The Romans had gods and goddesses for everything, even minor deities for minor matters. Security, however, was not a minor matter, and its gods figured prominently in the Roman pantheon. Securitas was the Roman goddess who was the personification of security. Janus was the god of doors and thresholds, gates and locks, whose symbols were the key and staff, to open doors and to ward off those not entitled to enter. His temple in Rome was open in times of war for those who would pray to keep enemies outside. It was closed in times of peace, but Roman policies being what they were, was rarely closed.

Content This is a presentation of Roman security hardware from my own collection, with illustrations of several important related items from other sources. This is not intended to be a survey of all known types; descriptions, images and analyses mostly refer to owned items. There is a bare minimum of literature citations and copyrighted material is used only in reference. Although many web references are appended, I have not made links to all sites mentioned. Any inaccuracies of fact or interpretation are my own. Some items presented are Byzantine, and this is noted wherever possible. The attributions of dealers are of variable quality, and should be accepted cautiously. It is concluded by Vikan (Security in Byzantium) that on the basis of specimens of keys and lock plates found, slide key locks did not transfer appreciably to Byzantium.

Origins.  It is usually stated by lock historians that Roman locks were derived from Egyptian and Greek sources.  Well, maybe so, but metal locks were used by the Etruscans (,  Mysterious Etruscans), (ancientworlds) so Occam's Razor would have it that the concepts and some of the technology would have been inherited by the Romans from their former masters. For example, the Etruscan demon goddess Vanth is usually depicted with some of her attributes of snakes, torches, boots, chiton and keys. At right is an image from an Etruscan sarcophagus of Vanth bearing on her left side the key to Hades, fastened by an unbreakable iron chain about her neck. Note that this is a key to a warded lock with a rotary mechanism.

I have not yet found any images of Etruscan locks, but here is a quote from (Dennis): "Above this tomb, in the higher part of the mound, were discovered three very small chambers, one of which was unrifled, and contained . . . . a hoe, a key and part of a lock of iron." 

And here is another quote from (Dennis): "Next in antiquity come the warded locks in ordinary use. Whence these had their origin is unknown; they seem, however, to have been used by the Etruscans, as rude specimens of them have been found in the excavations opened on the sites where the cities of Luni and Populonia arose three thousand years ago. In the Etruscan Museum of Volterra, a wonderful collection very little known abroad, there are some locks of this description, which seem, however to have never fallen under the observation of writers on ironworks."

On the terracotta sarcophagus of an Etruscan woman named Seianti Thanunia, dated to the second century B.C.E, she appears to be wearing a key ring.(Marshall)

Fabrication. The metal of preference for small artifacts was bronze. Brass (orichalcum) was known and used also to a lesser extent (Rehren). It was highly regarded and even used for coins (numismatics) and even for the recently discovered sceptre of Maxentius (societasviaromana), but bronze predominated for most small articles.

 The casting and working of bronze is, and was, not a high tech business and was widely practiced in the ancient world. For a detailed discussion of Roman bronze casting, see Strong & Brown and Needham. In histories of the lock it is always pointed out that the Romans did not invent them, but that they did develop metal locks with complex mechanisms, and spread the various types all over the map of the Mediterranean region and much of Europe. Just who did the actual development of any particular feature will of course remain forever unknown. Metal work was labor intensive and expensive, and it is likely that security products that involved metal work were pretty much for the small elite classes who had costly possessions to secure. No doubt most Romans made do with locking devices no more sophisticated than a simple wood bar for the door, and perhaps a latch lifter, as did the poor in more recent times.

All the same, it's evident that metal working was a large and widespread industry. Coins, of course, were made in the millions, but other small items such as the fibula or garment fastener, strigil, stylus, buckle, handles and ornaments for horses, chariots, shields, belts, containers and furniture are plentiful. A great deal of artistry went into the design of such metal artifacts, and skill into their fabrication. It's clear that a lot of talent was wasted in menial, poorly rewarded tasks, but that's a modern concept that may have been unfamiliar to the Romans.

Most of what is known or said of Roman locks and lock making is based on archaeology, deduction and speculation, since we are troubled by the lack of written records of crafts and craftsmen. Those Romans who were literate and could afford security hardware wouldn't have been very interested in the mechanics of its production, and certainly not in the efforts of lower classes trying to make a living. No doubt locksmiths were members of trade/craft guilds, called collegia, but I've not examined that relationship.  For discussions of Roman Collegia, see such articles as Long and Haywood. It's worth noting that collegia did survive into the Byzantine era, where they were called somateia or systemata.  They evolved into organizations similar to the medieval guilds.

We know very little about the methods that artisans used in antiquity. All we really have are the locks themselves, and mostly fragments at that. To most classical writers, such matters would have been beneath contempt! If you have a noble world-view, you don't think much about all those sweaty guys in their squalid workshops making locks and keys, much less write about them! Nevertheless, we can see here (right) a crew of Romans doing blacksmith work.

