Proceedings 74        8 June 2002        Oxford

European Gunpowder 1250-1600: from Oriental Curiosity to Critical War Material: a paper by Dr. Bert Hall, of The Institute for the History and Philosophy of Science and Technology, University of Toronto, Canada.    

———————    

As every schoolboy knows, gunpowder is made of three things: saltpetre, sulphur, and powdered charcoal. Six parts of saltpetre and one part each of sulphur and charcoal will make the best proportions.Fortunately for the health and bodily safety of schoolboys, simply stirring the three powders together in a mixing bowl will make a gunpowder that is very hard to light, and once lit burns with a sputter and fizzle rather than exploding. Making good gunpowder is rather more difficult than it seems.

Like our schoolboys, once medieval Europe learnt abut gunpowder from China in the middle of the thirteenth century, it found that good gunpowder was rather harder to make than it seemed. How did Europeans learn to make their own gunpowder? How did they learn to make it burn with a bang rather than a fizzle? How did they learn to gather and prepare the critical raw materials from indigenous sources? Usually, historians ask questions about origins or effects. This talk is about the development  of a technology from the mid-thirteenth century to the mid-sixteenth.

Thanks to the work of the late Joseph Needham and his colleagues at Cambridge University we now know for certain three things about gunpowder that affect how it appeared and developed in both east and west.

First, gunpowder is not extremely old in China. The earliest written formula for gunpowder dates from 1044 A.D., during the Song Dynasty. This is a nice date, in that it makes gunpowder's mid-thirteenth-century appearance in both Islam and Europe more readily understandable. We do not know just how gunpowder came westward, but the time scale no longer seems unreasonable.

Second, Needham shows that gunpowder was a by-product of Chinese alchemy and its concern with nitrate compounds. Nitrates had been known and partly purified in China for hundreds of years before the Song, from the Han at least. When gunpowder was invented, it came from a body of very old knowledge. 

Finally, despite myths to the contrary, the Chinese appear to have adapted gunpowder to  military purposes and to have invented a number of new weapons: incendiary arrows, "fire lances" (primitive flame-throwers), percussion and incendiary bombs, rockets and cannon of different sizes. Most of these were imitated in the West.

This means that gunpowder came to the West in the thirteenth century under the rubric of a  "modern" military technology, in a manner very much like twentieth-century "technology transfer" projects. Yet we know that transferred technologies often fall on fallow ground. Technologies require deep structures of knowledge and extensive cultural institutions to support them. Medieval Europe lacked the essential knowledge for gunpowder to grow, i.e. how to make cultivate and purify nitrate salts.

This was a critical deficiency, because gunpowder is 50-75% saltpetre. Saltpetre provides the oxygen essential for combustion; it is the oxidizing agent. You can make gunpowder without sulphur, and you can get way with any number of things to provide the carbon, but you cannot make gunpowder without saltpetre.

The earliest European work to record a workable formula for military gunpowder is the Liber ignium ad comburendos hostes, attributed to "Marcus Graecus,". The gunpowder recipe portions are datable to 1275-1300. Liber ignium includes four recipes for gunpowder, three of which have nitrate contents ranging from 66.5% to 75%. How could thirteenth and early fourteenth century Europeans meet this kind of demand? Europe's geology and meteorology do not favour naturally occurring nitrate deposits (although there are a few such sources), and so for the most part, Europeans were in a position of utter dependency on Asian sources for this new material.

Perhaps this accounts for some of the delayed effects of early firearms. In fact, it is extremely difficult to find any military action in all Europe whose the outcome was influenced in the slightest way by firearms before about 1400, and in its own way, this is an astonishing fact. In Asia, at the siege of Khaif'ng-fu in 1232, masses of firearms were in use, at least according to Joseph Needham, while in the West, for nearly one hundred fifty years after gunpowder was introduced, it played a marginal role. In 1346, preparing for war, Edward III's agents  purchased in London 912 lbs of saltpetre and 886 lbs of sulphur "ad opus ipsius regis pro gunnis suis." Normally one would want about 4-6 times as much saltpetre as sulphur, but plainly there was not so much nitrate to be had. Prices for nitrates were high, and the source of imported supply was always the same ñ Venice ñ meaning that "Chinese snow" travelled the same trade routes as drugs and spices, and commanded comparable prices.

