mining and industrial
activity may have destroyed it. In any case, minor Roman roads may be narrower and have
a wide variation in form, making denite identication difcult.
HER are aware of this work. A digital copy of the full report, from which the extract above
is taken, is available at https://altogetherarchaeology.org/reports.php
- 284 -
Roman Roads in 2022
Oxfordshire
Geophysical Survey of Roman settlement at Blacklands, Swalcliffe Lea, Cherwell,
centred on SP 3903 3861
Work by Abingdon Archaeological Geophysics
Magnetometry took place on 0.93ha of land as part of further investigation following eld
walking that took place to the north east. The investigation supported the Roman evidence
that was found during eldwalking and identied multiple positive anomalies that have
been attributed as possible buildings, an enclosure, ditches of various functions, a potential
trackway or road and walls which may have been used as a boundary. Due to the Roman
nds found in the vicinty they have tentatively been dated to the Roman period.
HER EOX7230, Unpublished document: Abingdon Archaeological Geophysics. 2021. Blacklands,
Swalcliffe Lea, Oxfordshire: Geophysical Survey
Magnetometry Survey of Possible Late Prehistoric to Roman Settlement at Symmetry
Park, Ardley. Centred on SP 5572 2881.
Work by WYAS (West Yorkshire Archaeological Service)
A magnetometry survey was conducted in two zones that covered an area of c. 30 ha.
Multiple anomalies were uncovered which represent settlement activity in the vicinity. A
banjo enclosure as previously identied through cropmarks and aerial photography and
identied again through this magnetometry. A rectilinear enclosure 62m x 43m is observed
along with a square enclosure that is 15m x 15m. Two other square enclosures are observed.
Many ditches are observed in the area that appear to be of multiple phases. A possible
boundary ditch extends for a considerable distance. A long double ditched feature extends
for 200m and is 15m wide and may resemble a trackway. A cluster of pit-like anomalies may
be part of the wider agricultural landscape in conjunction with the other features.
HER MOX 28025. WYAS. 2021. Land at Symmetry Park, Ardley, Oxfordshire: Geophysical Survey.
Somerset
RR45b, Beacon Hill-East Cranmore
Lidar analysis by E. Mouelhi
Lidar analysis has identied a possible 3.6 km stretch of the Charterhouse-Old Sarum road
(RR45b). A linear agger earthwork diverts from the course of the Old Frome Road just east of
Beacon Hill Woods, roughly at ST 6456 4576. Heading south-east, the road crosses Chelynch
Road just south of Long Cross Cottage. From ST 6625 4491, its course is mostly preserved by
eld boundaries. This line is only a slight deviation from the broadly parallel, projected
course, marked by a dashed line on the Ordnance Survey. Lidar cannot trace the road with
certainty beyond ST 6780 4426, near Cranmore Cottage, so eld work must be carried out to
try and plot its course eastwards.
Somerset HER: 47580.
- 285 -
Roman Roads in 2022
RR45b, Leighton
Work by Cotswold Archaeology
Related to the above entry, trial trenching by Cotswold Archaeology near Leighton (ST
697 432), across the projected course of RR45b, found no traces of a road. A geophysical
survey of the site carried out in 2018 similarly found no road (Somerset HER: 42451). The
projected course may be incorrect, or alternatively the road has been ploughed out in
the excavated area and possibly across much of the eastern Mendips, where in contrast
to the western part of the plateau, RR45b does not leave much in the way of earthworks,
save for some stretches where its course may be preserved by eld boundaries.
Somerset HER: 47334. Report: Kowalska, A. (2022) Porr Slab Track Factory, Trinidad Works,
Wanstrow, Shepton Mallet, Somerset: Archaeological Evaluation.
RRX033, near Banwell
Work by Oxford Archaeology
In advance of the laying of electrical cables for the Hinkley Connection Project, Oxford
Archaeology excavated a previously unknown Roman market town near Banwell, which
was featured on the BBC’s Digging for Britain. The town is thought to have been situated
at an important road junction and may have been a stop on the fabled Charterhouse-
Uphill road (RRX033). Full details will become available when the excavation report is
published.
