Chemistry, Chemical Reagents, Physics, and the History of Manuscripts

Please Note: This is not a complete article on how the physical sciences canbe used to help us in textual criticism. This is an extremely broad field, with referencesscattered in journals of many fields and (as far as I know) no general manual. Ihave pulled material together from a lot of sources, but this is just a collection ofnotes, not a comprehensive summary of the field.

Sections: Chemical Reagents* Paints and Pigments* Carbon Dating* Spectroscopy* Isotope Analysis* Detecting Forged Manuscripts

Chemical Reagents

Old manuscripts can be extremely difficult to read. The most obvious examples are palimpsests, but even a manuscript's upper writing can fade.

Today, scholars have excellent tools for dealing with such problems (notably ultraviolet photography, though there are many other techniques in use). That wasn't so in the past, but the desire to read the manuscripts was just as great. In consequence, scientists developed a number of chemicals for trying to bring out faded or eradicated ink. The first ink restorer seems to have been oakgall (gallic acid or, technically, trihydroxybenzoic acid, C6H2(OH)3COOH), used as early as the early seventeenth century (possibly earlier), but much stronger chemicals were eventually discovered. Some of the reagents used in the nineteenth century include ammonic sulphydrate, potassium nitrate, potassium bisulfate, and Gioberti tincture -- successive coats of hydrochloric acid and potassium cyanide (!).

Supposedly (according to E. Maunde Thompson's An Introduction toGreek and Latin Paleography, p. 65), the "most harmless [reagent]is probably hydro-sulphuret of ammonia." Similarly, M. R. James wrote that"ammonium bisulphide... unlike the old-fashioned galls, does not stainthe page." Which mostly tells you howdamaging the others are. Hydro-sulphuret of ammonia is a strong hair dye, withacid properties. It is certainly capable of damaging manuscripts.

If you somehow talk someone into letting you use this gunk on an old manuscript,be sure to dab or pat it onto the parchment. Do not paint it (which can cause theink to smear) or spray it (which will apply more than you need).

The problem with these chemicals is that, although they can bring out the writing in the short term, they destroy the manuscript in the slightly longer term. They can cause the ink to blot and the parchment to decay. (As a result, there was a brief period during which scholars applied their glop, photographed the results, and washed the chemicals off. Somehow this doesn't seem much better than leaving it on the manuscripts.) Among New Testament manuscripts, this sort of defacement happened notably to C (though it is not clear whether Tischendorf, who is frequently blamed for it, was guilty; other scholars seem to have been the primary culprits). The problem is especially bad when multiple chemicals are applied -- as was done, e.g., to the manuscript of The Poem of the Cid); not only does this damage the parchment, but it also renders ultraviolet photography less effective. Ian Michaels, in his introduction to the Penguin bilingual edition The Poem of the Cid, tells us on p. 15 that "the reagents have not only blackened the folios where they were applied but also appear to have eaten through the parchment in the worst affected places; they have also left a flourescence which greatly reduces the effectiveness of ultra-violet light." The chemicals used were apparently ammonic sulphydrate plus, in a few cases, "yellow potassium prussiate" and hydrochloric acid.

Chemical "enhancement" of manuscripts is now strongly frowned upon, and has effectively stopped -- having been replaced by much less damaging techniques. Unfortunately, there are instances of the use of chemicals as late as the 1920s; many manuscripts which survived the Middle Ages have now been permanently damaged by more modern scholars who generally did not learn much as a result of their vandalism.

It's interesting to note that some of these chemical reagents were known long ago. Pliny the Elder was perhaps the first to describe an invisible ink. Of greater significance, perhaps, is a remark by Philo of Byzantium, who refers to an ink of nutgalls which could be developed with what we would now call copper sulfate. Since many ancient inks contained nutgall, Philo deserves credit, in a sense, for the first method of "developing" palimpsests.

Paints and Pigments

Chemistry can be a powerful tool for textual criticism in its paleographic aspects -- specifically dating and verification of manuscripts. Spectroscopy and other tests can reveal chemicals contained in inks or paintings without damaging the manuscript. And if a manuscript contains a chemical not in use at the time it was thought to have been written, well, that implies a problem. This line of argument has been used, e.g. to implicate 2427 as a forgery, since it probably contains Prussian Blue, a dye not invented until the eighteenth century, well after 2427's alleged date. The problem with such arguments is that they depend to a strong extent on our knowledge of history of chemical use; there is currently a major argument about another chemical, titanium dioxide, thought to be modern but now found in small amounts in ancient inks.

(Incidentally, while Prussian Blue is the most famous, and mosteasily detected, of modern colours used to fake artifacts, it is notthe only one. The infamous Piltdown Man hoax of the early twentiethcentury involved bones of a human being and an orangutan being jumbledtogether, broken up, filed -- and stained to make them look like a match.Some of the staining was done with a commercial paint, Vandyke Brown.Vandyke Brown is not as chemically unique as Prussian Blue,but it will surely be evident that million-year-old fossils didn'tgo around staining their teeth with paint manufactured around thebeginning of the twentieth century!)

Another recent surprise came when a technique called Raman spectroscopy was used on the British Library's King George III copy of the Gutenberg Bible. According to a (non-technical) article in Renaissance magazine (issue #45, p. 18), the inks used to illuminate that printed book (which of course is contemporary with some late manuscripts) included cinnabar for bright red (as expected), carbon for black, azurite for blue (not a surprise, though some blues use lapis lazuli), calcium carbonate (chalk) for white, malachite for olive green, and verdigris (copper ethanoate) for dark green. Those were no surprise. More notably, the Göttingen copy was found to contain anatase and rutile, which had been regarded as modern compounds. This may be the result of contamination, but it may be a hint that we may still have more to learn about ancient inks.

Some pigments can be detected simply by the way they decay over time. Anexample is paint using white lead. White lead was prepared by exposing elementallead to the fumes of vinegar (acetic acid) to create lead acetate(also called lead (II) ethanoate, Pb(CH3COO)2) and sundryhydrates. Often the work was done in the presence of animal dung to make morecarbon dioxide available and speed the reaction.

Lead white was a delicate white, much liked both for a wall covering andfor detailed paint. Often it was mixed with other pigments to produce pale shadessuch as pink. Sadly, it is unlikely to look pink any more. If exposed to hydrogensulfide (a common by-product of gas lighting and especially of coal-burning),it reacts to form lead sulfide (PbS), which is black. The decay of white lead hasbeen known for centuries (Cennino mentioned it in the early fifteenth century),but it was such an excellent white that it continued to be used -- there was nogood alternative until zinc white was discovered after the manuscript era, andthere was no good and cheap alternative until titanium white came along in theearly twentieth century. So if you see a paintingof someone's face which looks as if it had been expected to be pink, but now looksdark brown or black, odds are that it used white lead. Unfortunately, the use ofwhite lead isn't very useful as a dating method; its preparation was firstdescribed by Theophrastus (372-286 B.C.E.), and it was widely used by theRomans. (Information in this paragraph primarily from John Emsley,Nature's Building Blocks: An A-Z Guide to the Elements, corrected edition,Oxford, 2003). It was still being used as a pigment as late as the time ofEngland's Queen Elizabeth I (died 1603); her extremely heavy makeup was saidto be founded upon white lead. (I can't help but wonder if it shortened her life.)

Speaking of pigments and makeup -- keep in mind that most manuscripts withilluminations were painted before the invention of oil paints (the best of thenatural oil bases, linseed oil, was known from the eighth century but reportedly wasnot used for painting until the fifteenth). Earlier paintswere almost like wet versions of pancake makeup, using materials such as egg whiteor fish glue to attach the pigments to the page. This affected how they were laiddown, how they were mixed, and how they survived. Many illuminated manuscriptsseem to be in a rather un-modern style. But this may have had more to do with thepaint than the painter.

It's worth keeping in mind that the different pigments needed to be used withoils than with other binders. Oil has a different index of refraction than theothers, meaning that colors actually changed when mixed with oil rather thantempera or glair. Ultramarine in oil is darker and less attractive, verdigris losessome of its opacity. On the other hand, the so-called "lake" colors,which aren't worth much in the older binders, gained in vibrancy with oil.Again, though, this is not likely to matter in a Biblical illuminated manuscript.

Table of Common Chemicals Used in Ancient Inks and Pigments

Please note: This list isn't even close to complete; I'm adding chemicals as Ilearn of them.

Also, without listing them in the table below, we should probably list the twomost common components in ancient paints: Egg white and egg yolk. These were not usedfor color; rather, they were binders, holding the pigment to the page. Egg white isusually refered to as "glair"; egg yolk was used to make tempera paint. (The name"tempera" being derived from the verb temperare,"to temper," since thetempering agents served both to bind the paint sometimes to change its appearance.They also affected the transparency of the pigment -- how much the background showedthrough.) Both binding agents were common,tempera probably more so. In general the pigments were ground into a fine powder,then dissolved in a smallquantity of water, which was then mixed with the egg yolk. Once the yolk fully dried,it proved quite stable; even water affected it only slightly. Yolk contains enough oilto be rather like oil paint, except rather duller and less reflective. (Tempera infact is said to retain its color and stability better than pigments in oil, exceptfor the problem that it hardens to be more fragile than oil and so is more subject todamage.)