Let us consider for a moment another action touted as showing the effect of gunpowder weapons, the French attack on the English fortification at St-Sauveur-le-Vicomte in 1375. This is famous because one lucky French cannon shot made its way through the window of the bedchamber belonging to the English commander, Sir John Chatterton, It rattled about smashing furniture and eventually crashed through the floor. As Chatterton was lying in his in bed when the shot struck, he seems to have been somewhat unnerved by the experience, and he began to parlay with the French.

A closer look at the siege, however, suggests how limited firearms still were at this date. The French sources mention only four guns that actually fired at St.-Sauveur (the English had none at all, by the way). The French were shipping 32 extra guns to the siege site when the English surrendered. Where we can track the cost of powder for the siege,  it was expensive; the supplies procured for use at St-Sauveur cost on average 10 sous (one-half livre) per pound. At those prices, we are dealing with a constricted supply, 100 lbs here, 200 lbs there, and  powder had to be used sparingly. Only thirty-one lbs of powder accompanied those thirty-two extra guns en route to the siege, less than a pound apiece. Money was not the issue; supply availability was. In the end, Chatterton made a good bargain, negotiating a surrender bonus amounting to 55,000 gold francs (!), and on 3 July, 1375 the English garrison marched out of St-Sauveur, collected its gold, and went on its way unmolested. In other words, the French could afford the cost of bribing the English to surrender, but no amount of money could buy enough powder to force the issue.

The choking of saltpetre supply limited the size of firearms. As well, lots of gunpowder was needed per shot; the ratio of one pound of powder per pound of shot appears frequently in inventories of the fourteenth century, a factor nearly three times what later practice demanded. Thus suggests low saltpetre mixtures, but it exacerbated the supply problem nevertheless. Guns were made smaller than they needed to be. Between 1382 and 1388, the Tower of London accounts mention 87 new guns  for which payment was made by the pound avoirdupois. The largest gun was nearly 737 pounds, but most of the others weighed between 320 and 380 pounds. The cost of guns themselves was not exceptionally high, about 4 English pence per pound, and they could have been much larger if the cost of production alone were all that mattered. But the cost of powder ñ and its simple unavailability ñ meant that fourteenth-century  firearms were usually rather small.

This picture begins to change radically around the turn of the fifteenth century. The underlying cause was  the development of saltpetre "plantations" as they are euphemistically termed, stone-lined pits where rotting organic waste was composted under controlled conditions to yield crude saltpetre. The earliest known references to cultivated saltpetre are from the area around Frankfurt am Main in the late 1370s. The effect of boosting domestic production was to slash gunpowder prices and expand availability dramatically. By the 1420s in France, gunpowder cost half what it did in the 1380s,  and by the last quarter of the fifteenth century, French gunpowder prices were less than 20% of what they had been a century earlier, a level where they came to be fixed by law. In Frankfurt, prices for saltpetre itself fell even more rapidly, from 41 florins per hundredweight in 1381-83 to only 16 florins in 1416 and 9 or 10 florins by 1440. The price series data for England are incomplete, but they show a similar trend.

As costs went down, supplies increased, despite soaring demand. Whereas fourteenth-century documents speak of purchases in the hundreds of pounds weight, fifteenth-century  accounts mention tens of thousands of pounds bought, sold, or in storage. Preparations for the anticipated siege of Calais in 1406 involved the purchase of some 20,000 lbs of gunpowder. In 1413 a Parisian dealer sold 10,00 lbs of powder and raw materials to John the Fearless of Burgundy. In 1421-22 in Paris alone there were 10,000 lbs of  poudre ‡ canon on hand, together with the material to make 10,000 lbs more, plus (significantly) nearly 8,000 lbs of saltpetre that was surplus to anticipated needs. Saltpetre was once rare; now it was in surplus stores.

