HeritageGateway - Home

Login  |  Register
Site Map
Text size: A A A
You are here: Home > > > > Gloucestershire County Council: Historic Environment Record Result
Gloucestershire County Council: Historic Environment RecordPrintable version | About Gloucestershire County Council: Historic Environment Record | Visit Gloucestershire County Council: Historic Environment Record online...


The Roman road known as the Foss(e) Way runs from Exeter to Lincoln, with sections running north-east and south-west from Cirencester.
County: Gloucestershire
District: COTSWOLD
Parish: CIRENCESTER
NGR: SP 02 01
Monument Number: 6491
HER 6491 DESCRIPTION:-
The road known as the Fosse Way has two sections that runs northeast from Cirencester and southwest from Cirencester. The entry that follows includes both sections of the road: a general road description, followed by the northwest section of Fosse Way described generally and then by parish moving south, and finally the southwest section of Fosse Way described generally and then by parish moving south.
General Description
The Fosse Way (Exeter - Lincoln) was probably military in origin, and is followed for most of its course by modern roads. {Source Work 403.}
The south west section of the Fosse Way is followed by the course of the modern road (A333) until Jackaments Bottom from where it is followed by the County boundary. The Fosse Way running northeast from Cirencester to Alcester, Warwickshire, is closely followed by the modern A429 in Gloucestershire and commonly used as the parish boundary. {Pers comm J Isaac, 1986.}
Lengths of agger off the course of the Modern road are visible at Stow-on-the-Wold and in the parishes of Moreton- in-Marsh and Batsford. Sections that have been examined both north and south of Cirencester do not indicate massive construction, and side ditches where visible are close to the agger. The junction with the Rkynild Street (HER 6666) has been located north of Slaughter Bridge. Disuse of the Fosse Way as a major through route in the Saxon period is indicated by burials in the agger near Bourton, (see HER 6561 Fosse Way - section northeast from Cirencester). {Source Work 403.}
It is thought that the Fosse Way was constructed slightly later than Ermin Street, based on the way they relate to Cirencester. The road is likely to have been constructed initially by the Roman army and subsequently maintained. Its course can be traced through the county as modern roads tend to follow its course, although there are deviations, as on the hill into Stow-on-the-Wold, where a short stretch of the original road has been uncovered for many years, badly overgrown and the sign barely visible. {Source Work 158.}
Margary’s designation for Fosse Way is Roman Road 5c. {Source Work 5926.}
The planning of the Fosse Way is thought to have been carried out in stages, with W St Clair Baddeley thinking - in his Presidential Address to the Cotteswold Naturalists' Field Club - that the road northeast of Cirencester was completed before the extension of the road to Bath and southwards. {Source Work 10134.}
Northeast section
Northeast section of Fosse Way national grid references:
NGRs - SP 0397 0500 - SP 0500 0735 - SP 0720 1000 - SP 1000 1378 - SP 1100 1500 - SP 1500 1974 - SP 1522 2000 - SP 1897 2500 - SP 2000 2980 - SP 2004 3000 - SP 2133 3500 - SP 2240 3760
The northeast section of the Fosse Way leaves Cirencester by Querns Lane and Lewis Lane, and runs northeast for 1 mile to Hare Bushes, where Akeman Street diverges to the east and the Fosse Way runs due north. From Hare Bushes Lodge to Bourton-on-the-Water, the road runs on an immense agger around 5 - 6 feet tall and often 8 feet on the lower side. The short northern alignment ends at Baunton Downs, near Ragged Hedge Covert and from here the main alignment was sighted to a point near Stow, although a 2.5 meter deviation in two short alignments was made to avoid combes Winterwell Valley. At the last of these combes, near Calmsden, a zigzag was made to ease the crossing and the old agger is still visible beside the present road (NGR 406070 208290). The road is continuously followed by parish boundaries until past Bourton. At a combe just past Northleach, this (parish) boundary carries on while the road now bends and here the agger can be seen undisturbed in a grass field, some 40 feet wide and 2 - 3 feet high (NGR 411770 215980). Half a mile beyond this is another deviation, again on the high ground. At Slaughter Bridge (NGR 417100 222200), Rycknild Street (HER 6666) branched off to the north-west. After Bourton, the road is no longer raised and, past Stow, the alignment is again to the north, towards Moreton. Near Broadwell, a portion of the old road foundation (a layer of stones set on edge) is exposed within iron railings (NGR 419430 227320). Beyond Donnington, the road is again raised and generally straight as far as Moreton. Beyond the railway (HER 15488), the course of the Fosse Way lies on the north-western side of the present road for half a mile ( as shown on aerial photographs) before turning to the northeast. {Source Work 5926.}
Batsford
The Fosse Way (north-northeast from NGR 420500 232800) is largely destroyed, but is visible in places as a low spread ridge or as cropmarks, continuing its former alignment north from Moreton-in-Marsh. The agger, 7.5 meter across and 0.3 metres high with contiguous side ditches (deepest on the eastern side), is preserved for 68.5 metres before destruction by a railway cutting (HER 15488) at NGR 420570 232970. After some 457 metres it crosses an ancient road, which meets it at right angles, then veers to the east. Some 274 metres further northeast, where it adjoins Dorn (HER 361), the road is seen as a low spread ridge. {Source Works 4638, 403, 3140, 3149, 3246 & 3502.}
1928 - (NGR 421880 236360) - Four pairs of piles 1.2 metres apart, thought to be the remains of a Roman bridge were found during the widening of Knee Bridge on the Fosse Way in 1928. They were 2.4 metres below the present water level, embedded in silt and clay in a position, which does not correspond with either the 19th century bridge or the earlier pack-horse bridge. No finds were made which would help to date the piles, two of which were presented to the Corinium Museum. There are no visible remains of any earlier bridge structures beneath the 1928 bridge. {Source Work 902, 1353 and 862.}
1973 - Sections of the Fosse Way were discovered outside the Bathgate to the west of Cirencester during excavations in April 1973. (HER 4975).