Glair does notmix readily with water, and had to be whipped to reach the proper texture -- and thishad to be done in a special container to prevent contamination of either the glair orthe container. Interestingly, scribes found that adding a little ear wax made glaireasier to prepare and work with. (Just what you wanted to know, right?)

In later centuries, gum arabic (acacia gum, a complex compound) came to be usedas a binder as well, either in mixture with or as a substitute for glair or tempera. This was a key ingredient in "watercolor" drawings. Properly, gum arabic was made from acacia plants -- although sometimes the gums of other sorts of trees were used. From what I can tell, though, this was rare in New Testament manuscripts; watercolors tendto be applied in broad strokes and with pale colors that simply aren't suitable forsmall illuminations!

More common as a binder was size, or (as we now call it) gelatin. This was made from skin and other body parts of dead animals, and probably contained other materials as well. It was especially useful for binding blue pigments, most of which did not work well withtempera (when a binder other than tempera was used, the result was often called a"distemper"). Size-based glues were also good for binding gold leaf. (Properly speaking, "size" can refer to many materials; the name means "seat," because size was used to seat, i.e. to hold, materials in place. In some contexts, it was used to slightly raise the pigment above the levels of those around it. For these purposes, any material, such as plaster, could be a size, but the best were things like gelatin or gelatin with honey. Sadly for the noses of scribes, size was stickiest and most useful when it had been allowed to rot before application.

If someone ever invents an easy test for egg yolk, it will be a good way of checkingfor forgeries; starting in the fifteenth century, it was gradually replaced by oilbases, with egg no longer in use in the sixteenth century and after.

Although both glair and tempera were used in manuscripts, Daniel V. Thompson,The Materials and Techniques of Medieval Painting, tells us that of the variousbinders, "glair and gum [were] chiefly for book, egg yolk chiefly for panels,lime chiefly for walls, and size and oil for woodwork."

It's worth noting that different regions had relatively standard palettes ofpigments, meaning that you can sometimes use the colors in a manuscript to identifywhere it came from -- e.g. early British manuscripts tended to use a palette of red(from red lead), yellow (from orpiment), and green (from verdigris), without a blueor purple color. Pliny the elder asserted that classical Greek painters used onlyfour colors, black, white red, and yellow (so Philip Ball, Bright Earth, p. 15;on p. 68 he hints that this might be by analogy to the four alleged elemets ofearth, air, fire, and water),although we cannot verify this and do not know which actual pigments were used;Ball points out that 29 different pigments have been found in use at Pompeii.

Also, there in many schools of art, it was considered improper to mix colors --perhaps because mixed colors were less bright than pure colors; also, of course,the colors might react with each other. Furthermore, the ancients did not have anything resembling color theory; it wasn't until the seventeenth century that Newton produced a color theory of light, and the idea of complimentary colors and the color wheel is a nineteenth century invention. Lacking these, mixing colors would inherently have beena hit-or-miss process -- not even a matching recipe was much help when certain colorwords could refer to hues as diverse as yellow and blue! It was not until the invention of oil paintingthat mixing colors became commonly accepted. The lack of ability to produce intermediateshades of course limited the level ofdetail possible, but it does mean that many manuscript illustrations were particularlybright and striking.

Also, although it doesn't matter to us, it's worth noting that pure colors will look the same whatever sort of sensor observes them. This will not necessarily be true of mixed colors, which depend on the way the human eye senses colors. So if a non-human species ever observes a painting done with mixed pigments, it might not look the way it does to us.

Note: There are so many colors in this table that I've put together an index by basic colors, below. In this list, the capitalized word is the name under which it alphabetizes, e.g. "green Ochre" is filed under Ochre; "Naples yellow" files under Naples.

Color Pigments
blackBone black, Lampblack, Wine Black
blueAzurite, Egyptian Blue, Indigo, Lapis lazuli, Turnsole
brownBistre, Sepia
greenBuckthorn, Chrysocolla, Green earth, Honeysuckle green, Iris green, Jade, Malachite, Mixed green, Nightshade green, green Ochre, Sap green, Verdigris
orangeBistre, Realgar
purpleArchil, Dragon's blood, Turnsole, Tyrian purple (note that the Greek πορφυρεος and Latin porphyria or purpure do not refer specifically to a violet color but to any dark red or violet or even to blues; drying blood, e.g., was purple, so when something is called purple in an ancient text, it is not a definite description of color.)
redCinnabar, Dragon's blood, Dyewoods, Hematite, Kermes, red Lead, (red) Madder, red Ochre, Realgar, Whelk red
 silver Silver, Tin
 white Bone white, Calcite, (Gesso), white Lead, Shell white, Silver
 yellow Aloe, Bile yellow, Buckthorn, Celandine, Gold, Indian yellow, yellow Lead, Mosaic gold, Naples yellow, yellow Ochre, Orpiment, Saffron, Weld

The color wheel below probably isn't very accurate on your monitor, but it will give you some idea of the relative colors of some ancient pigments.

Example fallback content: This browser does not support PDFs. Please download the PDF to view it: Download PDF.

