by Ross King
Leonardo was fascinated by the possibilities of working with oil paint. His notebooks contain numerous references to different types of oils. He seems to have experimented with making oils from walnuts, linseeds, mustard seeds, juniper needles, and the gum of cypress trees. His studio must at times have resembled an apothecary’s shop or alchemist’s laboratory, with oils oozing from presses and bubbling in pots. He even tested out ways of getting rid of the “evil smell” of certain nut oils by mixing them with vinegar and reducing them over a flame. He had no end of advice on the subject: “And if you want the oil to be good and not to thicken,” reads one of the instructions in his treatise for painters, “put into it a little camphor melted over a slow fire and mix it well with the oil and it will never harden.”12
Compared with egg tempera, oil paint was still, relatively speaking, a new and somewhat experimental technique. Vasari mistakenly claimed that oil painting was invented in the first half of the fifteenth century, but in fact the process of using oil as a vehicle for pigment was described as early as the twelfth century in a Latin treatise by a German monk known only by the pseudonym Theophilus Presbyter. However, painting in oil achieved no widespread currency until improvements in the technique were made in Flanders in about 1410 by Hubert van Eyck. The court painter to the duke of Burgundy, van Eyck perfected a method of suspending his pigments in a mixture of linseed and nut oil blended with resins and lavender oil. His renown for this innovation was such that the bones of his arm and hand were preserved like relics near the church in Ghent where he was buried beneath the sobering inscription: “I was called Hubert van Eyck. I am now food for worms.”13
Making generalizations about medieval Italian painting is difficult due to the fact that fewer than 5 percent of all works painted in the fourteenth century have survived.14 However, anecdotal evidence suggests that oil painting was known in Italy at least since the early fourteenth century. In 1325, a pupil of Giotto named Giorgio d’Aquila was provided with a supply of nut oil when the duke of Savoy hired him to paint a chapel at Pinarolo—but when the oil failed to meet Giorgio’s standards, he sent it to the duke’s kitchen to be used for cooking.15 Van Eyck’s innovations entered Italy by means of Antonello da Messina, who worked for a time in Bruges, and whose tomb in Venice (where he died in 1479) proudly recorded that “he was the first who conferred splendour and durability on Italian painting by the mixture of colours with oil.”16 Antonello supposedly passed the secret to Domenico Veneziano, who, according to one of Vasari’s taller tales, was murdered by a rival—none other than Andrea del Castagno—who was jealous of his painterly feats. In Vasari’s version, Castagno gained Veneziano’s confidence by entertaining him with carefree evenings of serenading pretty girls with lutes. After Veneziano finally divulged the secret of oils, Castagno beat him over the head with a lead pipe.17
Alas for the story, Castagno died of the plague in about 1457, several years before Veneziano. There is probably no more truth in the story that Giovanni Bellini tricked Antonello into revealing the secret by turning up at his studio disguised as a Venetian nobleman who wished to have his portrait painted, then watched carefully as Antonello went to work.18 However apocryphal these stories, there is no doubt that throughout much of the fifteenth century the technique of mixing pigments with oils—of finding both the right oils and the right mixture—was a jealously guarded workshop secret.
Verrocchio, who worked only in tempera, did not concern himself with this newer technique. Leonardo, though, was interested in oil paint even as an apprentice. The face and curly head of the kneeling angel in Verrocchio’s Baptism of Christ—widely accepted as Leonardo’s handiwork—were done in oil rather than tempera. For his portrait of Ginevra de’ Benci, done in the mid-1470s, he used both oil and tempera paints, but a decade and a half later, in the portrait of Cecilia Gallerani, he used oils exclusively. The superiority of oils for capturing certain optical effects, together with Leonardo’s proficiency in the technique, appears to have been recognized by the members of the Confraternity of the Immaculate Conception, whose 1483 contract with Leonardo and the de Predis brothers repeatedly stressed that both the figures and the landscape background of the altarpiece should be done in oils.