1994 - A topographic survey and air photo transcription was undertaken by the RCHME of the Roman small town at Dorn. The Fosse Way survives as a band of metalling between 8 metres and 10 metres wide, with intermittent traces of flanking ditches. One of the main east-west streets of Dorn makes a junction with the Fosse Way beyond the eastern defences of the town. {Source Work 6076 and 4638).
1998 - A site visit was made to Dorn by A. Douthwaite of GCCAS in relation to the Monuments Protection Programme. The ditch of the Fosse Way was still visible as an earthwork, but very degraded. {Source Work 484.}
Moreton-in-Marsh
The Fosse Way passes through the parish of Moreton-in-Marsh. {Source Work 403.}
Longborough
The Fosse Way passes through the parish of Longborough. {Source Work 403.}
Broadwell / Donnington
The faint line of the Roman road is visible between NGR 41955 22765 and NGR 41973 22840 on RAF aerial photographs. {Source Works 863 and 862.}
The Roman road (between NGR 4195 2276 and 4197 2282) follows a line north-northeast from Stow-on-the-Wold and in places forms part of the western boundary of the parish. The Roman road fabric, which consisted of pitched stones of local oolite was exposed in 1931 and 1936, immediately east of the present road. {Source Works 403, 1389 and 1412.}
1931 - (NGR 41943 22732) - Old road foundations were noted in November 1931, partly exposed on a bank within one and a half feet of the eastern edge of the present road on a bank, around one and quarter miles north of Stow. The foundations consisted of a line of stones, set on edge, on trimmed bedrock. {Source Work 1412 and 862.}
1936 - A stretch of the Fosse Way was located on 10 July 1936 during road-widening between Stow and Moreton. A length around 128 metres long and 0.9 - 1.2 metres wide was uncovered on the east side of the present road, around half mile north of Stow and just below the brow of the hill, where the road begins to descend to the Moreton valley (NGR 4194 2273). The road was constructed of pitched stones (local oolite) set on edge on the natural subsoil. A section was removed and the remainder was covered by a Modern concrete curb. {Source Works 862, 1389 and 10543.}
1987 - (NGR 41943 22732) - A site visit by J. Isaac (GCCAS) followed a telephone call from Mrs M. Sheppard (Moreton-in-Marsh Local History Society) regarding remains of the Foss Way Roman road. She recalls several years ago seeing a stretch of Roman road visible alongside the present A429, railed off and with a plaque recording it as part of the Foss Way. From her conversations with local people she thinks that the railings were removed some time in the 1960s. The verge along this stretch of road is now very overgrown but she thinks she has identified the location. A small hole has been excavated about 30cm in diameter and 20cm deep revealing a stone surface, which may well be the surface of the Roman road. The location is approximately 70 metres south of the junction of the middle one of three minor roads to Broadwell turning east from the A429. It is on the eastern verge of the road close to a sycamore tree. There is a medium to long term threat to this stretch of Roman road from road improvements to the A429. However, in view of the dangerous location of the site and the need for controlled excavation the decision was taken that the location of the site would be noted in the GCC HER and that further work would be done when/if road improvements were undertaken (Pers. Comm. M. Sheppard).
This is presumably the same stretch of road referred to on the Ordnance Survey record map {Source Work 862} as "Roman road found here in 1931 by St Clair Baddeley" according to H. E. Donovan in 1936 and marked by J. E. Titmus sometime after 1955. (Pers. Comm. J. Isaac, 1987).
Stow-on-the-Wold
The Fosse Way (NGR 41899 22543) is represented by the agger, 30 feet wide, up to 1.5 feet high and probably ditched on both sides, extending for 250yds (228.6 metres) in the cemetery to the southwest of the town. {Source Work 403.}
1980 - The agger at NGR 41899 22543 is as described in Source Work 403, although it is only marked at the southern end of the cemetery. The ditches are not obviously visible, although a slight change in profile suggests their former presence. The agger mound is used for graves. {Source Work 470.}
1980 - At NGR 41899 22543, the Fosse Way agger is unploughed. {Source Work 305.}
Bourton-on-the-Water / Lower Slaughter
The Fosse Way (between NGR 4137 2175 and NGR 4172 2222) is generally overlain by the modern road and follows a northeasterly course for 3.5 miles on the parish boundary. The road crosses two streams and the River Windrush; at the latter (Bourton Bridge) there was a paved ford and a footbridge. Local deviations appear to be for topographical reasons. There is a slight deviation at the eastern entrance to the Romano-British settlement at Bourton Bridge (HER 345); the main alignment being rejoined a corresponding distance southwest of the bridge. The road is followed by the parish boundary and the agger was constructed of gravel layers on pitched stones, but 260 yards northeast of the ford, the road construction was of small stones and yellow silt on a bed of rammed slag. A kerb was also found in the settlement area where also at lest one building impinged on the road. The only observed width is 22 feet. {Source Work 403.}
1958 - (NGR 4167 2217) - Part of the Fosse Way was lowered near the Coach and Horses Inn, Bourton-on-the-Water, a small excavation recorded a section of the Foss Way and revealed eight Anglo-Saxon burials with associated finds (HER 2624), indicating disuse of the agger. The Foss Way at this point runs on a causeway across heavy clay between gravel spreads. {Source Works 1227, 4638, 403 & 709.}
1958-9 - (NGR 41605 22095) - Widening and rebuilding of the eastern side of Bourton Bridge revealed the remains of a paved stone ford and small stone piers, presumably for a foot bridge. The Roman ford under the present bridge over the River Windrush, had a surface of large pebbles with, in two places, two courses of rough stone blocks. The presence of a footbridge on the eastern side of the ford was deduced from the discovery of small piers built of cemented rough Oolite rubble. {Source Work 1227, 4638 and 403.}