Common NameChemical FormulaColorComments
Aloecomplex organicYellow Seemingly not used as an ink or dye in itself, but it could be mixed with something else to give a yellow color -- e.g. it might be mixed with egg yolk and a metal (mercury, silver, tin) to give the impression of gold.
Archilcomplex organicPurple A purple color derived from a lichen. The name comes from Latin rocella, the name for the lichen, via oricella and older English orchil. It was used primarily as a dye rather than as a pigment.
Hydrous copper carbonate
Blue Often found with malachite; the two are close chemical relatives. (When they were created first artificially, the workers had trouble getting consistent color results; it turns out that which compound the reaction produced depended on the temperature.) A very hard mineral, which required much grinding before it could be used as a pigment (and which, even when ground, retains its crystalline structure). It also tended to be found as medium-sized grains in a matrix of sand, malachite, and other compounds. This meant that it had to be prepared by grinding and then separating with water, soap, and other chemicals. The result was a very deep blue if the grains were the right size, but a paler color if they were ground too small. (Grinding always affects the hue of a crystaline pigment, because the shape of the surface affects reflections, but this was more noticeable with azurite, perhaps because it was so hard to grind. There are actually paintings which use azurite for two different colors by changing the way it was ground.) The large particle size meant that it often needed a different binder (size rather than tempera), and it had to be laid on in thick layers -- a difficult technical challenge, and one that meant that varnish did not mix well with azurite. Also, it could easily chip off. And those large grains meant that varnish could affect it and cause it to discolor. It might also darken if exposed to sulfur or acids, or undergo a chemical change to malachite, making the blue turn green. Because it is so difficult to prepare, it quickly went out of use when modern blues became available (the fact that available supplies were becoming exhausted may also have played a role, or just the fact that it didn't work especially well in oil paints). Moderns may also find it listed as "blue bice" or "blue verditer"; this too is copper carbonate, but prepared chemically rather than from natural deposits. A truly pure azurite will be a very deep blue, but because it is often mixed with malachite, it is likely to look blue-green. Indeed, in early times the mixed material was sometimes called "verde azzurro," "blue-green," or "acquamarine" because it was used to create sea colors.
Bile Yellowcomplex organicyellow Made from the gallbladders of various vertebrates, with fish being the most preferred; tortoises were also a common source. It was ground with chalk or vinegar or other compounds. It was used for golden shades, primarily in Greek regions.
Bistrecomplex organicorange or brownMade by burning the roots of certain resinous trees. It is not a very stable color, so it probably was not often used, but because it was not developed until the fourteenth century, it can be used to date such manuscripts as do use it.
Bone Blackcomplex organicBlack Lampblack was probably the best black available to the ancients, since it was almost pure black carbon, and hence very black plus the particles were very small and made a good even ink. But unless one was willing to burn candles or oil solely to produce ink, the supply was limited. Among the substitutes sometimes used was bone black -- charred bone. Although this contained only about 10% carbon, with most of the rest being calcium phospate plus a few percent of calcium carbonate and other things, it was a very good black if properly prepared. However, it was trickier than lampblack because the bone had to be very finely ground to be usable, and heated very carefully to prevent it burning away. Typically it was more of a brown than a black color as a result.
There are stories of human bones being used to make bone black. I know of no verified evidence for this, but it is certainly possible. The only question is, why would anyone do it when other bones were available? Certainly such ink would not have been used by, e.g., Jews.
Bole(varies?)Pink and others A pink clay, commonly said to be from Armenia but apparently found in many places, described as "soapy," so it was not a good pigment in itself. It was, however, a good substance to mix with size to give the size a color. Pink bole was apparently the most popular, but there were red, yellow, green, and white boles; the term "bole" refers to the texture, not the color.
Bone Whitecomplex organicWhite Although white lead was the preferred pure white in ancient paintings, it did not mix well with some colors. When a mixed color was needed, bone white might be used. The bones were heated to a high temperature in a fire, until they turned white, and then ground. It was not easy to work with because it was pasty, but it was combined with verdigris or orpiment to produce stable mixes of those colors.
Buckthorncomplex organicYellow or Green The juice of unripe buckthorn was very occasionally used as a yellow ink, or mixed with a blue to produce a green color. However, the color was very unstable, so it was rarely used in manuscripts and, as far as we know, not at all in other sorts of paintings. Riper buckthorn might be used to produce a Sap Green (which see).
Calcium Carbonate
Calcium carbonate occurs in a wide variety of forms. There are three crystal forms (calcite, aragonite, and vaterite, though the last of these is very rare and the second unstable and tends to decay into calcite), and is even more common in non-crystalline form as limestone and chalk. Pure forms are usually white or clear, though impurities can cause it to take on almost any colour. It often is found as part of other rocks (see the notes on lapis lazuli). The form found in pigments is typically chalk, used for white paint or to change the brightness of mixed pigments. It was not often used with oil pigments, except as a translucent effect, because it is almost transparent in oil.
Carmine(See Kermes)
Charcoal(See Lampblack)
Mercury Sulfate
Red or red/brown Sometimes called vermillion. Ores usually found in Spain, Italy, the Balkans. A very vivid red, but rather dangerous to deal with because it was relatively easy to liberate the mercury -- and mercury is toxic, and some of its compounds even more so. Another name occasionally used for it is minium, although that name is more frequently and more properly used for red lead (Pliny referred to cinnabar as the best form of minium and to red lead as minium secondarium, second-rate minium, or "false sandarach" presumably because it was cheaper but not as bright as cinnabar, but Gerber, who eventually became a more important authority, reserved the term minium for red lead). In early times, cinnabar seems to have been mined (generally meaning that it was not very pure, and indeed, artists were warned to buy it as stones rather than already ground, as apothecaries might adulterate the ground form). Later, purer forms were created from elemental mercury (often derived from natural cinnabar!) and sulfur.
Although known to the ancients, it has been claimed (I do not know how accurately) that it became popular only in the Middle Ages, under Moorish influence. Supposedly, in early times, it was as expensive to cover a page with cinnabar as gold -- but I have never heard of a cinnabar/vermillion codex.
Although it is a beautiful orange-red, cinnabar was not very suitable for use in mixed colors such as purples because of its orange tint, so other colors such as the so-called "red lakes" had to be used for mixing. And it doesn't look very good in oils, so it became much less popular as oil painting took over. And, finally, although it is stable in conditions of low light, if exposed to bright bluish light, the red cinnabar can convert to black "metacinnabar," so the color could darken and lose its vibrancy over time; it is likely to look brown rather than bright red. It was said to be particularly subject to degradation if used in fresco.
Celandinecomplex organicyellow A mix of mercury and egg yolk produced a silvery sort of ink, which could look like gold if a yellow color was added. Apparently the juice of the celandine plant was often used to supply this yellow. However, the color did not last long and will usually be faded.
ChrysocallaCopper Silicate, CuO3Sigreen One of several copper silicate pigments (see also Egyptian Blue), this one is greenish and is a common modern pigment but in the past was apparently used primarily as a binder for gold; it was sometimes called "gold glue."
Dragon's Bloodcomplex organicreddishA purple-red resin-based compound, difficult to identify because it looks much like other reds, but almost certain to be old (it is no longer sold). It reportedly came from the Arabian Peninsula and beyond, from a shrub called Pterocarpus draco, (others suggest that the source is the rattan palm Calamus draco; evidently the color name and the plant name are closely linked). In either case, it is an oriental plant, so it was probably more common in eastern manuscripts.
Its primary purpose is said to have been to add a bit of color to metals, e.g. to make gold look a little more reddish.
It appears a few alchemists referred to cinnabar by the name Dragon's Blood (and, to add to the confusion, to call Dragon's Blood "Indian cinnabar") which might explain how the name arose, but I know of no artists who call cinnabar by that name.. In later years, Avicenna was reported as saying Dragon's Blood came into existence as the result of a battle between a dragon and an elephant, in which the elephant sat on the dragon's tale and made it bleed, whence the name.
Dyewoodscomplex organictypically red Dyewoods are a large group of colors derived from the ashes of burnt wood. The most familiar sort was brazil wood -- and, yes, the name for the wood came before the name of the country; Brazil, the story goes, was so named because it became a major source of dyewood. The way the ashes were treated would affect the color and permanence of the pigment, and are probably too complicated to be worth discussing here. They became popular relatively late -- apparently Brazilian red did not become really widespread until the fifteenth century. Brazilian red is, however, very subject to fading, especially if exposed to light -- more so, it is said, than European dyewoods; this might perhaps be used as a hint about a manuscript's history.
Egyptian Blue Copper Calcium Cilicate, CaCuSi4O10Blue This has been called the first artificial pigment, although the name "Egyptian Blue" is modern. It is not known how it was prepared in ancient times, but it is used in many Egyptian monuments, and the color survives to the present day. I have found no reports of it being used in Greek or Latin illuminated manuscripts, but it might someday turn up in documents from Egypt.
Gesso CaSO4-1/2H2O
and other
(White) Not really a pigment; gesso is a substrate, used to attach pigments or (often) gold leaf; it is a combination glue, colorant, and surfacer. Later, when canvas replaced wood as the usual surface for paintings, it was often used to smooth the rough surface of the cloth. The basic ingredient is slaked Plaster of Paris -- hydrated calcium sulfate, usually purified. It was often prepared by heating gypsum to drive out the water, then rehydrating it to achieve the exact desired consistency. (This could be thick and stiff or thin and runny, depending on the particular purpose in mind. A thin gesso, for instance, shrank upon drying, meaning that it achieved a better "fit" to a particular surface)
There were other ingredients, however. Many mixes include significant amounts of white lead for color (up to 25%). Gesso that was to be used as a substrate for gold leaf often had Armenian bole or another reddish earth added (so it would be less noticeable if the gold rubbed off. The red was due to iron oxides). Sugar might be added as a dessicant, and gum to make it cohere better. Finally, water and egg glair (made from egg whites) would be added, the former to moisten the mix so it could be applied to the page, and the latter to make it stick. When applied to manuscripts (as opposed to walls or the like), it was applied with a pen, then allowed to harden. The overall effect seems to have been rather like water-based correcting fluid. After it had dried, a layer of gold leaf might be burnished on top of the gesso (usually after waiting at least a day).
Sometimes gesso was placed behind other pigments as well. It is likely that this was done to brighten the pigment -- it would not change the color of the painting, but it would cause it to reflect any light which passed through. The effect was a bit like painting over a mirror.
For more on gesso, see the discussion of binders above.
Folium(See Turnsole)
GoldAuGolden Gold was used to represent, what else, gold; it was either applied as a thin sheet (gold leaf) or ground up to use as an ink. Obviously this was done only in the most expensive manuscripts. When the patron could afford it, however, it was usually used both for golden shades and for yellows.
Ground gold was rare for another reason besides expense: because gold is so soft, it was very hard to grind; the particles tended to stick back together. It helped to mix the gold with salt or honey, and grind the mixture, then wash away the intrusive material, but no technique known to the ancients could make the particles small enough to make a good suspension, so gold ink had to be applied in a thick layer, using a lot of gold. This made gold leaf a better economic proposition: you could cover a lot of page with leaf for the equivalent of the cost of a little gold ink. And the leaf looked better and shinier.
Eventually another method was developed for writing in a golden color, known as mordant gilding: a sticky material was used to write on the parchment, then gold leaf pressed on it. The gold would stick to the glue and not to the parchment, so the writing would be gold and the rest of the leaf could be reused. But it was hard to produce attractive writing this way; the boundaries between parchment, glue, and gold were often not very sharp, and the adhesive might get smudged. Some glue mixes were colored red to try to make this less obvious, but the method was never really perfect. And the mordants were often water-sensitive and gradually affected the gilding; the response to this was oil mordants -- in effect, sticky varnishes -- but I believe this technique was not developed until quite late.
