The most abundant items of Roman security hardware seem to be keys and lock bolts. Keys were used mainly for doors, chests, boxes, caskets, cupboards and padlocks. Less often they were used for ceremonial or decorative purposes, such as matron keys, jewelry items and votive offerings. 

It is alleged that some warded keys were worn by women as symbols of household authority, as "keeper of the keys". If true, such would be difficult to identify as having served that purpose.

Ornamental items would be represented by keys 394 and 486, where they are attached to hair pins. The following passage (LEG III AVG) describes another usage: "rings for fingers and thumbs were worn by men and women, sometimes several at a time.  They were made from gold, silver, bronze, iron,  lead or glass." A working key made of lead is so absurd that it guarantees some non-functional use. I can do no better than reprint a quote from Westropp; "Lead rings, set with intagli, of early date and good work are sometimes to be met with, but they are exceedingly rare.  It is evident that these leaden rings, mr. King writes, in their time passed for massy gold, a deception favored by their weight and ductility, and not to be easily detected when encased in a thick envelope of gold leaf, of which they often retain the trace". Another reference to the finding of a lead key is in Curlesnewstead, Chpter 15. The examples in this collection show no trace of gilding.

A more likely use for a lead key is as a votive object. It is known that many ordinary articles were reproduced in lead for use as votive offerings, usually in miniature. These include axes, mirrors,  plaques, furniture, amphora, figurines, cups, etc.  I suggest that ring key 336, rotary long key 472 and pin tumbler long key 513 were made for this purpose. As remarked in the introduction, there was no lack of gods to whom one might appeal for safety and security.  

Door locks were sometimes made of wood, usually with iron keys, but sometimes of bronze. Such locks have not survived, but keys have been found. The latch lifter was a primitive key, a very simple metal, bone or ivory shaft with a hook or a couple of teeth on the end. In use, it was to pass through a hole in a door and fit into a latch to lift and move it. The

Romans did not need to invent the latch lifter; it was used by other early peoples. Key 372 is a Celtic example. Such keys were not really intended for serious security, but more as a convenience in lifting the inside bolt. A variety of latch lifters was found at Vindolanda.

Key 200 In addition to iron, some bimetallic keys have been found, with bronze handles cast on heavy iron shafts. Figural lion bronze/iron door keys are especially rare and desirable; for example, see the key below from Augusta Raurica (Riha) labeled Basel Lion. This is the finest example I've ever seen. My own copy is nice, but lacks the elaborate, richly detailed modeling of this key. Below the Basel Lion are two more superb door keys that we can only admire, since they are in museums. The first is a horse head key from the Getty (flickr) and the second a molossian hound head key excavated from Mevaniola (archeobo).

Key 299 This was obviously a custom made assembly, for someone who knew exactly what he wanted. In fact, it was probably made by an in-house smith, as in an isolated villa. Villas in Britain, for example, tended to be completely self-sufficient. Both door key and ring key are iron.

Key 300 Fairly typical iron door key. They tended to be long and heavy, and many survived in spite of centuries of rusting away.

Key 315 Another very nice composite which has survived very well, considering.

Key 366 A key for double security! This is the key, or rather latch lifter, for a primitive type of wooden lock, inserted through the door, rotated and drawn into the bolt. The knife on the other end was of some comfort to the householder returning home through the dangerous streets of the Roman city at night!

Key 378  Quite similar to 366, but with the more conventional loop for hanging.

Key 368  Another elegant composite.  It almost seems that while appreciating the appearance of bronze, the Romans distrusted its strength for heavy duty door operation.

Key 376  Drilled rotary key with concretions

Key 379  Huge, massive double ended iron military key. Looks like a key, anyway, but I don't know why it was useful to have two ends, since they appear to be substantially the same.  Possibly it also had some secondary use such as a heavy duty fastener, carrier or grappling hook?

Key 381  This heavy bronze key shows considerable shrinkage, caused by improper gating in the casting process.

Key 382  Very heavy and strong, likely for a door but perhaps for an extremely massive chest.  Excellent condition for iron.

Key 385  Iron with two parallel rows of three teeth, which is quite a lot for this type

Key 393  Not completely sure of the use for this; alleged to be military and could be for a padlock.  That side notch looks like it might be for spring compression and the slots for wards.

Key 397  Unusual design, the only one I've seen with a second row of teeth at a right angle.

