Change Style Print Page RSS Feed

The Roman Capital of South Scotland

Tons of Stuff about Trimontium Fort






THE FORT - SECOND PHASE - THE DOMITIANIC FORT (after the Emperor, Domitian)











The fields behind the Stone, to right and left, up to the foothills of Eildon Hill North on your right (a Bronze Age tribal capital of the Votadini, twinned with Traprain Law in East Lothian) contained a very large Roman fort surrounded in the second century by a 20' high stone wall, backed on the inside by an enormous 'wedge' of earth ie a 'rampart' 40' across at its base, and 'fronted', on the outside, by a set of three ditches (the barbed wire of the ancient world), the first of which was 20' across and 10' deep. Behind you, to the North, was an 'annexe' or settlement, leading down to the river, surrounded itself by a smaller ditch and rampart, where people lived and worked at all sorts of trades under the protection of the fort. Similar 'annexes' existed elsewhere: to the West, in the field with the telegraph poles, leading down to Newstead (where, in from the turn of the road, and removed a little from the military atmosphere, stood the huge half-timbered mansio - motel/admin/trading centre - and the fort bathhouse, with its curved concrete roofs - in the Spring see in the field the dark patch of the dumped coal and charcoal, even after 2,000 years) ; to the South, spreading up the low green ridge to the left of the Eildon, (part-industrial estate/market town, part field system) ; and to the East, along the road on which you are standing, stretching beyond the field to the railway viaduct line and beyond.

(As you walk East you will come to a gap in the hedge on the left - the first century North Gate; over the hedge to the right you may be able to see, in a good dry summer like 1996, the yellow marks of the buried streets and buildings of the fort, as if from an aerial photograph, such as in the Museum. Still walking on, at the point when you can see the three bridges - railway, 18th century road, and modern A68 - the hollow on your left is the first Roman military amphitheatre to be identified in Scotland (by Dr Lonie of Newstead in 1991) and, at present, the most Northerly and Westerly in the Roman Empire. Between the little 18th century bridge and the A68 bridge there was a second century stone Roman bridge described in 1743 as a 'famous bridge' there was a second century stone Roman bridge. The road mound has been found: the bridge stones, in the water, if any are left are more elusive. - carrying Dere Street across the Tweed. The first century crossing may have been by ford or bridge to the West of Newstead on the flat land North West of Millmount Farm, where aerial photographs show a temporary camp.

The whole complex covers some 370 acres. The landowners are Col Younger of Ravenswood, Melrose and Lord Devonport of Kirkwhelpington, both Trustees of the Trimontium Trust.


The fort at Newstead, situated on a bluff on the South bank of the river Tweed, commands the Tweed valley.

Eildon North Hill and Trimontium Stone It was a key defensive site throughout the Roman period and was the hub of Roman roads in Scotland. Of these 500 miles of Roman roads only one, so far, has produced a milestone, found at Ingliston near Edinburgh, and giving the distance in Roman miles from the roads HQ - TRIMONTIUM, (perhaps originally 'castra trium montium' - the camp of the three hills or place of the three peaks or Triple Mountain). See the Newstead information shed.

Trimontium is the name given to it in Ptolemy of Alexandria's second century map and in the list of ancient place names, the seventh century Ravenna Cosmography. It is taken to refer to the three Eildon (pronounced 'Eeldon' ) Hills - Eildon Hill North; the Mid Hill; and the Little or Bowden Eildon - all Bronze Age sites and landmarks visible from all directions. (There is also the Little Hill beside them, the vent of the volcano of long ago.)

It was no accident that the Romans placed their South of Scotland HQ beside such a landmark. The 1986 excavations on Eildon Hill North indicate that there is a 1,000 year gap in the occupation of the hill between the end of the Bronze Age and the Roman Iron Age ie that the Romans found the hill unoccupied, placed their signal station (still visible with ditch and bank) on the Western tip of the hill and encouraged some reoccupation of it.