Yet another trick for applying gold was to use a mordant and some sort of rough material such as ground glass, and then rubbing it with a lump of gold; the rough material would cause the gold to flake off and stick to the mordant. This was "attrition guilding," but it probably wasn't very common, because it was complicated and you got a rough page that wouldn't be good for the facing pages.
Note also that gold ink was not very visible against ordinary parchment; it really only worked on parchment died purple or some other dark color. So gold ink, already expensive in itself, required unusually expensive parchment as well. All in all, very un-economic.
There is an important bit of terminology about applied gold. Gold leaf isn't just a thin layer of gold -- a thin layer of metal is a foil (as in aluminum foil, tin foil). Foil is about as thick as a sheet of paper. If the layer of metal is even thinner than that, it is leaf. When gold was applied to parchment, it was leaf gold, not foil gold. Many other metals, however, might be applied as foils, because it was much easier to beat gold out to leaf thickness than to prepare tin or copper as such thickness.
In general, gold leaf would have gotten thinner as time passed. This was not for economic reasons; it's that the techniques for making it improved. Very thin gold leaf easily sticks to things, so it had to be prepared on a special sort of parchment called "goldbeater's skin," which was not developed until the Middle Ages.
When gold was used -- either leaf or ink -- standard practice was to burnish it after application; this assured its smoothness, and hence a golden reflection; if this was not done, the reflection would be yellowish but not metallic-looking. Unburnished powdered gold was sometimes used as a yellow pigment, but the wastefulness of this will surely be obvious. Burnishing in manuscripts at this time was usually done with a tooth, with the teeth of carnivores preferred because of their hardness, although in larger drawings, where the gilt area was large, a smoothed stone had to be used. Hematite was often used for this purpose.
Occasionally gold was overlaid with paint, and the paint tooled or scraped off in elaborate patterns. This was known as "sgraffito" (the general technique of mixing patterns was "damask").
Green earthFerric and ferrous oxides plus silicatesGreen Often called terra verde or terre verte, with the same meaning as "green earth." This is said to have been the most common green pigment in the middle ages; it was a by-product of iron mining. As a pigment, it was a rather dull green, of varying hue, from olive-green to apple green.
Its chemical composition varies, being mostly a mix of minerals, glauconite and celadonite. Ferrous oxides (Fe2O3; see Hematite below) and silicates seem to be the most common components. Celadonite, which is a light green, was most strongly associated with Verona and northern Italy; glauconite, which was more yellow or olive, was brought from the Czech regions. Little research seems to have been done on distinguishing where and how the various green earths were used. Being dull colors, they were more likely to be used for backgrounds than for primary features of an illustration.
HematiteFe2O3Red As a mineral, hematite is usually black, so it isn't often used as a paint, but it streaks a streak plate with red, and so is used to make red inks -- it was the usual red ink in Egyptian papyri.
Honeysuckle Green(complex organic)Green One of many greens made from plant juices, this recipe probably originated in an Arabic-speaking country, for many of its early names seem to be debasements of the Arabic word. The green probably derived from the chlorophyll in the plant's leaves. It was not very common.
Indigo(complex) BlueGreek Ινδικος (although this is a tertiary meaning; the primary meaning is "India.") One of the earliest known permanent dyes, found in both indigo plants (from Asia; the name "indigo" evidently derives from "India") and woad plants (known, e.g., in Britain), although the concentration in woad is far less than in indigo plants (which apparently led to early protectionist measures as dyers who used woad tried to block importation of indigo). The blue is the color of "blue jeans," which are colored with indigo. It was also used as a medicine, being a powerful astringent. The chemical is complex (if I counted right, it has three sodium atoms, thirty hydrogens, 35 carbons, three sulfurs, two nitrogens, and nine oxygens; there are four benzene rings, one modified benzene ring, and three NaSO3 groups). Nonetheless it has been synthesized by moderns -- there are even bacteria which have been modified to produce it. It is not as rich a blue as lapis lazuli or the copper compounds, and suffers from the fact that it is not very opaque (it is a far better dye than pigment) but was used because it was more available. Although the supply even of woad was rather limited -- it was a plant, but one that tended to use up the soil if cultivated. In an era before fertilizers were widely known, a few years of cultivating woad would leave a field no longer capable of sustaining a crop.
Interestingly, although it is a better dye than pigment, it is easier to prepare as a pigment (dying with indigo involves a complex chemical stew, and the indigo for a time is transparent; it is not obvious how dyers learned how to do it); Roman shields are said to have been colored with a paint based on powdered indigo.
For another indigo-based color, see Mixed Green.
Indian Yellowmagnesium euxanthateYellow or yellow-orange Apparently known from ancient times in India and surrounding regions. It is usually stated that it was made from the urine of cows fed on mango leaves -- a practice now outlawed as it is hard on the cows. It should be noted, however, that Victoria Finlay tried to investigate this process in the part of India where the product was made, and couldn't find anyone who knew anything about it. It did, however, smell pretty bad, so there is probably some truth to the story. The name in the Middle East seems to have been purree or something similar (puri, peori). Whatever the original source, the pigment can now be made with magnesium and euxanthic acid (C19H16O10). It reportedly did not make its way to the west until the nineteenth century. Thus its presence in a manuscript illumination of a Latin manuscript would indicate a very late date, although a Greek or Syriac manuscript might have contained it earlier.
Iris Green(complex organic)Green A rare but very attractive green, made from the juice of iris flowers plus alum. It was commonly used for manuscript illumination in the fourteenth and fifteenth centuries, but rarely if ever before that.
Iron pigments(see red ochre, yellow ochre, green ochre under Ochre)
Jadeusually jadeite, sodium aluminum silicate, NaAl(SiO3)2; sometimes nephrite, Ca2Mg5Si8O22(OH)2 (many colors, but often green) The informal name "jade" is used for two different minerals, both found in large enough blocks to be suitable for sculpting, but attractive enough to be considered a semi-precious stone. Jades range in color from white to green to black; green jade is probably the most desirable, and was used in eastern countries to produce the pigment known as "spinach green." I gather that this form was occasionally ground up to produce a green pigment, although this was more common in the east; I do not know of instances of green jade in western illuminated manuscripts. But my knowledge is far from complete!
Kermes (complex)red This pigment is believed to be the one referred to in Genesis 38:28; indeed, the word "crimson" is said to derive from kermes. The word "kermes" itself reportedly being from Sanskrit "kermidja" "(made) from a worm" -- a fitting name, because kermes was not prepared from a mineral but from the bodies of small red insects which inhabited evergreens along the Mediterranean. This or something similar can also be found under the name "cochineal," although this color (which can sometimes be more purple than red) is not very stable under light -- especially if it is not treated with alum to fix it. (Complex organics such as Kermes are rarely as stable as simply mineral dyes, and the most common uses for kermes and cochineal and other insect-based red colors in painting was in "lake" pigments where the kermes supplied the color and the lake made it opaque; such pigments were almost an invitation to fading.)
Despite its name giving rise to an English word, Kermes can't have been common outside the Mediterranean (although it is said to have been used in the Lindisfarne Gospels), because it took a tremendous number of bugs to produce a relatively small amount of pigment. (This is also true of cochineal, which came from bugs on the knawel plant. Cultivation was complex -- you had to pull up the plant, which was a perennial, during a window of about two weeks, pull off the insects, replant the plant, and then process the bugs.)
Supposedly the kermes was extracted from the insects by drowning them in vinegar or killing them with vinegar fumes. (Modern kermes insects, it is reported, are not easily killed this way; perhaps they have evolved an immunity to acetic acid.)
Interestingly, the Americas also boast a pigment, carmine, derived from the bodies of insects (both insects reportedly being of the cochineal family). In recent years it has become a major commercial product in Latin America, where there are farms of prickly pear cactus set up to support the insect. This is a much more successful industry because the carmine insect produces more of the chemical than the kermes bug. The name carmine is also derived from kermes. Carmine and kermes are not the same chemical, but they are closely related. The proper name of carmine is 7-D-glucopyranosyl-3,5,6,8-tetrahydroxy-1-methyl-9,10-dioxoanthracene-2-carboxylic acid. Not too surprisingly, this is sometimes called by the shorthand "Carminic acid." Kermes has as its active ingredient "Kermesic acid," which is chemically similar but not as complex (both have the same structure of benzene-like rings at one end, but there is a chain of carbons and hydroxyls at one end, and the chain on carminic acid is four links longer).
It will perhaps make some readers less than happy to realize that cochineal is an approved red food coloring, used e.g. in certain candies.
Other names for insect-based reds are the Latin vermiculum and English grain/greyn, which although it sounds like it refers to a cereal is in fact a name for crimson. There is at least one Middle English reference to "crimson in grain" as a dyed cloth. There is also "Polish cochineal."
All these names probably go back, in some form, to kermes, which seems to have had many names in antiquity. Because the bugs involved resembled berries, Theophrastus called it "κοκκος," which became Latin "coccus" -- a term used by Pliny, who however also called it "granum," "grain," which also went into English as another name for scarlet (the word "scarlet" itself was once a term for a cloth type that came to be applied to a color instead). Jerome called kermes insects "baca," "berry," but knowing it came from an insect, also used "vermiculum," "little worm," whence "vermilion."
Some have said that the Hebrews used Kermes to dye the curtains in the Temple. This obviously is beyond proof, but it is not unreasonable that Solomon would spring for such an expensive but excellent color. (And expensive it certainly was, since it supposedly took about 150,000 insects to produce one kilgram of the color!)
There are some amazing stories about kermes being used for tribute in Roman times, and about wild escapades by nations and individuals trying to gain a part in the kermes or cochineal trade. Those interested may refer to Victoria Finlay, Color: A Natural History of the Palette. This book has information on some of the other colors listed here as well, but the kermes story is perhaps the most adventurous....
Those who study alchemy may find references to a kermes mineral, but this is certainly not the same substance, although it is far from clear what inorganic compound is being referred to.
LampblackCarbon, C Black Pure carbon is one of the purest blacks known, and as such was used both in ink and in paint to supply blacks and mixed tones. Its use reportedly goes back to ancient Egypt, at least 1500 years B.C.E. It has always surprised me that so many inks did not use lampblack -- but poor monasteries which used mostly natural lighting might have a limited supply, and lampblack -- being a non-acidic suspension -- would not cling to the parchment as well as more acidic inks. The problem of limited supply could sometimes be solved by creating carbon by other means, e.g. bone black. Also, gum might be added to bind it to the parchment, or a gall ink might be included to burn it in to the surface of the material. Because it was thin and did not mix very well, lampblack was rarely used in mixed pigments, and only occasionally in pure blacks. It was primarily an ink, not a pigment.
Lampblack was not the only ancient form of carbon. Charcoal is also basically carbon. Many sorts of charcoal were made, from vine branches to ivory chips. Some were nearly pure carbon, others had substantial impurities. These might affect both the color (or, at least, how black the black was -- willow charcoal, for instance, was notably gray rather than black) and the graininess. Thus lampblack was clearly the best ink, but a charcoal might be the best pigment for a particular purpose.