RR513(x), Cothelstone-Chawleigh
Lidar analysis by Dr. C. Smart
A westward continuation of the Enmore-Cothelstone road RR513(x) has been identied
by Dr Chris Smart. From Cothelstone, a section of the road has been identied cutting
across Lydeard Down (ST 144 334 - ST 128 331), but from here its course is unclear. It is
possible that the road is continued by two further sections, one southwest of Elworthy
(ST 083 345 - ST 064 330 (incorporating a holloway, Somerset HER: 42623)) and another
traced to near Morebath, in Devon (ST 028 304 - SS 966 250). However, given the
apparently differing trajectories of the latter two sections with the Lydeard Down
section, there must have been a sharp deviation, if all were parts of the same road, as is
postulated. Similarly, the Morebath section would have had to turn sharply and cross
the Exe back into Somerset, if a small earthwork at SS 894 242, near Sowerhill,
represents its continuation. The earthwork is believed to be part of a road that can be
traced for 20 km into Devon to SS 745 115, east of Chawleigh. It may also be the case that
a branch of RR513(x) joins up with a road previously identied at Clayhanger (Devon
HER: MDV122173).
Somerset HER: 42842.
- 286 -
Roman Roads in 2022
Shropshire
RR6b, Watling Steet (West), Wroxeter, (The Roman city of Viriconium),
Grid Reference: SJ 563 083 (1st watching brief) & two trenches (2nd watching brief) (T1 at
SJ 5643 0854; T2 at SJ 5641 0848)
Archaeological watching-brief conducted by Nigel Baker Archaeology
- 287 -
Roman Roads in 2022
The location of Topsy Cottage (watching Brief No 1) and the water pipe repair trenches T1 and T2 (watching Brief
No 2) within Wroxeter - Roman Viroconium. The ditch surrounding the Roman City is clearly visible in the lidar.
© Paul Smith 2023. Base lidar data is © Crown Copyright 2020.
Two separate watching-briefs, conducted by Nigel Baker, have revealed the metalling of
RR6b Watling Street (West) within the scheduled area of Wroxeter Roman City (SAM SA
32).
The rst watching-brief was conducted on the installation of a new water main pipe off
the modern road frontage to each end of Topsy Cottage, Wroxeter. A deviation from RR6b
which bent south-west to the River Severn crossing 200 metres SW of Topsy Cottage,
whose oldest component, a sandstone rubble built cottage of probable 18th-century date,
lies exactly on top of the Roman road. The water-pipe trench up the drive was dug
through metalling laid in 1982 straight onto strata of Roman date.
The ner details of Topsy Cottage’s immediate environment are provided by two
excavations. The rst was a trench dug approx. 18m to the south of the cottage in 1972-3
by the late Dr John Houghton to check the line of the road seen on air photographs. The
trench located the street as a series of metalled layers already more than a metre deep by
the 2nd century. In the later Roman period, the road continued to be re-metalled.
Close to the cottage, the dark topsoil visible in the new 13m water main pipe trench,
diminished in thickness and disappeared replaced by a deposit of extremely hard gravel
rising to the underside of the modern drive roadstone 20cms below the present surface.
Here a deposit of hard, iron-stained, yellow-brown gritty gravel was visible. The gravel
deposit, which could be seen in section to be composed of numerous tip lines or re-
metalling episodes, including a band of ferruginous sand, occupied the whole of the
trench prole to within c.10cms of the surface under the 1982 drive roadstone – clearly it
was superuous in this area as the gravel, with a substantial ferruginous component
throughout, was so hard as to almost have defeated the mechanical excavator. This was
initially confusing as there was no separation between the Roman road metalling and the
1982 road metalling; a plastic supermarket food label was in fact recovered from the
surface of the Roman road. On closer inspection, the distinction between the 1980s grey
roadstone and the underlying yellow gravel became completely clear.
- 288 -
Roman Roads in 2022
Composite photo of the main trench north section at 12m-16m from the modern road,
showing the slope of the street metalling (left) and the possible roadside ditch (right).
© Nigel Baker Archaeology
Further examination of the trench and its sections on the conclusion of the machining
showed that the basal deposit of rm khaki silt encountered between the modern road
and the gravel deposit had been cut by a negative feature, lled with dark soil and some
sandstone rubble, bounding the gravel: this was interpreted as a roadside ditch, there
being little doubt that the hard gravel deposit represented the Roman road known from
previous work on the property and immediately to the south.
The principal interest of this watching-brief lies in its addition to the existing body of
work on and around this property, in particular, the course and character of the Roman
road, Watling Street. The 2022 exposure of the Roman road does add to knowledge, in the
sense that its eastern edge was established – which it could not be in the earlier
excavations – and was found to be approximately ve metres to the east of where it was
expected, though running on the anticipated NNE-SSW alignment. Its width in Dr
Houghton’s trench to the south was about 10 metres.