Key 434 An unusual specimen in several ways. The bitting is perpendicular to the plane of the shaft, so that a little notch had to be added to enable the neck to be passed through the usual narrow slot. Also, the ring has a recessed area around it on one side, for no obvious purpose. There has been a little decoration added in the form of cast-in lines and a depression, scooped out of the wax model with a special tool. All this is a reminder of the great variability in Roman keys, which cautions us against making generalizations. Never say never or always, because every Roman locksmith was evidently pretty much free to invent and carry out his own designs.

Key 495  Unusual and crude. May at one time have had a hanging loop

Key 522  After removing loose rust and dirt from this one, what emerged was a type I've never seen before, a figural key in iron. It's so blurred by corrosion that all I can tell is that it's the figure of a person with arms folded. It would have taken a great deal of finishing work, and must have been expensive.

Door Keys

Long Keys, Pin Tumbler     

The purposes of many keys are not clear. The term "long" is merely a convenience and does not classify them by usage.  It is likely that some of the keys in this section are for doors and some for chests or cupboards.

Key 303 This is an interesting design, but could not be carried securely. Looks like it was meant to be hung up somewhere.

Key 314 This is one of my favorites: heavy, chunky, geometric.

Keys 213, 260, 261, 275  The iron key section, part 1 We can't even guess the original ratio of iron to bronze keys that were produced, but enough iron keys have survived to show that they were in common use. As a class, they seem much less ornate, and I suspect that they were cheaper. Functional, but not much prestige here!

Keys 202,214,259,321 Iron keys, part 2

Key 325 So far I haven't been able to think of a reason for the strange design of this one. It is unusually thick, and has a section of iron about 0.1" thick sandwiched between bronze top and bottom. No doubt the thickness of the shank is to allow it to be poured around an iron core without cracking, buy why should the bit section be continued in bronze?

Key 328  There are some unusual things about this one. For one, it's a folding key, first of this size I've seen. Also, the bits are opposite to those of all the bit keys shown here. It's alleged to be Byzantine, maybe that has something to do with it.

Keys 329,331 The Romans, like most everyone else up until very recent times, thought that everyday objects should be enhanced with as much decoration as could be afforded, or function would allow. We still haven't entirely abandoned that feeling, have we? The designs of Roman artifacts often included religious and architectural elements; remember the temple-furnace shown in the introduction? Keys 329 and 331 manage to work part of a temple facade into the design! I've seen keys in which this temple motif was carried out very elaborately. Figural Roman objects are very desirable, and command premium prices. On the other hand, see key 433 below which also has an openwork shaft.

Key 337 This is my smallest long format pin tumbler key, 1.460". It's unusual in that the pins were made by folding over the material, either the wax model or the bronze casting. Hard to say which, but I think it was in the wax. No idea why the fabricator would do such a thing. Clever, but I note that the bow is not very well done.

Key 342 Undoubtedly Roman, but a rotary key made to fit plate wards. An unusual configuration for an unusual lock, in my experience.

Key 343 Large, functional iron key, remarkable mostly for the curious flattened area along the shank. It was just smashed flat during fabrication, but I've no idea why this was done.

Key 350 Unusually large rotary key, with intact iron. Typical bow design for large bimetallic keys.

Key 360 Delightful figural with a ram head, a battering ram on a very tiny scale.

Keys 367, 371  Noticed that these have a similar decoration, cast-in bevels on either side of the shank. Otherwise, simply very nice examples of not-quite ring keys.  Wearing them on fingers would be uncomfortable.

Key 370  This is a type of long-format key that is not rare, but not so common either.  It is iron and strictly utilitarian. The pins are made separately and pressed into holes punched into the key blank. One protrudes a bit from the back, as I tried to show in figure 370B. I haven't seen any bolt that such a key might fit.

Key 398  Examination shows that it was bent in antiquity, and carefully done, at something close to a right angle, but why?  It must have been very awkward to use, and designed to move a bolt deeply concealed in its container.

Key 403 Rotary barrel key with a unique feature. It was found to have a bronze sleeve in the bow. This was done in antiquity and became visible only upon cleaning. I do not know how or even why it was made this way, and have never seen any other example.

Key 405 This rotary barrel key has an unusual design. Upon cleaning, the bit was found to be of bronze! It was covered on both sides with a thick layer of magnetic scale. This was not further removed because of the fragile attachment of the bit to the shank. The bronze core is barely visible in 405b and 405c. I have no idea why there was such a thick adherent coating on the outside of the bronze. It almost looks like there was an outer iron coating, which makes no sense to me.

Key 409 Another bent key, this one with about a 30 degree angle. The condition of the surface shows that the bend was made in antiquity.

Key 410 This was obviously made to operate a large and heavy lock, but even so was clearly much thicker and stronger than it needed to be! Perhaps it was a prestige item in some way; it alone in this group has a decoration: a little projection on each side. Decorations were usually reserved for bronze keys.

