Some Aspects of the Metallurgy and Production of European Armor
By Craig Johnson, first published in the Armored Proceedings Symposium Notes, 1999.
The metallurgical study of armour and weapons has taken some major steps forward in the last ten to twenty years. This information has been published in several journals and reports but in a fairly dispersed manner which is difficult for the general interest public to gain access to. I will here attempt to give an overview of where the current state of affairs stands and some ideas on what this will mean for the study of arms and armour in the future.
The analysis of the metal takes three major forms. The first being visual examination under magnification by a trained metallurgist to distinguish the crystalline structure of the finished product which can tell one a great deal about the life of the metal used. The second is a spectroscopy which may indicate certain components in the manufacture of the raw material, if the sources ore can be identified, and alterations to the material in the production of items. Finally, there are the physical properties of the metal, its resistance to penetration, and carbon content.
One of the main issues to arise from the metallurgical examination of period pieces is that steel was used for the construction of armor to a much greater extent than was previously thought. Many scholars up to a couple of decades ago were of the belief that the majority of armour was made from iron and not steel. The fact that steel was the material of choice for the majority of armor greatly affects the work process to make and maintain the armor. Iron was used for armor and weapons but it does not seem to have been the choice for even common objects and was reserved for the lowest quality munitions grade items. When the modern researcher comes across a piece of quality armor made of iron this can be seen as a red flag that the item may be a later reconstruction.
The "steel" used in the age of plate armor, say 1300 to 1650, was quite different from the homogenous refined material in use today. It was a very streaky steel that could vary from wrought iron to medium carbon steel in the same piece and often had a good deal of slag throughout. This was compounded by the fact that most large elements were constructed by forging a heterogeneous bloom, often by folding or layering, to equalize the variables of the material. The production of steel in this period is not fully understood and certain aspects of the processing are still under study, but recent research has uncovered much of what the making of armor entails.
The processes for turning iron ore into a workable material were known throughout Europe, Asia, and the Orient since antiquity. The product that was normally produced was wrought iron and later cast iron. Wrought iron never achieved a fully liquid state. A furnace held in the 1100°C to 1200°C (abt 2000°F-2200°F) range would allow the reduced iron particles of the ore to coalesce into a mass with the majority of the silicates liquefying and draining away as slag. What remained would be a spongy looking mass called a "bloom," the refining of which was accomplished by heating and hammering repeatedly to drive additional slag out and close any voids, resulting in a bar or plate of wrought iron. Iron produced in this way consists of large ferrite crystals and some slag inclusions and would have to be carburized to achieve a steely state. This carburization may have been accomplished in the furnace by lengthening the time the bloom is left in, tempreture increases or increases in the ratio of fuel to ore.
Cast iron was produced when a large furnace was taken to 1550°C (abt 2800°F)or the carbon content of the iron is raised as this will lower the melting tempreture of the iron, it being 1150°C at 2% carbo for example. If the iron reaches a liquid state it will quickly absorb carbon of 2% or more. This material, "cast iron," was then tapped off into a mold and cooled until ready for working.3
The problem was that neither of these materials was the steel the armorers were looking for. The wrought iron would have too little carbon and have to be carburized and cast iron would have too much carbon and have to be decarburized to achieve a steely state.
Carburizing wrought iron is often referred to as case hardening and has been known to metal workers from antiquity.13 The process is to take an iron or low carbon steel bar or plate and pack it in an organic material in a sealed vessel and then bake at high temperature (a red heat) for a long period of time. The organic material supplies carbon, which leaches into the item creating steel over time. One of the best materials for this is charcoal dust. There is another method where the bloom is left to bake in the furnace for a long period of time after it has formed and the carbon monoxide of the fire supplies the carbon to steel the iron. This process is very difficult to control and would produce a product highly variable in carbon content. This process seems to have been used on tools and edged weapons but little evidence exsists for use on armour.14 The first European mention of this process is in the 12th Century.3
An example of period carburizing of iron:
Giambattista della Porta, Natural Magick, Book XIII, Ch. IV, 1558
"Take soft iron armour of small price, and put it into a pot, strewing upon it [soot, and organic powders to supply carbon], cover it, and make a good fire about it: then at the time fit, take the pot with iron pinchers; and striking the pot with a hammer, quench the whole herness red hot in water; for so it becomes hard ... But, lest the rings of a coat of male should be broken, and flie in pieces, there must be strength added to hardness. Workman call it a return. Take it out of the water, shake it up and down in vinegar, that it may be polished and the colour be made perspicuous: than make red hot a plate of iron and lay upon the same: when it shows an ash colour, cast it again into water, and that hardness abated, and it will yield to the stroke more easily: so of a base coat of male, you shall have one that will resist all blows."
