When recently unboxing a rare book, I was struck by the bright green cloth of the binding. Immediately after I washed my hands and grabbed a pair of gloves. Recent research has confirmed the presence of arsenic in pigments used during book production, leading to changes in handling and access guidelines for many institutions. One popular colour in the Victorian period known to contain arsenic was emerald green.

Emerald Green

Emerald green (copper acetoarsenite) was first created around 1800, in an attempt to improve upon Scheele’s green (copper arsenite) which was prone to discoloration. Emerald green was later commercially produced in Bavaria in 1814 and during the Victorian period (1837-1900) it became a popular pigment to colour textiles, apparel, wallpaper, toys, and books. On books, it has been found in illustrations, endpapers, paper labels and onlays, textblock edges, and on both paper and cloth covers. The most hazardous of these applications is bookcloth. The documentation for bookbinding procedures is vague, as the process was a closely guarded industry secret. In general, we know that the bookcloth used to cover books in the Victorian period was not made from dyed cloth. Rather the cloth first had a coloured starch slurry scraped into its weave, giving it some colour and making the cloth less likely to have adhesive squeeze through when it was applied during bookbinding. After this the bookcloth was backfilled with a mixture of colorant, starch, and other fillers which both further coloured and coated the cloth. This process is what makes emerald green bookcloth particularly dangerous. These layers of starch and colour are friable, meaning they are prone to flake and dust off, so when handling there is the risk of possible inhalation or ingestion. Arsenic is known to cause long term health problems such as skin irritation, respiratory complications, and cancer. This means no arsenic levels can be considered “safe” when examining bookcloth.

Victorian were aware of the risks arsenic posed to their health. In fact, they often purchased arsenic as insecticide and rat poison. In 1848 Alfred Taylor’s On poisons, in relation to medical jurisprudence and medicine identified the dangers of arsenic. To attempt to mitigate poisonings in the U.K. the Arsenic Act of 1851 required sellers to record to whom they sold arsenic, the quantity, and the supposed use, but it did not prevent the use of arsenic in household products. Nevertheless, emerald green remained a popular colour. As the Victorian period continued, more debates and warnings about the pigment's use emerged as it continued to harm people’s health. Gowns and wallpaper were the most common subjects of concern. The presence of arsenic in ball gowns was often satirized in publications, such as this “Arsenic Waltz” depicted in the British weekly magazine Punch. In Frank Leslie’s Illustrated Newspaper, along with its own skeletal dance, it describes how arsenical colour is often laid on loosely and turns to dust easily. This is in line for what we know about bookcloth production and its current risks to our health. The first written warning of the danger of wallpaper to human health was from Dr. Gmelin in 1843 and by 1863 the Ladies’ Sanitary Association and the medical department of the Privy Council were investigating arsenical greens used in wallpapers, artificial flowers, ballgowns, toys, and book covers. Henry Carr’s Our Domestic Poisons reports arsenic’s danger to public health and describes a child in Troy, N.Y. who used a book cover as a watercolour palette then went into convulsions because they had ingested some of the cover’s dye made wet by the paints. Despite these warnings from various groups, economists and industrialists cautioned against any ban because the industries involved were highly profitable. The use of emerald green finally ceased at the end of the nineteenth century when superior aniline dyes, also called coal tar dyes, appeared on the market. It may be tempting to judge Victorians for the use of emerald green despite the risks, but history gives us the opportunity to examine our own lives and see our similarities to the past rather than our differences. Is this discrepancy any different than our use of microplastics, forever chemicals, and fossil fuels which we know are harmful to our health and our environment? Do we still not have the same debates between public good and corporate interests?

The Poison Book Project

The first study into arsenical book bindings was published in 2019. In it, the green paint on the bindings of four books housed at the University Library of Southern Denmark and the Smithsonian Libraries in Washington D.C. were tested. Initially examining the books to decipher the manuscript fragments beneath the paint, upon analysis they found the paint to contain arsenic. These books were from the sixteenth and seventeenth centuries, and the green pigment on all four books was found to be a mixture of orpiment (a yellow arsenic sulfide) and indigo. Also in 2019, the Poison Book Project by the Winterthur Museum, Garden & Library and the University of Delaware began to investigate toxic pigments in Victorian-era publisher’s bindings. In the first test of over two hundred books in a variety of colours, they found arsenic, chromium, lead, and mercury, with lead being found in over 50% of the nineteenth century bindings tested. Chrome yellow (lead chromate), for example, was found mostly in books from the 1880s to the 1890s and was used to create bindings in green, yellow, orange, and brown. However, unlike emerald green, chrome yellow stays adhered to the bookcloth and does not offset any heavy metals onto the hands. The Poison Book Project then looked only at green books to focus on arsenic. The Winterthur tested books from its own collection, from the Library Company of Philadelphia, as well as a couple from second-hand booksellers. From the 406 tested, 39 green cloth bindings were found to contain arsenic. The Project used multiple testing methods to identify emerald green.

The first tests were conducted using X-ray fluorescence (XRF) spectroscopy. This equipment can determine the chemical composition of materials by directing X-rays at the sample. When an x-ray reaches an electron close to the nucleus that electron is ejected and the atom becomes unstable. An electron farther from the nucleus will then move closer to stabilize the atom. This stabilization creates a photon of light, or florescence, which is unique to each element and so can be used to measure which elements are present. In the case of the green books tested by the Poison Book Project, the presence of arsenic and copper was detected. XRF can identify toxic elements while causing no damage to the book and there are also portable options available, preferable for some institutions as the books do not have to be moved. This method is enough for most institutions as it confirms the presence of potentially hazardous elements and handling guidelines can be updated. However, the Poison Book Project was looking to determine the specific pigment and therefore used raman spectroscopy to confirm the molecular structure of the copper and arsenic to be copper acetoarsenite. Like XRF, this method is non-destructive and can be done with portable instruments. New testing methods are also being established as more institutions begin to test books in their collections. For instance, visible near-infrared spectrometry (vis-NIR) has been shown to effectively detect the presence of arsenic. However, some institutions and private collectors do not have access to such testing, in which case visual cues and general trends can be used to assess risk.