I don't know whether Roman civilian locksmiths made custom products or maintained stocks of prefabricated locks, keys and caskets from which selections could be made. Probably both, I would guess.

We can only speculate on the marketing of locks & keys. However, it is reasonable to suppose that they would have been sold over the counter in a shop typical of the small business such as that shown below (Middleton). For a nice discussion of Roman shops, see therthdimension, from which is the following quote: "Roman shops were generally small, with the family living above or behind the shop. . . most shops closed down at dusk, when shopkeepers would secure their shop's shutters to the pavement with strong padlocks." The only known example of locksmithing premises is the Locksmith's House (Casa del Fabbro) in Pompeii. I believe that the owner of this house was more likely a building contractor or locksmith's patron. For a discussion of small tradesmen and their shops at Pompeii, see Johnston  who remarks that "locksmiths, goldsmiths and other craftsmen had the necessary equipment and sold their own goods."

And finally, I'll show that famous image from a locksmith's grave stone, now in the Aquileia museum, that actually shows Roman lock makers at work! Note especially the completed door lock on the lower right. The bellows and tools: hammer, tongs and file, are immediately recognizable, and remain much the same for the next 1500 years and more. Notice also the design oŁ that little furnace, made like a miniature temple.

I believe there is a great amount of literature that has been published that refers in passing to Roman security hardware that I've not yet found. For example, an interesting reference by

Alexander Adam (Adam, 1835) shows a group of door fittings (below). Adam also makes the interesting statement "It appears that the locks of the ancients were not fixed to the panels of the doors with nails like ours, but were taken off when the door was opened, as our padlocks ". I don't know if this belief is still accepted.

Collecting A word about the collecting of ancient locks. Since locks and lock making were and are regarded as minor crafts rather than art, archaeologists normally record them only in passing. Many locks and keys remain in the collections of local museums near their find spots, whether displayed or not. Most locks and keys appearing on the market at this time have been found by individuals with metal detectors, and appear with little or no provenience. Fortunately, these are regarded as mere "collectibles", having relatively low monetary values and relatively low interest to the general public. National governments of the places of origin do not regard them as national treasures, place few or no restrictions on their export and certainly have no interest in their repatriation from other countries. We are blessed by neglect and indifference and may sail serenely below the radar of nationalist art advocates.

Since the recovery of Roman security hardware is (very distantly) related to archaeology, let's note in passing two terms that collectors will encounter. Provenance has been defined as "the history of successive custody of a particular item or collection". Provenience is simply a record of the place where an artifact was found (find spot). Normally, neither will be available for small collectibles such as I discuss here. Provenance will become increasingly important for special items of high value as they are documented and pass from owner to owner, but doesn't appear to me to be very important at this time. Still, better keep all that paperwork and don't entrust it all to digital files, either!

I do have somewhat mixed feelings about the collecting of Roman locks and keys. At this time, most Roman security hardware is probably in private hands and progresses rather quickly from private excavator to dealer to collector. Archeological institutions have codes of ethics which require archeologists to have nothing whatever to do with unprovenanced materials. Consequently, archaeologists deplore collectors and the collecting of antiquities: they are regarded as encouraging looting and the consequent destruction of sites that would otherwise have provided valuable information about ancient cultures. I agree with that, but it is also true that a great number of finds of locks and keys with suitable pedigrees languish in museum vaults and are unavailable to most collectors and authors. In my opinion, collectors who publish or allow access to their material can perform a considerable service to students of technology and history. They provide places where many examples of their obsession are gathered together, permitting a comparison and survey of the subject. The other side of that coin is that collections with casual protection and security provide places where many artifacts can be destroyed simultaneously! This can be a disaster when one-of-a-kind specimens are involved, as is often the case with Roman materials.

Although the provenience may be lost, this has not been regarded by collectors as a great problem in the special case of Roman security hardware since the technology of their design and production seems to have been uniform throughout the empire. However, I have to admit that metal artifacts without context do not allow dating, and provide no information on the historical development of this hardware. Although such information is very widely scattered, I believe that given sufficient interest and resources, it would be possible to write a history of the development of Roman locks from archaeological information. It has not been done.

A rotary key similar to several examples that will be shown later (224,279,etc.) has been found by Soren and James at Kourion, Cyprus, in a context that can be securely dated to the earthquake of July 21, 365 CE.  Of course the finds at Pompeii and Herculaneum provide another fixed point for the dating of styles, but we do not yet have access to all the security hardware recovered there. Any volunteers to take on the history project?  It would probably be most suitable for a European with better physical access to museum collections. How about a doctoral thesis?