And as gunpowder became more available, guns increased in both number and size. When Henry V decided to renew the French Wars in 1415 sources comment on how heavily equipped with firearms Henry's expedition was when it sailed from Portsmouth in mid-August.  The French chroniclers emphasize the "unheard of size" (inaudite grossitudinis) of Henry's weapons, as well as their noise and smoke, during the siege of Harfleur.  Henry's success at Harfleur in 1415 (which led to the Battle of Agincourt and one of Shakespeare's best dramas) was the first in a series of victories as the cities of Normandy fell to his guns from 1417 to 1422. The powder expenditure required to reduce Normandy to obedience would have been considered ruinously expensive by the standards of an earlier age. Each firing of, let us say a 400 pounder, would have taken some 60-120 lbs of gunpowder, to say nothing of the powder consumed by the smaller firearms. Had Henry V been born a generation earlier, his military exploits would necessarily have rested on a very different technological basis.

The first half of the fifteenth century is also the golden age of giant guns, those eponymous cannon that still grace public squares or museum rotundas: "Mons Meg" now in Edinburgh, firing a stone shot weighing about 550 lbs, "Dulle Griet" or the "Great Bombard of Ghent," shooting over 750 lbs, and the biggest surviving gun, the "Pumhart von Steyr," in Vienna, designed for a stone shot weighing 1530 lbs. These guns made sense, in that the energy needed to break down walls with stone shot and still relatively weak gunpowder demanded a large missile, and large missiles were now feasible given the new availability of cheap gunpowder. Yes, they were difficult to transport, and slow to fire, but they worked well enough when brought to bear on a recalcitrant town.  ILLUSTRATIONS 1 & 2: GIANT GUNS

The transformation I have just described is the first step in the changing of European warfare at the end of the middle ages. It rested on a technological breakthrough, the cultivation and refinement of nitrates from decomposing organic matter. We really don't have any detailed descriptions of how the process worked; for that we must wait until the sixteenth century. I live in the hope that someone, somewhere, will uncover an earlier treatise describing "saltpetering", and that we may be able to understand if the European method resembles the earlier Chinese methods. For the moment, however, all we can point to are  the effects. As saltpetre came to be more and more a domestic product, the product of a peasant industry based on  a seemingly limitless supply of animal manure and urine, gunpowder became cheaper and guns larger and more numerous. This is the sort of "economic" effect we are trained to expect from technology, and there seems to be nothing controversial about it. This is not the case for the next technological step.

Nitrates are produced by bacterial action as part of the long chain of events that convert once living tissues into their chemical components. Nitrates are readily soluble and are easily transported throughout the bio-web. They serve as the main ingredient in most commercial fertilizers, and they are what people who compost the waste in their allotments or gardens are trying to foster in their heaps.  Nitrites and nitrates are formed chiefly through the combined action of two genera of bacteria, nitrosomonas and nitrobacter. Both nitrosomonas and nitrobacter are also very fond of calcium or magnesium carbonate. This makes sense, because calcium, and magnesium, make up some 95% of the exchangeable cationic complement of arable soils in temperate climates. Thus, when nitrates are leached out of the composted manure, calcium nitrate [Ca(NO3)2 ] and magnesium nitrate will predominate.

In later centuries, it was the practice to mix a solution of partly refined saltpetre with potash [K2CO3] to produce potassium nitrate saltpetre. Mixing any carbonate salt into the crude saltpetre solution will form calcium carbonate (CaCO3), a highly insoluble salt that is the principal constituent of limestone, or magnesium carbonate [magnesite: MgCO3], which is nearly as insoluble. The carbonates will precipitate and leave potassium nitrate in solution. A clear technical description of the process appears in Vannoccio Biringuccio's Pirotechnia ca. 1540. But how did saltpetre and gunpowder makers cope with this problem at an earlier date? And what would the effects have been if earlier "saltpetre" were mostly or even partly made up of calcium nitrate?