General Roy, surveying Scotland after the 1745 rebellion, placed the Roman fort at the village of Eildon, to the South East of Newstead. It was not till the Waverley Railway Line was being laid in the 1840s and Roman artefacts began to be found by the navvies in cutting through the South Annexe that its location became clearer. It was another sixty years later, towards the end of a busy two decades of Roman fort-finding and excavation in Britain (eg Birrens 1896), that James Curle of Melrose, a solicitor by profession, was given permission by the Society of Antiquaries of Scotland to excavate. This took from 1905 to 1910, proved a sensation at the time because of the quality of the artefacts found, and was recorded by him in a magisterial work published in 1911 and entitled 'A Roman Frontier Post and its People'.

In 1947 Sir Ian Richmond undertook a corroborative excavation with the aid of German prisoners of war. For forty years fieldwalkers from Selkirk - the Mason brothers, J Walter Elliot, Jack Cruickshank and Caroline Cruickshank - gathered evidence from the field surfaces, including many intaglios (soldiers' rings with semi-precious stones). Aerial photography by Professor J K S St Joseph of Cambridge and the Royal Commission in Scotland (G S Maxwell et al) also took place on an annual basis. Bradford University (Drs RFJ Jones and Simon Clarke) were involved in summer excavations from 1987 to 1998, (full report expected in 2006/2007) including the Newstead Project, which attempted to study Romano-Celtic interaction in a 50 sq km area around the fort with parallel excavations at native sites, and in the rescue excavation along the route of the third phase of the Melrose Bypass, which the local authority succeeded, after two public inquiries, in putting through the old railway cutting in the South Annexe, which had been returning to nature since the closure of the railway in 1968, and where, far from there being little to find because of the work of the navvies, forty major archaeological features were recorded, including four wells to add to the 107 wells which Curle had found and which have been the glory and the enigma of Trimontium since his time. Some of the artefacts in these easily dug, stone-lined wells or pits in a high water table area may be rubbish discarded when the fort was abandoned. Some certainly represent votive offerings to appease the gods of the underworld and they range from priceless chased sports helmets, to carpenters' tools, offcuts of tents, and animal heads, given by people obsessed by the spirits of the natural world around them who carefully sealed off these entrances to the underworld when they were filled - and dug more.

There are no upstanding stone remains at Trimontium today but guides on the Trimontium Walk point out the features that can still be seen in the fields, including the swell of the ploughed-out rampart, and the amphitheatre. It is an almost tangible story.

The 'Newstead' artefacts form the greater part of the national collection which is on view in the Early Peoples section of the Museum of Scotland, Chambers Street, Edinburgh. Apart from the Trimontium Museuem in Melrose Square there is a small collection of finds in a room of the Commendator's House in Cloisters Road, entry to which is gained by visitors to Melrose Abbey.


The earliest Roman remains on the site are a series (eight at the last count) of large marching camps (the ditches still visible after 2,000 years as dark lines in aerial photos). They represent temporary stopping places for tented armies on the march ie for the first garrison before they completed the fort or, after it was occupied, for visiting troops passing up or down the line, who could not be accommodated in the fort itself. They are of varying sizes, two being 40 and 50 acres in extent, and one being called by Curle 'The Great Camp'. Trimontium may have been a gathering place for armies eg under the Emperor Septimius Severus in 208-10 when he campaigned in the North of Scotland and may have wintered at Cramond.

There are many periods or phases of fort construction at Trimontium. The first in 80 AD probably built by the Ninth Legion from York ( not 'lost', according to the latest evidence) during Agricola's northern campaign leading up to his victory at Mons Graupius (in Aberdeenshire?) in 83 AD, was about 10.5 acres in extent, contained wooden buildings, and was defended by a rampart of earth only, built up on a foundation of cobbles, and with two V-shaped ditches, 9' wide and 3' deep approx, in front. In shape it was an 'irregular' fort in that it departed from the standard playing-card shape layout. The lines of the rampart in each quarter are staggered, so that people approaching the gates in each side must do so at an angle, thus exposing themselves to side fire. Other Agricolan forts in Scotland display similar characteristics, but the overall idea seems to be exceptional (the work of one engineer?) and examples are rare. See outline drawing. The West Annexe, an enclosure defended by two ditches, seems to have been the first extra-mural development.