Identification and Handling

The Winterthur has created a bookmark with common shades of arsenical cloth bindings and arsenical paper. They offer this bookmark for free upon request, printed by the institution to ensure colour accuracy. This in conjunction with other trends can help in the identification of emerald green. Through their testing the Winterthur has also developed an Arsenical Books Database containing books that they, other institutions, and private collectors have found to contain arsenic. From this database the Winterthur was able to establish certain trends common among arsenical bookcloth bindings. The first clue is the bright green colour of the cloth binding often with gold or blind stamped decoration and gilt edges. The spine can also appear brown as it is more exposed to air and oxidizes over time. These books were mostly printed in Britain or North America primarily between the 1840s and the 1860s. The books range in size and sometimes have embossed patterns on the cloth. These guidelines are helpful but come with the understanding that books may fall outside these generalizations. For example, excess bookcloth may have been stored by binders and used on books published outside this twenty-year window. The identification of arsenic in paper bindings is a bit more difficult. They have found arsenic in paper from 1816 to 1899, in various shades of green, on matte and glazed paper, both plain or printed. The Winterthur does still provide some examples of emerald green on paper on its bookmark. The bookmark together with the general trends found in books with arsenic bindings allows for some level of certainty that can help an institution with planning. In our case, we found three books in our collection that likely contain arsenic in their bookcloth when combining the bookmark, the database, and the overall trends. All three books were printed in London between 1854 and 1859. They all also have some level of gilt decoration on the covers. These match the general trends we see in emerald green bindings. Another edition of Trees, Plants, and Flowers is also already in the Poison Book Project database. While not as certain as testing, erring on the side of caution allows us to ensure the safety of patrons and staff through the application of additional standards of handling and care. 

When an arsenical book has been suspected or identified, care and handling guidelines must be updated to keep patrons and staff safe. As mentioned previously, there are no “safe” exposure limits for arsenic so safety measures emphasize limiting contact, especially through inhalation or ingestion. Emerald green found on illustrations or labels are low risk but bookcloth and textblock edges must not be touched directly. To mitigate all risk, patrons should first be encouraged to use digital copies. If that is not possible, nitrile gloves must be worn (unusual for rare book handling) and hands must be washed after handling, especially before eating or drinking. Books must be kept on hard surfaces as the pigment can migrate into porous materials like book foam, and the surface must be wiped down after use. Library staff must also follow these instructions, because they are at a higher risk for health issues as regular interactions can result in the gradual buildup of arsenic in the body. Conservators should work under a chemical fume hood as arsenic migrates and offsets in a larger concentration when exposed to moisture, a localized heat source, or acids. Certain strains of mold can also digest copper acetoarsenite and release toxic arsine gas. All of this means proper storage is essential for the safety of staff. Books suspected to have emerald green bindings should be pulled from circulating collections and stored together with clear signage. They can be stored in something as simple as a polyethylene zip-top bag. This simple solution allows for monitoring of potential pigment fragments, it is easy to label, it keeps water out, and its visual cue encourages extra caution. These books should also be added to any disaster plan; it is essential for safe handling during recovery especially if the disaster involves water.

Since the Poison Book Project began, multiple institutions in Canada and abroad have conducted analyses of their collections. The result of this research is a greater understanding of safe handling and care guidelines for Victorian-era cloth bindings as well as an ever-growing database of book bindings found to contain emerald green. It has also helped spawn the new field of bibliotoxicology, an interdisciplinary study of the toxic elements within books, that tries to balance access with health and safety concerns.

Written by Taylor Tryburski.

Sources

"Arsenic Waltz." Punch, or, The London Charivari. London: Punch Publications Ltd. 62, (Feb. 8, 1862): 54.

Bell, Kim, and Robin Canham. “Pages of Poison: Identifying 19th Century Arsenical Green Books at Queen’s University.” Canadian Journal of Academic Librarianship 11 (2025): 1–28. https://doi.org/10.33137/cjal-rcbu.v11.43087.

Delbey, T., Holck, J.P., Jørgensen, B. et al. Poisonous books: analyses of four sixteenth and seventeenth century book bindings covered with arsenic rich green paint. Heritage Science 7, 91 (2019). https://doi.org/10.1186/s40494-019-0334-2.

Gil, M. Pilar, Elizabeth Henderson, Jessica Burdge, Erica Kotze, and William McCarthy. “Detecting Emerald Green in 19thC Book Bindings Using Vis-NIR Spectroscopy.” Analytical Methods 15, no. 47 (2023): 663–69. https://doi.org/10.1039/d3ay01329d.

Poison Book Project

Poison Book Project-Safer Handling & Storage Tips

Poison Book Project-IDing Arsenic Bookbindings

Tedone, Melissa, and Rosie Grayburn. “Arsenic and Old Bookcloth: Identification and Safer Use of Emerald Green Victorian-Era Cloth Case Bindings.” Journal of the American Institute for Conservation 62, no. 1 (January 2, 2023): 1–12. https://doi.org/10.1080/01971360.2022.2031457.

Tedone, Melissa, and Rosie Grayburn. “Editorial Introduction to the ‘Bibliotoxicology’ Special Issue of Studies in Conservation.” Studies in Conservation, 2025, 1–2. https://doi.org/10.1080/00393630.2025.2456899.