To Dye For: Inside the Vast Library That Stores the World’s Rarest Pigments

Narayan Khandekar is opening and closing cabinet doors, pulling out vintage jars and pointing out bright powders, semi-precious stones, and other materials as he tells me about his favorite artists' pigments. Here is Tyrian purple, a pigment made from mollusk secretions that was once so expensive even royalty struggled to afford it. Here are metallic flakes originally designed for car finishes, used by 20th-century artists like Richard Hamilton to make paintings shine—"which I think is kind of amazing, actually"—and over there is a yellow weld pigment used by Dutch painters such as Vermeer. He highlights a sample of lead tin yellow, a pigment that fell out of popular use in the mid-1700s and wasn't rediscovered until the 1940s. He picks up a vintage jar filled with an orange powder. This particular pigment, he says, is light-sensitive, "so it starts off this very bright orange, and then it reacts with light and it darkens. So you often see things that look like they're a browny-chocolatey color, but in fact they might have been orange to start with," he explains.

We're on the fourth floor of the Harvard Art Museums, inside a lab at the Straus Center for Conservation and Technical Studies that looks out upon a skylit atrium through crystal-clear glass. Behind its transparent walls, visible to the public below, is an assemblage of art supplies arresting enough to give the paintings downstairs a run for their money: rows upon rows of jars filled with a vibrant rainbow of every hue imaginable. This is the Forbes Pigment Collection.

Khandekar is director of the Straus Center and the keeper of the pigment collection as well as its counterpart, the Gettens Collection of Binding Media and Varnishes. A quick-to-smile man in round spectacles, he has a passion for colors that extends to his dapper clothes, which today include a royal blue suit that complements the blue-striped shirt and bright orange socks that peek out from under his pants. He's the type of wonk who can get you really, really excited about seemingly unremarkable substances. He will talk at length about binding materials, which in his view don't get enough love, lumped in as they are with the much flashier pigments. But as he points out, binding mediums—the sticky substances that hold pigments together—are integral to painting. They affect the texture of a paint and how much light is reflected in the resulting color. In fact, whether they realize it or not, Khandekar says, most people describe types of paintings by their binders: oil, tempera, watercolor, acrylic.

As he moves through the pigments, he puts down the jar with the light-sensitive orange powder he had been telling me about and moves on to the next container that catches his eye—then pauses in the middle of a sentence. "That jar that I picked up was vermilion, so—I've just got to wash my hand before I do anything," he says, already halfway across the room. Vermilion is made of ground mercury sulfide, a toxic chemical compound.

For most of human history, artists couldn't just run to an art supply store and buy a tube of paint. They had to make their own, using powdered pigments mixed with tree resin or another type of binder, like egg or oil, that would congeal the color into a paste capable of sticking to canvas or plaster. Together, the Forbes and Gettens collections are one of the most renowned archives of art materials in America. They include almost 3000 samples from across the world and throughout history, from ochres sourced from the ruins of ancient Pompeii to Day-Glo paints used in 20th-century Pop Art, dangerous materials like vermilion, and brand-new colors created only a few years ago.

But those flashy powders in antique jars aren't just aesthetically pleasing. They can tell us a lot about how art comes into being, and why the art we love looks like it does. They're a window into history, both at Harvard and in the wider world. And they're a vital tool in protecting art from the march of time.

The pigment and binder collections got their start almost a century before Khandekar arrived at Harvard. They were the brainchild of Edward Waldo Forbes, an influential museum director whose name the pigment half of the archive now bears. The son of Bell Telephone Company co-founder William Hathaway Forbes and, on his mother's side, the grandson of Ralph Waldo Emerson, Forbes is the reason Khandekar's job exists at all.

Born in 1873 on a private island off the coast of Cape Cod, Forbes led a fairly typical life for a wealthy 19th-century heir. He attended the elite Milton Academy outside Boston before entering Harvard, where he became an avid student of the prominent art historian and cultural scholar Charles Eliot Norton. Not long after graduating in 1895, Forbes, like many other young men of his position, decamped to Europe, where he dedicated himself to studying art, with a particular eye toward the Italian Renaissance.

While living in Rome, he became determined to bring the best classical paintings and sculpture he could afford back to the U.S. His collection began with what some might consider a questionable financial choice: Madonna and Child with Saints Nicholas of Tolentino, Monica, Augustine, and John the Evangelist, purchased from a Roman warehouse in 1899. The 15th-century work was more than a little worse for wear, with paint that was blistering and, in some places, missing. It was while overseeing the painting's years-long restoration—he would eventually commission Italian, English, and American experts for the job over the course of more than a decade—that Forbes became fascinated with the science of preserving works of art from deterioration.

