top of page

Roman Collyweston

Historically we all know about some of Collyweston's History what with the Palace and the link to Kings and Slating, but recently I discovered the Time Team episode that was filmed not far from Collyweston at Bedford Purlieus woods and it got me thinking....


If Romans were as close as Thornhaugh to Collyweston then is there any chance that we may have more Roman activity closer to Collyweston….or even in Collyweston?


This started an investigation to see if Collyweston had any trace of Roman history.


During this investigiation I will try to explain my findings starting with a brief overview of pre Roman life in Britain before looking at how Romans arrived into our area.

I will attempt to explain the Roman activity found in the Collyweston area and provide some context as to where the Collyweston area fitted into Roman activity through the Roman period of occupation.


Finally we will find out how Roman influence in Collyweston ended.


How and when the Romans arrived to our area?

Before the Roman invasion of Britain lets just set the scene of what life was like. Britain today was referred to by the Romans as Britannia. 

Britannia was made up of a series of regional Iron Age tribes. The local tribe covering our area was the Catuvellauni. Their area of control covered Lincoln in the North to St Albans in the South. The catuvellauni capital was St Albans. 


These Iron age tribes were often in a state of war with each other. The division and subsequent conquering of Britain was made easier because of the regional factions and tribes.


Claudius appointed Plautius to lead his invasion of Britannia in 43, in support of Verica, king of the Atrebates and an ally of Rome, who had been deposed by his eastern neighbours, the Catuvellauni. The army was composed of four legions: IX Hispana, then in Pannonia; II Augusta; XIV Gemina; and XX Valeria Victrix, plus about 20,000 auxiliary troops, including Thracians and Batavians. Legio II Augusta was commanded by the future emperor Vespasian. Three other men of appropriate rank to command legions are known to have been involved in the invasion: Vespasian's brother, Titus Flavius Sabinus, and Gnaeus Hosidius Geta appear in Dio Cassius's account of the invasion; Gnaeus Sentius Saturninus is mentioned by Eutropius, although as a former consul he may have been too senior, and perhaps accompanied Claudius later.



















On the beaches of northern Gaul Plautius faced a mutiny by his troops, who were reluctant to cross the Ocean and fight beyond the limits of the known world. They were persuaded after Claudius's freedman and secretary Narcissus addressed them. Seeing a former slave in place of their commander, they cried "Io Saturnalia!" (Saturnalia being a Roman festival in which social roles were reversed for the day) and the mutiny was over.








The invasion force sailed in three divisions, and is generally believed to have landed at Richborough in Kent, although parts may have landed elsewhere (see Site of the Claudian invasion of Britain). The Britons, led by Togodumnus and Caratacus of the Catuvellauni, were reluctant to fight a pitched battle, relying instead on guerrilla tactics. However, Plautius defeated first Caratacus, then Togodumnus, on the rivers Medway and Thames. Togodumnus died shortly afterwards, although Caratacus survived and continued to be a thorn in the invaders' side.


Having reached the Thames, Plautius halted and sent for Claudius, who arrived with elephants and heavy artillery and completed the march on the Catuvellaunian capital, Camulodunum (Colchester). A Roman province was established in the conquered territory, and alliances made with nations outside direct Roman control. Plautius became governor of the new province, until 47 when he was replaced by Publius Ostorius Scapula.[4] On his return to Rome and civil life, Plautius was granted an ovation, during which the emperor himself walked by his side to and from the Capitol.[5]


Between 43 and 47AD the Romans under the command of Plautius took control of cambridgshire, northamptonshire and lincolnshire. They fortified ground won with garrison defences as they expanded north. Two major garrison towns in our area were established at Durobrivaue and Great Casterton.