Let me try to answer the latter question first. Unlike other salts, which must be removed from the crude saltpetre before one can make gunpowder, calcium nitrate will do no direct harm. If the saltpetre contains common table salt [NaCl] for example, any gunpowder that uses it simply will not burn. Calcium nitrate, on the other hand is a splendid oxidizing agent; gunpowder made with calcium or "lime" saltpetre will combust brilliantly. But it will also spoil very quickly in storage. Ca(NO3)2  has a voracious appetite for atmospheric moisture. Moisture absorption is always a major headache for gunpowder authorities; if more than about 4% water (by weight) is absorbed the whole batch of powder is unusable.  The principal reason potassium nitrate is preferred over other forms of saltpetre, is that KNO3 is  the least hygroscopic of all the common nitrates. Potassium as such contributes little or nothing to the thermochemisty of gunpowder's combustion, but potassium nitrate makes a dramatic difference in the shelf life of the product. At a guess, the shelf life of gunpowder made with calcium saltpetre would be a matter of weeks, months at the most, whereas potassium saltpetre gunpowder in Royal Navy stores was expected to last several years at a minimum, even at sea.

Looking at the available sources or the fifteenth century, we certainly get the impression that gunpowder was not very long lived. There are numerous references to spoilt powder, countless recipes for restoring moisture damaged powder, and myriad recipes for improving, refining or "purifying" gunpowder, many of which can be interpreted in modern terms as attempts to minimize the calcium content. But did any of these anticipate the sixteenth century by adding potash and reacting crude saltpetre to potassium saltpetre?

The earliest evidence for a positive answer is from about 1280, when the Syrian writer,  Hasan al-Rammah Najm al-Din al-Ahdab, whose Treatise on Horsemanship and Stratagems of War contains a recipe for using a wood ash or charcoal additive in making barud (saltpetre).   Al-Rammah's text was discussed in detail by nineteenth-century Arabists, mainly in France, and the text is available in a French translation. There can be no doubt that the Treatise on Horsemanship as a whole is genuine, and it is remarkable testimony to al-Rammah's knowledge of Chinese practices in regard to the preparation of gunpowder and firearms.  Unfortunately the recipe for purifying saltpetre with ashes is plainly defective in the available extant copies, and its exact meaning is extremely unclear. J.R. Partington read al-Rammah's recipe as a potash process making potassium saltpetre, and Partington's authority as a chemist makes his a powerful endorsement, but Partington did not read Arabic, and I have not been able to learn that he consulted any text other than the garbled French translations. Indeed, even in Partington's version of the text, al-Rammah's recipe would not yield potassium saltpetre. There may be some garbled form of "purification" of saltpeter which al-Rammah's recipe is meant to express, but it is clearly not a method for making potassium nitrate, even though Partington assumes that it is. Moreover, even if al-Rammah can be read as recommending potash, this still does not prove that Europeans followed his advice.

European evidence for the use of ashes in saltpetre-making is present, but often in confused and technically improbable recipes. One of the best of a bad lot  dates from about 1428-29 and is found in a medical treatise now in the Germanisches Nationalmuseum in Nuremberg. Its specifically recommends adding cinere clavellato to the composting organic material. Cinis clavellatus is "dyer's potash," although putting it in the compost is not where one would want it. Another good piece of evidence  comes from 1474 in Winchester's municipal accounts. John Vanderson and Andrew Gowdrych  were paid for "asshez for ... to fyne Ö saltepetre," as well as for vats and firewood. Unless they were simply cheating the town council, their invoice item for ashes suggests, although it does not prove, that saltpeterers were gaining skill by the 1470s in refining crude saltpetre into potassium saltpetre.

Taking one consideration with another, my own belief is that the fifteenth century sees a gradual improvement of the saltpetre situation throughout Europe. But especially the early decades of the 1400s the overall state of the market was not at a high level of technique regarding refinement. In those decades, most, though perhaps not all, gunpowder would have been subject to storage problems.