Key 427 This is a remarkable production. The Romans made many iron keys, but a key shape as intricate and precise as this one looks extremely difficult to make by forging and finishing with hand tools. As for casting, the Romans did not attain the temperatures required for the melting of iron, except occasionally by accident. When that happened, they actually threw the product away as useless! (Collingwood). This specimen is extremely well preserved, although slightly over cleaned.

Key 433 This key would seem to say that openwork shafts are just whimsies of the locksmiths who designed them. The divided shaft is actually an outline of the more familiar solid shaft that flares toward the bottom. Coincidentally, the key outline is in the shape of a modern keyhole

.Key 434 An unusual specimen in several ways. The bitting is perpendicular to the plane of the shaft, so that a little notch had to be added to enable the neck to be passed through the usual narrow slot. Also, the ring has a recessed area around it on one side, for no obvious purpose. There has been a little decoration added in the form of cast-in lines and a depression, scooped out of the wax model with a special tool. All this is a reminder of the great variability in Roman keys, which cautions us against making generalizations. Never say never or always, because Roman locksmith were evidently pretty much free to invent and carry out their own designs.

Key 439  The pins look like they were made to operate a pretty heavy bolt. But it looks like the bolt had finally got stuck in the open position, and the key was bent trying to move it forward.

Key 466  Sometimes it's hard be sure what a key could have been used for. This is an example of a door type key that's really too small for a wooden door lock, and awkwardly shaped for a pin tumbler lock. I'd have to guess it was for a cabinet lock.

Key 488  Composite bronze/iron key, showing the same splayed end bits as key 397.  Don't know how this feature worked.

Key 498  Another composite bronze/iron key. Nice, but how did it work? How do you lift pins with big dimples? I'm missing something here.

Key 505  Iron, made in a thick shape more typical of bronze keys.

Key 510  Composite figural key, iron and bronze, showing a man's face in the handle.

Key 513  Lead key. I think this was a votive key, certainly not a usable article. See introduction for discussion, and see keys 336 and 472 for other examples of lead keys.

Key 516  Unusually long pins on this one

Key 517

Long Keys, Pin Tumbler

Rotary keys  Locks using rotary keys were first developed by the Romans. The keys are instantly recognizable to us; they resemble those in common use until the last century. I'm reproducing (a little later with the rotary key images) a drawing from Biasiotti that is the clearest I've seen of the functioning of a Roman lock operated by a rotary key. Rotary keys were also usually warded. They were contemporaneous with pin tumbler locks and keys, as shown by this example from Pompeii at left (Gilman).  All rotary keys found at Pompeii have hollow stems, turning on a key post (archive).

Key 225 Despite the iron stains, this is a bronze key

Key 228 Many keys are found with this little nib on top. Zara suggests that the purpose of this is to push or turn away a keyhole guard.

Key 224 The "shepherd's crook" design. This one is bronze, but the design is also found in iron keys

Key 306 In addition to its size, this one is remarkable in that it has a carefully made chain attached to a leaf shaped plate, somewhat like a key tag. This is just like the tag shown before in chest lock group 4940.

Keys 326, 354 Stylish little keys with concave bow loops and barrel ornamentation. They're virtually identical except for a slight difference in size. Alleged to be Byzantine.

Key 327  Byzantine

Key 356 I guess this delightful little figural qualifies as a cartoon. I've seen only one other like it.

Key 359 Extraordinarily chunky and heavy, built to last!

Key 365 I found this key interesting because the bitting is so remarkably similar to that of keys made in very recent times. I'm sure there are still locks around that could be opened by this key.

Key 386  Roman keys were developed with very complex shapes to operate high end locks.  Key 386 is an example where the shank detours around a ward before ending in a perpendicular set of bits. Beside it is shown a better example from the French Ministry of Culture site. Next are four examples of an even more elaborate design, where there are two sets of bits set at angles. These types are extremely rare.

Key 422  Note that this key is a near duplicate to that shown by Ward, figure 3, key E, aside from the exact placement of the slots.

Key 426  An iron rotary key, very much in the modern style, with a comma shape bit and a cut in it to pass a ward. So far I haven't come across a lock plate or escutcheon made specifically to fit a key of this shape.

Key 442  Most such keys are flat. Seems a bit illogical to make the bow and shank perpendicular to the bit. It automatically makes the key more bulky.

Key 448  Large and pretty elaborate, almost too large for a chest, perhaps a door key. With keys having such elaborate bitting, one always wonders if it's not partly for show, and the wards are not really so complex.