Durobrivae was a Roman fortified garrison town located at Water Newton in the English county of Cambridgeshire, where Ermine Street crossed the River Nene. More generally, it was in the territory of the Corieltauvi in a region of villas and commercial potteries. The name is a Latinisation of Celtic (or more accurately Brythonic) *Durobrīwās, meaning essentially "fort (by the) bridges".[1]


During the Iron Age Britain was divided into distinct tribal areas. The area lay between the Catuvellauni to the south and the Corieltauvi to the north, with the Iceni to the east. The origin of Durobrivae is said to have been as a vicus attached to a "pre-Flavian" fort, established about half a kilometre to the east of modern Water Newton, between the Claudian invasion of 43 AD, and the beginning of the reign of Vespasian in 69 AD.[2] Its first historical mention is in the Antonine Itinerary of the late 2nd century.[3] Archaeology of the Roman period shows that Durobrivae was then the production centre for a fine tableware known as Nene Valley Colour Coated Ware, from the 2nd century to the 4th.[2] In Anglo-Saxon times, local settlement came to centre on Medeshamstede, now known as Peterborough.



Great Casterton - The Romano-British settlement at Great Casterton developed in the protective lee of a Roman auxiliary fort which was built on the north bank on the River Gwash during the early Claudian campaigns of the propraetor Aulus Plautius c.44AD. The defences of the fort, which lay to the north-east of the settlement, enclosed an area of around 6 acres (2.4Ha). The fort is large enough to comfortably house an auxiliary cavalry ala, or perhaps a cohors milliaria equitata, a mixed unit of cavalry and infantry a nominal one-thousand strong. Whatever the actual garrison, military occupation is thought to have ceased sometime around 80, presumably due to the unit being withdrawn for use in the campaigns of the propraetor Gnaeus Julius Agricola.



What Roman Settlements are there in and around Collyweston?


Base Collyweston Hill - Enclosure (?) (unlocated, but perhaps around SK 9903) was recorded by William Stukeley in the mid 18th century. He examined what he called a Roman Camp on Collyweston Hill, N. of the village, which was '200 ft. square'. The enclosure was apparently bounded by a rampart the S. side of which was 'very intire' and the N. side partly so. The ramparts on the E. and W. were 'thrown down and ploughed over'. There was no trace of a ditch. The Romans called a simple rampart wall an agger; at this date great height was not necessary. (Surtees Soc., (1885), 54 and 55). No remains are now visible on the ground.


Collyweston Quarries - Quarries (SK 999029, TF 000030 and 004037, etc.), covering large areas on limestone N. and E. of the village. These are the remains of quarries worked from at least as early as the Roman period for the well-known Collyweston Slate, used widely as roofing material. 

According to the grid reference the location is in the field to the back of Des Knapps old house and near the pocket park.

The slate is a fissile sandy limestone at the base of the Lower Lincolnshire Limestone and was either worked in open quarries or mined from small shafts. Remains of both types of workings are visible. The blocks of limestone were laid out on the ground, with the bedding planes vertical, and kept watered to prevent drying out. They were then split into thin layers by hand, often helped by frost action. 










Bedford Purleus - Iron ore production and potential villa

In the early 19th century the antiquarian Edmund Tyrell Artis came across the remains of a Roman site in the woods near Peterborough. He claimed to have found Roman statues, buildings and burials, surrounded by evidence for ironworking, but he only ever published drawings and a map, and over time his site was lost, until being rediscovered in 2005 by the Forestry Commission.


An earthwork survey and evaluation trenches by Northamptonshire Archaeology (NA) identified a range of buildings and a possible courtyard (interpreted as a courtyard villa), a series of large quarry pits, and an enclosure with evidence of ironworking.


Time Team aimed to follow up NA’s findings with further trenches. The range of buildings turned out to be fairly basic and utilitarian in nature – no sign here of painted wall plaster or other ‘high status’ elements. However, a raised platform in the south-east corner of the courtyard produced not just painted plaster but also box flue tiles from a possible hypocaust system; this may have been the site of the villa’s bath-house.








