Short URL for this page:

[image ALT: Much of my site will be useless to you if you've got the images turned off!]
Bill Thayer

[image ALT: Cliccare qui per una pagina di aiuto in Italiano.]

[Link to a series of help pages]
[Link to the next level up]
[Link to my homepage]

[image ALT: link to previous section]
This webpage reproduces a chapter of
Archaeological Handbook
of the County of Gloucester

by George Witts

published by G. Norman, Clarence Street
Cheltenham, n. d. (1883)

The text is in the public domain.

This text has been carefully proofread
and I believe it to be free of errors.
If you find a mistake though,
please let me know!

[image ALT: a blank space]

Archaeological Handbook of Gloucestershire


Akeman Street

Or Akman Street, seems to have been a name given to that portion of the Foss Way which runs from Cirencester to Bath.​a It has been seriously described by some writers as an Anglo-Saxon name, signifying the Sick Man's Road, referring to the principal highway to the celebrated hot baths of Aquae Sulis (Bath). Jackment's Bottom, the point at which the Ikenild Street leaves the Foss Way, four miles south-west of Cirencester, probably has the same derivation as Akman.

Buggilde Street

This ancient road runs from the Ryknield Street, near Bidford, in Warwickshire, and proceeds by Honeybourne and Weston-sub‑Edge, ascending the Cotswold Hills near Saintbury and Willersey Camps; then, leaving Broadway Tower on the west, it runs along the ridge of the hills above Snowshill, with its round barrows, above Upper and Lower Guiting, with barrows on either side of it; then on by Wagborough, and joins the Foss Way near Bourton-on‑the‑Water. At its junction with the Foss there is a fine Roman villa. (See Bourton Villa.) The first mention of this road known to me is in a Saxon Charter, printed in Kemble's Codex Diplomaticus, No. 61, dated 709, in which certain boundaries are described as "running along Bugghilde Streete, on Stanihtan Hyll." In another Charter, dated 967, it is called "Bucgan Streete." At Weston-sub‑Edge it is still known as "Buckle Street." It passes near the village of Buckland, also by Brockhill and  p110 Buggyhill, now called Beckyhill. The road is raised in several places near Kineton Thorns, and where so raised it has a uniform width of nearly eight feet.

See "Transactions Bristol and Glou. Archae. Soc.," vol. IV, p213.

Condicote Lane

This is a perfectly street portion of a Roman Road, leading from Condicote Camp to the Cheltenham and Stow Road (called in an old charter the "Via Regia"). In several places the road is raised three feet above the adjoining lands, and where so raised has a uniform width of about thirteen feet. Probably this road extended further in each direction; towards the north it ran by Hinchwick Camp, and towards the south to Salmonsbury Camp, at Bourton-on‑the‑Water. A glance at the map will show that its direction is pointing directly to the latter place, though little or no trace of its course can now be found.

Cribb's Causeway

This is a local name for a portion of the Western Trackway which ran from Carlisle to Exeter. Cribb's Causeway seems to have commenced at Almondsbury. Leaving that portion of the West Trackway called the Ridgeway, it proceeds by Knole Park Camp to Henbury, thence by Blaize Castle and King's Weston Camp to Abone, crossing the river Avon into Somersetshire near Sea Mills.