It is suggested that the ala Petriana, the biggest in the country, provided the cavalry wing stationed at the fort at this time.

6. THE FORT - SECOND PHASE - THE DOMITIANIC FORT (after the Emperor, Domitian)

About 86 AD the Agricolan fort was extended to 14.5 acres and its defences strengthened. The two ditches were infilled and replaced by a single huge ditch, 20' across and 12' deep. The earth rampart, again on a cobble and rubble base, was now 43 ' wide and 28' high (including the palisade on top). The buildings, though still wooden, had stone foundations - to last longer. As with phase I, there is little evidence for the arrangement of buildings and streets within the fort.


During the occupation, settlements, perhaps military in origin but subsequently of Romanised natives, presumably with some form of their own local government, developed all round the fort - to the North (discovered in 1996) and not yet fully measured, but busy with trade and artisan activities; to the South (14 acres eventually; a market township astride two roads - one of the first century, one of the second - coming up to the South wall of the fort; an industrial estate; an agricultural area, leading to the outlying field system which had large U-shaped drainage ditches, as opposed to V-shaped protective ditches); to the East (20 acres; the main entertainment area for the troops; large residential houses for the merchant entrepreneurs; a bazaar for travellers along Dere Street, the later name for the main North-South Roman road, crossing the Tweed at Leaderfoot); and to the West, where the first bathhouse was built (later much extended) and the mansio, a huge half-timbered building, traditionally regarded as a motel for official travellers, and recently suggested to be (perhaps in addition) an official trading station placed outside the fort, where local dignitaries could make their council tax arrangements with the Revenue Department of the Roman State.

The annexes were defended by ramparts of piled-up earth, and by ditches, similar in dimensions to the fort ditch. The West, South and East annexes (and presumably the North also) seem to have had inter-connecting gateways


The traditional view is that the first period of Roman occupation lasted from 80 to 105 AD. The Romans gave up trying to have a Tay-Forth or Clyde-Forth frontier line of forts. "In the early years of the second century the costs or risks of maintaining the outposts (in South Scotland) appear to have exceeded the advantages, and they were abandoned. For nearly forty years thereafter the lower isthmus (of the province of Britannia) formed the north-western frontier of the Empire, and the tribes of Scotland were left to pursue their respective ends without the hindrance or help of a resident garrison." G S Maxwell, 1989.

The Emperor Hadrian's visit to Britannia in 121-122 AD resulted in the building of the Hadrian's Wall frontier. Again the traditional view is that Trimontium was re-occupied about 140AD and became a support centre to the rear of the (Forth-Clyde) Antonine Wall, when Hadrian's successor Antoninus Pius brought an army back into Scotland. Trimontium may well have been re-occupied some time before that as an outpost fort, North of Hadrian's Wall.


After twenty or even forty years' absence the earth rampart immediately required an additional 'dump' of earth to strengthen it, and the fort ditches seem to have been hurriedly recut, with the inclusion of branches as an easily available additional deterrent. The buildings may have reverted to wooden construction in this 'precarious' phase. Extra-mural settlement, with a market place, existed in the South Annexe. There was capacity for 1,500 troops, not counting the South Annexe civilians. Like the initial first century Agricolan phase this second century re-establishment was probably brief.


After the building of the Antonine Wall in the early 140s AD, Trimontium, no longer an outpost beyond the front line, but a support fort in the rear, had its garrison reduced and the fort rebuilt in stone for permanence, by men of the Twentieth Legion. A 6' thick red sandstone (?) wall was built in front of the earth rampart, but the size of the fort area was reduced to 10 acres by the building of a new stone West wall (the 'Reducing' Wall) INSIDE the fort to the East of the existing West rampart, and leaving the Western portion of the actual fort as a new Industrial Annexe. The 'Reducing Wall' itself had no rampart or ditches - these still remained on the Western edge of the new industrial area. Three relatively small, widely-spaced ditches enclosed the fort and its new internal Industrial Annexe.