But Forbes never planned on keeping his art to himself. On the advice of one of his Harvard friends in Rome, he decided to loan his growing collection to the newly established art museum at his alma mater—the Fogg Museum. Soon, Forbes would go on to do even more for the museum. In 1909, he became its director.

Forbes was determined to expand the Fogg's collection, but he worried about how artwork would fare in the facility. Temperatures, moisture, and lighting varied widely between parts of the building and throughout the seasons. The humidity levels, in particular, were unpredictable and inconsistent, causing wood, paper, and canvas to expand and contract, lifting and cracking the layers of paint above them. One of the early victims of the poor environment was the then-partially restored Madonna and Child with Saints. It began to develop blisters in the paint almost immediately after it arrived at the Fogg; Forbes once described some of them as being "almost as big as a soup plate."

Forbes soon realized that to understand how to properly restore and protect works like his Madonna and Child with Saints, not to mention the rest of the art in the museum, he had to understand their components. Over the next few years, he became obsessed with the materials that went into art, including the pigments that created the colors. As Khandekar explains it, Forbes "wanted to understand how these works were made, what they were made of, what was original, what was a later restoration, what was a forgery."

Around the same time, Forbes also began teaching in Harvard's fine arts department, where he brought his zest for technical analysis into the classroom. "Just as a man who undertakes to know about swimming should be able to swim," he said, art historians should know how art is made. He asked his classes to reproduce paintings using Old Masters' materials and methods, and would buy heavily damaged work, restored paintings, and sometimes even forgeries in his quest to present his conservation students with real-life technical problems. As part of this effort, he started collecting examples of the raw materials used by classical artists.

His first collection was devoted to pigments used by his favorite Florentine painters of the 14th and 15th centuries. Forbes started by gathering the ones described in Cennino Cennini's famous 1437 guide to painting, The Craftsman's Handbook, buying jars of sought-after pigments like ultramarine blue, acquiring chunks of raw pigment materials like azurite and malachite, and planting madder root to make the red pigment known as rose madder lake. When one of his student researchers began trying (and failing) to make amber varnish like the one used by Renaissance oil painters, Forbes started collecting varnishes, too. By the 1930s, he was hunting down resins, seeds, gums, and other ingredients from all over the world, bringing them back to Harvard for study. He traveled and corresponded regularly with art suppliers, gem merchants, the Department of Agriculture, and anyone else who could help him obtain examples of pigments and binders.

Alongside his collecting, Forbes aimed to turn the Fogg Museum into what he called a "laboratory for the fine arts," an institution where scientific analysis and research could guide conservation as well as curation. And so he gave the sciences a permanent home at the museum, establishing the Department of Conservation and Technical Research—which became the Straus Center for Conservation and Technical Studies in 1994—as the first conservation department in the U.S. To help guide this endeavor, he hired chemist Rutherford John Gettens, the first scientist ever employed by an American museum. Gettens analyzed the physical and chemical properties of the materials Forbes collected, and accumulated his own stash of varnishes and binding mediums.

Creating an entire department for conservation meant that collecting pigments became more than just a personal hobby: It was now a central part of the museum's mission.

Unlike Forbes, Khandekar was interested in the science of art from the beginning. He started off studying organic chemistry at the University of Melbourne in his native Australia, and it was during school breaks that he discovered his love of art, going to the National Gallery of Australia to see the museum's collection of Lichtenstein paintings while visiting his parents in Canberra. "I wanted to understand them in a way I could appreciate," he says, "and that was through materials."

His chemistry career began with studying marine sediment, which he says isn't as different from art materials as you might think—they both involve lipids and carbohydrates. "It sounds like a big jump," he says, "but if you look at it this way, you're analyzing paint samples instead of sedimentary mud."

In fact, some of the materials in the collection he oversees make mud seem glamorous in comparison. They underscore just how unappealing the reality of making art can be, and just how much work history's artists had cut out for them even before they broke out their brushes.

Until the advent of modern synthetic pigments in the 18th century, painters had to rely almost exclusively on the colors the natural world had to offer. That meant pigments frequently came from sources that today we might consider pretty gross. Some were made with mollusk secretions; others were made with urine (both human and animal), blood, and feces. Bone black was made from charred animal bones. Kermes, a red dye used by ancient Egyptians and medieval Europeans alike, was made by crushing up shield lice that lived on oak trees. And those are some of the tamer examples. In the 18th century, Turkish merchants sold one of the brightest reds around, dubbed "Turkey red," which was made in "a tortuous process," as author Kassia St Clair describes in The Secret Lives of Color, that involved mixing the roots of madder plants with sulfonated castor oil, ox blood, and dung.