Ermine Street

Most antiquaries are agreed in opinion that this was originally a British Trackway. Its name is thought to be derived from Ερμης (Mercurius), who is said to have presided over the highways. The road runs from Spinae, now called Speen, near Newbury, probably to Abergavenny and St. David's. It enters Gloucestershire about three miles north-west of Cricklade, through which place it passes. It then runs between  p111 Siddington and Preston to Corinium, now Cirencester; thence by Stratton, leaving the Bagendon Earthworks and Combend Villa to the east, and, passing near the villages of Elkstone and Brimpsfield, it reaches Birdlip. Here it is protected on the south side by a strong mound, and many Roman coins and other antiquities have been found when excavating for the foundation of houses near its course. After leaving Birdlip it descended the escarpment of the Cotswold Hills, but not on the line of the present road. It turned sharply to the west into Witcomb Wood, through the present lodge gates, and after pursuing its westerly course for nearly a quarter of a mile it turned round to the north, its old course being still visible, descending the hill by a far better gradient than the present road, and sunk several feet below the surface. It crossed the line of the present road near the sixth milestone from Gloucester, and rejoined it at the bottom of the hill. Leaving Witcomb Villa and the great camp on Cooper's Hill to the left, it pursued its course direct to Wotton, and thence, turning to the west, it entered Glevum (Gloucester). In lowering the road at the point where the Bristol and Birmingham Railway crosses it, the Roman pavement was discovered eighteen inches below the present level; the pavement was so well constructed that the engineers had to resort to blasting operations to assist in its removal. Entering Gloucester by the North Gate, and leaving it by the West Gate, it crosses the river Severn and Alney Island, and, passing Highnam, it leaves Huntly to the north and Longhope to the south, finally leaving Gloucestershire three-quarters of a mile east of Mitcheldean Road Station, on the Hereford and Gloucester Railway; thence it runs to Ross, Abergavenny, and probably to St. David's.

Forest of Dean Roman Roads

These are so numerous that it will be impossible in this work to describe them all. In that invaluable little guide called "A Week's Holiday in the Forest of Dean," published  p112 by Mr. John Bellows, of Gloucester, it is stated that nearly every road in the Forest of Dean is a Roman Way, and that the precise date at which these Forest of Dean Roads were made cannot be fixed, but that we have clear evidence that it was considerably earlier than the close of the 1st century. A portion of the Via Julia, running a little to the west of Welshbury and Little Dean Camps at Tibb's Cross, with the original Roman pavement still preserved, was examined a few years back by the members of the Cotteswold Field Club. In describing the continuation of this road near Blackpool Bridge, the guide above referred to says that a good deal of the paving is still visible, it being perfect for 100 yards up to Blackpool Bridge. The bifurcation made by the Roman engineers is distinctlyº traceable, by which, while the main road went straight on through the brook, a branch of it passed over the bridge to enable the traffic to be carried on in flood time. The pavement is seven feet ten and a half inches in width, and consists of cubes of conglomerate or millstone grit eight or ten inches square, with margins or kerbstones five inches wide and from ten to twenty inches in length. In many parts of the Forest of Dean the same kind of pavement is visible. Any one walking from the Speech House Railway Station to the Speech House will notice it in several places. The reason, no doubt, for there being so many Roman roads in the Forest must be attributed to the extensive mines worked by the Romans in this district, and in the position it occupies, with Glevum (Gloucester) on the north-east, Blestium (Monmouth) on the west, and Lydney and Caerwent on the south. In some places the margin stones which keep the central cubes in their position are in their turn supported by a line of heavy blocks firmly bedded in the ground in such a way as to bear against the points of the kerbstones. An excellent example of this exists on the road from Little Dean to Newnham, opposite the New Zealand Inn. The principal highway which passes by the Forest of Dean, and which I call the Gloucester branch of  p113 the Via Julia is further described under the Head of "Via Julia." The next in importance, a portion of which has been described in detail, comes from the north, by Mitcheldean, Welshbury Camp, Littledean Camp, Blackpool Bridge, and probably Lydney, with its fine Camps and Villa. Another important road leads through the forest from Littledean, by Cinderford Station to Coleford and Monmouth. Another runs from Mitcheldean, by Drybrook, to Coleford. There are others leaving the Via Julia — one at Westbury-on‑Severn, by Flaxley; another from Newnham to Littledean, and there are two lines of Roman Roads from Littledean to Mitcheldean. The above are the most important Roman Roads in the district; but, as I before mentioned, nearly every highway in the Forest of Dean was originally Roman, and, possibly, many of them British.