What the saltpetre-maker could not cure became a problem for the powder-maker. In order to understand how gunpowder developed we need to look for a moment at how gunpowder is made by professionals, if not by schoolboys. Because gunpowder is a "mechanical mixture" of three solids, it is incumbent on the gunpowder-maker to get the particles of the three ingredients as close together as possible. It won't do to have six little bits of saltpetre over here and four particles of sulphur over there, with a bit of carbon somewhere else. Ideally, any cubic millimetre of gunpowder should contain the requisite 6 parts of saltpetre, and one each of sulphur and charcoal. In practice, the powdermaker achieves this by mixing his ingredients in a mortar under the repeated blows of a pestle.

The shearing action of the pestle head on the side walls of the mortar is important in consolidating or, as the trade calls it, "incorporating" the ingredients. The stamping-and-grinding process must go on continuously for many hours. Surirey de Saint-RÈmy, writing in 1702, recommended a full twenty-four hours under the stamp for the best quality gunpowder, and he noted that the resultant mix should be as fine as women's face powder, i.e. ground talc. While we may take this as a counsel of perfection, there is no reason to assume that this was merely a later refinement of gunpowder-making; fifteenth-century evidence suggests many of the same practices were already in place.

ILLUSTRATION 3: INCORPORATION (TOWER FWB)

To facilitate incorporation of the three ingredients, they are also wetted while in the mortar under the stamp. Not too much liquid is added, just enough to make the whole contents of the mortar into a pasty mass. The point of this seemingly obsessive attention is to force the small particles of saltpetre and sulphur to adhere to the micropores in the pulverized charcoal. (And this is also the reason that certain species of wood make better gunpowder charcoal than others; it has to do with their porosity.) When small amounts of water were added to the mix during the later stages of grinding, saltpetre, by far the most soluble of the three ingredients, would partially dissolve and coat the inner micropore surfaces of the insoluble charcoal, carrying sulphur, also poorly soluble, along with it.

Now this seemingly small technical detail is actually of paramount importance, for it led directly to what is called in the terminology of the trade "corned" gunpowder. "Corned" means "granulated" in the sense of being shaped into small kernels. In later practice, the pasty mass of wet gunpowder was compressed and forced into special moulds create gunpowder granules of homogeneous composition and shape. In the Renaissance, the uncompressed paste was forced through sieves or perforated plates into somewhat irregular grains, a process the modern culinary arts refer to as "ricing" a dough.

"Corning" is of  such great importance because of the way gunpowder burns. In effect, each grain of gunpowder burns from the outside to its core, and that means that the size of the gunpowder grain has a powerful influence on the burning characteristics of each gunpowder charge. In effect, grain size acts as what the physicists like to call a "Goldilocks Effect," neither too much nor too little, but just right. If grain size in a gunpowder charge is too small, the burning slows down; if grain size is too large, burning slows down. For maximum burn speed and thus maximum gas pressure and maximum ballistic effect, you need just the right size of gunpowder grain or "corn."

The effects of  wet mixing and corning were remarkable. On one hand, it made possible the development of the first effective portable or shoulder arms, the Hackenb¸chse or arquebus. Corned powder was so powerful that it was just the thing for small arms. Within thirty years of corned powder's appearance, the ancestors of the musket had already been invented. These weapons are intelligible in terms of the new powder, which made it possible to generate supersonic missile velocities in small calibre weapons.

ILLUSTRATIONS 4 & 5; ARQUEBUS IN ACTION

On the other hand, large artillery was slower to benefit from corning. The new gunpowder posed a challenge to existing guns, whose walls were often too weak to safely contain its higher pressures safely. Exploding guns were a constant threat. Adapting to the new gunpowder meant redesigning artillery along lines never before seen. Barrels grew longer, and they had to be cast from more robust materials, usually bronze. To work optimally, these guns fired shot made of high density cast iron rather than stone. Those artillery designs appeared in the second half of the fifteenth century and accompanied the French, for example, on their descent into Italy in 1494. ILLUSTRATION 6: FRENCH ENTERING NAPLES They were the recognizable prototypes for all smooth bore artillery down to the time of Napoleon and the American Civil War. Eventually, powdermakers learnt to use the Goldilocks Effect to good advantage; powder for big guns was made too large-grained to burn at maximum speed. A compromise was reached between ballistic performance and safety. But I am getting ahead of myself.

