The Evidence for Roman Chain Mail

I've briefly blogged about Roman chain mail before - Roman Chain Mail - but several new discoveries were announced this year, upping our known examples and I wanted to go through some of the others.

For centuries our only information about Roman chain mail came from it's depiction in art, for example this detail of the Ludovisi Sarcophagus.

The Ludovisi Sarcophagus dates from the mid 3rd century AD, and is sometimes linked to Hostilian, the son of Decius who was himself briefly emperor in AD 251. Although the sarcophagus depicts a Roman general fighting barbarians, Hostilian did not fight the Goths, but rather died of the plague at the age of 21.

Decius and his elder son Herennius Etruscus died the same year fighting the Goths, so the iconography might have been better suited to them. Decius' body was never recovered, so the sarcophagus could not be his. I'm not sure what became of Herennius' corpse, but presumably it did not make it back to Rome as he never seems to be suggested as a potential occupant of the sarcophagus.

Whoever it belonged to, the Ludovisi Sarcophagus is a fine depiction of the long chain mail shirt worn at the time, with sleeves that finish above the elbow.

The artistic evidence supports the idea that the lorica hamata was worn by auxiliaries and Gauls or enemies: Trajan's Column is the best known example, with a fine depiction on the base of chain mail amongst the discarded arms and armour, presumably the enemy's (third photo here; very short sleeved, and to the upper thighs). Roman soldiers in chain mail are also shown on the column. These are the earliest securely dated depiction I have been able to find of chain mail in a Roman context.

As people may have worked out by now, Adrian Goldsworthy is a very good friend, and we rarely disagree. He believes that chain mail is shown on the late Republican so-called Altar of Domitius Ahenobarbus, and wrote this in his The Complete Roman Army book. I have included the two details from it below, but I am not sure; it may be scale armour. And it seems that Roman soldiers rather than auxiliaries are represented. My feeling is that the armour is too schematically depicted for us to be certain either way.

Adrian also sees as an example of 'chain' mail the Augustan Vachères Warrior, but I feel that it shows a lorica squamata - the warrior is in any case a Gaul who would have been an auxiliary and not a citizen.

Adrian also included this metope from Trajan's monument at Adamklissi as an example of auxiliaries wearing mail. I agree with him on this one, and this ties in with the evidence from Trajan's Column:

As a quick disclaimer, Adrian and I have not discussed this ... yet; but I assume we will now at our next monthly lunch. The point is that there seem to be no certain depictions of chain mail in Roman art before Trajan's Column.

Moving on to actual parts of chain mail found in excavations.

Pat Southern and Karen R. Dixon [The Late Roman Army, Yale University Press, 1996, pp. 96-7] give an example of a fragment of lorica hamata or chain mail in their book on the later Roman army which was found in the 19th century - their fig. 15, to the left here.

It was found in Thorsbjerg Bog in Germany, where deposits date from around 100 to 500 AD, and I have not been able to find a date for it. The bog is in an area of Germany never conquered by the Romans, and it is unclear where the chain mail was made since the finds include Danish, Germanic and Roman items. From the drawing the construction looks looser than Roman examples, and closer to the much heavier chain mail the Vikings adopted.

This fragment of chain mail was found at Dura Europas, as town on the edge of modern Syria which in Antiquity repeatedly changed hands. This fragment was found at the Temple of Bel (source) in the 1930s.

More recently a whole skeleton wearing mail was excavated at Dura Europas (JSTOR). His context shows that he was mining under Tower 19 in AD 256 according to research by Dr Simon James. He is now believed to have been a Roman soldier rather than a Sassanian (photo below). The chain shirt is shorter than the one on the sarcophagus, ending at the waist, but the sleeves are longer and cover the elbow.

A few more examples are given by Southern and Dixon, which I will go through one by one.

The Vimose Bog in Denmark, whose finds date from the Iron Age to the early Middle Ages, and which includes "a few coats of chain mail" according to the preliminary publication by

The local museum at Vimose has an exceptionally good web site, and includes photos of the suit to the left and a drawing of the chain-mail belt below. The museum identifies the chain-mail as similar to Roman chain-mail worn by German auxiliaries from the 3rd century onwards, and notes that in Scandinavia only leaders wore it due to its expense.

At this point, I would like to mention Nick Papa. He is one of the guards at the Acropolis, who makes his own chain-mail, and was kind enough to share his knowledge with me. Although chain-mail is time consuming to make, it is possible for a man to make it himself - unlike bronze cuirasses - and to repair and adjust it himself with relatively simple tools. The main expense would have been the metal. He uses just pliers once he has formed the rings.