The gateways were re-positioned, eg the new South Gate - and a new second century road coming through the South Annexe to it - was established 70m East of the previous gate and the first century road which led to it (and still remained in use as a street). These can all be seen on the fort plan. The busy South Annexe, with its marketplace, had a hundred 'stripe' buildings (up to a size of 100 sq.m each) fronting on to these two roads - shop in front, with workshop and living quarters behind that, and a rubbish dump out back. The East Gate was rebuilt in stone; the line of the main road out was moved some ten yards further South; further settlement took place in the East Annexe; the orientation of the fort was to Dere Street and the East. Apart from the reconstruction of the bath-house, the Western annexe became a backwater, compared to the South and East Annexes.

The garrison - 1,000 strong, housed in twelve barracks in the Eastern part of the fort - was, for a time, a specialised cavalry unit, the ala Augusta Vocontiorum civium Romanorum, a tribe originally from Spain and the South of France. The Vocontii were a crack regiment and both horses and men were magnificently equipped. In the centre of the fort stood the principia, the HQ building, with two granaries, horrea - one on either side. The commander's house sat South of these buildings, and North of them another 'monumental' building, perhaps the the Deputy CO's house.


The Antonine Wall was in use for only twenty years or so, and may have been given up for a few years within that period because of 'military difficulties'. The repercussions were felt at Newstead; major changes took place in the arrangement of the fort. The 'Reducing Wall' was demolished; the wall round the fort - the 'Curtain Wall' - was extended round what had been the new internal Industrial Annexe; this area was now given over to long narrow buildings with timber sills on stone foundations ie barracks, to house a much bigger garrison. Outside the fort, in the Western annexe the emphasis was on reduction and defence, as is shown by the building of an earth rampart round the bathhouse, and the cutting of a ditch between the bathhouse and where the mansio had been. The extra-mural population began to decline; Trimontium was now a front-line fighting base, instead of a centre for industrial production and distribution.


The final period of occupation at Trimontium lasted from the abandonment of the Antonine Wall in the 160s AD until the army withdrew, probably about 185AD. The major changes now included the demolition of many buildings in the Western area of the fort and their replacement by small hut-like buildings; and the demolition of the HQ building and its replacement by the so-called 'cross-hall ', which created , with the granaries, an open courtyard facing West ie a new open-plan HQ perhaps facing in another change of fort orientation. The barracks in the Eastern part of the fort seem to have changed to 'chalet' form and may signify not only housing for each soldier and his family but a massive reduction in troop numbers.

The extra-mural population was now in decline. The South Annexe may by now have been deserted, since field ditches are found in what was a settlement area. Civilians may well have moved inside the fort defences.

Trimontium changed from an occupied stronghold to but another former 'frontier post', where visiting troops in the buffer zone North of Hadrian's Wall might stop on a regular tour of inspection.

Some thirty years later, Scotland was still important enough for the Emperor Septimius Severus to invade it again in the campaign of 208-9 AD. He may well have encamped at Trimontium on his way to Fife and the North, where he is thought to have fought the Caledonians. After probably over-wintering at Cramond, he returned to York (Eburacum) and died there in 210AD. Thereafter, his two sons, Geta and Caracalla reverted to the policy of containment of Scotland by means of a buffer zone, in the Borders and Lothians, of tribes paid to be friendly (note the Traprain Law, East Lothian, treasure of 'hack' silver, given by the Romans to the chiefs of the Votadini - the same tribe as had occupied Eildon Hill North - and found in 1919 by A O Curle, brother of James Curle).