Turning to gunpowder, how did corning evolve from earlier powdermaking practices? It has always been assumed by historians, myself included, that the oldest way of making gunpowder was to incorporate the saltpetre, sulphur and charcoal under the pestle in a dry state. Only much later, everyone thought, did wetting the powder in the mortar become a common practice. It is now possible to date the change from dry incorporation to wet incorporation with unusual precision: shortly before 1411. This is a stunningly early date by the standards of the older literature and I need to look closely at the evidence.

ILLUSTRATION 7: CGM 600

A nameless codex in the Austrian National Library, dated 1411, contains the following recipe for "plain gunpowder" [simplified translation]:

If  you want to make a plain gunpowder . . ., take four pounds of saltpetre, well purified, and one pound of sulphur and one of charcoal.  Crush these together [in a mortar] along with good wine in which you have dissolved some camphor, and [afterwards] dry it out in the sun. Without camphor, the gunpowder will go off and quickly spoil, but when you add camphor it preserves all kinds of powder, and also makes the gunpowder strong and fiery (brunstig, "burny").  *

The directions in the recipe are unequivocal. Add wine to the mass of ingredients to be incorporated and you will make better gunpowder. By the way, this is a "plain" or "simple" gunpowder because it does not have any of the additional additives that the age considered made for really supercharged gunpowder, like sal ammoniac (ammonium nitrate) as does another recipe in this text. This was just plain old gunpowder.

This is a wonderful text, and I am grateful it was recently published. This is the text that forces us to distinguish between wet mixing and corning, and it challenges us to see how the former led to the latter. Obviously, you cannot granulate a dry, powdery mix; in that sense, wet incorporation precedes corning. But the 1411 recipe shows no sign of shaping the pasty mix into grains in any way whatsoever. Neither does any other part of this manuscript. We are left to imagine what the powdermaker wants to do with the mixture we last saw drying in the sun. I would like to come back to this question in just a minute.

First though, please look more closely at the reasons given in the  recipe for adding the liquid. The emphasis is largely on preserving the gunpowder from spoilage. The fact that it makes gunpowder stronger and "more fiery" is an added bonus. We know, of course, that the wine and camphor are not really needed; pure water will do just as well. But for the anonymous author, the camphor is the crucial ingredient because it preserves the powder. "Without camphor," he insists, "the gunpowder will go off and quickly spoil," but  he adds, "camphor preserves all kinds of powder." The wine (whose water content, we know, is what is really important) is in the author's mind merely a solvent for the camphor. (By the way, this is also true: camphor is far more soluble in alcohol than in plain water.)

Why should there be such an emphasis on preservation rather than on improving the ballistic aspect of the gunpowder? I have already suggested the answer: 1411 lies in the period when inferior homemade saltpetre was flooding the European market, making life more difficult for the powdermaker. The hygroscopic qualities of calcium nitrate reduced the shelf life of gunpowder significantly by making it even more prone to spoilage from the damp. Camphor, on the other hand, was a dry resinous substance, quite the opposite of water; it was "dry" and "cold" in the humoral theory of elements. What better means to rebalance the problems of gunpowder's unfortunate tendency to go off by becoming wet? The fact that camphor is also mildly flammable was probably another count in its favour.

If, as I am arguing, wet-incorporation was originally intended as a means of enhancing the shelf life of gunpowder, then the next step towards corning stems from the same motive. In the so-called Feuerwerkbuch or "Book of Fire Works," whose earliest versions date from the around 1420 (the earliest dated redaction is 1428), there is a recipe for "lump-powder"(Knollenpulver):-

Moisten a mixture of saltpetre, sulphur and charcoal with good wine vinegar. And grind the mixture together with a wooden mortar, and make it so moist with the vinegar that it will clump together, and then roll the lumps as large as you want them to be. Then take a glazed round deep saucepan or pot or a copper bowl and press the lumps in wet just as you would force a cheese into a pot; then knock it out onto a board -- it will come out easily

ILLUSTRATION 8: MAKING KNOLLEN

The text tells us to air-dry these Knollen either in sunshine or in heated sheds, depending on the season. The text does not suggest making small, shaped grains, but instead speaks of the dried product as Knollen, lumps. At least one manuscript illustrates the process quite unequivocally; the powdermaker shapes little loaves of gunpowder.