Oil paint had a number of distinct advantages over egg tempera. Most obviously, pigments blended with oil looked richer and glossier than those mixed with tempera. Oils allowed an artist a greater range of color values and tonal gradations because they could be blended either on the palette or (as Leonardo’s smudging fingerprints attest) on the pictorial surface itself. An artist could also vary the ratio of oil to pigment, resulting in a wide range of consistencies, anything from thin, translucent layers of color to dense impastos. He could heat his oil before the pigment was added to create a more viscous paint, or he could reduce the viscosity by adding camphor or other oils. Unlike tempera, oils did not change color as they dried. They also dried more slowly, permitting the artist to rework his painting by making as many corrections as he might desire—one of the many other properties that undoubtedly appealed to Leonardo.
As the legends of espionage and murder suggest, a painter working in oils required a great amount of tradecraft. He needed to understand the specific properties of his various pigments: which ones worked best in oil and which in tempera. He also needed to decide which oil—linseed, walnut, poppy—to mix his pigment with, and whether he wanted to heat the oil over a flame or by standing it in the sun. To these ends Leonardo, with his pestles, mortars, and bubbling pots, conducted his various experiments.
Painting with oil had allowed Leonardo to capture the startling visual effects that were winning him his reputation as a painter: the moody half lights and misty atmosphere of The Virgin of the Rocks, the soft-focus facial expressions in his portraits of Ginevra de’ Benci and Cecilia Gallerani. His aptitude and experience in oil rather than fresco led him to consider painting his Last Supper in a technique different from the usual one: that is, by working with oils on a dry wall.
This method was new and largely untested, though its effectiveness had been affirmed by the architect and polymath Leon Battista Alberti. In his De re aedificatoria (On the Art of Building), written in the 1440s, Alberti offered instructions for fresco painting to which he appended an intriguing coda: “It has recently been discovered that linseed oil will protect whatever colour you wish to apply from any harmful climate or atmosphere, provided the wall to which it is applied is dry and in no way moist.”19
Leonardo owned a copy of Alberti’s work, one of the most famous architectural treatises of the fifteenth century.20 More to the point, he would have known the works that Alberti was talking about. Alberti was almost certainly referring to the murals of Domenico Veneziano, who in the late 1440s (around the time Alberti was writing) worked with oil paints on a dry wall in the Portinari Chapel in the hospital of Santa Maria Nuova in Florence, an experiment repeated in the same chapel a few years later by Veneziano’s supposed rival and murderer, Andrea del Castagno.21
Although neither painter’s work survives, Leonardo would undoubtedly have seen them (he had a bank account at Santa Maria Nuova) and appreciated their novel technique. He must also have known the mural that Paolo Uccello and Antonio di Papi painted using oils in the refectory of the Florentine church of San Miniato al Monte. This mural, a Crucifixion scene done in the mid-1450s, does not survive, but documents from San Miniato are clear about the oil, which the pair used “even though they were not obliged.” The men in charge of the decorations were so pleased with the result that the painters received a bonus.22
The examples of these painters, together with Alberti’s statement and his own innovative nature, would have given Leonardo the inspiration and confidence to tackle his Last Supper by using oils on a dry wall. In fact, he was prepared to go one better, mixing his pigments not only with oils but also with egg yolk to create an “oil tempera.”23 In essence, he took tempera paints and mixed emulsifying oils into them. The process was so novel—there is
no known precedent—that he must have spent a good deal of time beforehand experimenting in his studio in the Corte dell’ Arengo, trying to perfect the right recipe.