The second watching-brief was undertaken in January 2023 on emergency water-main
repairs in Watling Street, to the south of the English Heritage Guardianship sites. Two
trenches were observed. Roman road metalling appeared to be present in both trenches
to almost the full depth of the excavations (1.2m) and in T2 was certainly present to
within 30cms of the modern carriageway surface. Trench 2 was probably coincident with
the water-main Lining Hole 10, excavated in 1999 (White in White, Gaffney and Gaffney
2013).
Trench 1 was located on the west side of the carriageway and grass verge of Watling
Street, 118.20m south of the southern boundary fence of the west-side Guardianship area.
Excavation had started before the watching-brief commenced. The spoil from the earlier
(unobserved) excavation work was examined and found to contain some Roman roof-tile
fragments; these were noted but not recovered.
Trench 2 was placed 75m south of trench 1 (at a distance of 186.65m from the south side
of the west Guardianship area boundary) and took the form of a 2m by 1m trench along
the road. Removal of the tarmac immediately revealed grey roadstone-type ballast or
MOT which was instantly recognised by the STW contractors as the ll of a previous
repair trench. This indeed proved to be the case and the excavation proceeded entirely
through this light-grey trench backll, down to the failed repair, which was rectied. The
ground was saturated throughout with water and the trench had to be pumped out as
work progressed.
In the course of the work in Trench 2, the grey MOT ballast fell away from the sides of the
original cut, revealing in the east section a line of large, rounded cobbles, up to 25cms in
size, at a depth of 30cms below the carriageway surface (see Fig 3 above). These were laid
over very coarse, rounded, orange gravel in a red-brown gritty sand matrix. All appeared
to be material composing Roman Watling Street (west). No artefacts were recovered.
The watching-brief results are consistent with those of the 1999 water-main relining work
and show that, where investigated in Trench 2, Roman Watling Street still lies beneath the
modern road – it has not yet started to diverge westwards away from the modern road line
- 289 -
Roman Roads in 2022
– and that stratigraphy consistent with Roman road metalling survives extremely close to
the modern carriageway surface. This is consistent with results further down the road to the
south, Watling Street on the Topsy Cottage site being buried at a depth of about 0.8m below
the eld surface north of the cottage, reducing to about 0.6m on the south side of the
property (Baker, 2022).
Watching brief 1. Published as ‘An archaeological watching-brief on Watling Street at Topsy
Cottage, Wroxeter’ by Nigel Baker. July 2022. Available from Shropshire HER as ESA9909.pdf
Watching brief 2. Published as ‘Wroxeter Water Main WB Report 1-23.pdf.’ Available from Nigel
Baker Archaeology
Staffordshire
RR18c - Ricknild Street or Rycknield Street. Bean’s Covert in Branston, to the south
of Burton upon Trent. From SK 228 217 to SK 230 219
From Dr Mark Knight, Cultural Heritage Ofcer, with the Transforming the Trent Valley Scheme.
A preliminary report from the U3A was received by the Transforming the Trent Valley
Scheme regarding a short section of RR18c at Bean’s Covert in Branston, to the south
of Burton upon Trent. This was helpful in proving that the route of the road long held in
local folk memory was a demonstrable fact, and the scheme was able to target their work
- 290 -
Roman Roads in 2022
Trench 2, looking south-east, showing the surface of large stones as it was rst
exposed. © Nigel Baker Archaeology
accordingly. Volunteers from the Scheme removed scrub and bramble from the known
course of Ricknild Street in Bean’s Covert.
Once cleared of bramble, the road was clearly apparent to the eye, the top of the curve of the
agger or convex road surface standing some 45cm higher than the level of the surrounding
land surface, and approximately 8m from side to side. The adjacent ditches or fosses, either
side of the road are also visible in places, although very silted due to accumulation of soil and
leaf litter over the centuries. The segment of road is 232m long and does not appear to have
been damaged by ploughing. It was part of an historic deer park owned rst by Burton
Abbey, and following the Dissolution, by the Paget family, later the Marquises of Anglesey.
RR573b(x), Branston/Dunstall and onwards towards Rocester?
