A Roman Mystery - the Lost 9th Legion Content from the guide to life, the universe and everything

A Roman Mystery - the Lost 9th Legion

0 Conversations

Copyright Reading Museum (Reading Borough Council).

There is some mystery, and considerable speculation, regarding the fate of the Roman 9th Legion VIIII1 'Hispana'. Its story was made famous by the novel The Eagle of the Ninth by Rosemary Sutcliff. The truth, however, is perhaps even less tangible.

Originally, the 9th Legion arrived in Britain around 43 AD as one of the four legions chosen to conquer the new province, and seems to have disappeared while on active service. Opinions are divided as to the fate of the 9th Legion.

Some favour the explanation that around 117 AD it was wiped out by local tribes north of the yet to be built Hadrian's Wall. There is little evidence to suggest that this was the case, apart from the fact the Legion's name disappears from the military rolls, and the discovery of a memorial to a soldier of the 9th killed 'in a great battle'. Compelling perhaps, but not conclusive. A further piece of evidence found in Silchester was an alleged military Eagle, a standard said to be carried by the 9th. More of that later.

Others favour the concept that the 9th Legion may have survived its tour of duty in Britain and saw action on the Rhine frontier, then later in Parthia, where it seems to have then disappeared completely, around 160 AD.

A Brief History Of The 9th Legion (VIIII)  'Hispana'

Formed in 58 BC, the battle standard of the 9th Legion was a bull. The Legion was awarded the title 'The Hispana' after action in Hispania (Spain), during the campaign against the Cantabrians (25 - 13 BC). A favourite of Julius Caesar's, fighting with him through all the Gallic Wars, they also stayed loyal to him in the Civil War against Pompey. The Legion was at the battles of Dyrrhachium and Pharsalus in 48 BC, and after the disaster of the Teutoburg Forest in 9 AD, was sent to reinforce Germania, then later moved to duties in Pannonia (Austria, Hungary and Serbia).

At the time of the British invasion, the 9th Legion was under the command of Caesius Nasica. Four legions were sent to Britain:

  • The 9th Legion (VIIII) 'Hispana'
  • The 20th Legion (XX) 'Valeria Victrix'
  • The 14th Legion (XIIII) 'Gemina Martia Victrix'
  • The 2nd Legion (II) 'Augusta'

These men were battle-hardened campaigners from all over the Roman Empire. Some history of the Legions shows that the sort of encounters they would have had before the British invasion would have made them well suited to the campaign. Three of the Legions had fought together before: the 2nd, 14th and 20th. The 9th Legion 'Hispana' had not fought alongside the others, but was more than capable of the task before it.

Locations In Britain Associated With The 9th Legion

The 9th Legion has left behind solid evidence of being in the Britain, and its route through the country can be mapped through the following locations;


The name Eburacum2 has an alternative interpretation as 'the land or estate of Ebracus or Ebros', and in the histories written by Geoffrey of Monmouth it is stated that the founding king of York was Ebracus. York was founded in 71 AD, initially a fortress constructed by the 9th Legion, it was built at a suitable point close to the Rivers Fosse and Ouse, and at a site where a bridge could be constructed. The area had a ready supply of timber for the erection of a fort also, positioned on a sandstone outcrop, and protected by the rivers to the south and east and substantial masonry walls. The settlement built around the Legionary fort thrived. Many of the streets were paved, and houses were built mostly from stone, as were many fine municipal buildings; including baths, theatres and temples to the town's Gods. It was from this base that the 9th Legion kept the peace in Northern Britain.

The settlement of York was also sited at an important junction in the Roman road network, and eventually had connections to the following forts and settlements to the north of the city:

  • North-east to Derventio (modern Malton), a distance of 17 miles.
  • South-east via Ermine Street to Petvaria (modern Brough-on-Humber), 28 miles.
  • South by road and ferry to Winteringham: the southern ferry crossing point of the river Humber.
  • South-west via Ryknild Street to Calcaria (modern Tadcaster), 10 miles.
  • North-west to Isurium Brigantum (modern Aldborough), 15 miles, to Dere Street and the roads north of the wall.

From York, the Legion Garrison could march out and access the north on the newly established route of Dere Street. The journey north progressed through the following established Roman settlements and forts;


The settlement of Aldborough is interesting as it shows evidence of the presence of the 9th Legion, provided by the following inscription found impressed on an excavated tile:

LEG VIIII HISP – Property of Ninth Legion Hispanic.
This indicates that the 9th Legion constructed buildings on the site. The town itself later developed into a city defended by stone walls, earth ramparts and ditches, with strong gateways. There were paved streets, a basilica, several temples and altars, a mansio – one of a number of privately-owned accommodation hostels established along the routes – and a bath house. The housing was of good quality, stone and timber-framed construction. How much of this the 9th started is unclear, however, the vast majority was completed much later after the 9th Legion left the area.