This of course illustrtates that the process was understood, but does not prove this was done to mail.
Cast iron needs the reverse process to be usable as a steel product. The material needs to be reduced in carbon content and this is usually accomplished by passing air through a stream of the liquefied iron onto a charcoal hearth, known as a finery, or "Walloon Furnace". There is some conjecture that such a process was being used in the pre Alpine valleys as early as the mid 13th century. By the 15th C. the Northern Italian production centers were casting iron cannon, which may indicate the technology was available in previous centuries to manipulate liquid iron. This also may have been the process used in Styria. The method for decarburisation by forging together cast and wrought iron and allowing carbon diffusion to take place resulting in a steel material is described by Biringuccio in Pirotechnia (1540) though this would be a difficult proces to control. Some have thought this maybe the "Brescisian Process" but this was probably a finery of some sort that was misunderstood by Biringuccio.14 The "Brescisian Process"may have been a source for the large amount of steel needed by the Milanese armorers.6
Such a finery processes would produce very heterogeneous steels, especially when folded and stacked in the early workings of the raw material. This folding and layering gives the armor its characteristic banded appearance under the microscope. If this process had been carried further one would arrive at a nearly homogeneous piece similar to the technique used by Japanese sword smiths.6
The production of raw steel may well have been one of the main limiting factors on the development of armor in the Middle ages. The ability to produce pieces of appropriate size and quality to make plate armor on any type of scale would almost certainly have dictated what was feasible for the armor industry to accomplish as plate armor was developed.14
The attribute that made steel so desirable to the armorer was its ability to be heat treated to create a harder object better suited to resisting the penetration of weapons and missiles. The iron/steel that the armorer had available to work with was very pure in make up. There was no attempt it seems to alloy the material. X-ray analysis of a Helmshmied piece of very high quality falls within "... typical compositions for medieval charcoal-smelted bloomery irons, containing hardly any alloying elements."4
Typical Elements Present
Some munitions armor of the later part of the 16th and early part of the 17th century has a similar makeup but a greater degree of P. This is probably due to the increase of the finery or indirect method of steel production and the use of coal as a heat source for mass production of low end armor. This result may occur as well from the reworking of old armor into more fashionable styles, which records show occurred in the Thirty Years War, for example. Other than this most armors have inclusions of Fe Si (silicate slag ) and smaller proportions of Ca, Al, Mg, and Mn as illustrated above. These are all typical of bloomery slags and exact methodology can not be pinned down as the result would be generally the same.1
The material's ability to achieve hardness could only be improved by the amount of carbon present and the use of heat treating. The carbon content of the steel the armourers had to work with varied from .02 to .8% 6 which equates to low and medium carbon steels today. This would not be consistent throughout the plate. The highest quality material did come close but almost all of the material examined so far exhibits a streaky and inconsistent in nature. This inconsistency would not allow heat treating to achieve a consistent hardness throughout any given piece and would have been the chief challenge to an armorer of the period attempting to produce hard armor on a consistent basis.6 The high quality of raw material, as seen in the case of the court armories in Innsbruck and Greenwich for example, helped heat treating tremendously.
The Business of Making Armor
The production of armor varied in organization from production center to production center. The strength of local governments and guilds, and the supply of raw materials all played important factors in how the work of the armorer was accomplished. The medieval guild system of craftsmen was not set on any Europe wide model, but rather on how the strength of the particular group in each locality was able to control the type and amount of production and how much political control was allowed or won by the guilds. In many areas the overriding force in what was allowed to occur was the local council or government rather than the guilds. These governments were often run by the merchants who bought and sold wares as opposed to those concerned with production.