Work by Transforming the Trent Valley Scheme led by Dr. Mark Knight
Volunteers conducted resistivity surveys on the proposed line of the so called Via Devana,
where it is thought to cross the River Trent on the Walton/Drakelow boundary in
Derbyshire and the Branston/ Dunstall boundary in Staffordshire. Roman roads are
notoriously difcult to locate using geophysics, even if the route of the road is known. The
Via Devana was no exception and we were frustrated in our search, with a possible result
rather than a denite one. Our survey was focused on the west bank of the River Trent and
it is possible that the road material has been robbed out, or buried beneath millennia of
- 291 -
Roman Roads in 2022
The agger of RR18c visible in the lidar. © Paul Smith 2023. Base lidar data is © Crown Copyright 2020.
- 292 -
Roman Roads in 2022
Lidar image of the ‘new’ road from the Trent cossing, over Rykneild street RR18c and onwards to Rocester. Base
lidar data is © Crown Copyright 2020.
ooding and silt build up, alongside the dredging of the river during the middle decades of
the 20th century, with deliberate silt deposition on the river banks.
Further to the above;
Lidar survey by David Ratledge and Neil Buckley
Lidar suggests the road approaching the Trent from Leicester may have crossed the river
making a junction with Rykneild Street (RR18) and then continued on. Rocester in
Staffordshire may be a possible destination. A lot of this is overlain by modern roads giving
a direct route but not as surmised between Colchester and Chester.
See Leicestershire for what may be a continuation of this road and Derbyshire for
possible fort site on the east bank of the Trent
Suffolk
RR34a, Long Melford, Suffolk, TL858454
Work by Long Melford Heritage Trust (LMHT)
Long Melford Heritage Trust (LMHT) have conducted a number of small excavations in
recent years in an area known as Stoneylands, close to the known Roman settlement at Long
Melford, Suffolk (LMD172). Stoneylands was long suspected to contain Roman
archaeological remains, and a parch mark which often appears during dry summer weather
has been interpreted as a possible Roman road leading westward from the settlement in the
direction of RR34a, along the upper Stour Valley towards the Roman settlement at Wixoe.
Two excavations in recent years by LMHT, LM354 and LM362, have proven the existence of
a Roman road surface, which shows evidence of having been a causeway leading across
marshy land, liable to ooding in the winter, to a crossing of the Chad Brook. Evidence of
possible votive offerings was found in the vicinity of the Chad Brook crossing, along with
some evidence for there having been a wooden bridge structure during the Roman period.
It seems likely from other work to the east of the Long Melford settlement that this road is
in fact a continuation of RR34a, which Margary had assumed crossed RR33a in the vicinity of
High Street, to the north of Long Melford. A recently discovered unrecorded Roman road
from Holton St. Mary, as below, also appears to join with these routes at the Roman
Settlement.
HER Report No. SSF60100. Archive: Suffolk County Council Archaeology Service. Published
Reports: LM354, LM362
RR344(x), Holton St. Mary to Long Melford, Suffolk, TM 069 363 to TM 865 451
Work by Colchester Archaeological Group (CAG)
As recorded in the of Itinera, Vol. I (2021), a cropmark of RR3c just south of the modern A12
appears to show a distinct spur leading off from RR3c in a north westerly direction towards
Holton St. Mary. Since that time, members of CAG have been tracing the route of the spur
- 293 -
Roman Roads in 2022
using mainly desk based techniques such as Google Earth aerial imagery and lidar image
analysis. In late 2022, Apple Maps also released a new set of aerial images for the area which
showed up several cropmarks in more detail than those captured in 2018 on GE. Using these
techniques, CAG have been able to trace the route of the Roman road with a high degree of
certainty, all the way to the known Roman settlement at Long Melford. Extrapolating the
route through Long Melford appears to show it linking up with the supposed Margary route
RR34a westwards towards Wixoe.
Beyond the short stretch of “tramline” cropmarks at Holton, only faint lidar traces of the
road can be seen until the road almost certainly follows the route of the modern road down
Sulley’s Hill towards Shelley. Near Shelley Hall however, lidar and cropmarks show a distinct
agger crossing the Brett Valley and passing to the south of the hall. The road then appears
to follow a cut through an unrecorded earthwork behind Shelley Hall which shows traces of
bivallate defences, although it appears on a 16th century map as a rabbit warren. It seems
possible that this may have been a prehistoric mound which has been re-purposed over the
years? In the vicinity of Shelley Hall, the road would have crossed RR330 heading south from
Bildeston, and evidence suggests that this may in fact have been a single junction only, as no
traces have been found so far for a continuation of RR330 towards Stratford St. Mary as was
postulated by Margary.
- 294 -
Roman Roads in 2022
Route of the previously unrecorded road from Holton St. Mary to Long Melford, Suffolk. Base lidar data is ©
Crown Copyright 2020.