If you follow the route of Dere Street north from York, you will reach the area of the wall after a journey of around 88 miles. This is a three day journey for a Legion travelling at a standard pace, over easy ground or roads.

The route north from York would have probably have been as follows;

Cataractonium (modern Catterick)

'The Waterfall Town', founded around 70AD.

Morbium (modern Piercebridge)

Founded around 125AD.

Vindomora (modern Ebchester)

'The fort on the end of the hill', was at first a marching camp, then a fort was built by the 5th Cohort of an unknown legion, between 80 and 180 AD.

Corstopitum (the Stanegate Fort, Corbridge)

'The valley of great noise', the fort that constructed here was founded in 79 AD.

Onnum (modern Halton Chesters)

'The rock' was constructed by the 6th Legion around 190 AD.

At this point the route enters the area that was to become north of the wall, and was hostile territory.

Habitancum (modern Risingham)

This fort was built in 189 AD to support the Antonine expansion north.

Bremenium (modern High Rochester)

'The roaring stream', a turf and timber fort was constructed here in 80 AD.

Marching Camps

Camps were built by a legion at the end of a day's march to provide shelter for the troops while moving through hostile territory. Chew Green and Cappuck were built by the 20th Legion, around 80 AD.

Trimontium (modern Melrose)

Trimontium, 'the fort at the foot of the three Eildon Hills', was built in 80 AD and garrisoned for around 100 years. At its largest the fort was 60,700 square metres and the garrison comprised 1,000 infantry and 500 cavalry. In all, the fort extended to 80,937 square metres of fortified areas and enclosures, grouped around a parade ground. A large settlement developed to support the garrison, and this featured several shrines as well as a military amphitheatre. The water supply for the fort and town came from over 200 wells dug around the site. All this survived, and appeared to thrive in hostile territory for a considerable time.

To the west of Dere Street, was another route north into Caledonia from the settlements of Carlisle and Stanwix. It is likely that this would have come into the area protected by the 9th, as the nearest support Legion, the 14th, was 180 miles to the south in Wroxeter, Staffordshire.

Uxelodunum (modern Stanwix)

'The River Fort', also known as Petriana, was the main fort on Hadrian's Wall (estimated size 32,420 square metres) and was built by 20th Legion, around 122 AD. The 20th was based in a legionary fort built by the 14th Legion in 58 AD, at Viroconium Cornoviorum (modern Wroxeter)3 from 88 AD. This was the largest fort on the Wall and was placed at the point where the course of the Wall moved from the north bank of the River Eden to the southern bank. It appears that Stanwix was built on a ridge of high ground clear of the northern edge of the flood plain to defend the side of the Wall north of the river. The fort also was positioned to defend the end of River Eden bridge supported by Carlisle in guarding the southern end. It is worth noting that the garrison fort at Uxelodunum was built just 5 years after the possible incursion of the 9th into Caledonia. The tribes in the area do not appear to be have been actively hostile for some years.

Luguvalium (modern Carlisle)

Sometimes confused with Stanwix, which was the fort to the north of the Stanegate road. This was at first a timber fort built around 72 AD. Later enlarged in 122 AD, it was garrisoned to provide support to Stanwix Fort. This fort was situated a mile south of the wall and a road ran north from Carlisle to the Wall and Stanwix fort. The town of Luguvalium provided a crossing point for the western end of Stanegate Roman road, passing through Carvioan, Chesterholm and ending at Corbridge. There are three crossing points on the Wall in this area and local roads ran north to Neatherby (Castra Exploratorum) 12 miles. This meant Carlisle was well placed as the local trading centre. The settlement was on an important trading and supply route as goods came up from the coast via the port of Ravenglass (Glannoventa). Other roads also entered the town, from York via Catterick, and the west coast road from Ribchester. The traffic meant trade would have been good, and the town became very prosperous with paved streets, stone housing, municipal buildings, trade buildings and warehouses.

There is only evidence of three Legions (2nd, 5th and 20th) being engaged in the construction of Hadrian's Wall. The fourth Legion could have been held in reserve to reinforce the other Legions, or deal with other trouble spots. This was the other route into Caledonia that the 9th could have taken if it went north in 177 AD.

The home of the 9th Legion was ostensibly York, and the area that was to become Hadrian's Wall was in the Legion's patrol area. As the 9th took no part in the construction of the wall, it could indicate that it was busy elsewhere in Britain. They may have been moved to duties elsewhere in the Empire, or had indeed fallen in battle with the Caledonians in 117 AD. However, compelling as it might appear, this is not evidence of, or proof of, either a posting or annihilation.

So What Happened In Caledonia?