Condition The preservation and recovery of ancient locks and keys is largely a matter of chance, but amazing survivals are possible under the right conditions. Bronze persists fairly well in the soil. Iron and wood depend on special conditions. For these, there must be a location were there is little oxygen, which can be in the presence of organic matter such as peat or in mud or other waterlogged soils. The Thames mud is celebrated for finds of artifacts from all periods. The abundant leather and metal ware found at Vindolanda are a tribute to the preservative effects of waterlogged soil. As we will see, some few locks are even known with associated surviving wood and wood impressions. The locks recovered from Pompeii and Herculaneum are a special case, and have provided a great amount of material that is still being analyzed and published. We are greatly indebted to Alberto Biasiotti for his work with this material, and for the use of several of his images.

Ancient locks and keys of iron are relatively few, due to rapid corrosion in soil containing water and oxygen. Those that have been recovered are usually in poor condition, and it is not uncommon to find bimetallic iron/bronze keys in which the iron is nearly completely gone, leaving a key handle. Another problem with iron is that the Romans could not achieve temperatures high enough to melt and cast it. All iron artifacts had to be forged and finished with hand tools. Surviving iron keys and lock parts are not as pretty as bronze, and therefore not so desirable to collectors.

Collectors greatly prefer that ancient bronze security hardware be uniformly coated with a smooth greenish patina, acquired during many centuries of corrosion in the ground. This is an attractive, hard, protective layer of a complex copper hydroxycarbonate. In addition to its appealing color, it is to some extent a badge of authenticity. Keep in mind, though, that a high gloss on a green surface is likely due to a protective film of wax. Specimens covered with rough mineral concretions or that have not developed the outer green coating are less desirable. The second preferred coating is the thin reddish cuprous oxide layer that forms next to the metal surface. Complete removal of all corrosion layers is considered a fault, to be avoided if possible. Just doesn't look appropriately old!

For an interesting analysis of the patina of a bolt being attacked by bronze disease, see Medal-Project.

Conservation. Collectors of metal locks and keys, and especially ancient ones, are usually faced with the problem of conservation: that is, preserving them from any further deterioration. It's not quite enough to pay a lot of money for an ancient item and put it into a display case, because you should be aware that unwanted changes can still take place. To preserve not only appearance but also value, articles should be stabilized. For starters, just wash them with water, detergent and a soft brush. It's amazing how many pieces will reach you that are just plain dirty, and need to be cleaned up before that first careful examination.

          Bronzes. Ancient bronze locks and keys recovered from soils often have embedded chlorides, and exposure to air leads to a progressive and catastrophic reaction known as bronze disease. Spots and areas will turn into a fluffy, bright green powder (cuprous chloride). If the bronzes are in contact with others, they can even pass the process to them! If the artifacts have been recently excavated, this may already have started or even develop while they are in in your possession. Examine items carefully when you get them, and if you find any specks of such corrosion, you really need to do something about it. The article will be rapidly pitted and eventually totally destroyed. Treatments for the condition are available. I advise collectors to do an online search on "bronze disease" and select the treatment that seems most reasonable.

I can only describe the process I use myself, which is the benzotriazole (BTA) treatment method. I can't really recommend it for general use, since BTA is hazardous: a suspected carcinogen. If you do decide to use it, it must be handled very carefully, avoiding inhaling it and using disposable plastic (polyethylene) gloves, containers, funnels, bags, etc. BTA is a white powder which can be purchased online. It is dissolved in ethyl alcohol to make a 3% solution. The powdery greenish material is first brushed off the article and it is soaked in this solution for one hour, then washed and dried. The BTA does not remove the corrosion, but bonds with it, forming an impermeable barrier and halting any further activity. Since there's no apparent change in the surface, a record of treatment should be kept. Just a dab of nail polish on the i.d. tag will do it. For a more in-depth description of the method see such sites as

Once stabilized, items should be coated with microcrystalline wax. Renaissance Wax is preferred for this purpose, and may be obtained from many sources. Consult your favorite internet search engine. Renaisssance wax was a development of the British Museum in the 1950s. It is a synthetic blend made from petroleum, and is preferred to commercial natural waxes. It was found that the finer crystal structure forms a tighter barrier to gas diffusion: water, oxygen, pollutants. It is non-yellowing and acid free. Commercial waxes are much cheaper, but this is definitely not a place to economize. When you acquire a genuine piece with a lovely glossy green patina, you can bet that it came from someone's collection, and the gloss comes from a protective wax coating. Waxing deepens the surface color and improves the appearance. It can be removed with mineral spirits (paint thinner) if in the future someone decides he doesn't want it. I've tried this and it does work. If you do your own photography, you should also be aware that waxing will change the image slightly. Whether this is significant is a question only you can answer.