Key 469  Raw casting of a rotary key, still attached to its gate. Alleged to be from 100 CE. The other key of the pair was broken off.  Don't have much comment on it; speaks for itself. It was a very, very poor casting, all right, but why wasn't it melted down? Bronze scrap had value.

Key 472  Here's another lead key!  I found faint traces of cast-in decoration at one spot.

Key 499  A big, heavy specimen, perhaps a door key.. Allegedly found in Britain

Key 503  At 1 inch long this must have been a casket key for sure.

Key 518  Iron key with hollow shank

Key 528  Bronze with two uncommon features: flattened for easier wearing & handling and designed to be worn as a pendant

Ring keys are the most abundant and popular type that are collected today. We have been told often enough that Roman ring keys were worn on the finger. It could certainly have been done in most cases, but I think that the ring was primarily a device for hanging. Sometime I must try wearing a ring key for a day. If worn so that the shank and bits were on the outside of the finger, I suspect it would catch on everything and flop around something fierce! If worn on the inside of the finger, I can picture two citizens shaking hands and locking keys!  Or did they shake hands in those days? No, but they did clasp hands on certain important occasions, as suggested by the ring shown at left.  Many ring keys are known which are too small for any human finger. I believe that most often, ring keys were hung on thongs around the neck or suspended from a girdle. There is physical evidence (muzarp) that keys were sometimes attached to a belt, and even a key that was worn on a bronze armlet (Cook)

It would be interesting to know how long the use of the pin tumbler key persisted. There's no doubt much information on this question that I have not yet found.

Many ring keys advertised for sale are listed as "wearable" with a modern ring size given. However, if you should do this, you should be aware that the patina will soon wear off, decreasing the attractiveness and value of the piece. Lots of Roman rings can be found that are quite wearable and do not have this problem.

My classification of keys into "ring" and "long" is pretty arbitrary. From the sequence shown here, it's clear that the length of the shank varies a lot, and certainly as keys become longer, the likelihood of wearing one on a finger becomes smaller.

There may be an organizational scheme for Roman keys somewhere that I haven't yet found. Therefore, to bring some sort of order into the chaos of ring keys, I've arranged my own in the following categories:

1. Normal tumbler ring, shank in the same plane as the ring

2. Shank up tumbler ring, shank perpendicular to the plane of the ring

3. Rotary ring

4. Folding ring

5. Warded ring

This is a selection of the ordinary or "standard" ring keys that are available at this time. There may be duplicates somewhere in the world, but each of these is an individual, differing slightly from all the others. Comparison of such keys with bolts that have been found shows that not all the decorative cuts and slots in the bits were used in the mechanism. They were partly for display and prestige.

Keys 309, 388 Despite their nondescript appearance, these are rare finds, bimetallic ring keys. 309 is brass/iron and 388 is silver/iron. Much of the iron has been corroded away, but we can see the joints where the metal was poured into the iron bit section. If there were one available in very poor condition, it would be interesting and worth while to section it to determine exactly how the join was made. Nondestructive testing would be preferred, of course. However, I really think key rings that display only stubs where the bits should be are just bimetallics with the iron all gone. But how did they shape that little well in the iron into which the brass or silver was poured?

Key 392  Another composite bronze/iron key, with unusually well preserved iron part, and a nice circular pin design as well.

Key 335 Another iron key, exceptionally well preserved

Key 336 This one is made of lead and is a rare specimen! The bitting has been broken off, apparently during excavation. There is cracking and splitting and many pits filled with lead oxide, but the survival is pretty good.

These seem to have been less popular, judging from the number now available. Why? If they were actually worn on a finger, they would be much less awkward.

Key 281 Very tiny, gold plated. No idea why such a thing would be made!

Key 396  This was represented as a child's ring key because of the small diameter, 0.565".

Key 504  Iron

Key 302 Very attractive specimen, with a faceted band.

Key 313 Decorated with cast-in indentations.

Keys 201,211,232,250,292,293,296,365

Folding ring

All the Byzantine ring keys shown from the Menil collection (Vikan, 1980) are folding. However, not all folding ring keys are Byzantine. Examples have been found at Pompeii, including that by Price shown at left.

I must agree with most commentators that the purpose of folding was convenience in wearing on a finger, if the owner elected to do so. However, I personally feel that a hinged ring flopping around on one's finger would be a considerable inconvenience, and I don't think the question is settled yet. All folding keys are for rotary mechanisms, and all but 247 are also barrel keys.

The connection of shaft to ring was made by simply crimping the pre-cast ends of the shaft over the ring. Usually the seam is still visible. In some cases, such as key 418, the ends don't even meet. Crimping was probably done hot.