In fact, the Milan area, the major production center in the early period of plate armor, was noted for its lack of any guild-like control and was run on the basis of many small shops and family operations working together in a fairly fluid environment of short term partnerships and cooperative deals to complete specific orders and projects. This allowed certain families to become quite wealthy and may have contributed to the fact that Milan seems to have provided the seeds, in the form of armorers, to start many of the other major European production centers.7
The Missagalia family is a good example of this fluidity and ability to oversee the entire process of armour making. They not only used subcontractors in armour fabrication but also were involved in mining and smelting as well.14 The profession of armor maker diversified and specialized as it advanced from helmet and mail making, which were the only two recognized distinct crafts in the early 1300's and often belonged to the same Guild. Only in the latter half of the 14th century do we see a distinction and separation between them being made. In Cologne, for example, the Armorer's Guild did not split from the Mail makers until 1399.7
Milan's strength as a power house of armor making in the first half of the plate armor period probably was enhanced by the freedom from Guilds allowing small and large shops to combine in short term contracts to fill specific orders and respond to the demands of the market. This ability created some very rich armorers and also allowed for a tremendous amount of production. In preparation for the Battle of Maclodio, 1427, Milan was able to supply 4000 armors for cavalry and 2000 for infantry within just a few days.7 The difficulty of studying and testing armor from the early period of plate development is the scarcity of examples. The earliest examples date from the last half of the 14th C. with 24 coats of plate from Wisby (1361) and the 1370 harness of the Vogt of Matsch in Churburg. The earliest horse barding to survive is from Italy, 1450 and resides in Vienna City Museum.7
The Milanese armorers were renowned and sought after. In fact, the first armorer known by name was Aramano Rubei, recruited in 1232 to start a mail making operation in the city of Vercello. In the 15th C., 24 Milanese armorers were recruited to France, 11 going to the area of Lyon.7 This cross regional dispersion of armorers was a factor in almost all the major production centers from Augsburg and Innsbruck to Greenwich.6 The movement of these craftsmen would have helped the process of dissemination of technique and practice and the true secret of the master armorer would have been in the details and experience of the individual, especially in the heat treating process.
The Northern European shops were much restricted by the Guild system when compared to the Milanese and Brescian areas. The example of Nuremberg illustrates the restrictive control on an individual armor shop. There, to achieve the title of master armorer one had to indenture for four years as a journeyman and be examined and licensed in each component of armor, i.e.. helmet, gauntlets, and arms. He could then be examined for a new component of armor once a year. Thus, many years of work and experience would have been needed to legally produce a full harness in a Nuremberg shop. Furthermore, each shop was restricted to 1 master, 2 journeyman (4 after 1507) and 1 apprentice. The merchants would also check all product for quality before allowing the Nuremberg mark to be placed on it and allowing export and sale of the item.6 Such tight control sometimes limited the amount that could be made, as in 1362-3 when an order from Emperor Charles IV strained the resources to complete 1,816 armors. This is in marked contrast to the free wheeling environs of Milan, where restrictions on the number of employees were not seen till the very end of the effective plate armor period when sales were slow and the armorers banded together to control the market as best they could to prolong business.7
In Augsburg, to become an apprentice one had to submit a full armor to be judged as a trial piece, then four years of apprenticeship was followed by four years of being a journeyman before one could be judged on mastership.7 There, the Master armorer was not restricted in the number of journeymen he could employ; at one point there were 22 masters and 35 journeymen producing in Augsburg.15 Greenwich, as an example of an unrestricted armor shop, was set up to run with 1 master, 1 clerk, 1 yeoman, 9 hammermen, 3 millmen/polishers, 3 locksmiths, 2 laborers and occasional apprentices.6
Even with the limitations set by the guild and government control large amounts of armour and weapons were produced. As an example in 1295 Frederic the Lombard assembled for the fleet of Philip the Fair in Bruges 2,853 helmets, 6,309 round shields, 4,511 mail shirts, 751 pairs of gauntlets, 1,374 gorgets and 5,067 coats of plates.7 Even with conservative estimates, this equals thousands and thousands of hours in labor and would need an impressive "industrial complex" to produce and deliver such a large order in a efficient way.
Another example is the armor store of Francesco Datini who imported armor to France from Milan. An inventory in 1367 showed 45 bascinets, 3 chape de fer (iron hats), 10 cervelles, 60 breastplates, 20 cuirasses, and 12 mail shirts in stock.7 If this equates to any type of average for an armor seller to have in house, we are contemplating a huge production capacity for the industry of armor making and one that points to wide spread sources of production, not just a few great centers which supplied the entire need for armor.