Foss Way

This well-known road has been the subject of much difference of opinion among antiquaries as to its origin. For instance, the Rev. T. Leman says, "It was a British Road, running from the north-eastern coast of Lincolnshire, through several important British towns, to the great British port of Seaton, in Devonshire;" while Dr. Guest states, "The name Foss has given rise to some strange hypotheses; it has been supposed that the road was so called because it was one of the hollow ways which marked out the lines of ancient British traffic; but, in truth, the Roman character of the Foss is perhaps more decided than that of any other highway in the kingdom." I cannot help thinking that there was a British trackway on the line of the Foss, long, long before the days of Caesar, though no doubt the Romans improved it; but I will not stop to discuss that point now. The Foss Way seems to have started from Portus Felix, at the mouth of the Humber, passing Lindum (Lincoln), Ratae (Leicester), Benonis (Claychester), Corinium (Cirencester), Aquae Sulis (Bath), Ischalis (Ilchester), terminating at the port of Moridunum (Seaton), on  p114 the coast of Devonshire. It enters Gloucestershire at Dorn, near to Moreton-in‑the‑Marsh, passing close to Batsford Camp, and through the town of Moreton; it ascends the hills to Stow-on‑the‑Wold, leaving Swell Villa to the west, and descends to Bourton Bridge, where it is joined by Buggilde Street. Here there are many Roman buildings contiguous to its course, and the fine Camp of Salmonsbury is only half a mile distant. Regardless of steep hills, it now pursues its way to Cirencester, passing close to Norbury Camp at Farmington, and within half a mile of Northleach. It crosses the Salt Way near Stowell Park, and joins the Ikenild Street one mile to the north-west of Cirencester. Leaving this, it continues its course to Bath, leaving the county near Trewsbury Camp; but for several miles it forms the county boundary. In many places it is raised a considerable height above the adjoining district, with a deep ditch on each side. On its course from Cirencester to Bath it passes the Roman Station of Whitewalls.

Green Street

This has been described as a Roman Road, running round the southern side of Churchdown Hill, crossing the Ermine Street near Brockworth, and joining the Sarn Way near the foot of Cooper's Hill, at a small hamlet still called Green Street.

Green Way

There are two ancient roads called the Green Way in Gloucestershire, one running by Badgworth and Shurdington, and ascending the steep escarpment of the Cotswold Hills, the other leading from Norbury Camp, at Farmington, by Puesdown and Shipton Olive to the great Roman Station at Wycomb, near Andoversford, and continuing its course, probably, along the hills to Cleeve Hill and Nottingham Hill Camps.

Ikenild Street

This is so called from the Iccĕni or Ikkens. It runs from the eastern side of the Island, and passes Kirklington, Woodstock,  p115 and Stonefield, crossing the river Evenlode to Wilcote, and so to Ramsden, a little beyond which village it can be traced at Whitty Green; but from this point, by Astally, and so through the fields to Bradwell Grove, it is scarcely visible. Nevertheless, there it is again to be seen holding a straight course into Gloucestershire, passing by Coln St. Aldwyn to Corinium (Cirencester), and joining the Foss Way one mile before reaching that town. After leaving Cirencester it runs on the same line as the Foss as far as Jackments Bottom, passing near Trewsbury Camp. It then branches off to Rodmarton and Cherington,º at both of which places there are Roman Villas. "It then traverses the turnpike from Tetbury to Hampden, passes a house called the Star and Garter to Chavenage Green, from whence it is an obscure horse-way through the fields to the Bath Road (which it crosses about a quarter of a mile before the separation of the Rodborough and Frocester Roads). It then descends into Lasborough Vale with a kind of sweep, and winds up the opposite hill to regain its course, having, as usual, tumuli for a direction on each side. It passe the inclosures by the edge of the valley in which Bagpath Village is placed, tending towards a vast tumulus on the brow of the hill, close to the road leading to Dursley and Rodborough." It then runs along Symond's Hall Hill to the ridge above Wotton-under‑Edge, where it descends the hill and all trace of it is lost. It probably continued by Elbury Hill to Cromhall, where there is a Roman Villa, and thence by Tytherington and Abbey Camps to Elberton, with its ancient camp, and Aust, where it crossed the river Severn, and joined the Via Julia, near Venta Silurum (Caerwent).

Oakle Street

A short road, only a mile and a half in length, forming a junction between the Ermine Street and the Gloucester branch of the Via Julia, four and a half miles west of Gloucester.

 p116  Patch Way

This appears to be a local name of probably a Roman road leaving the ancient Western trackway at Almondsbury, at the point where Cribbs Causeway and the Ridgeway join, and proceeding nearly due south by Patchway Green to Filton, and thence by Horfield to Bristol, crossing the main line of the Via Julia before arriving at that town.