Keys 230, 247  Two of the most elaborate ring keys I have are also folding. This is in accord with the Byzantine style.

Key 230 origin not stated.

Key 247 is from "the Balkans"

Key 253 is iron and quite plain. Alleged from Syria

Key 266 also alleged to have been found in Syria

Key 338 is also iron, but sufficiently corroded that it will never fold again. Origin not stated

Key 347 is ideally designed for suspension and impossible to wear as a ring. There would have been nothing to gain by the folding feature and I think in this case it was fashionable rather than useful. I suggest that since such keys were normally folding at the time, this one was made that way by custom rather than for usefulness.

Keys 353, 377  Identical keys of silver & bronze, respectively. Ornamental, and I think unlikely to be useful.

Key 353 is from Lincolnshire, which would make a Byzantine origin unlikely, though not impossible. The outer ring surface is octagonal. Iron.

Key 390   This is really getting a bit too large to wear on a finger, I think

Key 418  The ends of the shaft are marked to simulate a hand with fingers grasping the ring. Remarkably, there are no shoulders on this ring to keep the rest of the key from sliding around it.

Warded ring keys are supremely attractive to collectors because of the variety of intricate designs. Here again, I suspect that these were to some degree prestige items, and that not all that maze-like metal area was functional. They have been called "matron keys", but I'm not yet convinced. (see W. Jones and S. Bury)

Key 233 This is the most ornate of warded ring keys that I've seen, and the only one to bear an inscription (inverted in this image, I think). I'm embarrassed to admit that I cannot translate it Anyone?

Key 232 I expect little Marcus got spanked for stepping on this one!

Keys 304, 308 Don't know quite what to make of these, very delicate.

Keys 267, 333 Of course one is tempted to think of these as Christian symbol rings, but they're attractive just as designs.

Keys 345, 346 A little more decoration than normal

Key 364 Very odd design; I think made for decoration rather than function.

Key 419  Very strange key. I suppose it to be for a warded mechanism, which I'd like to see!

Key 438  This key is so delicate that I'm convinced it was meant for an ornament, not actual use in a lock.

Normal tumbler ring keys

Shank up tumbler ring keys

Padlock Slide Keys     

Slide keys were used to compress the springs in padlocks with long cases and barbed spring mechanisms. Those in this collection are of iron. Iron slide keys are surprisingly rare, no doubt because of corrosion in the ground. They are entirely functional: undecorated and crudely made, with a simple form that is little different from those used in modern times.

Key 361 As has been pointed out earlier in the discussion of padlocks, this key is eminently suited for opening a lock with a keyhole and open slot in the top. The right angle is necessary because a vertical key shaft would be blocked by the shackle.

Key 362 Made for a keyhole at the rear end of the lock. This was forged from a strip of iron, showing evidence of folding over at the suspension hole and in the front section of the shaft. The I-shaped front end was more carefully made, probably by filing.

Key 373  This was purchased as a "Roman tool", and so it is, a tool for opening a barbed spring padlock.  The lock must have been unusual though, very long and narrow with a relatively small spring assembly.

Keys 384, 391  The shapes of these keys show that there were wards to be bypassed.

Key 450  Interesting example of the variability in workmanship in making the same product. This is much like 362, except for the straight shank and the more careful finishing work in folding and forging to make the bow.

Key 473  The most elaborate example found so far, showing a complex system of wards.

Key 482   Curious iron slide key with a thick round shank and delicate end pieces to compress the springs.  Those slender end fittings are like the bronze Viking key 425, made some 800 years later. It's probably not a coincidence, since the Vikings are said to have borrowed from Roman designs.

Key 496  Complex keys of this type are sometimes sold as Roman or Byzantine. They are in fact for Tibetan slide key padlocks.

Key 512  For a two-spring side by side arrangement. The smith cared about this one to give it a little decoration of scored lines, which is unusual for this type of key.

Key 519  Flattened grip and obviously for a bottom key slide

Key 527  This was alleged to be Roman, but I'm not at all sure of its origin. At first glance, it does resemble all the padlock iron slide keys shown here. However, it is actually for a screw key lock, and a large one judging from the depth and sturdiness of the threads. It has an interrupted thread as shown in 527 B, allowing it to pass a pair of wards. I don't recall seeing that particular feature before in a screw key. As far as I know, neither the Romans nor Byzantines had screw key locks. Wikipedia comments that locks with such a mechanism were "contemporary with smokehouse padlocks and originating in the Slavic areas of Europe."  If true, that makes them modern.