Lapis Lazuli [lazurite: (Mg,Fe)Al2(PO4)(OH2)] and others BlueUnlike most common pigments, lapis lazuli is not a simple compound. Rather, it is composed of lazurite mixed with calcite and often small amounts of other compounds such as pyroxenes (including iron pyrite, or "Fool's Gold"). Lazurite, the key component, is a deep blue compound, sometimes called ultramarine, although the name ultramarine originally applied to lapis lazuli itself. The name "ultramarine" was given because the material had to be imported from far away, often over the sea. (It should be noted that lazurite does not refer to a specific mineral but to several closely related minerals with different amounts of sodium, sulfur, and calcium. It was not until it was artificially reproduced that lapis lazuli became a truly standardized color.) In natural lapis lazuli, the calcite generally serves to lighten the color of the rock; lapis lazuli is always blue, but how dark a blue depends on the exact nature of the mix. Sometimes an artist would go so far as to use chemical processes to purify it, although this was much more common in later eras than early; it wasn't until the late middle ages that dark blues from lapis lazuli became (relatively) common. It was not until the late 1820s that chemists found ways to create artificial lazurite.
The stone is considered a semiprecious gem, and the best source for lapis lazuli is said to be Afghanistan, so there is some geographic variation in its frequency of use -- it is said that it is more commonly used in Italian manuscripts, much less so in books from northwestern Europe. A manuscript which uses it extensively is most unlikely to be from Germany or Scandinavia, e.g. It was rare enough that there are reports of it being scraped off manuscripts for reuse. Indeed, it is said to have been the most expensive pigment known to the ancients other than gold; a writer complained in 1508 that it cost 100 florins per pound. (For this reason, there was a significant tendency to use it in conjunction with gold, to make a really expensive manuscript.) It has been hypothesized that Michelangelo's unfinished painting "The Entombement" was left incomplete because no one could find or afford the required ultramarine paint.
In addition to its rarity, lapis lazuli required very complex handling in order to extract the desired color. Modern methods involve materials such as linseed oil that the ancients probably would not have used, but they likely had some sort of grinding-and-kneading procedure to extract the color; the impurities needed to be controlled and it couldn't be ground too small or it wouldn't be as deep a blue. This was work enough that it might have added still more to the price.
Byzantine illuminations using lapis lazuli have been dated as early as the seventh century. In the late middle ages, it was largely displaced by smalt, which is a less attractive blue but was cheaper and had better hiding power.
More recent preparations from lapis lazuli would try to separate the blue crystals from the calcite and others to produce a more intense blue. This is not a good indication of date, however, because the effects will depend on just how pure the original stone was.
Modern ultramarine blue pigment, which different sources date to 1814 or 1829, is chemically identical to lapis lazuli, and will look much the same at a casual glance -- but because the lazurite is prepared artificially, the particles are much smaller than in lapis lazuli paint; microscopic examination can often distinguish them. In addition, early ultramarine often contained some sulfur which made it look somewhat more purple. Ultramarine was considered a blue for (dark) skies, as opposed to azurite (occasionally referred to as aquamarine), which was for sea blues.
Because it was so rare and expensive, ultramarine would become the standard color in medieval illuminations for the clothing of the Virgin Mary; almost all illustrations show her gowned in blue, and it was considered somewhat scandalous when a later artist showed her wearing red.
It is said that lapis lazuli would sometimes degrade to a gray and develop splotches, but no chemical mechanism has been discovered for this and it happens only rarely; I suspect some sort of impurity.
Red Lead Lead tetroxide, or Lead (II, III, IV) oxide, Pb3O4 RedThere are actually two red oxides of lead, lead oxide, PbO, or litharge, and lead tetroxide, Pb3O4, sometimes known as minium (although that name is sometimes, confusingly, used for cinnabar; see the note there. Vitruvius and Pliny called lead tetroxide sandaraca or sandarach. The name "stupium" was also used, and "miltos," plus "sinopus" by confusion with red ochre. The consensus among those who described all the colors is that "minium" should be used as the name for red lead. The name also gave rise to the verb "miniare" meaning "to work with minium" and even, less directly, to our word "miniature").
Lead tetroxide is what is usually know as red lead, and was prepared by heating white lead. Lead oxide was made by simply heating lead metal and allowing it to oxidize. I suspect, since ancient red lead is sometimes referred to as orange, that it was sometimes a mixture of oxides, since another crystaline form of PbO formed yellow crystals (see yellow lead).
Red lead was used notably in the Lindisfarne Gospels to produce a sort of red shadow for the outlines of letters. It is said that over 10,000 dots of red lead were used on one of the pages of that gospel (beginning of Luke, where every letter is outlined in two or more rows of dots). It was known as a poison at least from the time of Nicander of Colophon (second century B.C.E.). For some reason -- perhaps because of its dangers -- use of red lead tended to decline over time, with vermillion becoming a more popular replacement. If you ignore its poisonous aspects, it is an excellent pigment, with a clear red color, good hiding power, and small particle sizes that make it easy to mix and apply. As a result, in addition to its use as a pigment, it has sometimes been used as a primer, binding to the page (or other materials such as metal), with a second, harder-to-appy, pigment being laid over it. The Romans seem to have used it extensively as a wall paint, which can't have helped their health. (In medieval times, litharge was actually used as a medicine, applied to England's King Henry IV. He was already sick, which is why they applied it, but perhaps it's not surprising that he ended up dying young!
Red lead, like white lead, could turn brown or black when exposed to air; this was particularly a problem in frescoes, but it happened in books, too, unless the lead was varnished or otherwise protected.
White Lead Lead acetate: lead (II) ethanoate, Pb(CH3COO)2 WhiteGreek ψιμυθιον. The preparation of white lead is described above. It is used both as a white paint and as a whitener (that is, it would lighten the shade of other paints). Unfortunately, it decays over time to toward dark shades (the result of exposure to hydrogen sulfide), producing artwork which often looks very strange (since, the lighter the original color, the darker the final shade). It was sometimes referred as ceruse. It was known as a poison at least from the time of Nicander of Colophon (second century B.C.E.). Modern paints based on white lead often contained mixtures of compounds, perhaps 70% white lead and 30% lead hydrate. White lead has another interesting use: It can sometimes by employed to reveal when a painting has been corrected. It was the original "white-out," used to paint over a mistake. And, because lead is very heavy and stops x-rays, these corrections will show up as a blank area if a painting is x-rayed. Titian's "The Death of Actaeon," for instance, shows that the artist re-did the image of Actaeon himself. Whether there are manuscripts which have been re-painted enough for this to matter I do not know.
White lead's opacity -- which made it a good correcting medium -- was another reason it was so widely used; the other whites (bone white and chalk) were translucent. It was not until lake pigments came along (which consist of an opaque material with little color with which a translucent pigment could be mixed) that there was a good alternative to white lead for opaque whites.
White lead also had the advantage of being cheap -- one of the cheapest pigments available; there is a price list showing it costing only one-one hundredth as much as azurite.
Yellow Lead Lead chromate: PbCrO4 YellowAs with red lead, there are two compounds which might be known as yellow lead. Today the term seems to be used primarily for lead chromate, or chrome yellow. I suspect, however, that the usual form known to the ancients was lead oxide, PbO. (Supposedly chrome yellow and the related chrome orange were not manufactured until the early nineteenth century. Chrome yellow was actually a mix of lead chromate and lead sulfate, with the shade of yellow determined by the mix.) Note that this lead oxide is chemically identical to the lead oxide form of red lead. The difference is in the structure of the crystals, with one form being converted to the other by heat. This would theoretically allow the possibility that red pigments would turn yellow over time, but since the heat of conversion is in excess of 400° C, this would not have happened extensively. Yellow lead oxide was apparently known as massicot, and perhaps as giallulinum.
Red Madder
complex organicred Derived from the root of the madder plant, Rubea tinctorum. It was used primarily as a dye, not a pigment, but sometimes could be mixed with other colors. One of the chief difficulties with madder was that it took a lot of processing to convert the basic plant matter into a useful colorant. The active chemical was alizarene (called "alazarin" when produced by industrial rather than natural means, as is usual today), C14H8O4, an orange-red. Alizaren was the first natural pigment to be artificially replicated, so its use in a drawing, while not proof of a forgery, is indicative. The color is not likely to be used in an early Western manuscript since it was not widely used until the late Middle Ages (Europeans were said to have learned of it after the Crusades); its heyday was the seventeenth to nineteenth centuries. It was used far earlier in the east. It has some tendency to fade if exposed to sunlight.
Malachite Cu2CO3(OH)2
copper carbonate hydroxide
greenFrequently found in the same deposits as azurite (which see), but more common. The properties are similar -- e.g. it is hard and requires much grinding to use. It is, however, more chemically stable -- azurite sometimes turns to malachite, causing blue pigments to turn green, but the reverse does not happen. A more modern formulation has been sold under the name "mountain green," but it is no longer used as a pigment in the west because it does not work well with oil. As a pigment, it had the difficult characteristic that it could not be ground too finely, because it lost its color. But it was so common that it appears the Egyptian word for green, vatch, is the same as the word for malachite. See also verdigris.
Massicot(See yellow lead)
Minium(See red lead; also cinnabar)
Naples Yellow Lead antimoniate:
Orange-Yellow A synthetic yellow first found in the late middle ages, and used primarily in pottery glazes rather than paint. But it was known early enough that it is possible it would have been found in some late manuscript illuminations -- the Chaldeans used it as a pottery glaze during the time of the Babylonian Empire, and the Egyptians also knew how to create it, but it does not seem to have been common as a pigment until much later. (The way the Babylonians made it is unknown; no other ancient people seems to have adopted the method.) The eye pigment known as khol is also antimony-based, but does not seem to have been used in manuscript illustrations.
Mixed Green(complex)Green There were many combination greens used by medieval artists. But, for some reason, the combination of orpiment (yellow) and indigo/woad (dark blue) was so popular that it was often prepared as a distinct color rather than being mixed on the spot. So it perhaps should be considered its own pigment.
Mosaic Gold(tin sulphide)Yellow The name is a rendering of Latin "aurum musicum." It is yellow but does not really resemble gold; some versions look more like bronze. It is apparently not entirely clear how it was first synthesized (the preparation is complex, requiring tin and sulphur plus catalysts and heat and pressure); it seems to have come into use in the thirteen century. It does not seem to have been widely used in practice although there are several manuscript descriptions of it.
Nightshade Green(complex organic)Green One of many greens made from plant juices. It was not very common, and seems to have originated in the thirteenth century. The green was probably from the chlorophyll in the plant.
Green Ochreimpure Iron (III) oxide hydrate(green) Although one sometimes finds references to green ochre, this is not a pure substance; it is yellow ochre with impurities, typically magnesium or aluminum cilicates. See also "Green Earth."
Red Ochreanhydrous iron (III) oxide: Fe2O3 red/brownSee under yellow ochre.
Yellow Ochre Iron (III) oxide hydrate: Fe2O3 • H2O Yellow Yellow ochre, and the related red ochre (the anhydrous form) are among the oldest pigments known to humanity; there are neandertal grave sites with sprinklings of red ochre, and others with red ochre stones -- it is widely suggested that the bodies of the dead were coated with ochre, and the stones used as grave goods, although this is controversial. It is certain that red ochre is still used as a body pigment today by some peoples, and was valued by some tribes of Australian aborigines because of its use in drawing. It is also sold as red chalk. Red ochre, and other red iron oxides, have long been used as pigments, e.g. "Venetian red," as initially sold, was an iron oxide pigment.
The more common form of ochre is, however, yellow ochre, which is the hydrate, sometimes known as limonite; red ochre will in time turn to yellow in the presence of water, so red ochre is found mostly in dry environments.
Both forms of ochre are used as pigments; yellow ochre has the advantage over orpiment (another common yellow pigment) of being much safer to handle. But the color is not as brilliant, which is why orpiment is also used; for small drawings, yellow ochre just wasn't yellow enough to give a clear color.
Red ochre, which is pure iron (III) oxide (Fe2O3) can be prepared from yellow by baking the water out of it. It is a reddish-brown color, also used in pigments. For brighter reds, however, something such as cinnabar must be used.
