Siôn Parkinson
As part of an ongoing project at the Royal Botanic Garden Edinburgh (RBGE), I am researching the olfactory heritage of fungi in botanical collections:, across herbaria, historical mycological books (particularly those published before 1850), and in the field. My work focuses on how fungal smells are recorded, represented, or ignored in natural history archives. While working in RBGE’s Library and Archives, where I am based, I began noticing something that fell outside the scope of smell, yet still felt curiously related. It started while handling a rare set of volumes by the naturalist and illustrator James Sowerby, books that seemed to be producing images of their own.
James Sowerby was a British naturalist and illustrator active in the late eighteenth and early nineteenth centuries. He is best known for his richly illustrated works on natural history, including fungi, shells, and mineralogy. His Coloured Figures of English Fungi or Mushrooms (1797–1809) was published in three volumes and contains over 400 hand-coloured plates. It is one of the most ambitious and visually distinctive mycological publications in English. The set held at RBGE is rare and well preserved. Although these are printed books, the illustrations have not remained confined to the pages they were printed on. In several volumes, the mushrooms appear to have reproduced. Pigment has transferred from one leaf to the next, forming ghosted impressions and partial mirror images. These ‘counter-images’, as I am calling them — impressions formed through pressure, pigment, and time — appear as faint mirror-doubles or, more unusually, as layered bleed-throughs that generate entirely new fungal forms (Figure 1). They are not part of the original illustration, but they are very much of the book.
Figure 1. Counter-image in James Sowerby’s Coloured Figures of English Fungi, with bleed-through from Plate 1 and ink transfer from Plate 2. Left-hand page shows a composite imprint formed across multiple leaves. Digitised by the Linnean Society for Siôn Parkinson, 2025.
.
These marks are most apparent when seen in person. Sowerby’s Fungi is usually bound in three volumes, but the order and composition vary widely across collections. In some, the text pages are gathered at the front, followed by the illustrations. In others, text and plates alternate. Most often, the illustrations appear on the right-hand page, or recto, though this is not consistent. The intensity and clarity of the counter-images also vary, depending on how the books have been are bound and the conditions in which they have beenare stored. In the first volume held at the RBGE Library, there is visible staining from plate 9, where the large umbo of a depicted cap bleeds into the next page, forming a blurred but recognisable shape. This is not an isolated case. I have recorded similar phenomena in volumes held at the British Library, the Linnean Society of London, the Natural History Museum, and the Bodleian Library in Oxford. Most recently, I confirmed the same in a rare set at Dumbarton Oaks, a Harvard-affiliated garden, museum, and research library in Washington, DC.
At the British Library, several volumes carry stamped impressions on the reverse of the illustrated pages, but these are not caused by fugitive ink from the plates themselves. The marks are from the Library’s own rubber stamps, once commonly used by public institutions to prevent the theft and resale of individual pages. This practice has declined, not leastpartly because it can obscure or stain the material it was meant to protect. Even so, the transfer of pigment between pages — whether by time, pressure, or misapplied stamp — remains a feature of the book’s ongoing material life.
Online, however, these marks are often rendered invisible. On the Biodiversity Heritage Library (BHL) website, a digital archive of public-domain scientific literature, the pages following image plates are routinely labelled ‘Blank’ (Figure 2). This description is not always accurate. Even on screen, many so-called blank pages bear evidence of image transfer: faint outlines, smudges, and patches of bleed. These are not blanks; they are secondary surfaces. In some cases, the book appears to have printed itself. Digitisation cannot replace the physical encounter with a book in hand; still, it can register something of the physical book’s strange afterimages, the signs of its own material undoing.
Figure 2. Screenshot showing page labelled ‘Blank’ on the Biodiversity Heritage Library website, from a digitised copy of James Sowerby’s Coloured Figures of English Fungi. Visible transfer from the preceding illustration.
.
I first began noticing these marks while looking (or sniffing) for something else. My initial interest lay in how smell might be suggested visually in historical mushroom illustrations. In comic book parlance, a ‘wafterton’ is the wavy line used to depict odour. I had been curious to see if similar marks appeared in late eighteenth- and early nineteenth-century botanical illustration. What I found were stains.
Without quite intending to, I found myself shifting from one sense modality to another: from smell to touch. What these counter-images share with another method of mushroom representation — the spore print — is their mode of production: both are made through direct contact. To make a spore print, one removes the stalk of a mature mushroom and places the cap, gill-side down, on a sheet of paper, often black and white to test contrast. Covered and left overnight, the cap releases its spores, producing a radial image. This technique is used to distinguish between species whose visible features may otherwise be misleading. A spore print is a textbook example of what the philosopher Charles Sanders Peirce called an index: a mark produced by direct contact with its source. Like a footprint or a fingerprint, it bears the physical trace of the object — the mushroom — that made it.
The images that emerge in Sowerby’s volumes serve no such purpose. They do not clarify. Rather, they smear, distort, and recombine, generating new, unauthorised fungal forms. These counter-images, too, might be called indexical. They are made by pigment pressing against paper, by one page leaving its trace on another. But the resemblance to indexical signs goes only so far. In some copies, pigment has passed through multiple leaves, layering shapes and colours into accidental compositions. Such images are not merely impressions; they are accumulations. They are slow, iterative forms produced through centuries of storage, handling, and humidity.
Writing at the end of the nineteenth century, Peirce distinguished between three kinds of signs: icons, which resemble the thing they represent, like a portrait or a botanical illustration; symbols, which are arbitrary and learned, like letters or numerals; and indexes, which bear a physical or causal connection to their source. Spore prints fall squarely into this third category. So too, in a more unruly way, do Sowerby’s counter-images.
But if a spore print stabilises the mushroom into a form that may be identified, named, and compared, then the counter-image destabilises. The counter-image, by contrast, multiplies, blends, and generates new forms that resist taxonomy. These novel prints are notneither authorial nor scientific. T; they are authorised, instead, by the book itself. The book, not Sowerby, becomes the inadvertent or passive artist. The images may serve no taxonomic function, yet they carry aesthetic value all the same.
It was while discussing these qualities with library staff at RBGE that the term ‘grey literature’ came up. Grey literature is the classification given to pamphlets, internal reports, and other fugitive documents that fall outside commercial publishing. It occurred to me that these fungal afterimages, unofficial and uncatalogued, might warrant their own category. Not grey literature, but brown.
I have begun testing this loose typology, which I call ‘brown literature’, in an ongoing essay project and in conversation with colleagues at RBGE, the National Library of Scotland, the Linnean Society, and other artist-researchers attentive to the material traces left in botanical books and archives. The term is provisional, yet it has proved persistent. These counter-images are not officially authored, nor are they illustrated in any formal sense. They lie somewhere between accident and intention, residue and record. They complicate how we define both image and object. Smell may belong here too, not as metaphor but as a kind of bibliographic off-gassing, something emitted: a slow effusion from paper and pigment, glue and binding. The lingering odours and faint transfers (often mislabelled as ‘blank’ in digital archives) resist capture. As such, brown literature, remains tied to the book itself. Understanding it requires proximity and a multisensory, in-person encounter.