To get even more stomach-churning, take the example of "mummy brown." To make it, actual mummies were dug up and shipped to European apothecaries, who proceeded to grind them into powders for artists, as well as for medicines designed to cure all manner of ills. Later, the color was available in commercial paint tubes (which, before collapsible metal tubes were invented in 1841, were made of pig's bladder). Paint manufacturers continued to make mummy brown up until the 1960s, when, as one company told Time magazine, they ran out of mummies to make it. Harvard currently holds two tubes of the stuff, as well as a few small mummy fragments—that is, body parts—used in the manufacturing process.

Other artist's materials were treacherous, containing dangerous substances like arsenic, cadmium, and mercury. That meant artworks could pose very real dangers to their creators, as some of the pigments in Harvard's collection demonstrate. The browning label on an antique jar of realgar, a red pigment made of an arsenic sulfide mineral, is covered with a red-bordered sticker that reads, in green handwriting, "Poison!" A similar label appears on a corked jar filled with the yellow pigment orpiment, a naturally occurring mineral that's about 60 percent arsenic by weight.

Painters were not unaware of the dangers of their colors. At a time when an artist's palette was limited to the shades of the natural world, it was a compromise that some were willing to accept in exchange for the brilliance of the color in question. Renaissance painters regularly used orpiment despite knowing the risks it posed to their health—in ancient times, the pigment was even used as an assassination tool. ("This color is really poisonous," Cennini cautioned. "Beware of soiling your mouth with it, lest you suffer personal injury.") Nevertheless, it remained a popular pigment until the 19th century, though artists used it sparingly. For some, the beauty simply outweighed the dangers.

For decades, the pigments and binders at Harvard remained an exceptional resource for art conservators, but the average art lover wasn't aware of their existence. When Forbes retired from his position at the Fogg in 1944, the museum lost its main champion for the conservation program. The pigment collection and other scholarly materials "fell victim to benign neglect," as Francesca Bewer, a research curator at the Harvard Art Museums, writes in her book on the Fogg, A Laboratory for Art. The department went without a staff scientist for decades, and few new pigments were collected.

The pigment and binder collections also stayed largely hidden from public view until only a few years ago. But in 2014, Harvard combined the Fogg Museum with two other university museums to create the Harvard Art Museums, renovating and expanding its facilities. In the process, the collections got a more visible place in the museum, behind the glass walls of the Straus Center's lab space on the fourth floor. And it was Khandekar's job to figure out how to display the collections once they were reintroduced to the public view. "I spent somewhere between three and four months arranging all the pigments," he says. In their current form, they're aligned like a color wheel, the bottles fanned out with yellow in the center, blue and purple bookending either side. Some of the raw materials used to make the pigments, like the blue mineral azurite, sit on display underneath.

Still, Khandekar is more than just an arranger of colorful artifacts. He is the modern-day heir to the science-driven institution that Forbes and Gettens created.

Over the years, all art suffers from wear and tear, even if it's well-cared-for. Just like old books become brown and musty, paintings decay, their materials reacting with each other, the light, the climate, and other factors. As a painting's colors fade and change, it no longer reflects the artist's original vision. Modern conservation techniques can't keep artworks frozen in time—nor is that what most conservators want—but they can shed light on what paintings originally looked like, and what museums can do to keep them looking like that for as long as possible.

To take one example, the colors in many of Van Gogh's paintings have changed significantly over the centuries. His bright-yellow sunflowers have turned brown, and his reds have faded to the point that you may not even realize they were there in the first place. Though the walls in his painting The Bedroom were originally purple, the red pigment used to mix the color has all but disappeared, and only the blue pigment shines through. The artist knew the paints he used wouldn't be stable over time, but chose them for their vibrancy anyway. "Paintings fade like flowers," he famously wrote to his brother Theo. He wasn't kidding—the once-pink flowers in his still life Roses have now turned almost entirely white.

Contemporary artists have to deal with fading paints, too. In 1962, Mark Rothko was commissioned to paint murals for a Harvard dining room. The murals were only up for a little more than a decade before light spilling in through the room's floor-to-ceiling windows caused the paint to fade dramatically.