1966-7 - Excavations by H O'Neil in 1966-7 have shown details of the road structure and disturbance, its relationship with structures of the Bourton Bridge settlement (HER 345), a road diverging towards Salmonsbury (HER 342) and a junction with the Ryknild Street (HER 6666) immediately north of Slaughter Bridge at NGR 4171 2222. Sixty yards northeast of the ford at Bourton Bridge, a branch road led east-southeast, but the actual junction with the Fosse Way was not observable. Road widening at NGR 4162 2212 located the western verge of the Fosse Way, three road surfaces and pottery were found and just north of the ”bakehouse” building (HER 345), foundations of the Roman road were exposed due to a bend in the present road. A length of 24 feet was exposed, constructed of pitched stones and layers of gravel. The alignment of this section tied in with that exposed in 1958 near the Coach and Horses Inn, around half a mile to the northeast. {Source Work 4638, 1227 and 403.}
Turkdean
For details of the excavation at Legore Hill, see Northleach with Eastington below. {Source Works 2951 and 403.}
At a combe just past Northleach (NGR 41177 21598), the (parish) boundary carries on while the road now bends and here the agger can be seen undisturbed in a grass field, some 40 feet wide and 2-3 feet high. {Source Work 5926.}
The modern road is now re-aligned, omitting the bend which is shown on the tithe map {Source Work 425.}, however the agger of the road survives to the west of the present alignment. {Source Work 5295.}
Northleach with Eastington
The Fosse Way forms the north-west boundary of the parish. {Source Work 403.}
1964 - The agger of the Fosse Way on the top of Leygore Hill (NGR 4118 2159) was excavated by HE O'Neil in 1964 and proved to be 20 feet wide between the ditches. A terraceway to the east of the modern road (NGR 4117 2161) represents the original course of the Roman road down the hill slope to the north. The road crosses the valley bottom at a narrow point some 150 yards from the present east wall of the road where there were the traces of slight abutments of a narrow bridge. {Source Works 2951 and 403.}
1983 - (NGR 44114 21549 ) - In October 1983, during the construction of the Northleach Bypass and after excavation at the Roman settlement nearby (HER 2583), the contractors removed the Fosse carriageway down to the bedrock. For about 200 metres on either side of the point where the new road was to cross, the Fosse Way ran at a height of about 1.25 metres above the adjacent fields. The build-up of hardcore and soil is presumed to have taken place within the last 200 years, possibly when the road was cut down through rock at the edge of the plateau half a km. to the southwest, in order to improve the gradient. The make-up of the raised part was seen in a long section along the centre of the road when only one carriageway was in use. The following sequence was noticed at about 30 metres south of the new roundabout and this is representative of the length in question, where the greatest amount of overburden was removed by machine. At 1.4 metres, the machines had cut down into a limestone bedrock. Above this, between depths of 1.15 metres and 0.95 metres, there was a dark brown compressed clayey loom, completely free of bones or man-made objects - similar to the brown clay, which sometimes overlay the rock band on the excavation, where it was taken as a geological deposit. Between 0.95 metres and 0.75 metres, there was a confused stony layer above a level of tile stone. The surface of the latter was smooth and above it were occasional loose worn stones. This may have been the disturbed Roman surface. Above the 0.75 metres level there was a brown soil layer 0.20 to 0.25 metres thick, which could have been a former soil level. The top 0.50 metres consisted of 0.25 metres of crushed rock and limestone 'sand' and hardcore with tarmac above. The section has similarities to that found and recorded by H O’Neil at Leygore Hill, 750m north of this point. {Source Work 2099.}
Coln Saint Denis
The Fosse Way forms the western boundary of the parish of Coln Saint Denis. {Source Work 403.}
A brief note of the Foss Way in the valley of the Coln is given by Sawyer in the Proceedings of the Cotteswold Naturalists Field Club. He notes that the Foss Way crosses into trhe valley at Foss Bridge and that is was well-known that this of Roman period date. {Source Work 2984.}
Cirencester
The problems relating to the Roman road network on the eastern side of Cirencester have been summarised in sources. {Source Work 5642.} and {Source Work 4926.}
There is clear evidence for a bridge at the Verulamium Gate (HER 644). {Source Work 158.}
In 1955, Margary suggested that the Ermin Street from the south was the first to be constructed across the site of Corinium, as the town layout was based upon it. He states that the main alignment of the northeast section of the Fosse Way, if continued, would have reached Ermin Street close to the Gloucester Gate, but instead it runs south at Baunton Downs to meet Ermin Street at Preston Bridge. He suggests that this alignment was planned as a whole, with a later spur which formed the short axis of the Roman town, with the amphitheatre at its western end. He considered that the southwestern section of the Fosse Way was a later addition to the road network around Cirencester as its main alignment (Tetbury Road) met Ermin Street c.400m to the north (see HER 6491). He considered that if the two alignments of the Fosse Way met directly, then this cross roads with Ermin Street would almost certainly have become the focus of the town. {Source Work 4926.}
In 1965, Wacher referred to Margary’s suggestion that the original alignment of the northeast section of the Fosse Way was directed on the (future site of) Gloucester Gate, whereas Kenyon, in 1948, had noted that it was directed towards the (future) site of the Verulamium Gate. Wacher suggested that that this route was directed to a gateway on his proposed early fort at Leaholme, and that the subsequent diversion from this alignment was to prevent through traffic from the Fosse Way to Ermin Street from entering Corinium. He considered that both the southwest section of the Fosse Way and Akeman Streets were later additions. {Source Works 4926, 5642 and 4781.}
Margary’s analysis was made before the road excavations west of the of the Bath Gate, and before a fort (c. AD50-65) was discovered towards the southern end of the Fosse Way, near Axminster, Devon. This had its long axis on the Fosse Way, indicating that the road was contemporary or earlier. This revived Collingwood’s view and would mean that the Fosse way was built as a whole. This would mean that the southwest section of the Fosse Way directed to the northern end of the Corinium is an early road, built long before the amphitheatre and connected directly with the northeast section of the Fosse Way. {Source Work 4926.} On GIS, the alignment of the Fosse Way between the A419 and Bathgate has been extrapolated from the known orientation of the road at Bathgate.