The enclosure to the west of the villa was confirmed as being industrial in nature, by the identification of an iron ore-roasting floor. A trench through one of the quarry pits found evidence of iron ore extraction, the disused quarry being subsequently used as a dump for domestic waste from the villa.


One of the finds from Bedford Purlieus was a colour coated cup which now sits proudly in the Royal Ontario museum in Canada.

A43 Opposite Manor House - Roman settlement (?) (SK 99560228), S. of the village on a W.-facing slope at 280 ft. above OD on limestone. Roman pottery is said to have been found in an old quarry (OS Record Cards) and pottery and roof tiles have been noted on adjacent land.








Collyweston Bomb dump - 

In Autumn 1953 a widespread Roman site was uncovered by contractors Wilson, Lovatt and co at the old bomb site located in Collyweston Wood. The site lies within 3 miles of Ermine Street and 5 miles of Durobrivae.

The discovery at this site of large quantities of stone and pottery were reported to Mr J.C.P Langton who was Lord Exeters Agent. After a small scale excavation by Mr J.C Stephens of the same office, a corner of a rectangular building was unearthed. At this point the chief inspector of ancient monuments was informed and work was started in October for 1 month on the site. A remarkable complex of buildings was found. These comprised of a group of dry-stone walled buildings, including two rectangular, one circular, one octagonal and one hexagonal, was excavated after partial destruction. 

The summary of buildings was:

Two temples side by side with 2 small buildings immediately adjacent.

A square temple with 2 small buildings adjacent.

A square temple in a rectangular enclosure with 3 smaller buildings. There were also a number of other smaller buildings scattered around.

A square temple with 2 small buildings beside it.

Another square temple

A square temple in a very large enclosure in an irregular hexagonal shape. There were 3 other buildings in line.

An octagonal building on the west side of a basilica with a large diameter of 32ft. It appears to have been monopteral although it has a little porch. Monopteral means Round and without a cella; consisting of a single ring of columns supporting a roof; said especially of a temple.

A considerable amount of pottery including some beautiful peaces such as a large bowl with a dolphin pattern, others with sea horses from local famous potters. Its important not to under represent the shear large quantity of vases, sherds, jars. There was also some beautifully decorated broochs and pins. The pottery found ranged from the 1st to the early 4th centuries.


Roman temples (TF 00730058) lie in Collyweston Great Wood on rocks of the Upper Estuarine Series, at 275 ft. above OD.


Tixover Grange - The villa lies in the parish of Tixover, Rutland, in the grounds of Tixover

Grange, and is 250 metres north-east of the Grange.Excavations were conducted in 1958 and 1959 by the University of Leicester Department of Archaeology. The first published reference to Roman finds at Tixover seems to have been the comments of the Reverend Edward Trollope who, when writing about a Roman building at Apethorpe more than a century ago, reported that part of a hypocaust was found. Much later tesserae were found in 1912 in the Grange.s In 1932 S. E. Winbolt conducted limited excavations, and found two geometric design mosaic pavements, walls, pottery and coins, but the results of the work were not fully published.6 Fortunately, a teenager, Graham Webster, kept a notebook of what took place, and this included sketches of the mosaic floors. Further Roman objects were said to have been found during the construction of the sewage disposal plant in the 1940s.

The major part of the villa seems to be under the present spinney and on a small area of terrace gravel which overlies limestone of the inferior oolite series. The main axis of the villa lies NNE-SSW and it seems probable that it faced east where the ground slopes gently downwards to the River Welland 200 metres away. One hundred metres south of the villa is a natural spring. 


Roman floor level can have been only just below the present ground and much of the building has been destroyed by cultivation and tree roots. To the north of the spinney this destruction was almost complete,although a number of possible wall foundations, one-course high, were found. It is interesting to note that the only early pottery from the 1958/9 excavations came from north of the spinney in trench VI.