Leonardo’s approach to his mural would therefore deviate from that of Montorfano, who was working on his fresco of the Crucifixion a little more than a hundred feet away. Leonardo created an entirely different surface on which to paint. Once his first coat of plaster dried, he covered it with a thinner, slightly granular layer of calcium carbonate mixed with magnesium and a binding agent probably made from animal glue.24 Once this preparation layer had dried, he added an undercoat of lead white: a primer coat, in effect, to seal the plaster and enhance the mural’s luminosity.25
Leonardo must have known that this strategy of using a lead white primer was a risky one. Made by combining strips of lead with vinegar and horse manure, white lead was the most widely used pigment in history, and for many centuries it was the white of choice in European art. Its use was inadvisable in frescoes, however, since as it oxidized it transformed into lead dioxide and turned a brownish color. Evidence of this transformation could be seen in Cimabue’s frescoes in the Lower Basilica of San Francesco in Assisi, where the whites darkened with disastrous consequences, inverting the original contrasts. Leonardo must have believed the pigment would be stable if mixed with oil or applied with a binder, such as egg tempera, or if it were added to a dry wall rather than to wet plaster.
There was another risk to using lead white, especially in such large quantities. Lead is, of course, poisonous. Its toxic nature had been known at least since the time of the ancient Romans, when the architect Vitruvius noted that it was “hurtful to the human system”: plumbers, he noticed, had a “deep pallor.”26 Lead poisoning was an occupational hazard for painters. Raphael and Correggio may have suffered from it, while James McNeill Whistler was certainly a casualty: he fell ill in 1862 while painting The White Girl. Van Gogh, who used to lick his pigments (lead has a sweetish taste), was probably yet another victim. The madness and melancholy for which so many painters became notorious was attributed by many—such as Vasari, who himself suffered periods of depression—to their sedentary and cerebral lifestyle; however, it may in fact have had something to do with the poisonous pigment on the tips of their brushes.27 Few painters before or since have used as much lead white as Leonardo did when he painted in Santa Maria delle Grazie—though he appears not to have displayed any symptoms of lead poisoning.
Once this primer of lead white had been spread, Leonardo prepared to go to work with his paints, adding them to the snow-white, bone-dry wall. How exactly he began transferring his designs is not entirely clear. It is uncertain whether he used cartoons and, if so, how extensively, for no cartoons survive—but then cartoons, by their very nature disposable and inevitably damaged by their application to the wall, very rarely survived. In places Leonardo took a stick of red chalk and sketched outlines and patterns directly onto the undercoat of white lead.28 Elsewhere on the wall he made underdrawings by using a brush and painting freehand with a black paint. He also incised lines into the plaster, possibly using a cartoon. However, when the time came to paint he did not always follow these incisions, revising and improvising as he went along. Nonetheless, there are few signs of dramatic changes or hesitations. Nor are there any of the telltale pinpricks that would indicate the use of a cartoon.29
Once Leonardo found what he believed to be the right recipe for his “oil tempera,” he began adding his paints in successive layers, allowing one to dry before adding the next: a process impossible in fresco. He often used four or five separate coats of different colors to build shapes and create tones, sometimes starting with a darker pigment in the background and then augmenting and highlighting it with paler, semiopaque colors. When he painted the sky in the background, for example, he first used azurite (a mineral-based pigment) to which successive layers of lighter blue were added to give the illusion of depth. He also took special care with the flesh tones, using three or more pigments. For some faces he began with a base coat of lead white, followed by a black to which he added vermilion, then more lead white, and finally yellow ocher and yet another layer of vermilion.30 The layering of paint, the leisurely and deliberate process it entailed, the ability it gave him to rework areas—all of these things were characteristic of painting with oils on a wooden panel rather than on an immense plaster wall. He even used a series of translucent glazes to heighten the colors.31
Leonardo intensified his colors in other ways too. He used many pigments that were incompatible with fresco, especially bright blues and reds such as ultramarine, azurite, and vermilion. The frescoist was limited in his palette because mineral-based pigments were unable to withstand the action of the lime. They could be mixed with a binder such as egg white and then added to the wall, and indeed many painters worked in this way (Michelangelo later used extensive passages of ultramarine in his Last Judgment in the Sistine Chapel). However, the eventual discoloration of the binder meant that ultimately the colors would turn into (as one treatise cautioned) “ugly daubs.”32 The frescoist was therefore advised to stick to those pigments—the duller ochers and umbers—that were compatible with lime plaster and did not require mixing with a vehicle other than water.