The Tar Barrows (HER 13 and HER 14) lie immediately adjacent to the alignment towards the Verulamium Gate and may be Roman in date. If so, it is significant that this class of monument is frequently early in date and often adopts a roadside location. If this is correct, then it suggests that Akeman Street (HER 4508) and the Roman road at Kingshill Lane (HER 5963) are later. A Roman cremation was located during a watching brief related to the A419/A417 DBFO road scheme at SP 0322 0374). Although this is c. 300 metres away from the alignment of the Fosse Way towards the Verulamium Gate, it is located close to an alignment from Ragged Hedge Covert to the north gate of the Roman town, suggested by Margary (1973) as the main alignment of the road. The suggested alignments of the Fosse Way to the north or east gates of the Roman town do not appear to have been targeted in the Stage 2 evaluation for the A419/A417 DBFO road scheme. {Source Works 5642 and 4781.}
In 1998, Hargreaves suggests that the Fosse Way was built as a long distance road without reference to any military or civil settlement at Corinium. He suggests that there are three possible objectives of the alignment of the northeast section of the Fosse Way from Baunton Downs/ Ragged Hedge Covert. These objectives are a junction with the southwest section of the Fosse Way alignment projected north east, with a junction at Hare Bushes; the site of the future Verulamium Gate as suggested by Kenyon and a gate in the Leaholme Fort as suggested by Wacher and McWhirr. He argues that the first alignment is the most likely, as this would provide a through route for the original Fosse Way prior to the establishment of a fort or town at Corinium. This alignment is also compatible with the requirements of the road surveyors (see Source Work 4926 for further details). The anomaly of the hiatus of the Fosse Way at Cirencester, which does not exist elsewhere, would be removed if this alignment were accepted. He suggests that Akeman Street and the Fosse Way diversion south from Ragged Hedge Covert were contemporary, making a joint alignment to enter Corinium via the site of the Verulamium Gate. This diversion rendered the earlier alignments of the northeast and southwest sections of the Fosse Way and the junction at Hare Bushes redundant. He also suggests that the reason that the Fosse Way and Ermin Street crossroads did not form the focus of the town was that a more suitable site, on a gravel island, lay to the south. For further details see Source Work 4926.
1991 - (NGR 4037 2027) An evaluation excavation was undertaken by Cotswold Archaeological Trust (CAT) for the proposed Stratton/ Cirencester By-pass. Four trenches on the north and south sides of the Fosse Way established the presence of Roman, post-medieval and 18th century road surfaces, Roman road side ditches and 18th century quarry ditches beneath and adjacent to the modern A429. Large deposits of “road silt” on the northern side of the Fosse Way and deep colluvial deposits were found to overlie Roman roadside ditches on the south side of the road. {Source Work 2491 and 4771.}
1996 - (NGR 4037 2027) An excavation in relation to the A417/A419 DBFO road scheme was undertaken at Burford Road, on the alignment of the Fosse Way. Two long transects were located (in the area of the CAT evaluation trenches) in order to examine cross sections of the modern road and its precursors. On the southern side of the road, a substantial amount of ground lose has occurred, due to post-medieval quarrying. A post-medieval linear quarry also exists to the north of the road and has truncated the earliest surviving road surface here. A shallow quarry located in Trench 3 is almost certainly Roman in date, however, the excavations failed to locates any definite evidence for a Roman road underneath the modern one. Although the earliest surface in Trenches 2/6/4 is considered likely to have been Roman in date. The artefactual evidence suggests that the Roman road was in continuous use, without repair, into the turnpike era, with the resulting formation of a slight holloway in the bedrock. The Roman road in any case does not appear to have been of a substantial construction and there was no evidence of an agger, although this may have been completely eroded. {Source Work 5642.}
2000-2002 - Between December 2000 and June 2002, an archaeological watching brief at 33 Querns Lane, Cirencester, identified layers of metalled street surfaces made up of crushed sandstone and gravel. The layers were identified as street surfaces of Fosse Way, although no dating evidence was found to associate this section of street with the Roman period. The deposits are however similar to those seen in other area of the town and is in the area of the presumed line of the Fosse Way. The street surfaces are noted in trench 4 and 8. In trench 8, three phases of roadside ditch were identified running parallel to the road on the southeast side. The primary ditch was a wide cut ditch, the fill of whoch was cut by the second V shaped ditch. A tertiary ditch may also be present however it is above the latest level of the street surace noted during the excavations; it is likely that still later street surface were removed by later acivity. {Source Work 8789.}
Southwest section
Southwest section of Fosse Way national grid references:
NGR - SP 0207 0192; SP 0000 0020: SO 99744 0000; ST 9500 9562; ST 9454 9500; ST 9085 9000, ST 9000 8868; ST 8952 8795
North of Easton Grey (Wiltshire), to the southwest of Cirencester, the Fosse Way forms the county boundary for a distance of 7 miles. It survives as a green lane, except for a stretch now obliterated by Kemble Airfield. Over 7 miles the road is continuous and is usually very wet with no visible sign of an agger, although remains of metalling are visible at some points. The lane varies much in width, from 60ft to 18 ft wide. After Long Newton, beside an aerodrome (NGR 393000 193000), there is a distinct change in its construction and an agger around 27 feet wide and 2 feet high is visible. At Fosse Gate (NGR 394100 194300), after a short section of green lane, a very large agger, 33feet wide and up to 4 feet high, is visible; this continues up to 40 feet wide until Kemble Airfield. Just before Jackaments Bottom, the route becomes visible again, although there is some doubt about its route here. The green lane runs under the Tetbury branch railway (HER 3935), past the farmhouse to the Tetbury Cirencester Road, and then turns northeast to follow the course of the Fosse Way. The agger diverges east of this green lane and is crossed by the railway embankment east of the bridge, connecting with a lane east of the farm. As two parish boundaries follow these routes, it may indicate an early modification or bifurcation of the route here. From here to Cirencester the route is still a highway, at first with a deep hollow on its south side and then raised, generally by 4 - 5 feet, occasionally over 6 feet at low places, suggesting that originally there was a large agger. {Source Work 403 & 5926.}