The production process in the armor shop began with the purchase of iron and steel as raw material, charcoal for fuel, and leather for strapping the armor. In the Greenwich workshop for example a yearly budget of £21 for charcoal, £15 for Steel and Iron and £5 for Buff leather was calculated, while at the same time the Master Armorer received £17, armorer/millman £15, and apprentices £9 a year. Just for this shop that comes to £41 on a yearly basis for materials excluding gilding and such.7 Trying to compare prices is very difficult for the period when armor was being made as there were no set exchange rates, but a rough comparison can be seen on the prices set by the same shop from their books for 1540. A Breastplate £1, Coat of plates £4, Harness for Field £8, Undecorated Garniture £12.6 (a notable exception to the purchasing of raw steel to work is in Milan were the plate makers seem to have prefabbed certain elements at the finery stage and sent the armorers rough shaped elements. In fact, in one case a doctor in Milan was brought to court by the armorers on charges of buying these prefab elements, cleaning them up a bit, and undercutting their prices with inferior product. The Merchants who sat on the council found the doctor not guilty as it was good for business.)15
The vast majority of armor produced throughout the period of armor making would have to have been for "off the rack" purchase. Those who ordered a fitted suit would pay a premium. Comparing the prices listed above for Greenwich armors, which were probably fitted, with the price paid by Henry VIII when purchasing 1200 full harness in Cologne for £451, or a little over 7s. a harness in 1539, one can see the vast range armor prices covered.
To take the measurements of a person armorers were known to travel great distances, have special clothing fitted to the customer to use as a model, travel with a semi-completed suit for a trial fitting and, in the case of a Spanish monarch, have wax copies of his limbs made and sent to his favorite Augsburg armorer.7
The metallurgical evidence can tell us a lot about how the armor was actually worked in the shop. The elongated slag inclusions, seen in most samples of armor tested, indicate that the main portion of forming the individual elements of armor was done as hot work. The undistorted grain structure of these pieces also indicates that a minimum of cold work was done on the armor.2 The work which was done cold, as illustrated by many period depictions, would have been detailing and sharpening of decorative elements such as flutes. The hammermen and smiths would construct the different parts and fit the pieces to one another. This probably included some test fitting and partial assembly.
The items would then move on to the polishers, who would take the rough and black pieces and grind them by hand or wheel to a polished surface. This would need to be done after the heat treating in some cases, or at least touched up after the heat treating as that would discolor the metal. The armor would then be assembled if it was to be bright. If it was to be painted, gilded, etched or blued that process would need to be done before final assembly. Most commentaries which have addressed the topic of the armor process in the shop say little about who actually did the assembly. It has probably always been assumed that it was the armorers themselves. However, there are some records that indicate the polishers were the ones who actually finished putting the armor together.15 This must have been a significant task, as anyone who has attempted to assemble any portion of an authentic armor will testify that it does not lend itself to assembly line production. If the lames of a knee cop are assembled in the wrong order, which is easy to do as they all look relatively the same, the knee will not work. The fact that the polisher in many cases was a contracted separate shop and not part of the armory would add to the difficulty of keeping the right elements together and in the right order. This would have been less of a problem in shops like Greenwich or Innsbruck were the funding of Royal workshops allowed them to have all the trades on "staff".
The armors would be strapped out by individuals who specialized in this activity, 'finishers',7 though there are examples where this was done by the polishers as well. They were often connected to the leather workers and goldsmiths guilds as opposed to the armorers guild.15 The armorers would buy hinges and hooks premade from locksmiths for all but the most expensive suits. Those that would be done with custom fittings would probably be done in shops where locksmiths were part of the production team, again, as in Greenwich.7
Heat Treating Armor
The amount of knowledge we have gained on the physical aspects of armor and its heat treating over the past two decades has drastically changed our concept of what armor is and how it was made. Steel was the material of choice, not iron. Steel allows the armorer to heat treat his work to maximize performance and this seems to be one of the main attributes looked for in exceptional armors.
Carburisation, when done, seems to be an element of the steel production process and not something the armorer would be doing, though there are examples, such as the mail hardening process quoted previously, that indicate some carburisation was done after fabrication. It is surmised that possibly carburizing armor was used early in the period of plate armor but abandoned later. There are no clear cut examples where a plate armor item can be identified as having been case hardened after fabrication6,14, but a few where it may be a possibility, the Pembridge Helm being one.2 This practice, however, does not seem to be part of the armorers' repertoire on a normal basis and their shops' focus would be the fabrication of the pieces from stock material and the heat treating and decoration of the same.