Beyond Shelley, traces of the road appear at frequent intervals as tramline cropmarks and
faint lidar ridges, passing to the north of Polstead and then on through the cluster of small
villages named Calais Street, Whitestreet Green, and Stone Street just south of Boxford.
Cropmarks in this area show what might have been a small ladder settlement clustered
along the line of the road.
Northwest of Stone Street there is a clear tramline cropmark of width 14-15m, which is
consistent along most of the route, and this joins up with a short section of the modern
A1071 road from Ipswich to Sudbury. Cropmarks further northwest show the Roman road
departing the A1071 towards Newton Green, where various Roman nds are recorded,
including a villa site. Further on, a clear cropmark and lidar trace shows the road running
south of Great Waldingeld and crossing Valley Road, which lies on the route of RR322.
Based on a lack of evidence (also noted by Margary) for Valley Road being Roman, it is
suspected that the new route may have become confused in antiquity with Valley Road
(which is dead straight) and the Roman association became linked with that road. There is
an enormous amount of housing development taking place in this area so CAG have been in
touch with the local council to ensure that the new road is recorded in the HER as soon as
possible.
The newly discovered road then passes under Chilton Aireld, where it is rumoured there
was once a moot mound, now attened, and reappears as faint cropmarks and lidar traces
southeast of Long Melford. The road approaches Long Melford on a direct alignment
towards a known Roman settlement, and as noted earlier, could quite easily join up with
RR34c in the area of the settlement.
Rather strangely, the route of the road does not appear to coincide with any parish or other
boundaries, nor does it follow any modern hedge lines or tracks, save for the descent of the
modern road at Sulley’s Hill into the Brett Valley at Shelley, and the short section of the
A1071 west of Stone Street. This may possibly suggest that the road went rapidly out of use
following the withdrawal of Roman rule in the early 5th century. An abundance of cropmarks
around the Holton junction also suggests a possible Roman settlement, and it has been
suggested that this could be the location of Ad Ansam, although no work has been done so far
to investigate this further.
HER Reported to local HER ofcer, Suffolk County Council Archaeology Service Archive:
N/A. Published Report: N/A
Surrey
RR151, Ewhurst, NGR 508075 146077
From Matt Sparkes
The Rowhook to Farley Heath road RR151 extended NW from Stane Street RR15, the
alignment apparently heading for a Roman Temple about 7.75 miles away. The course for
most of that distance is well established. If the road continued beyond the temple, which
- 295 -
Roman Roads in 2022
would seem likely, it may have continued for approximately the same distance again to meet
up with the assumed London - Winchester route RR400(x), or further still to the London -
Silchester road RR4a.
The notes below were kindly provided by David Calow.
A relatively undamaged section of the Rowhook to Farley Heath Roman road has been found 14-
24cm below the ground surface near Ewhurst, Surrey, i.e. along the known course of the road. The
excavation was conducted by volunteers from Surrey Archaeological Society, Liss Archaeological
Society and Ewhurst History Society led by Chris Gibson. The alignment is almost parallel to that
shown on Ordnance Survey maps but about 20m west. The road is centred on NGR 508075 146077
and bears 33 degrees west.
The road at this location is well-engineered; 3.6m (12 pedes) wide with metalling about 20cm
thick and a camber of about 7cm. On Weald Clay, excavation showed traces of cuts at each side
suggesting a shallow formation trench. There was no evidence of drainage channels below or at
the edges of the road or of a primary layer of brushwood but the foundation has not sunk in the
clay and remains in good condition. Stones feather out an extra 40cm from the west edge and
there is a scatter of stones spreading up to 1.2m from the east edge. A 30cm slot was excavated
along the south side of the trench to expose the section and a recent land drain can be seen at the
east.
Road metal is 90% greensand and 10% chert from local sources. There are some larger stones up
to 200mm near the edges and centre and scattered occasionally throughout. The rest of
foundation is a mixture of hand broken stones between about 50mm and 75mm. The surface
shown in the photograph is a mixture of silty clay with fragments of broken stone between the
stones of the foundation. There is no apparent evidence of ruts or repairs suggesting this section
was not heavily used by carts.
A second trench 3m further north found the road surface has a well-preserved area of top
dressing of smaller stones up to about 25mm, although a large robber pit in the north of this
trench has removed all the road metal. Northeast of the road the stones extend an additional
- 296 -
Roman Roads in 2022
Surrey Archaeological Society
Rowhook to Farley Heath Roman road
RR151
November 2022
3.5m. The extension seemed contemporary with the road but the chert is more concentrated,
there is no top dressing and the layer of stones thins and becomes more sparse towards the east.