Port Way

This leaves Gloucester by the Eastgate, and proceeds in a south-easterly direction by Matson and Upton St. Leonards, where a branch road leaves the main line and ascends the Cotswold Hills by Prinknash Park. The main road goes by Kimsbury Castle to Painswick, and then, probably, by Stroud, Woodchester Villa, Nailsworth, crossing the Ikenild Street near Calcot Farm, and then runs nearly south, attended by a large number of tumuli, by the great Roman camp at Sodbury, the Roman station at Cross Hands, leaving Tormarton to the east and Dyrham Camp half a mile to the west. It probably ran by Cold Ashton,º Swainswick, and Little Salisbury Camps, and joined the Foss Way one mile north-east of Bath.


This is a local name given to that portion of the Western Trackway which runs from the Ship Inn, about one mile and a half south of Thornbury, to Almondsbury Hill, where it joins Cribbs Causeway and the Patch Way.

Ryknield Street

This was a British Road, stated to commence at the mouth of the Tyne, and, passing the Watling Street at Catterick, it proceeded thence by Aldborough, crossing to the Watling Street at Wall, thence through Sutton Coldfield to Birmingham, King's Norton, Alcester, Bidford, Sedgebarrow, Tewkesbury, Berry Hill, near Ross, and probably by Abergavenny to St.   p117 David's. This road is very difficult to trace though Gloucestershire, except from Sedgebarrow to Tewkesbury. It entered Gloucestershire somewhere between Childs Wickham and Evesham, and, running by Sedgebarrow and Beckford Inn, it probably crossed the river Severn at Tewkesbury; but whether it continued its course to Ross by Newent, or ran down the western side of the Severn to the Ermine Street, near Gloucester, I am at present unable to say.

Salt Way

This is an old British trackway, which runs from Droitwich through Worcestershire, under the name of the Salt Way or the Salters Way. It enters Gloucestershire near Ashton-under‑Hill, and probably ran by Dumbleton and Toddington to Hayles, though no trace of it can now be found. In Isaac Taylor's Map of Gloucestershire, dated 1800, the Salt Way is shown running through the villages above mentioned. There is also a barn standing on the line of this road near Elmley Castle called the "Salt Way Barn," and there is a field between Dumbleton and Toddington called "Salters Close." Near Hayles Abbey the road is distinctly to be traced in its ascent of the Cotswold Hills, and runs, attended in its course by tumuli, past Hawling, Salperton, and Hazleton, crossing the Foss Way between Northleach and Stowell Park; thence it proceeds by Crickley Barrow to Coln St. Aldwyns, where it crosses the Ikenild Street, and eventually leaves Gloucestershire at Lechlade on its way to the coast of Hampshire. It is mentioned in a Saxon Charter, dated 969, as "Sealt Street." Some have supposed them to have arisen from the traffic of salt from Droitwich, while others hold that the name simply implies the "Hill Way."

Sarn Way

This road ran from the Green Street at the bottom of Cooper's Hill through a portion of Witcomb Road (a branch of it probably leading to the Witcomb Roman Villa in  p118 "Sarundells"), joining the Ermine Street close to Birdlip. The pavement of this road can still be traced in places.