The modern colors raw sienna, burnt sienna (both named after the Italian town of Sienna where they came into widespread use), raw umber, and burnt umber are also based on iron compounds and are related to the ochres -- but they were not used in early paintings, which tended to avoid relatively dull colors such as browns; their use would be a likely indication of forgery.
Modern, synthetic ochres are apparently sold as "Mars pigments."
Although ochres are common, the impurities found in many of them made them rather poor pigments, because they lost much of their color when ground. So high-quality ochre was a special commodity. The best came from Sinope in Pontus, and so ochres were sometimes called "Sinopia."
Ochre has another significance to manuscript historians, in that (under extremely limited circumstances) it can be used as a dating method. (This works better for murals and other artwork with a permanent orientation.) As an iron compound, ochre responds to magnetism, and when freshly applied, the ochre will align itself with the north and south magnetic poles -- then will freeze in place as the substrate hardens. Since the magnetic poles wander about, the way the ochre points can sometimes be used as a dating method.
Orpiment Arsenic (III) sulfide: As2S3 yellow Known in Greek as αρσενικον, whence the modern chemical name "arsenic" -- a name thought to be derived ultimately from Old Persian. Probably the brightest and clearest yellow pigment known to the ancients -- so much so that the Roman Emperor Caligula allegedly tried to organize a project to turn it into gold; in later times it was called King's Yellow. The English name is said to be a distortion of Latin auri pigmentum, i.e. gold paint. Early orpiments were natural, but alchemists eventually synthesized it. We find early mentions of it in Egypt (Leiden Papyrus, third century C.E.) and Greece (Democritus, second century B.C.E.). It is frequently found with realgar, another arsenic compound, with a reddish tinge; mixtures might appear orange. Despite orpiment's brilliant color, it was somewhat hard to work with; it did not mix well with other colors (e.g. it hastened the process by which white lead turned dark, and could also cause other pigments such as folium and red lead to turn), and of course as an arsenic compound, it was fairly poisonous. The compound was found in Macedonia, Asia Minor, and Hungary, so it perhaps would be more common in eastern than western manuscripts.
Orpiment is not entirely stable if exposed to oxygen (the arsenic sulfide slowly turns to arsenic oxide), and if so exposed, it not only loses its color but ceases to attach to the page, so it is another pigment that might vanish in the course of time; some say it could also damage the parchment as this happened.
Alchemists sometimes referred to the arsenic compounds orpiment and realgar as the "two brothers," "two kings," or "two friends."
For another orpiment-based color, see Mixed Green.
Realgar Arsenic sulfide:
orange-red Chemically similar to, and often found with, orpiment, another arsenic sulfide. The name is said to be from Arabic Rahj al ghar. It seems to have been used primarily if not exclusively in the eastern Mediterranean and points east of there. It is very rare in manuscript paintings, although it was sometimes used to preserve glair. In the east, it is used in wall paintings as well as illustration. Like orpiment, it is poisonous -- if anything, even more so; it has been called the most poisonous of all ancient pigments. It is sometimes known as ruby sulfur, and was called sandaraca by Pliny, although this name is more typically used for red lead. Realgar is not very stable, especially under bright light; it will decay into yellow orpiment, which is why most instances of realgar paint look orange rather than bright red.
Alchemists sometimes referred to the arsenic compounds orpiment and realgar as the "two brothers," "two kings," or "two friends."
Pliny called realgar by the name "sandarach."
Rouen Green(See verdigris)
Saffron complex organic (C44H64O24) yellow A very delicate yellow, but chemically complex (the chemical diagram of the molecule takes up a whole page, and I won't swear I counted its components correctly); the ancients could not synthesize it, and had to rely on saffron plants. So it wasn't often used due to the high expense. (The pigment, which is also a spice, comes from a very small part of the flower; it takes great numbers to make a usable quantity of saffron -- supposedly 170,000 flowers to yield one kilogram. And, even in modern times, the stigma have to be separated out by hand, and even champions can only pick about one every two seconds. Plus the flower blooms only briefly, and withers if not processed quickly, making it almost impossible to mass-harvest it). But the fact that it was so expensive means that it is more common in manuscript illuminations (which were small, and given to rich patrons) than in paintings.
Saffron eventually came to be farmed in large areas of Europe, leading at times to collapses in the market. I suspect that it would be possible, if a sufficiently detailed chemical analysis were done, to tell classic oriental saffron from more recent European saffron -- there were at least three species of plant called saffron, with the "oriental" saffron considered to be the best. But I know of no work being done on this subject.
Salt Green(See verdigris)
Sap Greencomplex organicGreen Just what it sounds like: A green made from plant juice -- although not actually from sap; it was made from ripe berries of buckthorn plants. It was made by taking the juice of the berries, mixing with alum, and allowing to thicken. Since there were several species of buckthorn, the results varied somewhat, but the color was generally an olive green. It might be mixed with gum, but it was not rare for the dried juice -- which was thicker than most plant juices -- to be used as a paint without any binder. (Occasionally the juice was used without alum, but this color faded very quickly. Even with alum, it was likely to fade over time.) Today it is used mostly in watercolors.
Sepia complex organicbrown/black This is, to moderns, a confusing name, because sepia has become a name for a color rather than a pigment (e.g. we see "sepiatone" photos). The original sepia was derived from the "ink" of the squid, which it uses as a smoke screen to escape danger. The ink was collected and made into -- ink. Although squid ink appears black in water, when used on paper it usually appears brown, and rather transparent. As a paint, it is nearly useless, because it is not opaque, but some manuscripts are written with it, and some drawings sketched with it.
Shell Whitescomplex organicWhite Made from ground-up egg shells (dried in a flame) or oyster shells. It appears these colors were not used as pure whites, but were used to lighten other colors such as verdigris or orpiment/blue mixes.
SilverAgSilvery Silver, like gold, was applied either as a thin sheet or ground up to use as an ink. The ink form was more common, since it did not tarnish quite as rapidly as silver leaf, but tarnish was a problem with all uses of silver in manuscripts. Sometimes silver leaf was lacquered with yellow to try to imitate the appearance of gold.
Smalt(see smalt in the section on Chemicals Not Found in Ancient Manuscripts)
Spinach Green(See jade)
terra verde(See green earth)
TinSbMetallic Occasionally used as a foil or a ground ink component, presumably because it was cheaper than gold and did not tarnish as much as silver. Sometimes tin leaf was lacquered with yellow to try to imitate the appearance of gold, or a mix of powdered tin and a yellow would make a yellow metallic ink. A metallic green might be produced by covering tin with verdigris; a red glaze might also be used.
Turnsole complex organicblue or purple Made from seeds of plants in the Crozophora family, which had to be processed in complicated ways. Called "turnsole" because the plant turns toward the sun like a sunflower. It is a litmus-like chemical, blue, purple, or reddish depending on how much acid is mixed with it. (Litmus too being plant-based -- it comes from lichen.) The blue color of turnsole came out when mixed with alkali. It had some tendency to revert to purple over time, but rarely to go back to red; that required stronger acids than were common in ancient times. It was used primarily in manuscripts, particularly for a relatively light, transparent blue. The plant that gave rise to it was known by many names, such as "torna-ad-solum" or "morella;" also "folium." It became popular in the fourteenth century.
Tyrian Purple complex organicpurple This is a dye, not a pigment; it was derived from mollusk shells, and was used to supply the purple color in the togas of Roman senators and emperors. It was very expensive, and the limited supply was largely reserved for clothing (for the logical reason that it was one of the few reasonably permanent dyes known to the ancients). Even had the supply been greater, it is unlikely that it would have been used in paintings, because (like most dyes) it had little hiding power. But it has importance for students of manuscripts anyway -- it was the dye that made purple manuscripts purple. See also Whelk Red.
Ultramarine(see Lapis Lazuli)
Verdigris (various) green Verdigris (vert de Grece) is not the name of a particular chemical; it is what we call the green patina of reacted copper. There were at least three compounds called verdigris. The most common was probably copper (II) carbonate, CuCO3. This is the patina that usually forms on copper; it might hydrate to form malachite. Near the seaside, however, or where there is another source of chlorine, the patina might be primarily copper chloride, CuCl or CuCl2. These were not green, but the latter would hydrate to become copper (II) chloride dyhydrate (CuCl2•2H2O), which is blue-green. Finally, there is copper acetate, Cu(C2H3O2)2, which is the one of these which could be made artificially and quickly, by exposing metallic copper to vinegar fumes or hanging it over the lees of wine; Theophrastus and Pliny both describe how to make it, and Pliny called it "aerugo." Alchemists sometimes referred to it as "Spanish green." It is possible that there were other verdigrises as well, copper tartrate or copper malate (the latter might form if apple vinegar was used to make the verdigris).
All these forms of verdigris formed brilliant greens, but whether they were light-fast depended on the paint substrate. Da Vinci noted that it had to be varnished quickly if the color was to hold. It is said that no color, not even white lead, has deteriorated more in medieval paintings than verdigris. Verdigris is said to be stable in oil, but far less so in other media (so some sources; others think it was less stable in oil than in tempera, but the general sense is that it had to be guarded from air with varnish or something, meaning that it was not a good choice for manuscript illumination), although this would depend on the exact formulation used. Thus we sometimes find pigments which should have been green have turned to brown or near-black over time. (Ironically, in oil, verdigris was not very opaque and had to be mixed with other colors to improve its appearance. That, combined with its color instability, seems to have led to a rapid decline in its use once oil painting came in.) A second difficulty is that it could not be used with white lead; the two reacted quickly to destroy the colors of each. (At least, that was what was said at the time, although modern chemists can't figure out why this would be so.) Orpiment didn't mix well with verdigris either. A third is that, over time, verdigris could damage the parchment below it, leaving paintings with holes where green pigment would be expected -- although a small amount of saffron in the green supposedly could stabilize this.
It is said that verdigris in manuscripts has held up better than paintings, perhaps because it hasn't been exposed as much. Verdigris was often used to make illuminated initials in manuscripts.
Alchemists seem to have referred to verdigris as "Seed of Venus." I know of no such references among artists.
Sometimes a verdigris mix would be known under another name. "Rouen Green" was a name for verdigris made by coating the copper with soap before exposing it to vinegar; "Salt Green" was from copper coated with honey and salt before exposure.
Vermillion(see cinnabar)
Weldcomplex organicYellow The English name of the Reseda luteola, related to a garden plant called the mignonette. Mostly used as a yellow dye -- e.g. it was usually the color mixed with woad to produce the Lincoln Green you hear about in Robin Hood stories. Very occasionally used as a yellow ink, or mixed with a blue to produce a green color. However, the color was unstable, so it was rarely used in manuscripts and or paintings. It eventually came to be used in yellow "lake" paints, but this was mostly after the manuscript period; weld does not seem to have been commonly used until the fourteenth century, except as a dye (it was sometimes called the "dyer's herb" or "fuller's herb") -- although it was still used in that role until the twentieth century.
Whelk Red complex organicpurplish red A close relative of Tyrian Purple, to which it is similar in both source and method of preparation. Indeed, because dark reds were called purple in ancient times, whelk red is a purple, it's just not the purple (although the best purple was made of a mixture of chemicals from two different mollusks, so one might say that Whelk Red is part of Tyrian Purple.) Like purple, whelk red is a dye, not a pigment, made from mollusks (especially whelks, obviously). Its advantage over Tyrian Purple was that the shellfish involved were more common and widely distributed, making it somewhat cheaper than the true purple (although still very expensive). It was thus more likely to be used to dye manuscript pages. But it was also very variable (from a relatively pale magenta-like color to a darker purplish-red), presumably based on the exact source, so we often see (reddish)-purple manuscripts which show great variation in the color of the pages. See also Tyrian Purple.
Wine black complex organicblue-black Vitruvius reported making a black by burning dried dregs of win, and said that the finer vintages could also give a color reminiscent of indigo. Presumably the black is mostly from carbon, but I know of no details. If wine black was used for manuscript illumination, I have not heard of it.