Leonardo, however, clearly wanted his mural to have the chromatic panache of an altarpiece painted in tempera or oil rather than this more limited range of tones necessitated by fresco. He understood like no one before him the way colors interacted: how one color could affect, or was affected by, the one beside it. “The surface of any opaque body,” he wrote, “is affected by the colour of surrounding objects.”33 He realized, that is, the way colors change their intensity and hue depending on what colors surround them. For example, he noted that a red appears more intense if placed next to a white or a yellow rather than next to purple. He therefore discovered the law of complementary colors, observing that colors were the most intense if “surrounded by their strongest contrasts.”34 He was centuries ahead of his time with this observation. Only in the nineteenth century would the law of “simultaneous contrasts of colour” (as it was dubbed by the French chemist Michel-Eugène Chevreul) be developed further. In 1839, Chevreul devised a color wheel (now a mainstay of the school art room) that showed how intense, vibrant contrasts could be achieved by juxtaposing colors 180 degrees apart—a perception ultimately leading to the pointillism of Georges Seurat and then the violent, clashing colors of the postimpressionists.35
One of Leonardo’s other observations on the relativity of color is notable: “A shadow is always affected by the colour of the surface on which it is cast.”36 Arguably, the implications of this observation—that shadows are not black but rather tinged with color—were not understood or explored until it became one of the key perceptions of impressionism, neatly defined by an American critic as the “blue-shadow idea.”37 Leonardo even had a recipe for making blue shadows, a pigment composed of verdigris mixed with lac, a gummy reddish substance derived from insects.38
In any case, Leonardo’s choice of technique and materials guaranteed that the friars of Santa Maria delle Grazie would eat their meals in the midst of a dazzling show of color.
Leonardo’s accounts show that, faced with a major project, he purchased his pigments in bulk. A few years later in Florence, while working on another mural, The Battle of Anghiari, he bought two pounds of cinnabar, four pounds of a yellow pigment, six of a green one, twenty pounds of German blue (made from the mineral azurite), and forty pounds of linseed oil in which to suspend his pigments. The total cost was 120 lire, or 30 florins, the annual wage of a low-paid manual worker such as a female weaver.39
The pigments for The Last Supper, a mural smaller in size than The Battle of Anghiari, would presumably have been roughly half that sum. Since the contract for The Last Supper has been lost, all details about the materials specified or the expenses allowed (for paint, plaster, and scaffolding) are unknown. However, Leonardo’s experience with the Confraternity of the Immacula
te Conception—when his entire salary went (so he claimed) on paints and other materials—must have persuaded him to give the contract some careful scrutiny.
The highest quality pigments came from Venice, and artists’ contracts sometimes made allowances for artists to travel from as far away as Florence and Siena to supply themselves.40 An artist could buy the raw materials for pigments (such as cinnabar for reds or malachite for greens) and make the colors himself by grinding them into a powder in the workshop. Alternatively, he could buy his pigments ready-prepared from a specialist purveyor. No records indicate where or from whom Leonardo purchased his pigments for The Last Supper, though for The Adoration of the Magi he bought prepared pigments from the monks of the convent of San Giusto alle Mura, who were renowned for their skill in manufacturing colors (other customers included Botticelli and Michelangelo).
Leonardo’s purchases from the monks of San Giusto alle Mura were one of the conditions of his contract. However, he seems to have preferred purchasing the raw materials himself and then—unsurprisingly—experimenting with his own mixtures and preparations. One of his recipes for a white pigment stated: “Put the white into an earthen pot...and let it stand in the sun undisturbed for 2 days.” A yellow pigment could be created, he noted, by dissolving ground-up orpiment, a yellowish mineral, with another mineral, realgar, in the corrosive solution known as aqua fortis (a mixture of nitric acid and water). A flesh tone could be made from grinding up reddish crystals from a place he called Rocca Nova, and a green from mixing verdigris and lemon juice. Alternatively, a “fine green” could be made from combining verdigris with either aloes, plant gall, or the herb turmeric.41