          Iron Artifacts. The greatest enemy of iron is of course rust. What to do without harming the article? I've debated this one with myself for a long while. Removing everything down to the bare metal is an option, but leaves the item with an unnatural appearance, in my opinion. Rusty iron can be cleaned electrolytically, but the apparatus is somewhat complex for the collector, and does remove everything I finally decided to just remove the loose surface rust, which can be done with a wire brush. I use a motor driven brush, or scratch wheel. The result looks good, and no useful information is lost. Afterward, the article should be given a coating of wax. You'll find that the final effect is that which you will have seen on iron artifacts in museums.

Caveat Emptor. "If it seems too good to be true, it probably is". The following observations are my own personal opinions, and I would not care to have to document them in detail. There has been a great amount of fraudulent Roman security hardware produced. I expect that despite care, there will be a few items in this collection that are fakes. I've removed all that are suspect, and will continue to do so. For information on eBay sellers that traffic in fakes, as well as some of their wares, see such web sites as collector-antiquities, ancientartifakes and ancientartifacts.

          Bronze padlocks. Figural "face" padlocks are expensive and suitable targets for fakery. They are relatively complex and labor intensive to fabricate, but some metal workers are willing to do it for a modest return on their efforts. Face lock covers are relative easy to cast, and merit attention. Bronze barrel padlocks with barbed spring mechanisms are extremely rare and expensive and should be considered carefully.

          Iron padlocks.  These aren't all that popular, and therefore expensive. I doubt that any faker in his right mind would bother with these, but never say never.

          Chest and casket locks. Fittings such as lock plates, handles and decorations are not likely to be profitable, and I am not aware of any problems at this time. Lock bolts are so inexpensive that there is no profit to be made. I have only a single example. Figural hasps are relatively easy to make and command good prices. I have found a few suspects. Beware.

          Keys. Pretty easy to cast, and many reproductions of ring keys are being made. For example, see such sites as (Stedmundsbury, Kimasajewelry), for products that are perfectly legal and legitimate. The only potential problem here would be the loss of the word "reproduction" as the items eventually pass from former owners to new. Silver keys especially have an excuse for a missing patina. Figural composite keys of bronze and iron are a problem. The Romans are said to have manufactured composites by inserting the iron forgings into precast bronze handles, and the fakers certainly operate that way. Some with very elaborate handles have recently come on the market. The iron is badly corroded and authentic, but those pretty handles are attached with green plastic adhesive/filler. Such items should be examined carefully with this possibility in mind. There are now also figural keys completely of bronze, best recognized by their complex sculptural features and pale, thin greenish patinas. And no undercoat.

Some general problems to be aware of.

          Exact duplicates. These immediately raise red flags, or should. They would be made in the same work shop, but that may not be so ancient! The point is that in order to justify the labor and expense of making a fake, it's likely that more than one would have to be made. Eventually they will all turn up on some market.

          Patinas. Thin, chalky yellow-greenish patinas should be examined carefully. Bronze can corrode in different ways, depending on its environment. However, as mentioned above, normally there is formed a thin, reddish layer of cuprous oxide next to the metal. A simple thin, transparent wash of green something should be suspect. Such a finish is common on the flood of elaborate, fake figural keys that is now appearing on the market, and is created by chemical dips.

For darker patinas, keep in mind that kits of verdigris paint are items of commerce. For example, see such sources as artmolds.  Very, very green coatings are cause for an alert. The following experience shows what can happen: an item arrived that had a few small patches of bronze disease, not at all unusual. It was put to soak as usual in a solution of BTA in ethyl alcohol, Before long, the solution had turned a brilliant green, and the item had lost its patina! Needless to say, it was returned. This is not a proof that the entire piece was a fake, but makes it very likely, and anyway, such "enhancements" are not ok! Wax removal, if any is present, followed by soaking in ethanol should be considered as a useful screening test.

It should be said, however, that there can be legitimate reasons for replacing or enhancing all or part of the original natural coating. It does happen that concretions and stains decrease the marketability of a piece so drastically that treatment is thought necessary.  I have seen genuine items with paint patches on them that are not, of course, mentioned in the description.

          Adhesives and fillers. Joints of every kind are positions to be closely checked. To simulate corrosion and age, fakers of items are fond of filling with plastic the points of attachment such as

 the ends of handles, hinges and shackles. The figure at left shows some such fillers, after an alcohol soak. They develop a color contrast and start to separate and flake off.

Parting shot.  The fakery industry has become so active and successful that in my opinion, it is best to avoid figural anythings. Figural keys and casket hasps especially should be considered guilty until proved innocent.

Use of this material. Text and images from this site with accession numbers are copyrighted. These may be downloaded for personal use. However, they represent a considerable investment in time and effort, and may not be published elsewhere in any form without permission. No permission will be given for commercial purposes.  Additional information, comments, constructive criticism and corrections are welcome, but keep in mind that I will not open any attachments or click any links unless I know you pretty well. I've made so many statements here that some of them must be wrong, so please let me know which ones they are! 


My address is