Chemicals Not Found in Ancient Manuscripts

If the chemicals listed above can be shown to be ancient, certain pigments were notinvented until after the manuscript era closed. If these colors turn up in a manuscript,the manuscript must be a recent forgery.

As with the list of pigments above,this is not an attempt to list every color created since the manuscriptera ended. Indeed, such an attempt would be misleading, because an individual artistmight have created a new color, or extracted one somewhere. The pigments listed hereare classes of pigments, identifiable by spectroscope, which are most unlikely tohave been used in early centuries because they are difficult or dangerous to make andunlikely to be found in isolation.

To put all this in perspective, a standard Impressionist paint palette of 20 colors included all of the following (those not known in ancient times marked *; those known only in the late Middle Ages marked **): *zinc white, lead white, *lemon yellow, *chrome yellow, *cadmium yellow, *Naples yellow, yellow ochre, *chrome orange, vermillion, red ochre, **madder lake, cochineal lake, *Scheele's green, *emerald green, *viridian, *chrome green, *cerulean blue, *cobalt blue, ultramarine (artificial), bone black. Thus, of these twenty, fully twelve were not available to manuscript painters. To be sure, the Impressionists generally weren't trying to fake Biblical manuscripts -- but the list shows how hard it would be to do if one weren't very careful. A nineteenth century fake, if it contains illustrations, will almost certainly be revealed by spectroscopy.

Nor is spectroscopy the only way to detect modern pigments. Even such a simple technique as microscopy can be quite useful -- e.g. it can detect the difference between ancient and modern ultramarine. In modern ultramarine, the lazurite particles are of a uniform size; in early ultramarine, no matter how carefully ground, there will be substantial variation in particle size.

Microscopy can sometimes give slight hints about the origin of a drawing, too, by giving us a look at contaminants (dust, etc.) in a painting -- e.g. there is sand stuck in the paint of some of Monet's beach scenes, so he presumably painted the pictures "on site."

Carbon Dating

The sciences have, over the last half century, given us many new ways to dateearly objects. The methods vary widely in both their accuracy and their side effects(e.g. electron spin resonance is largely non-destructive, but can be performed onlyonce), but the earliest and the best-known remain the methods based on radioactive decay.

The principle of radioactive dating is this: If you have a radioactive isotope, itdecays at a fixed fractional rate rate. (If you don't know what an isotope is, seethe section on isotope analysis.)If 20% of the original sample has decayed after athousand years, then in the thousand years after that, 20% of what remains will havedecayed (meaning that 36% will have decayed in that time), and 20% of the remainder afteranother thousand years (meaning that 48.8% will be gone, and 51.2% remaining).This is why we speak of radioactive half-lives: It is convenient to describe the time it takesfor exactly half of a sample to decay.

The general formula for radioactive decay is

N = N0eγt

Where N0 is the number of atoms of the material you start with, N is thenumber you still have after time t, and γ is the so-called decay constant, ameasure of the rate at which the isotope undergoes radioactive decay. A little algebraicmanipulation will show that the half life h is therefore given by

h = -ln(0.5)/γ

Or equivalently that


(With appropriate units, of course.)

Note what this means: If you have a sample of something containing a radioactiveelement, and seal it up for some period, you can determine how long it was sealed bytaking the ratio of the element and its by-products.

Alternately, if you have a sample which started with multiple isotopes of the sameelement, some stable and some radioactive, and you know the initial relative quantitiesof the isotopes, you can seal it up and wait for some years and again compare theratios, and on this basis determine how much of the radioactive isotope has decayed,and on this basis you can determine how long it was sealed.

There are many of these "atomic clocks." A popular one is potassium-40 andargon-40. Potassium-40 has a half-life of 1.248x109 years -- that is, one anda quarter billion/milliard years. It is thus very good for dating ancient rocks, sinceeven the oldest rocks still have a substantial fraction of their initial Potassium-40.

There are difficulties, however. Radioactive dating is only accurate to within about 5%of a half-life (sometimes more, sometimes less, depending on a lot of things including thesize of the sample. Take it as a rule of thumb). For potassium-40, that means a dating errorof ± 60 million years. That's no help dating a manuscript that was written some timebetween 100 C.E. and 1900 C.E.!