Possible explanations for the extension include the base of a stone dump for road maintenance, a
passing place or the entrance to an access track.
There was no pottery, charcoal or other dating material although three fragments of CBM of a
suitable colour and composition to be Roman were found in the road surface.
East Sussex
RR142, Glynde to Pevensey, TQ 6253 04406
Work by Chris Butler Archaeological Services (CBAS)
Recent work at Rattle Road, Westham comprising both geophysical survey and evaluation
trenching is seeking an alternative route to the Margary course down the holloway, Peelings
Lane near Pevensey. They are currently working on the SMS over part of the same road, but
they found a spur road that heads SW towards area of salt working last year. Both routes
appear to be constructed at same time (rather than the spur being a later addition) based on
the roadside ditch congurations for both elements. The post excavation report is due for
publication in one to two years, but CBAS should be able to provide more information.
RR142, Glynde to Pevensey,
Work by Swale & Thames (SWAT)
The same road as above was just about detected in a Swale & Thames Archaeology (SWAT)
site immediately to the NW of the above on Rattle Road which has just had its post
excavation assessment approved. This gives more credence to the entry above, the line runs
on to be Pevensey High Street
RR142, Glynde to Pevensey,
Work by Archaeology South East (ASE)
Further to the above, ASE picked up a metalled surface in the High Street close to Pevensey
Castle. These three separate sites taken together, suggest a more convincing line.
RR140/RR142, ‘With a Mag(netometer) on Farne Street’
Work by David Staveley
This work, published in a Sussex University publication and the subject of an on-line talk by
the author, traces the whole course of the Roman Road, that he calls Farne Street, from
Hardham Camp on Stane Street to Pevensey in East Sussex. This is RR140 as far as Bridge
Farm, Barcombe, then the continuation to Arlington seen by Mills, and thirdly the eastern
- 297 -
Roman Roads in 2022
part is RR142, which he regards as part of the same road. This has been a 10 years project,
walking the whole way with a magnetometer. The magnetometer only detected areas where
the road was built or mended with iron slag so most of his story is lidar. Without being
critical, the report is said to be difcult to follow because there are no maps or place names.
West Sussex
RR150, London to Brighton, Walnut Marshes, Crawley Down Road, Felbridge. TQ 36150
39228
Work by Archaeology South East
An evaluation and watching brief of the site in advance of residential development
identied the road running from the north-north-east and exhibited a distinct camber. Two
sondages were excavated into the feature, one in the south which was oblique and one in the
north which was approximately perpendicular to the road. What appeared to be wheel ruts
were observed particularly in the north of the feature. Both sondages revealed the method
of construction for the road. A layer of large stones [514]/[518] had been laid directly over
the natural ne grey windblown deposit [519] presumably to form a solid sandstone core for
the road. These stones were not laid with a ush upper surface perhaps in order to act as a
‘key’ for the overlying layers. Redeposited clay [512], [513] & [517] was laid over the stones
to form a camber.
In the north, a compact grey clay layer [516] capped [517] and several possible wheel ruts
were seen cut into this. In the south, the remains of a slag metalling layer [511] were
recorded. A single possible wheel rut was observed cut into this.
HER aware, WSHER2021-093. Land East of Walnut Marches, Crawley Down Road, Felbridge, 2019.
Archaeology South East, ASE Report No: 2019306
RR150, Land at London Road, Hassocks. TQ 30509 16504
Work by Archaeology South East
Although the possible route of a Roman road (Vine's Line) crosses survey areas, no
anomalies were detected by geophysics that correspond with its location.
HER aware, WSHER2022-024, Land at London Road, Hassocks, West Sussex, 2021, Archaeology
South East, ASE Report No: SUMO-03176
RR150, Land East of London Road, Hassocks. TQ 30509 16504
Work by Archaeology South East
The site lay on the course of a possible Roman Road, known as Vine’s Line, (named after a
local school master, who rst identied the possible route). The presence of outer ribbon
- 298 -
Roman Roads in 2022
development along this road was also considered a possibility. The most signicant
discovery at the site was that the purported route of the local Roman road shown on
Ordnance Survey maps did not run across the site, contrary to previous research.
HER aware, WSHER2023-009. Archaeological Investigations on Land East of London Road,Hassocks,
West Sussex, 2023, Archaeology South East, ASE Report No: 2022326
RR140, Archaeological Investigations on Rampion Offshore Wind Farm, Onshore Cable
Route.