When the Harvard Art Museums wanted to show the murals again in a 2014-2015 exhibition, they turned to pigment analysis to help figure out the best way of restoring them. Using chemistry techniques such as X-ray fluorescence (which can test whether a particular pigment has a metal like copper in it) and Raman spectroscopy (which allows researchers to compare a pigment's chemical composition to an established library of data), researchers ran tests on samples from Harvard and their own synthesized pigments. They were able to pinpoint [PDF] the source of the artist's crimson hues as the calcium salt of Lithol Red, which happens to be highly sensitive to light when it's made into a paint. Another sodium-based red used in the painting, by contrast, didn't fade.

As a result of that work, the Harvard researchers knew exactly which color was missing from the murals. Instead of restoring the paintings with more conventional techniques, they projected a precise pattern of colored light matching that red onto the canvas to return it to the brightness of Rothko's original design. The digital restoration was turned off each day when the museums closed, revealing the paintings' true state.

The pigment collection has also been used to authenticate works of art and ferret out forgeries. In 2002, a filmmaker named Alex Matter discovered 32 paintings in his mother's storage locker on Long Island. They were wrapped in brown paper that was labelled with scribbled notes describing them as experimental works by Jackson Pollock, a close friend of his parents. If that was true, Matter was sitting on a treasure trove of exceedingly valuable paintings worth millions of dollars. The art world couldn't come to a consensus on their legitimacy, however. Though the pieces were exhibited in places such as Boston College's McMullen Museum of Art, some experts weren't so sure they were authentic, arguing that they might merely be extremely careful replicas of the painter's style, not his original works. Upon viewing them, Ken Johnson of The Boston Globe wrote that "if they are not by the master, they are expert imitations."

To put the debate to rest, three of the paintings were sent to Harvard for verification. In 2005 and 2006, researchers compared standards data from the Forbes collection of pigments to samples of the three paintings. When that didn't turn up any matches, they turned to London's Tate, which had been collecting pigments during the latter half of the 20th century, after Harvard's own collection had ceased to expand. The British institution shared 250 pigments with the Straus Center, helping the Harvard conservators discover that some of the orange, red, and brown pigments used in the works weren't commercially available until after Pollock's 1956 death. The paintings, in other words, were copycats.

In addition to solving the mystery, the case was key to the evolution of the Straus Center. Realizing that modern pigments would be vital to its continued research, Harvard dedicated funding to expanding the Forbes collection once again.

These days, Khandekar—who became the Straus Center's director in 2015—spends part of his time gathering modern pigments to add to the historic collection. He tracks down samples of new colors, like YInMn Blue, created in 2009 at Oregon State University, and Vantablack, the world's darkest man-made material—the rights to which are, controversially, held by a single artist, British sculptor Anish Kapoor. (Khandekar also added the world's "pinkest pink" and a color called Black 2.0, both created by artist Stuart Semple in response to Kapoor's monopoly on Vantablack.)

It's a big job, because pigments come in and out of production all the time. "You have to really be active in keeping up to date with everything that's available—it's almost impossible to do," Khandekar says. He and his team stay in contact not just with contemporary pigment manufacturers, but people who recreate historic pigments, too, like the British pigment expert Keith Edwards, who has sent the lab samples of pigments he synthesizes himself based on historic recipes. In 2016, Edwards gave Harvard a sample of his blue verditer, commonly used in 17th- and 18th-century watercolors.

Sometimes artists also deliver pigments, like when the Turkish artist Aslı Çavuşoğlu gave Harvard a sample of Armenian cochineal, a red sourced from the Turkish-Armenian border, during a visit to Boston. Production of the pigment stopped in the early 20th century, and according to Çavuşoğlu, the Armenian researcher she received it from "is probably the only person who can still extract this red based on the recipes from the 14th-century Armenian manuscripts."

The Forbes collection has recently helped conservators analyze the colors and materials of an ancient Roman wall fragment, rare Chinese pottery, a 17th-century portrait of Philip III of Spain, and illustrated Persian manuscripts from the 14th and 15th centuries—just to mention a few examples.

Yet the collection's purpose goes beyond scientific analysis. It's also a teaching tool for the general public, even those who have no intention of studying conservation. The public display of the pigment and binder collections offers a rare look into the artistic process. "I don't think people think where pigments come from," Khandekar says. "People assume that color is there and available, but they don't think of where it might have come from."

As he later puts it, "In the same way that you teach a kid that milk doesn't come a carton, we're teaching people that pigments don't come from Dick Blick, you know?"