Some armour has been found in tombs, other examples on soldiers where they fell on the battlefield. These finds are preserved because they were thrown by 'barbarians' into bogs, presumably as votive offerings. Armour could and often was re-used, and handed down through the generations - Procopius in Wars describes the Franks on the Gallic frontier as dressing like Romans, and the evidence suggests that Clovis had a Merovingians army not dissimilar to the Roman one of his opponent Syagius of Soissons. The archaeological evidence suggests that other Franks also continued to use Roman military equipment, either captured off Roman soldiers, or passed down by ancestors that had fought in the Roman army as mercenaries.

The next example of loria hamata given by Southern and Dixon comes from Caerleon, but I couldn't find a photograph or details (mea culpa, etc). Caerleon was the Roman castrum of Isca, now believed to have been occupied as late as AD 380.

I did however find this photos of chain mail found in the Roman fort at Arbeia, which was the port that supplied Hadrian's Wall. The extremely well preserved suit was found in 1997 in a room of the barracks in the SE of the fort destroyed by fire (photo):

Prof  Michael Fuller has two photographs on his web site, including this one showing how well preserved the armour is:

In addition parts of lorica hamata were found associated with the Flavian fort at nearby Carlisle (Luguvalium) in 2000 (JSTOR).

The example of loria hamata given by Southern and Dixon from Rainau-Buch has well published by Bernhard A. Greiner in this 2008 book (PDF here): cat. no. 5.33, pp. 97-101, figures 130-7. It is a small piece of chain mail, as one can see in his image below.

Greiner's publication is well worth reading for anyone genuinely interested in the history of chain mail. The context of the find dates to AD 229-254, and the find spot in Germany.

The final example Southern and Dixon give is from Künzing in Bavaria, which I wasn't able to look up, but assume comes from the well known 3rd century hoard of military equipment.

As is clear, several finds have been made since the publication of Southern and Dixon, including two this year.

The first find came from a tomb in the Caucasus, Russia, and the announcement included the detail of the chain mail below: Treasure-Filled Warrior's Grave Found in Russia | Caucasus Necropolis | LiveScience

Although the necropolis, which was in use from the 3rd century BC, had been looted, this tomb dating to circa 200 BC was intact, as were the animals buried with him. The gold got the press, but the mail is always more interesting. Obviously the burial is not Roman, but it provides interesting evidence for the invention and use of chain mail which I will probably have to leave for another post. I don't think Valentina Mordvintseva has published it yet?

The second announcement came from Germany this summer, that came from the site of the Battle at the Harzhorn fought between the Romans and the Germans in the 230s (the battle is dated by coin finds at the site in Saxony):

Chain Mail Found Well Preserved On Ancient Battlefield - Science News

Although the suit of chain mail has been described as the "first" almost complete one, only photographs of fragments accompanied the news.

The mail was found at the edge of the battlefield, and the assumption seems to be that it was Roman and either left behind by a soldier or dedicated by the victorious Germans. Prof. Dr. Michael Meyer will be publishing it in due course.

The Romans did not invent chain mail, and the general theory seem to be that they got it from the Gauls, but it had been around much longer. The oldest example comes from a Scythian tumulus at Zhurovka, dated to the early 5th century BC by the red figure pottery which presumably obtained through trade via Olbia.

It is described in the 3rd century BC by Polybius (6.23.15), but only began to be used more regularly in the 2nd century AD.

Very few examples of actual mail survive from either Roman or Celtic graves, partly because metal was melted down, and because it corroded, so I thought it would be worth posting more examples I know of.

This example of Roman chain mail is on display in the museum at Carnuntum in Austria (photo; they give the following publication: Legionsadler und Druidenstab, F. Humer, 2006, ISBN 3854602294):

Some lorica hamata was excavated at Trimontium, the Roman fort at Newstead in Scotland, but the pieces are fragmentary and very much rusted together. The site has produced other types of armour and some spectacular helmets. Information about the chain mail is pp. 161ff here (PDF - no photo); the photograph comes from H.R. Robinson, The Armour of Imperial Rome. Arms & Armour Press, London, 1975 - this image is on p. 172.

Because of the unusual aspects of this chain mail, as see in the photograph below, it is sometimes classified as Lorica Hamata Squamataque, an unusual fusion of forms described in this article by

This Roman chain mail was also found in Scotland, in cauldon at Carlingwark Loch, Kirkcudbrightshire. The context was non-Roman, and well outside Roman territory so is 'native' Scottish.

 It is published in: H.R. Robinson, The Armour of Imperial Rome. Arms & Armour Press, London, 1975 - this image is on p. 172

Fragmentary Roman chain mail was found in a similar non-Roman Scottish context at Castlehaven.
This tiny piece of bronze mail was found at Lunt Fort, Bagington, Warwickshire, constructed during the revolt of Boudicca, and abandoned soon after around AD 80 (photo). It was in use again in the 3rd century.