Hence the need for shorter clocks. The half-life of carbon-14 -- radiocarbon -- is 5715 years,or alternately γ is -0.000121. And that is a short enough period to allow usefuldatings of almost any product of human civilization -- it was used, for instance, to demonstratethe the Shroud of Turin was from the medieval era, not the New Testament era. 5% of 5715years is about 275 years, so we can date any object made within the last 30,000 years orso with an accuracy of ± 150 years. Sometimes less than that, with the latesttechniques and a sufficient amount of source material.

Carbon-14 is formed in the atmosphere when nitrogen-14 is hit by cosmic rays, causing oneof the protons in the nitrogen atom to turn to carbon. The total carbon produced this way isestimated at seven killograms per year. That's not a huge amount, but at any given time itmeans that about 40 metric tons of carbon-14 is in circulation -- most of it as a chemicalcomponent of living things, where carbon is absolutely essential. And this is the onlysource of carbon-14; it cannot be found in rocks or anything that is not derived fromthe atmosphere, because all the carbon-14 inside the earth has long since decayed.

The key fact which follows from thisis that plants and animals only soak up carbon-14 from the atmosphere, orfrom other living things, for as long as they are alive. Once they die, the carbon-14 supplyis cut off. From then on, the quantity of carbon-14 can only decline, as individual atomsdecay back into nitrogen.

The rest of the carbon in the dead material is non-radioactive. It sticks around forever.So the age of a particular organic material can be dated by comparing the ratio of carbon-14atoms to the atoms of stable carbon-12 and carbon-13. (We should note that the originalmethod was developed by Willard F. Libby, and that it won him the 1960 Nobel prize,although the refinements made since are in manyways more important than Libby's original invention.)

Unfortunately, for the most part, testing for carbon-14 requires destroying a sample. Fortunately, the testshave improved dramatically over the years, giving greater accuracy while requiring lessmaterial. Today, if the object is a few thousand years old or less, a mere sliver of materialcan give a date within a few hundred years. Older materials are harder to test, because thenumber of carbon-14 atoms will be very small; the upper limit on dating is somewhere around30,000 years, and even that probably would require more material than we would like to spare.

For all its limitations, carbon dating has the tremendous advantage of being a dating schemethat is objective and (relatively) repeatable. It seems to me that it would be a great boonto textual scholars if some of the more important manuscripts were tested and dated, givingus a check on paleographic dating.

And, once in a great while, it can be done non-destructively -- if the sample is of known chemical composition, so that the expected amount of carbon is known. If that is the case, it is sometimes possible simply to count radioactive decays to know the amount ofcarbon-14 in the sample. It's just that this cannot be relied upon.


This is far too complex a subject to cover in detail, but it is a very powerful toolnow becoming available to textual scholars. Behnam Sadeghi, for instance, was able to usethe facilities at SLAC to determine much useful information about the copy of the Quranknown as San'ã' 1.

To vastly oversimplify, spectroscopy consists of shining a light on something and seeingwhat reflects back (or, if it emits light by itself, looking at the nature of that light).All elements and compounds have their characteristic spectrum -- thewavelengths of light they absorb and emit.

The reason for this was not known at the time spectroscopy was discovered, but itturns out to have to do with the energy of electrons. The rules of quantum mechanicsmean that an electron in an atom or molecule can possess only certain amounts of energy.It's like a ladder: You can only stand in the places where there are rungs. If the stepsof the ladder are 20 centimeters apart, you can't go ten centimeters up the ladder -- thereis no rung there. If you throw energy at an electron, it won't do anything until you giveit enough energy to move a step up the ladder, at which point -- pop! -- it instantly movesup a rung. Since this stepping-up always requires the same amount of energy, the electrons of a purechemical always absorb light of exactly the same color (since the color of light tells youjust how much energy is in the photons that make up the light).

And electrons don't like to stay high on the ladder. They have a strong tendency, afterbeing excited to the higher energy level, to give back the energy and return to the"ground" state. (The fact that physicists call it the "ground" stateshows how close is the analogy to a ladder.) When it gives up the energy, the electronemits a photon of light which has exactly the amount of energy it absorbed to moveup the ladder.

So, for instance, if you shine a white light (which contains photons of all energies)through a sample of sodium, the sodium will capture two different colors of yellow lightand leave all the rest alone. If you scatter the white light through a prism, you willsee a rainbow spectrum like this one:

White Light Spectrum

But if, before you pass it through the prism, you expose the light to sodium, whichabsorbs two wavelengths of orangish-yellow light, you will instead see this:

Sodium Spectrum

Note the two dark lines in the yellow region. This is the light that has beenabsorbed by the sodium. That pair of dark lines is unique to sodium; if you see thosebands in a sample of white light, you know it has been influenced by sodium.

Because every chemical has its unique spectrum, spectroscopy is an amazingly powerfultool. In the nineteenth century, e.g., it was used to identify the element helium inthe sun -- an element which was not discovered on earth until later. In the earlytwentieth century, spectroscopy allowed us to discover the expansion of the universe.The spectroscope has proved one of the most important scientific tools in the historyof chemistry and astronomy.

And that was with primitive spectroscopes. The equipment today is much better. Wecan (non-destructively) scan the ink used to write a manuscript. We can identify stains.With sufficiently high-quality equipment, we can even look at what lies under, say, apainting (this was done, e.g., with the Archimedes Palimsest).

Unless a manuscript is particularly important, it probably isn't worth going overevery stain and smudge to determine its chemical composition -- especially since thestains may well be later than the manuscript. But testing the ink of the original scribecould be informative. If it shows a signature of an unusual chemical, it might helpus localize the manuscript. We might also be able to work on the dating of variousink formulations.

Spectroscopy is also good at detecting forgeries, by looking at the materials inthe ink on a document. (See the section on Paints and Pigmentsfor how this has been used, e.g., with Prussian Blue.) These techniques are currentlybeing used on the small fragment known as "The Gospel of Jesus's Wife,"widely suspected of being a forgery, although as of this writing, the results havebeen inconclusive.

Among the new techniques of spectroscopy are:

Isotope Analysis

This is a relatively new technique for dating and (more importantly) locatingmanuscripts, although (like carbon dating) it is destructive.

As you probably know, an atom consists of a nucleus comprised of protons and neutrons(themselves made of quarks, but that need not detain us), circled by electrons. Theelectrons are what produces chemical behavior, and the number of protons in the nucleusdetermine how many electons an atom "wants" to have. So the number of protonsin the nucleus determines the element to which the atom belongs.

The number of neutrons, as far as chemical behavior is concerned,is irrelevent. A carbon atom has sixprotons. Most carbon atoms have six neutrons as well, but we find atoms with sevenneutrons, or even eight -- the version with eight neutrons is the carbon-14 used incarbon dating. Different atoms with the same number of protons but differentnumbers of neutrons are called isotopes, so-called because they're chemically thesame but structurally different.

The number of neutrons does not affect the chemical behavior in any way. Butneutrons have mass -- the isotope of carbon with six protons and six neutronsis lighter than the one with six protons and eight neutrons. This means that youcan separate heavier from lighter isotopes. The typical method of doing this is themass spectrograph or the centrifuge -- you take the atoms and, in effect, give thema push. The light ones will fly a little farther than the heavy ones. By countinghow many go a long way, and how many travel only a relatively short distance, youcan tell the ration of heavy to light isotopes.

This is basically the method used to create nuclear weapons by separating U-235(which is usable in bombs) from U-238 (which does not fission). However, separatingU-235 from U-238 is not a very efficient process. In enriching uranium, centrifugeswork on a compound known as uranium hexaflouride, UF6. The molecular massof UF6 349 if it has an atom of U-235, 352 if it has an atom of U-238.That's less than a 1% difference, and the centrifuging is a slow process that must bedone repeatedly to purify the U-235.

Isotope analysis is different. The usual method involves oxygen, particularlyisotopes O-16 and O-18, often in molecules of water. A molecule of water with in whichthe oxygen atom is O-16 has a molecular mass of 18 units, one based on O-18 has amass of 20 units -- a 10% difference. This is much, much easier to measure.

This technique is useful because climate affects the mix of isotopes. Water based onO-18 tends to sink lower than that based on O-16. The two may also form ice at differentrates. Based on facts such as these, one can sometimes use isotope analysis to determinethe date or location in which a material originated. I know this was used at least onceto determine that the parchment in a manuscript came from the Mediterranean basin.

Detecting Forged Manuscripts

There are many ways in which a textual scholar can detect a forged manuscript.Not all are based on science, but some are. The list below attempts to catalog most ofthe more common methods available to a scientific manuscript detective:

Another point to keep in mind: Modern painters buy manufactured paints; they go toart supply stores and buy a tube of Cerulian Blue or Cadmium Red or whatever. The squeezetube wasn't even invented until 1841; until that time, paints were usually stored in pigbladders, and usually had to be used quickly before they dried out. Duringthe manuscript period, artists rarely were able to purchase finished paints.At best, an apothecary would have the purifiedsource materials, which had to be mixed with a substrate. More often, the artist wentout and collected the materials himself, and ground and mixed them. Many of theformulae were secrets, which the artist kept to himself or passed on only to hisapprentices. These early formulations will not be as consistent as modern pigments.This has many implications which might be used to detect forgery apart from thechemichal hints above.