Within the spinney itself, dense undergrowth prevented detailed excavation, but a mosaic floor was uncovered in area XIV. Two mosaics were found in 1932, and sketched by Graham Webster. It is not possible to locate precisely the position of these mosaics, but in a note by Winbolt, now in Rutland County Museum, there is a clue. It says;"Remains of two mosaic pavements were recovered just inside the north fence of the spinney, from about five yards (i.e. 4.5 metres) from the east end of the spinney, extending westward for forty-four feet. These measurements would take us close to the position of the 1959 mosaic, and yet there are difficulties in connecting them. The detailed designs would not seem to be the same although there are some similarities between the 1959 floor and the western mosaic found in 1932.

The remains south of the spinney were more substantial, due to the fact that this part of the building was heated, and the foundations for the hypocaust were deeper than for the rest of the villa. In trench III the bottom of the wall was 1. 50 metres below present ground level, but the wall itself did not survive to this height throughout the trench. Only the

bottom courses of the footings survived in trench II.

Four coins are recorded from the site. Those mentioned in the published note are Faustina II, Postumus and two mid-fourth century coins. 10 In the same reference the pottery is described as being mostly Castor ware although second-century samian is also said to have been found. Winbolt's note in Onkham Museum gives a date range of c. A.D. 150-350.u 


A settlement situated so close to the great Nene valley factories was bound to have drawn on those prolific sources for most

of its common domestic pottery, and this is what we find. Some 70% of the pottery, so far recovered from Tixover, is in colour-coated fabrics, and all of these vessels are presumably of Nene-valley origin. Only three or four sherds of grey ware are present in this admittedly small sample, a situation which does not obtain on sites only a few kilometres away to the north.

Where so much is unstratified, we must be content with a definition of the date-range represented by these sherds. The pottery from south of the spinney is all fourth century in date, containing a number of distinctive flanged bowls of the form Gillam 230,'3 a very common type which emerged in the early fourth century, and lasted until the end of the Roman period.

The other colour-coated sherds from this deposit could all belong to the first half of the fourth century, but a date after A.D. 350 cannot be entirelyexcluded. Trench VI to the north of the spinney, however, produced pottery

of a much earlier period. This material includes ring-neck flagons and a few sherds of samian ware. The sample is too small for certainty to be attainable, but the date is likely to be within the period c. A.D. I 50 - 200.

The mosaics can be dated to the second half of the fourth century, but whether these floors were inserted into an already existing

building, or were laid at the same time as the building could not be determined.



The material from the early excavations is said to have found its way to the

Peterborough Museum, and there is a note in the accessions list to say that

tiles and tesserae were presented to the museum by Graham Webster in

1932.14 It is also reported that Oakham School Museum had some materiaI. 1 s

Finds from the construction of the sewage disposal tank were said to

have been reported to the British Museum, but it has not been possible to

trace these. 16

There were no coins or other small finds apart from a number of badly

corroded iron objects of indeterminate date.

All the material from the excavations, including plans, photographs and

notebooks, has been deposited at the Rutland County Museum, Oakham. 


Remains of two simply patterned pavements were brought to light in 1932,18

and part of what appears to have been a third was found in 1959 during the

excavations which are the subject of the present report.

One of the patterns was a rectilinear grid of red lines, each formed of

a double row of tile tesserae, with a background of white containing random

blue tesserae (fig. 4). The white squares enclosed by the grid measured 23

cm. b)Z .. 23 cm. A coloured sketch made in 1932 shows next to the pattern,

a surround of plain bands of white, blue, and red in one side, but it suggests

that only the blue and red bands were carried round the other sides. The

other pavement, also sketched in 1932,19 had a pattern of intersecting octagons,

each with a central square, executed in red lines on a white ground and

surrounded by a band of blue. A sketch of the remains uncovered in 1959

also records a rectilinear grid of red lines on a white ground, with a band of

blue tesserae three rows wide next to the pattern, and a band of white with

random blue tesserae ten rows wide between this and the wall of the room.