To harden steel, one needs to bring it above a critical temperature (usually a bright orange heat) for a certain amount of time and then cool it sufficiently fast, or quench it, to lock the crystalline structure in a form not usually created by a slower cooling process. Pure or wrought iron, a Ferrite crystalline structure, will not harden when quenched, and the majority of armor tested up to now have some form of heat treatment done to them. This indicates that the armorers tried to maximize their materials' qualities and it was the inconsistency of material and difficulty in controlling the process that separated the so-so from the masterful in creating hardened armors.
There are a couple of different methods one can use to attempt a quench. The full quench is where the steel is quickly cooled, creating a transformation to a 'Martensite structure' as fully as possible. This results in the steel being as hard as it can be and usually requires a quench in water for the materials it to achieve this state. This Martensite structure is a solid solution of carbon in iron which possess a distorted crystalline structure, this needs to be tempered to relieve stress and reduce some of the hardness by allowing some carbon to come out of the solution as iron carbide. A piece fully hardened could be tempered by heating to about 150°C to 260°C (300°F-500°F), to reduce brittleness and avoid cracking and breaks under use, by "relaxing" the martensite structure. The slack quench uses a slower quenching medium than water, such as brine, oil, boiling water, or some combination of materials which slows the cooling and results in a combined structure of Martensite and other constituents such as Bainite and Pearlite. Materials treated in this way will be harder than an air cooled item but not in need of a temper for performance as armor or a tool. This is the method used by the majority of armor makers. Interrupted, or timed quench is also a technique that was sometimes used. The item would be withdrawn from the quench medium before being fully cooled and the internal heat left in the item would temper the piece to some degree. The interrupted and delayed quenches are also called "slack quenching" today.
The detailing of this process and of the refining process discussed at the beginning, are just modern rationalizations of a completely experience based activity that was understood in only the barest physical scientific terms as we think of them today.2 The armorer working in a shop without consistent time keeping methods and accurate temperature control would need all the empirical knowledge and experience one could muster to create hardened armor on a regular basis. The raw steel varied in carbon content greatly, and if a high quality supply of consistent material was available it went a long way to creating a superior product, as in the example of Innsbruck.
The results of heat treating ran the full spectrum of achievable results. Many munitions armors were iron or unhardened steel, while even some of the early examples of plate armor pieces were achieving 75-430 VPH (less than 6-44 Rc). The Pembridge helm (bef. 1375) testing at 430 VPH on the surface and the Küssnach Coat of Plates(c. 1352) Item No. LM 13367 in the Swiss National Museum, Zürich averaging 390 VPH (about 40 Rc). The Braybrook Helm (bef. 1405)Royal Armories No. AL.30 with less than .1% carbon and left without heat treating, averages 108 VPH.2
Later in the 15th and 16th C. armorers achieved more consistent results, such as the Helmshmied family of armorers in Augsburg who between c.1480 and 1551 averaged 240-441 VPH (20.3-44 Rc) on 17 items of their work sampled.4 Lorenz Helmschmied created one of the most consistent and well hardened pieces yet tested in an Armet (c. 1492), No. 66 Churburg which averaged 525 VPH (abt. 50 Rc).3
One of the challenges yet to be fully understood is the deciphering of the physical techniques the armorer used to heat and quench the armor. As mentioned, consistent time and temperature control, while vital to such a process, would be a challenge in a period workshop. Furthermore, large pieces such as breastplates and complex elements such as greaves would have been prone to warpage in the quench. One theory is that iron braces were used to hold items to shape as they were quenched.8
One of the most interesting aspects of the development of hardened armor is that as it reached the point where armorers were able to consistently produce the items they begin to reduce and abandon the process. The North Italian makers stopped hardening armor regularly fairly suddenly around 1500-151014 and the Southern German production areas tapered off at the turn of the 17th C. The changing requirements of armor and the tactics it faced were probably the chief reasons for such a development.1 Another factor was the increased use of decorated armor -etching, gilding and bluing. These processes often involved heat and by the 1570's the German and Greenwich workshops were combining these processes with the tempering. It is likely that fashion won over hardness in Italy and makers left the armors soft to eliminate a step that might be nullified by over heating in these decorative processes.6
There are some interesting issues which this knowledge of the armorers' craft create for the researcher. How was the volume of steel needed to support the armour industry created with the technology available to them? The production capacity of the armor centers does not seem to be substantial enough to produce the numbers of armors which records indicate were ordered and delivered. How resistant were the hardened steel armors to contemporary weapons? How much reuse of out of date armor was part of the production process, especially for munitions armor?