The southwest Fosse Way leaves the Cirencester by a course which bears no relation to the other roads and must be regarded as a subsequent addition. {Source Work 65 and 862.}
Cirencester Parish
Excavations suggest that the Fosse Way diverted just west of the town to enter it via the Bath Gate. However the presence of 1st and 2nd century burials alongside the Tetbury Road, close to the town, indicates that a road was in existence here at that time. Thus the main road was diverted across the Querns to the Bath Gate in the 3rd to 4th century. Although there is no direct dating evidence, the construction of a roadside building (HER 651) shows that the diversion was in place before c. AD 280. {Source Work 295.}
Southwest of Cirencester, the Fosse Way follows a straight course, then curves to enter the Bath Gate. Excavation of a roadside building outside the Bath Gate (HER 651) indicates that this course was in existence before c. AD280. This may be a later development and originally the road may have continued along its original alignment, beneath the modern Old Tetbury Road. {Source Work 4781.}
In 1924, Collingwood held the view that the Fosse Way was planned as a unity and was designed and built as a whole. {Source Work 4926.}
In 1955, Margary considered that Ermin Street (HER 7542) from the south was the first to be constructed across the site of Corinium, as the town layout was based upon it. He considered that Ermin Street was built first, then the northeast section of the Fosse Way to Ermin Street, with a spur connecting it to Corinium, forming the shorter axis of the town layout, with the amphitheatre opposite its western end. Akeman Street (HER4508) connected with this spur and then the southwest section of the Fosse Way was constructed, probably after the building of the amphitheatre, as a later road from Corinium to Aquae Sulis. {Source Work 4926.}
In 1965, Wacher also considered that the southwest section of the Fosse Way (and Akeman Street) was a late addition to the road network around Cirencester. {Source Work 4926.}
Margary’s analysis was made before the road excavations west of the Bath Gate, and before a fort (c. AD50-65) was discovered towards the southern end of the Fosse Way, near Axminster, Devon. This had its long axis on the Fosse Way, indicating that the road was contemporary or earlier. This revived Collingwood’s view and would mean that the Fosse way was built as a whole. This would mean that the southwest section of the Fosse Way directed to the northern end of the Corinium is an early road, built long before the amphitheatre and connected with the northeast section of the Fosse Way. {Source Work 4926.}
In 1998, Hargreaves writes that the southwest section of the Fosse Way appears to run on its alignment to meet with Ermin Street, although trenching at NGR 40217 20220 inside the walls of Corinuium failed to find metalling of a Roman street on the alignment. It is not directed on the Leaholme Fort, nor is it directed to any terminus of the Corinium street grid, thus Hargreaves concludes that the road was planned as a long distance military road without any reference to any military or civil settlement at Corinium. Its terminus could hardly have been a 'T'-junction with Ermin Street without a continuation across it, a sighting point at Hare Bushes (north of Corinium) is seen through such a continuation. The three short alignments leading from the Bath Gate (which is on the street grid) are clearly a local adjustment made necessary to avoid the quarry on the site later adopted for the amphitheatre, when the need for a connection from the Fosse Way to the heart of Corinium arose. The earlier alignment (along the Tetbury Road) had lost its importance, as the diversion south to the site of the future Bath Gate provided more direct access to the town, this is seen in the lack of provision in the street grid of an entry into the town on this earlier route. As a long distance road, the southwest and northeast sections of the Fosse Way must have connected north of Cirencester (see HER 6561). {Source Work 4926 & 4781.}
1969-74 - Prior to development of the Querns area, a series of excavations was carried out between 1969-76. Before these excavations, it was assumed that a road linked the town with the amphitheatre (HER 401) and that the Fosse Way was on the same line as the former Tetbury Road. Following plans for the western relief road, a trench was dug close to the amphitheatre in 1969 across a pronounced ridge (thought at the time to be the road from the town to the amphitheatre). A metalled surface, probably 1st or early 2nd century in date, was located on the south side of the ridge and appeared to be aligned with the amphitheatre entrance. At least two further metalled surfaces were located on a different alignment and at a higher level. The latest had wheel ruts, which indicated that traffic was passing to the north of the amphitheatre and had a base of large stones. The earliest road surface identified in 1970 was much wider than the others, and again had a base of large stones. Further excavations were under taken in 1971-2 when sections were dug across the ridge. Two road surfaces were located, the first similar in character to that first located in the 1969 excavations. The second surface was a very worn metalled layer, heavily rutted, some patched, and of a different makeup, with a base of much larger stones. Over 1 meter of silt had built up on the northern side of this road surface, when this was removed, two courses of kerbing was located, which enabled the exact course of the road and its change in direction north of the amphitheatre to be plotted. Within the central section a small ditch and a thin metalled surface was recorded, on a different alignment to the main road surfaces. This probably represented an early road from the town to the quarries and amphitheatre. In 1973, a 22 metres stretch of road was located during topsoil stripping prior to the construction of the Querns Maternity Hospital. The road was about 6 metres wide and at least three surfaces were recorded.