Work by Archaeology South East
A ditch located in Site F, near Woodmancote Place, TQ 229 150, is situated c.900m north of,
and broadly parallel with, the projected line of the Barcombe Mills to Hardham Roman road
RR 140, which was supposed to cross the scheme at Oreham Common, although no trace of
it was identied in either the evaluation or SMS investigations. The series of shallow
interweaving linear depressions G33 and the gully-like feature G34 situated parallel to the
ditch’s north side are difcult to interpret either as agricultural furrows or drains, due to
their shallow depth and somewhat meandering form, and an interpretation as wheel-ruts or
erosion channels within a routeway seems plausible.
HER aware, WSHER2021-148. Archaeological Investigations on Rampion offshore Wind Farm,
Onshore Cable Route, West Sussex. ASE Report No: 2017402
Conrmation of the line of the Chichester-Arundel road RR153 using lidar
Work reported by Historic England
This resulted from a National Mapping Project that ran in parallel with the Secrets of the High
Woods lidar project. It is reported on in an HE Research Report Series no. 14-2016 The High
Woods from above NMP, which is available online at: https://historicengland.org.uk/
research/results/reports/14-2016 The section on the Chichester-Arundel Roman road is on
page 67 which at last consolidates Margary’s assertions of the route of this road. There are
other bits on Roman roads elsewhere in the report that may be of interest.
HER aware
Fine-tuning the line of the Chichester-Silchester road RR155
Work by the Chichester and District Archaeology Society, Cleverly S., Davies T. & Kallaway M.
Cropmarks on aerial photographs taken in 2018 reveal the zone ditches of the Chichester-
Silchester Roman road, just south of Lavant Primary School centred on SU 8539 0844, the
ditches appear to be about 16.5m apart. Similarly at Binderton SU 84938 10386 where in 2020
very faint traces of the ditches, 16.2m apart were visible in the resistance results, but not in
magnetometry. Geophysical surveys of the Langford Farm Roman villa at SU 84400 10640
and nearby Roman road by CDAS in August/September 2020 revealed the winged corridor
house identied from cropmarks on 2018 aerial photographs as were a possible aisled
building to its north-west and a large, rectangular building to its south. The line of the
- 299 -
Roman Roads in 2022
Chichester-Silchester Roman road, crossing the north-east corner of the same eld, was also
surveyed. Magnetometry was ineffective, but resistance worked well.
HER aware, CD10413, CD10414 and E1686
Chichester to Arundel road RR15, centred on SU 89130 06650
Work by Archaeology South East
Excavations along the line of a new sewer in 2018 revealed a short section of the metalling
and side-ditch of Stane Street RR15 (the Chichester--London road), close to where the
Chichester-Arundel road RR153 diverges from it. Nearby were a series of big early Roman
extraction pits that may have been dug for gravel for the road(s). Late Iron Age occupation
in the vicinity, which was cut through by the roads, seems to have continued into the Roman
period on either side.
A possible pre-conquest enclosure may have preceded the intensive Romano-British
occupation at MA7, which straddled the Stane Street and Old Arundel Roman roads. This
enclosure was subject to repeated modication during the early Roman Period when the
area alongside the roads was subject to an intensive sequence of building and consolidation
events spanning the later rst to second centuries. The excavations identied the south
eastern edge of the actual Roman road surface of Stane Street as well as a series of roadside
ditches and, probably, an area of quarrying used for its construction. A cobbled surface
exposed on Stane Street’s northwestern side may possibly have related to a later Roman
reorientation of the line of the road.
The Old Arundel Road corresponds to the western end of Margary’s RR153, Chichester to
Brighton, before it joins Stane Street. The recent aerial survey for the Secrets of the High
Woods project conrmed the presence of signicant sections of the road’s agger and side
ditches surviving as earthworks and cropmarks for over 8km of the total 15km distance
between Chichester and Arundel. The investigations at the south end of the site have
identied possible road quarry pits, below Roman cobbled hard standing surface on either
side of the Old Arundel Road, thereby presenting the rst excavated evidence for this road.
HER aware
Wiltshire
RR164, The Roman Road that just stopped.
From Paul and Rebecca Whitewick with supporting lidar work from David Ratledge
A blog by Paul and Rebecca Whitewick describing the course of this road is available on
YouTube at https://www.youtube.com/watch?v=lHIV4q11s6w. This informative blog
inspired David Ratledge to check out what was visible through lidar. The road appears to
make a junction with RR41 south of Lambourn. To the north it appears to be turning right
towards Oxford but may have taken a different course, and where did it cross the Thames?