Perhaps the question should be who happened to the 9th Legion? That 'who' was Gaius Julius Agricola, originally a general serving under the Governor Gaius Suetonius Paulinus, in command of the 20th Legion. He left Britain only to return as consul and governor of the province in the summer of 77 AD. His actions in the province may have had an influence of what happened to the 9th.

His first task was to finally subdue the revolt of the Ordovices tribe in what is now Wales, crushing the final resistance in Mona (modern Anglesey) and forcing the tribe to beg for peace4. Agricola then turned his attention toward the northern tribes.

Agricola's invasion of the north was some 35 years prior to the alleged demise of the 9th, however, it serves as an illustration of how an incursion into this hostile region5 was handled by the Romans. It could be that this act of pacification caused problems that could have led to the circumstances surrounding the disappearence of the 9th. However, the survival of the Garrison, and the settlement of Melrose, 60 miles north of the Hadrian's Wall until after 180 AD6, indicate the opposite.

Agricola chose 79 AD to launch an expedition into Caledonia, so during the early summer his forces were assembled. We now have to deal with uncertain estimates of the numbers involved. The Roman forces were said to have numbered between 17,000 and 30,000. This number is recorded by Tacitus, of whom it was said;

[he] never let the truth get in the way of a good story – Cassius Dio.

Tacitus, apart from being a historian, was also Agricola's father-in-law, so perhaps a little pro-Agricola enhancement was included in the history. Tacitus puts the numbers as follows:

  • Auxiliary infantry: 8,000.
  • Auxiliary cavalry, four squadrons: 4,000.
  • The 20th Legion: 5,000.

That number equates with the lower estimate of 17,000. Where the other 13,000 came from is anybody's guess, it could have been a second legion and another 8,000 auxiliary units.

This indicates that the Romans did not appear to venture into hostile territory either unprepared or short of manpower. So the idea of the 9th setting off in 117 AD to subdue the Caledonians with a force of only 5,000 legionaries is, to be blunt, amounting to criminal folly7, even if there was a massive military emergency. It is also worth remembering that in 60 AD, during the rebellion led by Boudicca, the 9th had been routed while trying to relieve Camulodunum.

As a result, during the battle and subsequent withdrawal the 9th Legion suffered a reported casualty rate of over 50%. The survivors of this battle would be familiar with the tactics of the native armies, and have respect for what they could achieve en masse. Tacitus has claimed that the Caledonian forces could have numbered as many as 30,000; not an enemy to approach outnumbered and unprepared.

The forces led by Agricola were supported and supplied by ships of the Classis Britannica, who patrolled the eastern coast as far north as the Moray Firth. The advance of the force was restricted to the relatively flat areas along the western cost, penetrating as far as Cawdor – as evidenced by a marching camp. So, to indicate a later incursion by the 9th there would need to be evidence of a string of marching camps, which would also provide proof of the route taken.

After four years of campaigning, Agricola finally brought the Caledonian forces to battle8 at Mons Graupius in 84 AD. The casualty figures given for the battle were:

  • Romans killed: 360.
  • Caledonians killed: 10,000, with 20,000 fleeing the field to hide in the forests.

This is an indication of the power of a Roman Army, even when outnumbered nearly 2 to 19.

So, What Happened To The 9th Legion?

Did the 9th Legion march north into Caledonia to meet annihilation at the hands of hostile tribes? There is little evidence that this was the case, apart from some odd, and interesting, facts;

  • The 9th Legion's name disappears from the Roman military rolls.
  • A memorial to a soldier of the 9th killed 'in a great battle' found.
  • A Roman Eagle found in Silchester, Hampshire, sometime after the disappearance of the 9th Legion.

Investigating these facts further, if the 9th Legion went north in 177 AD, it would have most probably have travelled along Dere Street, and on into Caledonia using the route taken by Gaius Julius Agricola 35 years earlier. It is evident that this route was taken, as it was familiar and went across easier country.

At Corbridge, 57 miles to the east of Carlisle, a box containing a quantity of iron articles clearly of a military origin was discovered. It appears to be of the correct period to correspond to the 9th Legion, as it contains pieces of Lorica Segmentata, legionary body armour. This again indicates military activity, although it is impossible to attribute the find to any particular unit.

The memorial found indicates that the 9th Legion was in active battle with tribes in the north, but it is difficult to accurately date this memorial. The Vindolana tablets10, excavated from a Roman fort, speak of the fighting spirit of the British warriors. These writing tablets are said to be contemporary with the construction of Hadrian's Wall however, so a 'great battle' between the 9th Legion and the local tribes before that date has not been recorded. Perhaps the memorial is being overly gracious.

It is strange also, that if the 9th Legion disappeared, it was not recorded either. It is also incredible that a vast opportunity for political point scoring was missed by the Imperial Senate. Again, a defeat of that magnitude in 'a great battle' had also appeared to have gone unrecorded in the military records.