Identical or similar patterns can be cited elsewhere in England, but in

the present context the most significant are those of other sites in the east

midlands. Both patterns are recorded at Haceby in Lincolnshire,20 a grid

pattern and an intersecting-hexagon pattern of the same class as that of

intersecting octagons, though elaborated with additional colours, are known

from Denton, also in Lincolnshire.2 ' A grid pattern formed part of the

mosaic of Medbourne in Leicestershire, 22 and an intersecting octagon pattern

is recorded from the 'Cherry Orchard' villa2 3 in the environs of Leicester.

A version of the intersecting-octagon pattern may also be noted in one of

the pavements of the villa at Great Weldon in Northamptonshire,24 the best

of which was almost a replica of mosaic no. 2 of the villa at Great Casterton,2s

in the same county as Tixover. The mosaic of Great Casterton was

dated by excavation to not earlier than A.D. 350/36526 ; those of Denton date

from c. 370. 21 It seems reasonable to surmise, therefore, that in this part of

Britain such patterns were current in the second half of the fourth century. 



Ketton - In 1902 several square yards of Roman tesselated pavement were discovered off the north side of High Street. This suggests that the area may have supported rural villas, with the main local settlement being at Great Casterton.


What was the purpose of these settlements, what was everyday life like?

Bearing in mind that pottery at the Collyweston Woods site has been dated from the 1st to the 4th century AD we assume that activity in and around Collyweston was busy with stone movement north through the old road to Wothorpe where it joined Ermine Street or transportation South through the drove to Peterborough.

Other important materials such as the Iron Ore production at Bedford Purlieus woods near Thornhaugh would also have bought additional activity to and from our area.

The discovery of the Rampart at the bottom of the valley near the river Welland near the bottom pond is also important as this would be placed on a route of importance. You don't put defences on routes that are not of importance. This route connected Collyweston with other villas at Tixover, Duddington and Ketton.




Whilst Collywestons Roman settlements and workings were not of the importance of the 2 main regional towns of Casterton and Durobrivae we can assume that Collyweston had considerable Roman activity. This could be attributed to the local stone slates which were distributed all around the area and exported through Peterborough Quay. 

The value of Collyweston and the immediate area was recognised with villas at Tixover,Ketton and Tinwell Road Stamford. Roman villas are common around the Northants/Lincolnshire/Cambs area however Collyweston was blessed with a large number of Roman Temples at the old Bomb dump in Collyweston Wood which were certainly an exceptional discovery.


The departure of the Roman Empire from Collyweston co-incided with the departure of the Romans from the British Shores.

The beginning of the end came with the death of Emperor Theodosius in 395 AD; his empire was divided among his two sons; Honorius took the East, and Arcadius had the West. While the Eastern Empire was thriving, the West was on the brink of collapse. By the beginning of the fifth century, Italy was under attack and Stilicho, the most powerful military presence in Rome withdrew the vast majority of legions in Britain. At the same time, Germanic raiders were attacking the Southern and Eastern coasts of England.


In 405 or 406 AD, the Vandals, Alans, and Suebi crossed the Rhine and caused chaos in Britain. Constantine III took charge of the troops in Britain in 407 AD and tried to establish himself as Roman Emperor in the West. The natives apparently expelled the Roman administration in 409 AD, and when they asked Emperor Honorius to help with the invaders in 410 AD, he told them to fend for themselves. This response marked the end of Roman influence in Britain.


By 425-430 AD, Britain and Collyweston was in no way, shape or form ‘Roman’ as villas had been abandoned, mosaic and fresco workshops had closed, and barter replaced money. London was in ruins by 430 AD, and Roman culture and organization had disappeared by 600 AD. Attempts to salvage the Empire in the West were in vain as the last emperor was deposed in 476 AD. Although many Roman cities in Britain fell into decay, others were expanded later on, and places such as Canterbury remain occupied to this day.

bottom of page