Feuerwerkbuch's Knollen  method was very similar to a technique used by the peasant fireworks makers of rural Spanish Galicia well into this century. They shaped the wet gunpowder into bolas  about 8 centimetres in diameter. These bolas were set out to dry in the sun. Writing in the 1470s, Francesco di Giorgio Martini advised his readers of a "secret" method to preserve powder on long expeditions: shape the wet powder into loaves, like bread, and let it dry. In the 1530s the Feuerwerkbuch  recipe was plagiarized by Franz Helm.  As late as 1628,  Robert Norton's The Gunner  has a similar recipe involving aquavit as the liquid. Kept properly, says Norton, these balls of gunpowder "will neither decay, nor waste by age."

Loaves of gunpowder would, of course, last longer in storage because they minimize the surface exposed to atmospheric moisture relative to the volume that they contain. Spheres and spheroids have minimum surface to volume ratios. In this text, as in the 1411 recipe, the powdermaker struggles to cope with the problem of spoilage, a problem that stems from poor saltpetre. In the case of Knollen, he does not resort to exotic water-repellents like camphor (homemade "good wine vinegar" is enough), but he relies instead on the shape of the product to make it store well.

Now let us return to the question I posed a moment ago: what do these recipes expect us to do with sun-dried gunpowder? Obviously, gunpowder has to be crushed back down to something like a granular consistency before it can be used in a gun, and at least one later copy of Feuerwerkbuch helpfully adds the advice to crush the Knollen of gunpowder before loading your gun. Interestingly, this text refers to the final product as gereden pulver; "grated" or "crumbed" would be the closest English equivalents. This would mean that the sun-dried gunpowder would be reduced to a pile of irregular crumbs before it was  loaded, and this most likely meant it was  forced through sieves to break it down into smaller grains.

The next step appears in its earliest complete description as given by Peter Whitehorne in 1562:  ILLUSTRATION 8: CORNING

The manner of cornyng all sorts of powder is with a Seeve made, with a thicke skinne of Parchement, full of little rounde holes, into whiche seve the powder must be put, while it is danke, and also a little bowle, that when you sifte, maie rolle up and doune, upon the clottes of pouder, to breake them, that it maie corne, and runne through the holes of the Seeve.

Whitehorne plagiarized much of his description of powder-making directly from Biringuccio's Pirotechnia, but he interpolated this original [?] passage into the chapter. We have earlier references to corning gunpowder using a colander or sieve from Franz Helm, in 1335, but Whitehorne seems to be the first to give us a full account.

The state of scholarship does not permit us to give a precise date when powdermakers stopped crushing sun-dried gunpowder and making little grains or pellets of wet gunpowder instead. My own research has turned up some indirect evidence suggesting this was the normal process by the beginning of the sixteenth century, but there is little to suggest controlled-grain corning was being practised in the mid-fifteenth century. I does not take much imagination to see how, if sheets or lumps of dried gunpowder were being broken down into crumbs by forcing the larger bits through a sieve, someone sooner or later would skip the drying stage and try forcing the wet paste itself through a sieve. This would produce fairly uniform wet particles of gunpowder, true gunpowder corns or grains, ready to be dried in the sun. But just when this began to be done remains a mystery.

At this point, someone is bound to ask "What about serpentine gunpowder?" 

"Serpentine," is a term dating from 1497 and used throughout the sixteenth century to mean a gunpowder "as fine as sand and as soft as floure, " to quote William Bourne's 1587 description. By 1627, when John Smith wrote his Sea Grammar, "serpentine" powder had become obsolete. "Your Serpentine powder in old time was in meale, but [is] now corned and made stronger, and called Canon corne powder." From this kind of evidence, English scholars in the nineteenth century concluded that serpentine powder was dry mixed, "[gunpowder] ground dry to a flourlike compound," to quote from a recent standard work. This is a deeply entrenched belief, one that has influenced my own work quite deeply, but one that I now believe is entirely wrong.