Notes on Weapons
There has been very little research into weapons of the same period and hopefully the next several years will see more efforts in this area. At this time some specific examples can be commented on but it would be difficult to extrapolate this to the whole as the sample is so small.
There has been one pole weapon examined, a 15th C. bill in the form usually described as Italian. The item was of no particular historical significance so the decision was made to bisect the piece in 8 places to learn as much as possible from the item. The construction method was a series of folding to create the form, but there does not seem to be any particular attempt to maximize any hardness of cutting areas or spikes. The carbon content was high enough to harden the item if so desired by the maker, but there is no evidence it was heat treated in any way to increase hardness and it was obviously not a goal of the smith. The cutting edges and top spike of the item are completely unhardened, registering below 8 on the Rockwell C scale.5
In the case of medieval and renaissance swords a little more is known. At the present time there have been about 20 such pieces tested and published, with the vast majority being constructed in a piled fashion and the edges being carburized and heat treated.9,12 The earlier Nordic pattern welded blades have actually been studied and tested far more extensively than their medieval counterparts. What has become clear is that the blade construction did not go from pattern welded blades to single homogeneous piece construction, but from pattern welded to piled construction using varying grades of carbon content pieces to achieve a hardenable sword edge.9 In fact there seems to have been an intermediate period, 9th to 11th C., where the pattern welding was a desirable decorative technique as sheets of the design were laminated to the surfaces of iron or piled blades.10 Many of the early swords, in fact some swords well into the 15th century, would probably not have had less flex than is normally thought. Instead, their soft noncarburized or low carbon cores would have been more prone to bending than flexing.
One sword tested had three distinct bands in carbon content ranging from .1% to .8% with the steely parts having achieved hardnesses in the 250-330 VPH (20-32 Rc) range.10 In eight swords sampled from the 11th to the 15th C., five were case carburized iron bars, two were iron/steel composite structures, and one was several pieces of steel welded together. The heat treating method included five blades slack quenched & tempered, two timed quenched, and one left untreated (an "Ulfbert" sword with an exceptionally high carbon content, which seems to have been made from crucible steel perhaps originating in the Middle East14).9
In the16th C. a different method of production is seen in some swords, using the technique of forging together layers or folding material to create a more homogeneous steel product. Of four 16th C. swords tested, two were constructed in this new manner; one was constructed by wrapping an iron core with a steel skin and heat treating; and the last was constructed in a piled configuration. The four swords were tested for hardness and fell in the range of 325-480 VPH (about 32-47 Rc). The sword with the 480 VPH was the steel wrapped iron core and the core hardness was 147 VPH (below 0 on the Rc scale, abt 78 on the Rb scale).12
The indications from these few examples is that the production of sword blades was a varied activity with several different methods being practiced and quite a range of results achieved. There also would seem to be a correlation in the development of better steels for armor and swords occurring at approximately the same times. Hopefully new research into the development of steel fabrication in the period under discussion will lead to a more detailed perspective on both weapons and armor fabrication and allow us to appreciate these items fully.
All hardness measurements are given in Vickers hardness scale (VPH) with their approximate Rockwell C (Rc) equivalents as listed in Hardening, Tempering, & Heat Treatment by Tubal Cain, Argus Books, England.
1. Williams, Dr. Alan R., Slag Inclusions in Armour Plate (1400-1640), Bloomery Ironmaking During 2000 Years. Seminar in Budalen, Norway, 1991.
2. Williams, Dr. Alan R., Four Helms of the 14th Century, JAAS, 198, Vol. 10, No 3, p 80-102.
3. Williams, Dr. Alan R., Fifteenth Century Armour from Churburg- a metallurgical study, Armi Antiche (Torino, 1986) 13,3.
4. Williams, Dr. Alan R., Augsburg Craftsmen and the Metallurgy of Innsbruck Armour, JAAS, Vol. XIV, No 3.
5. Williams, Dr. Alan R. & J. G. O'hara, The Technology of a 16th Century Staff Weapon, JAAS Vol. IX, No 5.
6. Williams, Dr. Alan R., & Anthony de Reuck, The Royal Armoury at Greenwich 1515-1649, A history of its Technology, Royal Armouries, Monograph 4, 1995, London.