1974-5 - Street metalling was recorded adjacent to Bath Gate (Site DX) at the time of its discovery in 1974-5. The road metalling was located just forward of the gateway and was interpreted as part of the FFosse Way leading out of Cirencester. The metalling in front of the gate was at least 1m thick and where sectioned showed evidence of at least ten surfaces. The full width of the road was not revealed, but the presence of kerbing on its north-western edge suggests that its total width may have been in the region of 9 metres. It was perhaps flanked by ditches. {Source Work 4926.}
1978 - In 1978, limited excavations were undertaken (by T. C. Darvill) at the Kitchen Garden of the former Querns House, prior to the building of an Ambulance Station. A road was located along the route of the Fosse Way diversion north of the amphitheatre. However it is suggested that its scale was inappropriate to it being the Fosse Way, also its construction was dissimilar to that found in 1973. Darvill therefore suggests that the Fosse Way probably ran north of this area and the long barrow. {Source Work 295.}
1995 - An archaeological evaluation (recorded in HER 16209) was carried out by Cotswold Archaeological Trust in November 1995 in connection with proposed tree planting within the Abbey Grounds. Eight 1 meter square test pits were hand-dug. Roman street metalling was encountered in test pit 7, which lay very close to the project line of the Fosse Way running northeast from Cirencester. {Source Work 3301.}
Kemble Parish
The Fosse Way forms the north-western boundary of the parish of Kemble. {Source Work 403.}
1997 - About a 100 meter length of the Fosse Way was observed during a site visit to Jackaments Bottom Farm, Kemble, undertaken by C. Parry on 15/01/1997 (centred on NGR 39647 19742). The road was revealed by ploughing in 1996. It is represented by a 1 meter wide by 0.1 metres high line of limestone fragments aligned roughly southwest to northeast across the valley floor. The feature corresponds with the former line of the Fosse Way shown on the 1st-3rd County Series Ordnance Survey maps. To the east the feature shown on the map appears as a cross-slope terrace visible under the grass. Presumably the ploughing disturbed the crown of the road, and the full width may be hidden beneath ploughsoil. {Source Work 84.}
1998 - The landowner has now covered the newly exposed road with hardcore, and the length in question is now used by vehicles (Pers. Comm. C. Parry 19/01/1998).
Papers relating to the enforcement notice, which did not proceed, for this length of track have been deposited in the site file (08/06/2001) {Source Work 484.}
Rodmarton Parish
The Fosse Way extends along the Rodmarton parish boundary at the south east. {Source Work 403.}
1939 - During the construction of the aerodrome, a number of roads were cut across the line of the raised ridge (agger). A section exposed at Culkerton Wood (NGR 3950 1956) showed alternating layers of limestone flags and gravely sand, the basal layer of flags appears to have been laid in a foundation trench. The overall width of the road was not established as a farm track intruded upon the line of the road on the Wiltshire side. {Source Works 1354, 403 & 5926.}
The Fosse Way forms the Gloucestershire/Wiltshire County Boundary (NGR 3950 1956). {Source Work 709.}
Long Newton Parish
The Fosse Way passes through the eastern part of the parish of Long Newton. {Source Work 403.}
1961 - The Bristol Water Works Company dug a trench along the middle of the Fosse Way from Long Newton to Shipton Moyne. About 200 yards long stretch was examined on the Long Newton boundary. The construction of the road was very slight. A layer 3-4 inches thick of small stones up to 4 inches long was covered by a 1 inches layer of small stone chips. The agger showed no signs of heavy wear or repairs. {Source Work 2906 and 403.}
Shipton Moyne Parish
2012 - Archaeological observation was undertaken by Border Archaeology on behalf of Bristol Water plc of open-cut trenching excavations for the insertion of a new water main, extending northeast from Shipton Moyne Pumping Station, along the line of the Fosse Way (a military highway of Roman origin) and through fields to the west of the road, terminating at the crossing of the River Avon (Tetbury Branch). The main phase of pipe-laying took place during March and April 2012, with a short stretch excavated in September 2012 with the granting of permission to undertake ground disturbance activity in the vicinity of badger setts. Trenching along the route of the Fosse Way revealed multiple modern road surfaces and three isolated sections of earlier road construction, which could possibly be of Roman origin, although it should be stressed that no artefacts positively indicating a Roman date were recovered from any of the road construction layers. Extensive evidence of disturbance by late post-medieval/modern road improvement works, land drains and service trenching were noted along the entirety of the route. The earliest feature identified was a northwest-southeast aligned ditch revealed in a section of trenching to the northeast of Fossetillery Farm beneath a sequence of deposits probably associated with the original construction of the Roman road. This feature was assigned an early Bronze Age date (c. 2,500-2,000 BC), based on the fact that a sherd of probable Beaker pottery (representing a distinctive kind of vessel characteristic of this period which appears to have formed part of a ‘cultural package’ that entered Britain at around this time) was found within the ditch.
In Section 1 to the northeast of Fosse Cottages and Section 2 southwest of Fossetillery Farm, the earliest evidence of road construction consisted of a thin bedding layer of small flat stones overlying an earlier pre-Roman soil horizon which was overlaid by a layer of golden-yellow gravels forming a make-up deposit for the road. To the northeast of Fosse Cottages, this make-up deposit was overlaid by a layer of stone cobbling but to the southwest of Fossetillery Farm it had been largely removed.
Section 3 to the northeast of Fossetillery Farm, a different method of road construction appears to have been employed. Although there appears to have been a bedding deposit of stones, similar to that encountered in Sections 1 & 2, there does not appear to have been an intervening layer of gravels at the interface between (1014) and the uppermost layer of road construction represented by (1013). This could be explained by the relatively shallow depth of the natural limestone bedrock in this area.
With the exception of the sherd of probable early Bronze Age date recovered from the northwest-southeast aligned ditch, very few stratified artefacts were recovered during this programme of archaeological observation.
No archaeological features were revealed across the fields parallel to the Fosse Way, at the northeast end of the archaeologically observed excavations, which terminated at the crossing of the River Avon {Source Work 12363.}
2008 - Gloucestershire NMP (Cotswold & Forest of Dean)
A short stretch of the Foss Way, between SP 2085 3391 and SP 2071 3359, was surveyed by the Air Photography Unit between October to December 1995. The agger was clearly visible as a negative cropmark running in a NNE SSW direction for 350m. Ditches less then 2m wide show intermittently along either side of the agger. (Collection UID 1030519, Event UID 1030518).