We now know that wet-incorporation began early in the fifteenth century. Why should dry-mixed powder have survived another 200 years? Worse yet, although we do indeed commonly find in English sixteenth-century arsenal inventories mention of "serpentine" gunpowder along with "grosse corne powder," and "ffyne corne powder," no equivalent  terminology  is found in Continental sources of the same period. Should we believe that English powdermakers ñ and only English powdermakers ñ practiced obsolete techniques for two centuries during the Renaissance?

There is a lengthy coda to the paper I have just presented in which I try to overturn the commonplace belief amongst English scholars that the answer to those questions is "Yes". I will not try to give you all the arguments today, but merely state the conclusion I believe to be obviously true: "serpentine" gunpowder was wet-mixed, then dried and ground down to a flour-like or mealy consistency. There were good technical reasons to do this. In essence, this was the English way of making good gunpowder for large cannons by minimizing the size of gunpowder grains. The English exploited the "Goldilocks Effect" by making cannon gunpowder too small in the grain ñ so it would burn slowly. The Germans and the French played the opposite side of the physical phenomenon, by making cannon gunpowder too big in the grain so it would burn slowly. The English method was subject to problems causing the powder to spoil in storage more readily than the Continental big-grained types, and eventually, in the seventeenth century, the English abandoned the practice of making "serpentine" in favour of big-grained, slow-burning gunpowder.

To conclude with a summary: As I have presented it, the history of gunpowder manufacture shows four main (overlapping) stages between 1250 and about 1600. In the earliest stage, up to about 1390-1400, saltpetre was rare and expensive, and gunpowder was made in small batches, the materials by being ground and  mixed as dry solids in a mortar. The guns using this powder were small, few in number, inefficient, and relatively unimportant for a military point of view.

In the second stage, which began at the turn of the fourteenth century, saltpetre became much cheaper and far more readily available, if somewhat lower in quality than before. This led to bigger guns and many more of them, and firearms began to influence warfare. The lower quality of this new saltpetre also led to efforts to make gunpowder less susceptible to the damp, paradoxically, by adding liquids to the mixing mortar and incorporating the powder while it was still wet. This produced a faster-burning, better-storing gunpowder.

Such gunpowders mark the beginning of the third stage, which emerged out of the second somewhat more gradually than the second stage came from the first. The fact that wet-mixed gunpowders had to be crushed anew before being used meant that they were all, in a very loose sense of the word, granulated powders, but with highly irregular grains. The burning characteristics of these powders meant that guns could be made longer with smaller bores, a movement we can see in the design of both artillery and shoulder arms. In fact, the fifteenth century sees the birth of the prototypes of  all firearms as they existed until the latter half of the nineteenth century.

In time, the irregular grains that were produced by the early recipes gave way to controlled-grain manufacture when the wet gunpowder paste began to be forced through perforated plates or sieves in order to make regular and uniform grains. This transition cannot yet be dated with precision, but probably took place around 1500-1525. It marks the fourth stage, when gunpowder could be made according to predetermined grain sizes. The English were slow to exploit the full potential of this technique, but they began to do so around 1600. Outside England, we find in the arsenal inventories of the 1500s the same divisions of gunpowders that persist into the nineteenth century: fine grained powders for mortars, small-medium grains for muskets, and coarse-grained powders for large artillery. Gunpowder had been mastered, and the European pupil had surpassed his Chinese preceptor.

*Wiltu ein slecht pulver machen nver von drein stucken. So nym iiij pfunt salniter, der vast gut sey und wol geluttert sey, und j pfunt swebel und j kol und stoz daz ab mit gutem wein, da gampfer innen gesoten sey, und derr daz an der sunn. Wenn wa [sic] nicht gamfer pey ist, daz pulver erwirt und verdirbt gern. Aber daz gamfer halt allez pulver auf und ist auch kreftig und prunstig in allem pulver, wenn man in darin tuot.