Still lots to work on.
- 300 -
Roman Roads in 2022
This work can be read in the RRRA Autumn 2022 newsletter at; https://romanroads.org/
Newsletters/Members/Newsletter%2023%20Summer%202022%20(1).pdf
Worcestershire
A possible Roman ford near Evesham?
Work by Aidan Smyth, Wychavon District Council Archaeological Ofcer
A recent discovery of what at rst appeared an engineered ford, and paved in a way very
reminiscent of classical Roman road paving has now been sentenced as a natural feature.
Aidan writes;
Unbelievably the wonderful feature turned out to be an incredibly rare form of geology from the
Triassic period. We are still waiting on the OSL as we think the Romans quarried it for the villa
nearby which is why the feature ends so abruptly. It took some investigating and is certainly one
of the most amazing things I’ve ever seen. It looked Roman, it measured as being Roman, it had
batted ‘kerbs’, but upon excavation there was no construction level. We then changed tack and
considered it without the knowledge of what else was in the landscape and reviewed it as a stand
alone feature, it slowly became obvious that it was natural.
So not Roman, but Romans may have quarried some of it to build the villa.
A report on the nding can be seen at; https://arkeonews.net/a-cobbled-ford-uncovered-
near-evesham-could-be-the- nest-roman-example-of-its-type-in-britain/?
fbclid=IwAR3IBf52o884AhLQo9BmU1iiFCd2iP9PIaP7ahGZjrAh9Q8TkwpIQvfUacE courtesy of
Arkeonews, including a BBC video clip, 18th April 2023.
- 301 -
Roman Roads in 2022
Image of what was initially thought to be a Roman ford near Evesham, courtesy of Aidan Smyth, Wychavon
District Council Archaeology Ofcer
- 302 -
RR180, Bartlett House, Birmingham Road, Bromsgrove, SO 9657 7164
Work by Worcestershire Archaeology
Previous work on this road, on the site of All Saints Garage nearby, (see Itinera Volume 1 p364
https://romanroads.org/Itinera/Vol1_2021/Roman-Roads-2020_Itinera2021.pdf),
suggested that it may survive to one side of the modern A38. However the site here
excavated is set well back from the A38 and no traces of the Roman road or roadside
structures were found, lying directly over natural ground. It is likely therefore that the
Roman Road or associated features such as roadside ditches did not extend this far west of
the current road, or that they had been truncated by the construction of the house.
HER aware, SWR26465
RR18d, Land at Battens Drive, Redditch, centred on SP 064 677
Work by Wardell Armstrong
Rycknild Street is shown crossing the site and the route of the road was still traceable on the OS
edition of 1979, following the realignment of Battens Drive, although had been redeveloped by the
edition of 1984. No remains of the Roman road are proven within the vicinity of the site, although
have been recorded elsewhere along this route. Roadside settlement is found alongside Roman
roads, as has been found to the south in Ipsley, where a late/post Roman building has been
excavated. The Blacksoils Brook runs just to the south of the Site (not through the Site as recorded
in the HER search) and is culverted under Battens Drive. Names such as Blacksoils are linked to
Roman settlement, which gives some contextual evidence of possible settlement in the area.
HER aware, SWR26353, from which the italicised quote is transcribed.
Yorkshire
RR28c Castleford to Tadcaster, North of Headley Lane, Stutton with Hazlewood, North
Yorkshire LS24 9BE. Centred on SE 4631 4168
From Archaeological Services WYAS via Heritage Services, County Hall, Northallerton
During top soil stripping in 2014 for a working areas for an overhead line the remains of part
of a Roman Road known as the Roman Ridge were recorded. Occasional patches of small
fragments of limestone and compacted stone dust were seen. This was thought to be the
remains of the Roman Road surfacing and inlled wheel ruts. A section cut through the
remains recorded construction deposits of the agger comprising a series of clay and silt
deposits, and a possible buried soil at the base of the section. There was evidence of a
ploughed down limestone surface and a roadside ditch. It is therefore likely that the section
was excavated through the south-eastern side of the road with further remains preserved to
the north-west.
HER: MNY40332. Published Report: None. Source: Knaresborough XD Overhead Line,
Knaresborough. Archaeological Watching Brief. Wells, M. Held by Heritage Services.
Roman Roads in 2022