SP 2078 3374. Part of the Roman road known as the Foss Way is visible as cropmarks on aerial photographs and has been mapped by English Heritage's Gloucestershire NMP. This section of the road is about 640 metres long, and is aligned north east to south west. It passes directly in front of the east end of the Roman town of Dorn, where short spurs of road appear to lead into the town via gaps in the rampart. {Source Work 4249.}
2010 Cotswold Hills/South Cotswold NMP
1517244 - A section of Roman road is visible as a cropmark on aerial photographs. The site is centred on ST 9604 9694 and extends between ST 9605 9695 and ST 9603 9624. The site comprises a pair of aligned linear roadways which measure 17 and 11 metres in length, and up to 7.5 metres in width. This site is a child monument of Foss Way (1164946), which extends between Bath and Cirencester. This site has been mapped from aerial photographs as part of the Cotswold Hills National Mapping Programme.
1517248 - A section of Roman road is visible as a cropmark on aerial photographs. The site comprises a pair of aligned linear roadways which measure 17 and 11 metres in length, and up to 7.5 metres in width. This site is a child monument of Foss Way (1164946), which extends between Bath and Cirencester. This site has been mapped from aerial photographs as part of the Cotswold Hills National Mapping Programme.
1517256 - A section of Roman road is visible as a cropmark on aerial photographs. The site is centred on ST 9564 9644 and extends between ST 9566 9646 and ST 9562 9641. The site comprises linear roadway which measures 68 metres in length, and up to 8 metres in width. This site is a child monument of Foss Way (1164946), which extends between Bath and Cirencester. This site has been mapped from aerial photographs as part of the Cotswold Hills National Mapping Programme.
1517252 - A section of Roman road is visible as a cropmark on aerial photographs. The site is centred on ST 9604 9694 and extends between ST 9605 9695 and ST 9603 9624. The site comprises a pair of aligned linear roadways which measure 17 and 11 metres in length, and up to 7.5 metres in width. This site is a child monument of Foss Way (1164946), which extends between Bath and Cirencester. This site has been mapped from aerial photographs as part of the Cotswold Hills National Mapping Programme. {Source Work 4249.}

Monuments
ROAD(ROMAN)
FORD(ROMAN)
FOOTBRIDGE(ROMAN)
BRIDGE(ROMAN)
Associated Finds
UNIDENTIFIED OBJECT(ROMAN)

Protection Status

Sources and further reading
862;Ordnance Survey;unknown;Vol:0;
65;Codrington T;1905;Roman Roads in Britain;Vol:0;
403;RCHME;1976;Iron Age and Romano-British Monuments in the Gloucestershire Cotswolds;Vol:0;
5696;Darvill TC;1984;
709;RCHME;1984-1985;Vol:0;
1354;Stevens CE;1939;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:61;Page(s):132-134;
497;Hancock J;1975;Vol:0;
3127;RCHME;1971;Vol:0;
297;Wacher JS & McWhirr AD;1982;Early Roman Occupation at Cirencester;Vol:0;
1035;Rawes B (Ed);1982;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:100;Page(s):258-66;
2906;Gracie HS;1961;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:80;Page(s):179;
1701;Witts GB;1879-1880;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:4;Page(s):199-213;
5926;Margary ID;1955;Roman Roads in Britain. South of the Foss Way - Bristol Channel;Vol:1;
4553;McWhirr AD;1981;Roman Gloucestershire;Vol:0;
295;McWhirr AD, Viner L & Wells C;1982;Romano-British Cemeteries at Cirencester;Vol:2;
4926;Holbrook N (Ed);1998;Cirencester The Roman Town Defences, Public Buildings and Shops;Vol:5;
4781;Darvill TC & Gerrard C;1994;Cirencester: Town and Landscape;Vol:0;
5134;Ordnance Survey;1878-1882;OS 1st County Series: 25 inch map;Vol:0;
5136;Ordnance Survey;1900-1907;OS 2nd County Series: 25 inch map;Vol:0;
5138;Ordnance Survey;1920-1926;OS 3rd County Series: 25 inch map;Vol:0;
484;Historic Environment Record;various;Vol:0;
6288;Esmonde Cleary S;1987;Extra-Mural Areas of Romano-British Towns;
9448;Holbrook N;1996;CBA RESEARCH REPORT;Vol:94;
9495;Walker G;1990;
1701;Witts GB;1879-1880;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:4;Page(s):199-213;
8789;Stoten G & Williams J;2002;
3248;NMR;1935;Vol:0;
863;RAF;1946;Vol:0;
470;Saville A;1976;Vol:0;
305;Saville A;1980;Archaeological Sites in the Avon and Gloucestershire Cotswolds;Vol:0;
1227;O'Neil HE;1968;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:87;Page(s):29-55;
1412;St Clair Baddeley W;1932;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:54;Page(s):385-387;
1389;Donovan HE;1935;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:57;Page(s):280-281;
2951;O'Neil HE;1964;PROCEEDINGS OF THE COTTESWOLD NATURALIST'S FIELD CLUB;Vol:34;Page(s):133-137;
3246;CUAP;1960;Vol:0;
6070;Clapham V (Ed);2000;Stow-On-The-Wold: Glimpses of the Past;
3149;CUAP;1964;Vol:0;
3140;CUAP;1959;Vol:0;
3502;CUAP;1963;Vol:0;
4638;Timby JR;1998;Excavations at Kingscote and Wycombe, Gloucestershire;Vol:0;
6076;Corney M & McOmish D;1998;
155;Margary ID;1967;Roman Roads in Britain;Vol:0;
5295;ADAS;1993;Vol:0;
425;Unknown;1837-1859;Tithe Maps and Apportionments for Gloucestershire;Vol:0;