There are some records indicating that former officers of the 9th Legion were still on active duties after the possible loss of the 9th. Perhaps the most notable former officer was the Governor of Arabia during 142 - 143 AD, Lucius Aemilius Karus.

Cohorts of the 9th are also recorded performing frontier duties during 121 AD, stationed near Nijmegen in Holland. It has been claimed these men were on detached duties when the 9th could have moved north and they missed the disaster, it is very possible however that these facts11 are subject to interpretation.

The Silchester 'Roman Eagle'

A piece of evidence indicating the loss of a Legion in Britain is a bronze Eagle, alleged to be military, that was found in Silchester. This has been claimed to be the lost standard of the 9th. Unfortunately, this appears to be a case of a wrongly interpreted evidence.

The Silchester, or to give it its proper name, Calleva Eagle was found by the Reverend J.G Joyce on 9 October, 1866. Reverend Joyce was engaged on the site excavations of the Roman town of Calleva Atrebatum near Silchester, Hampshire, between 1864 and 1878. What he is best remembered for is the discovery of this bronze Eagle. It is fortunate that the Eagle survived, as this treasure was discovered in what are probably the ruins of the town's basilica, which had been destroyed by fire.

This is unusual site, as it was abandoned after the Romans left Britain, and as such has allowed researchers to create the most accurate survey of any Roman settlement in Britain. The Eagle is cast in bronze, and the position of the talons indicates that it originally mounted on something. The wings were broken off in Roman times, and it appears it was repaired, however the wings were broken again and lost in antiquity.

Today the Eagle can be found in the Silchester Collection at Reading Museum, and according to legend was the inspiration for Rosemary Sutcliff's book The Eagle of the Ninth12 and the sequel, The Silver Branch.

This Eagle is not, however, a legionary standard. It is a civilian municipal symbol. Examination has revealed that the Eagle originally stood upon a globe, and was very likely to have been an integral part of a statue – almost certainly that of Jupiter, the Roman god. It has also been suggested that it was a Roman Imperial Gift to a local tribal leader.

The Final Verdict

It is probable that the 9th Legion 'Hispana' was wiped out on the battlefield. Whether it was in Caledonia in 117 AD or in Parthia in 160 AD, rather depends upon your own point of view. There is just not quite enough evidence to prove either school of thought is correct, however, the balance of academic opinion seems to be slightly in favour of Caledonia as the resting place of the 9th, and a battle site would be a useful find to prove it13. But if there were to be a public enquiry today it may well prove to be inconclusive.


There is another theory that has some merit, which takes the middle-way, but would explain almost all the circumstances of the other two scenarios. It is this; the 9th Legion were badly mauled in Caledonia, and the survivors and garrison units that stayed in barracks at York were stood down from active duty. Later, the 9th were re-established and sent east to Parthia, where they may have been annihilated around 160 AD.

With grateful thanks to the Reading Museum for permission to use their image of the Silchester Eagle.

1The practice of writing nine in Roman numerals as IX dates from mediaeval times. The Romans themselves always wrote nine as VIIII.2Modern York.3The settlement round the fort later developed into the Roman town of Viroconium Cornoviorum. 4And possibly destroying the Druid cult; a form of worship that disgusted the Romans. They must have been fairly awful to disgust a Roman, alas we shall never know as to whether the destruction was total.5And any other around the Empire.6The fort and settlement of Melrose was abandoned sometime after 180 AD, some 60 years after the alleged loss of the 9th. Archaeological examination does not seem to indicate the population were forced out.7And, as such, would be unlikely to have happened. Normal procedure alone indicates that a supporting force of auxiliary troops would have been part of any force heading north.8By attacking and destroying the grain stores of the tribes; it was fight or starve.9The final battle with Boudicca provides ample proof of the ability of the legions to slaughter the Celts; Gaius Suetonius Paulinus and a force of 10,000 men outnumbered 8 to 1, defeated Boudicca and her army of 80,000 in this area in 60 AD. Roman losses were around 400.10In particular tablet 164.11Like all the evidence for the 9th's disappearence.12The films The Centurion (2010) and The Eagle (2011) were in turn inspired by this book.13The site of the disaster of the Teutoburg Forest in 9 AD, has been found.

Bookmark on your Personal Space

Conversations About This Entry

There are no Conversations for this Entry

Edited Entry


Infinite Improbability Drive

Infinite Improbability Drive

Read a random Edited Entry

Categorised In:

Written by

Edited by

h2g2 Editors

Write an Entry

"The Hitchhiker's Guide to the Galaxy is a wholly remarkable book. It has been compiled and recompiled many times and under many different editorships. It contains contributions from countless numbers of travellers and researchers."

Write an entry
Read more