Scotland's Roman Remains - Lawrence Keppie - J Donald pub - £9.99) available at the Museum

The Romans in Scotland - G S Maxwell - J Thin pub. 1989 - £16.99

Roman Scotland - David Breeze - Batsford pub. 1996 - c £14


A Roman Frontier Post and its People - James Curle, Glasgow University Press 1911- collector's item

Inventory of the Ancient Monuments of Roxburghshire, 1956, 2 vols - Royal Commision on the Ancient and Historical Monuments of Scotland - collector's item

The Outpost Forts of Hadrian's Wall in the days of the Romans - Frank Graham, 1983 (illustrated by Ronald Embleton; available (c£2) from Vindolanda, Chesterholm, by Hexham

The Trimontium Story - J Walter Elliot - £1.50 - pub. The Trimontium Trust available at the Museum

Trimontium: A Roman Frontier Post and its Phases - Simon Clarke - £1 available at the Museum

Newstead 1996: The Northern Vicus and the Amphitheatre - Simon Clarke - £1 available at the Museum



Excerpt from report to First Public Inquiry, 1989

"The position of the fort will have been chosen following Agricola's third campaign, when much of Southern Scotland was overrun. The site commands the crossing of the Tweed by Dere Street, the main Roman road to the North, and affords a long view of the valley upstream. Other roads diverge from Newstead; Westwards along the valley; Southwestwards to Raeburnfoot and the Annan; and, in all probability, Eastwards down the Tweed. The fort thus lies at the hub of a road system, planned in relation to tribal territories. Dere Street has been traditionally regarded as separating the Votadini from the Selgovae (though Breeze and Mann suggest the Selgovae lay further Southwest). No other site is so well suited as Newstead to be both the military HQ in South Scotland, and an administrative centre, for the only Roman milestone from Scotland, that from Ingliston, shows that distances were reckoned from TRIMONTIUM.

One and a quarter miles to the Southwest of Newstead, the summit of North Eildon Hill is crowned by a hillfort , 39 acres in extent. The juxtaposition of the largest hillfort and the largest Roman fort known in Scotland, which can hardly be a matter of chance, offers a unique opportunity of studying the impact of the Roman invasion on the indigenous population. With the construction of the Roman fort, if the hillfort had not already been abandoned its continued occupation by the natives would not have been tolerated, as is shown by the presence of a Roman signal station inside the hillfort itself. Within a few miles of North Eildon lie some 20 to 30 small, enclosed or fortified 'native settlements', where excavation might reveal the response to the invasion, and also to the evacuation of the hillfort. The potential for History of the Roman fort at Newstead, both in itself and in relation to the archaeology of its neighbourhood is thus very great".


"Royal Fine Art Commission for Scotland, 9 Atholl Crescent, Edinburgh 26 May, 1989

Sir, - We have read with considerable interest William Chisholm's article "Digging in to protect the relics of our Roman past" (25 May), which arose from the recent (ie the first) public inquiry into the Borders Regional Council proposals for the final stretch of the Melrose by-pass.

Whichever route is chosen must endanger areas of archaeological interest which lie between Trimontium Roman fort on the lower slopes and Eildon prehistoric fort at the top of the hill above Melrose.

It is rare for a short stretch of road proposed through the countryside to excite so much public concern.

While it would be unwise to discuss matters which are in a sense sub-judice until the Secretary of State has reached his conclusions, the article may inadvertently have given some of your readers a wrong impression.

Dr Peter Addyman, director of the York Archaeological Trust, gave evidence not for the objectors to Borders Region's proposals, nor indeed for Borders Region, but for the Royal Fine Art Commission for Scotland.

In addition to Dr Addyman and Mr Robin Birley, other archaeologists gave evidence, including Dr David Breeze, of the Scottish Development Department, and Mr Roger Mercer of Edinburgh University, who spoke in favour of the Region's proposals.

J K St Joseph, a retired Cambridge professor who has spent a large proportion of his professional life investigating Trimontium, objected to the Region's proposals and favoured a route to the south which largely avoided the Roman encampments.

This Commission's task was to take a well-informed, objective and independent view on all aspects of the proposals, and in this it was greatly helped by Dr Addyman's expert advice on the likely archaeological damage which would be caused by the alternative routes.

On balance, the Commission's views coincide with those of Professor St Joseph and Mr Birley because of the proved significance and the potential public attraction of Trimontium.