The Netherlands have yielded both examples of lorica hamata squamataque; see the article by

from a barrow at Fluitenberg (photos left and below from
W A B VAN DER SANDEN, Fragments of a lorica hamata from a barrow at Fluitenberg, Netherlands, JRMES 4 , 1993, pp. 1-8; idem here in Dutch; and this article by Over stansen en klinken: De vervaardiging van een maliënkolder uit de IJzertijd gevonden te Fluitenberg (PDF here - the source of these and many more photos of the chain mail).

As well as differences in technique, one can deduce two rough dating criteria. Firstly in how the armour was attached: the 'Celtic' hooks, often in the shape of snakes, which held together the chain mail, as at Carnuntum, ceased to be used after the Flavian period. The second is the amount of the body covered by the mail: Republican lorica hamata seemed to be sleeveless, Augustan ones developed short sleeves, then the sleeves grew longer; in addition the earlier mails were waist length, the later ones much longer, although the evidence from Adamklissi may be interpreted as cavalrymen wearing longer mail in the reign of Trajan.

I will cover pre-Roman chain mail in another post ... and maybe some of the Roman chain mail I didn't mention.


  1. This is very interesting and useful. Thanks Dorothy. I am most interested in the Late Roman army where there is some debate as to whether the armour depicted on monuments is chain mail, or moulded cuirases of bronze or leather.
    A point on the Lunt fort at Baginton. My understanding is that the original fort was dated roughly to the period of the Boudiccan rebellion, but that it was re-used in the 3rd Century. Perhaps this view has been revised.
    Dylan (Bickerstaffe)

  2. Great post, Dorothy. I'm just wondering who now thinks that the dead officer in the mine under Tower 19 at Dura (pictured) is a Roman soldier rather than Sasanian. That would be quite a shift. Any published reference? Judith

  3. Dylan - thank you.
    You're correct about Lunt Fort, and I've amended that bit.
    re chain mail vs cuirases, I would have thought it was easier to tell them apart in art, but if you have links to specific examples I can try to help or ask Adrian

    Judith - I know Simon James initially said Sassanian in his AIA talk and AJA article, but his emphasis there was on the chemical gas element.
    From what I understand, and Adrian has discussed his research with him, he now is leaning more towards the soldier having been Roman ... and now has the soldier as Roman on his university web site http://www.le.ac.uk/ar/stj/dura.htm#destruc

    Adrian footnoted the info in The Fall of Rome, and is one of the most punctilious scholars in these matters. He cited for the soldier being Roman:

    S James 2005 'The deposition of military equipment during the final siege at Dura-Europos, with particular reference to the Tower 19 countermine', Carnuntum Jahrbuch 2005, 189-206
    S James 2004 Excavations at Dura-Europos, Final Report VII, the Arms and Armour, and other Military Equipment, British Museum Press, London
    James' (then) forthcoming book: Rome and the Sword.

    I know a 'scholar' has been claiming this research as her own, but ... it is not, nor has she seen the evidence herself. She's pulled this kind of crap before over other theories, and I prefer to rely on Simon James.

  4. Thanks for that information. I don't quite see that James says this on his website: "the countermine was still full of dead Roman soldiers when excavated", which need not exclude there being one Persian elsewhere in the mine; but I suppose it could be read either way. I'll check the other references asap and, if need be, update my Death-of-Dura-Europos post.

  5. IIRC, the Dura Sasanian warrior had a trident woven into the maille in copper alloy rings. This is typical Iranian heraldry.

    I'm afraid I don't have that book - would you mind summarising his argument please?


  6. Naddum, that sounds rather convincing but who knows ... I figure the Roman / Sassinian identification will keep changing, and one problem on the German frontier, at the other end of the empire, is that people wore enemy armour too.

  7. Hi Dorothy,

    One example of uncertainty as to whether a cuirass of metal or leather, or chain mail is depicted is the column of Arcadius in Istanbul. This is now destroyed, but I have Freshfield's publication of 16th century drawings (and there are other drawings in Oman, Byzantine Empire, and Bradley, The Goths), and took photos of surviving fragments built in to the wall of a 19th century building nearby.

    Best Dylan

  8. Latest on Celtic chainmail:


  9. Please, chain mail is a misnomer. The correct and accepted name for this type of armour is simpy mail.

  10. The largest collection of European mail images in the world.

  11. Jobs in South Africa, Online Job Search for Latest Jobs in South Africa


I do not moderate comments, but I remove spam, overt self-promotion ("read [link] my much better post on this") and what I consider hate speech (racism, homophobia etc).

Note: only a member of this blog may post a comment.