Via Julia

The course of this road has been the subject of considerable discussion among antiquaries; but I think the following may be looked upon as the most probable solution of the difficulty. The main line of the road started from Isca Silurum (Caerleon), the head-quarters of the second legion of the Roman army, and passing Venta Silurum (Caerwent), Aquaeº Sulis (Bath), continued by Cunetio, near Marlborough, to Caleverº Atrebatum (Silchester), which seems to have been the great military centre of the south of Britain. This main line had, however, an important branch leaving it at, or near Caerwent, and running by Chepstow and Lydney to Gloucester. The stations on the main line are thus stated in the Itinerary of Antonine. Ab Isca:— Venta Silurum, MP VIIII; Abone, MP VIIII; Trajectus, VIIII; Aquis Solis, MP VI. Bishop Clifford, in a paper read before the Somersetshire Archaeological Society in 1876, after repeating the idea that the Via Julia crossed the river Severn to Aust, states that the road ran from Caerleon to Caerwent, and thence to Sudbrook Camp on the banks of the Severn, crossing the river at that point to Abone, near Henbury, and Blaize Castle, in Gloucestershire; thence it continued its course to Trajectus at Bitton, and on to Bath by the high ground passing the Lansdown Camps. After leaving Bath, its course can be easily traced onwards to Silchester. By adopting this course, the distances given in the Itinerary of Antonine are found to be correct. All other suppositions as to its route commence by stating that the distances as given in the Itinerary are wrong; but there seems to be no reason to doubt the accuracy of the figures, with such a clear proof before us that they might be quite correct. The branch of the Via Julia leading from Caerwent to Gloucester is not mentioned in any Itinerary; but as it has been traced  p119 whole distance, the only difference of opinion can be whether it is correctly called part of the Via Julia. Leaving Caerwent by the east gate, it ran by the village of Crick towards Chepstow; here it made a considerable detour in order to cross the river Wye. The ancient track can still be traced descending the wooded banks of that river some distance north of Chepstow, near Piercefield Park. It then ran by Combesbury Camp to Tidenham, and thence by Woolaston and Alveston to Lydney. (See under Lydney for Camps and Villas.) Leaving Lydney it took an inland direction by Soudley and Littledean Camps, thence bending to the east to Westbury-on‑Severn. It then probably followed the line of the present road by Stantway and Minsterworth to Gloucester, forming a junction with the Ermine Street near Highnam Court, two and a half miles west of the city. Some details of this road are given under "The Forest of Dean Roman Roads." It is paved for a considerable distance, having an average width of eight Roman feet.

Western Trackway

This British trackway must have been one of the most important ancient roads in England, and one that has been much over­looked by antiquaries. It seems to have originated at Luguballium (Carlisle), and ran by Coccium (Blackrode, Lancashire), Salinae (Droitwich), Branogena (Worcester), thence by Tewkesbury to Gloucester (Glevum), though opinions differ as to which side of the river Severn the original road took between the last two towns. From Gloucester it pursued its southern course in a line nearly parallel with the present high road to Bristol, as far as Almondsbury, passing Hardwicke, Whitminster Inn, Stone, Abbey Camp, &c. After passing Almondsbury, it pursued the road now called "Cribbs Causeway," by Knole Park Camp, Blaize Castle, and King's Weston Camp, crossing the river Avon near Sea Mills, and thence pursuing its course by Uxella, near Bridgewater, and so on to Isca (Exeter), thus forming a direct road from the North of England to Devonshire.

 p120  White Way

This was without doubt a Roman Road, running due north from Corinium (Cirencester) by Baunton Downs, North Cerney Downs, and Chedworth Beacon to the celebrated Roman Villa at Chedworth, and probably to the Withington Roman Villa, and so on along the Withington Road to the Dowdeswell Camps, and the great Roman Station at Wycomb, &c., thus giving access to many important Roman Stations and Villas from the great centre, Corinium.

There are, doubtless, many other British and road Roads in the county, such as "The Jack Way," "Pig Street," "The Calf Way," near Bisley, "Chittening Street," "Letch Lane," near Bourton-on‑the‑Water; the "Port Way," near Upper Slaughter; the road from the Cross Hands, by Chipping Sodbury and Iron Acton; the road near Thornbury, leading from the Western Trackway to Oldbury; the road running from the Via Julia, north-east of Chepstow and Beachley; and others, which may have been of British or Roman origin, such as the Via Regia, running by Wycomb, Lower Swell, and Stow-on‑the‑Wold; the road from Minchinhampton to Cirencester; and, doubtless, a communication existed between the numerous Camps on the escarpment of the Cotswold hills, say from Leckhampton Hill Camp, by Crickley Hill Camp, Birdlip Camp, Cooper's Hill Camp, Kimsbury Camp, along Sponebed Hill to Haresfield Camp; and, lastly, the road from Gloucester, by Newent and Dymock, and so on to the north.


Thayer's Note:

a See, however, Codrington, Roman Roads in Britain, pp242, 379.

[image ALT: Valid HTML 4.01.]

Page updated: 13 Dec 20