2491;Rawes B (Ed);1992;TRANSACTIONS OF THE BRISTOL AND GLOUCESTERSHIRE ARCHAEOLOGICAL SOCIETY;Vol:110;Page(s):213-230;
4771;Walker G;1991;Vol:0;
5642;Mudd A, Williams RF & Lupton A;1999;Excavations alongside Roman Ermin Street, Gloucestershire and Wiltshire. The archaeology of the A419/A417 Swindon to Gloucester Road Scheme;Vol:1;
2099;Rawes B;1984;GLEVENSIS;Vol:18;Page(s):25-42;
486;Unknown;1986-1988;Terrier;Vol:0;
53;Burrow EJ;1919;The Ancient Entrenchments & Camps of Gloucestershire;
6878;Russett V;1989;
7932;Johnson J;1999;
51;Buckman J & Newmarch CH;1850;Illustrations of the Remains of Roman Art in Cirencester: The Site of Ancient Corinium;Vol:0;
8695;Various;1990-1991;Cirencester Urban Assessment Project record sheets;
3301;Bateman C;1995;Vol:0;
7408;Parsons J;2002;
4249;Historic England;Various;Vol:0;
10250;Google Earth;various;Google Earth;
2984;Sawyer J;1896-1898;PROCEEDINGS OF THE COTTESWOLD NATURALIST'S FIELD CLUB;Vol:12;Page(s):125-138;
10543;Donovan HE;1936;PROCEEDINGS OF THE COTTESWOLD NATURALIST'S FIELD CLUB;Vol:26;Page(s):105;
7271;RCHME;1990;
3880;RCHME;1996;Vol:0;
11232;Unknown;1991;
11233;Darvill, TC, Scaife, R, Timby, J & de Rouffignac, C;1991;
11234;Unknown;1991;
12363;Bunce A & Crooks K;2013;
10134;St Clair Baddeley W;1918-1920;PROCEEDINGS OF THE COTTESWOLD NATURALIST'S FIELD CLUB;Vol:20;Page(s):85-97;
15258;Various;2008-10;
15250;Various;2003-4;
14175;Historic England;2016;NRHE to HER;
14177;Ordnance Survey;Various;
14358;Stoertz C;2012;
14488;Smith A, Allen M, Brindle T & Fulford M;2016;The Rural Settlement of Roman Britain;
6075;Douthwaite A & Devine V;1998;

Related records
WARWICKSHIRE HER;MWA4759
HER   2204     A Romano-British settlement is located within the area of Salmonsbury Camp, an Iron Age Hillfort, Bourton on the Water.
HER   2583     Roman site located during the 1983 excavations along the route of the Northleach Bypass by GADARG, Hampnett.
HER   648     Roman Street System, Cirencester
HER   8898     Bathgate or Westgate of the Roman town of Cirencester.
HER   651     Roman extra-mural building west of Quern's Lane, Cirencester
HER   9891     C14 coin found on Foss Way
HER   3935     Disused railway, Kemble to Tetbury branch of the GWR.
HER   15484     Traces of the Roman road running from Moreton in the Marsh to Little Compton are visible as cropmarks, Moreton in the Marsh.
HER   361     Dorn is the largest of five Roman defended small towns which line the Fosse Way between Cirencester and Lincoln, Batsford.
HER   15488     Oxford, Worcester and Wolverhampton Railway, subsequently part of the West Midland Railway and the GWR.
HER   15388     Moreton-in-Marsh.
HER   2624     Series of eight Early Medieval - probably 6th or 7th century - graves excavated after ground lowering work in March 1958 by Mrs HE O'Neil in between The Coach and Horses Inn and the turning to Bourton-on-the-Water on the Foss Way, Bourton-on-the-Water.
HER   345     Bourton Bridge Roman settlement is a scheduled site located to the south of Slaughter Farm, Lower Slaughter.
HER   3988     Foss Way and Salt Way Intersection
HER   4944     Midland & South Western Junction Railway, C19 - C20, now disused, had own line from Andoversford Junction to Andover, later part of the GWR.
HER   9806     18th century milestone at Dentice Bushes, Baunton.
HER   4508     The archaeological and visible remains of Akeman Street Roman road start from the east area of Cirencester and are evident to St Albans (Verulamium).
HER   9823     Milestone on Burford road.
HER   644     Verulamium Gateway was located on the east side of the town defences.
HER   7542     The archaeological remains of Ermin Street Roman road.
HER   4508     The archaeological and visible remains of Akeman Street Roman road start from the east area of Cirencester and are evident to St Albans (Verulamium).
HER   401     Cirencester Amphitheatre, also known as the Bull Ring is a scheduled monument, Cirencester.
HER   6666     Ryknild Street Roman road, also known as Ryknild Way,Condicote Lane and Buckle Street.
HER   5963     Kingshill Lane, a possible Roman Road east of Cirencester.
HER   9222     Roman road recorded in Anglo-Saxon charter, Swell parish.
HER   27701     Margary's Roman road number 542. Amalgamated with HER 32858
HER   30311     Roman street surface observed on site of old railway line, Sheep Street
HER   4975     Mechanical excavation for foundation trenches reveals burials in 1973 on site of Christian Brann Building in Phoenix Way, Cirencester
SOUTH COTS / COTS HILLS NMP PROJECT;1517244
SOUTH COTS / COTS HILLS NMP PROJECT;1517248
SOUTH COTS / COTS HILLS NMP PROJECT;1517256
SOUTH COTS / COTS HILLS NMP PROJECT;1517252
CUAD;52956
NORTH COTSWOLDS NMP;1472778
SHINE;GC1333
HER   3876     Romano-British Occupation Site
HER   32858     Section of Roman road between Chavange Green and Lasborough, Beverstone
NATIONAL RECORD OF THE HISTORIC ENVIRONMENT;1164971
HER   342     A scheduled Iron Age bivallate hillfort (with evidence or Saxon and Roman occupation) known as Salmonsbury Camp, Bourton-on-the-Water.

Source
Gloucestershire County Council: Historic Environment Record Archive