It is billed as the largest fossilised poo in the world, the legacy of a sickly Viking, a find said to be worth as much as a mid-range car. The Lloyds Bank coprolite from York belongs on every list of oddities, which is exactly why it deserves a sober look. The 1983 excavation volume and the setting of the find do yield a good deal: cereal in the gut, an infestation by two kinds of worm, a damp and rotting town. Much of the rest rests on thin ground. This article separates the two, and says openly which text I could not consult myself.
The object itself is quickly described: a mineralised human stool, usually given as about 20 centimetres long and 5 wide. A coprolite is preserved dung, the word built from the Greek terms for dung and stone. It owes its name to an accident of address. It came to light during excavations in the cellars of a Lloyds Bank branch on Pavement, a street in the centre of York. Hence Lloyds Bank coprolite or, less politely, the Lloyds Bank Turd.
The vagueness begins with the measurements. Atlas Obscura says seven inches; an older piece from the “Uncle John’s Bathroom Reader” series says nine inches and half a pound. The 20 centimetres appear in many accounts that seem to copy one another. The original report presumably holds the reliable figure, but those are precisely the pages I could not read. I therefore treat the number as the customary one, not as a checked measurement.
Popular accounts like to tell of workmen building a new branch, stumbling on Viking finds and halting everything. The specialist reports sound less dramatic. York Archaeological Trust, founded that same year, cut four small trenches of about 3 by 2 metres in the cellars of the former York Coffee House, at the address 6–8 Pavement. The city’s Historic Environment Record states the aim: to expose as much as possible of the Anglo-Scandinavian buildings already met next door at 2–4 Pavement. So this was a targeted search. The Trust’s catalogue dates the work to 1972–73.
The yield was out of all proportion to the area. The Trust’s Jorvik site speaks of up to 9 metres of mostly Viking Age layers (a retrospective on the Trust’s history says up to 8), moist and peaty, in which timber, textiles and leather had survived. In the deepest cutting, Trench IV, as many as eleven phases of building and rebuilding could be told apart, and the authors place the sequence between the mid 9th and the late 11th century. According to the Trust’s catalogue the buildings seem to have been mainly domestic, with waste from leather, bone and antler working; volume 17/3 of “The Archaeology of York” adds jet and amber working, spinning and weaving. The article on leatherworking sets the leather in a wider context.
Four years later the big open-area excavation began nearby at Coppergate, the dig that made Jórvík famous. The coprolite does not come from there, though you will sometimes read that it does. It comes from the forerunner at Pavement. The mix-up is easy to understand, since the museum that shows it stands over the Coppergate excavation.
“Petrified” and “fossil” appear in nearly every headline. That misleads anyone who thinks of dinosaur bones. The piece is roughly a thousand years old and has seen no geological stretches of time. Researchers distinguish two routes of preservation for human coprolites, as Lisa-Marie Shillito and her colleagues explain in a review: drying out, for instance in desert caves, and partial mineralisation, which can occur in waterlogged, oxygen-poor conditions. The usual description puts the Lloyds Bank coprolite in the second group.
That has practical consequences. Dried samples often keep biomolecules better. A fully mineralised sample, the same review says, should hold no authentic organic molecules at all. Anyone hoping to read its maker’s genome from this piece should not expect much, and I know of no such study. For bran fragments and the tough eggs of intestinal worms, on the other hand, mineralisation evidently serves well enough.
Old York has no shortage of faecal matter. The Coppergate dig produced cesspits of the 9th to 11th centuries with the eggs of whipworm and roundworm, and in many English latrines cereal bran is practically a marker of human faeces. The trouble with such pits is that they blend the waste of many people, often over long periods. They say little about the health or the meals of any one person. David Smith, who assembled criteria for recognising cesspits in 2013, also points out that egg counts can hardly be converted into a quantity of faeces.
A coherent stool avoids the mixing problem. That, and not its size, is its real value. It records one digestive event of one human being. The window stays narrow, though: pollen, according to Shillito and colleagues, mostly leaves the gut after two to four days, in some cases far later. So the coprolite offers a snapshot of a few days from one life, not a life story. Shillito and her team also remark that reports on York’s latrines hardly ever speak explicitly of coprolites; the Pavement piece is the striking exception.
“He ate almost nothing but meat and bread” is the best-known claim. In that sharpened form I cannot trace it to any source I was able to read. The summary of the 1983 volume puts it differently: the coprolite gives the clearest evidence for cereals in the diet, and for a sensible reason. Charred grains occurred regularly in the layers, but only in low concentrations. Bran in a stool, by contrast, shows directly that cereal was eaten. Allan Hall, Andrew Jones and Harry Kenward dealt with bran and human faecal remains in a separate paper the same year.
The importance of meat is also attested for these layers, but on a broader basis: cattle bones were by far the most abundant mammal bones at both sites studied, and among fish bones collected by hand cod predominated. The volume also lists shellfish, birds, fruit and nuts. Whether Jones found a high meat content in the stool itself I cannot say without his text. David Jennings, in a look back at the Trust’s history, summarises the piece as pointing to meat and bread; that is the most reputable version of the claim available to me.
Even if it holds, it does not carry far. That someone ate little fruit in the few days before a bowel movement says nothing about their year. Season, stores and means cannot be read from it, and one person tells us nothing about the kitchens of a town. How varied Viking Age food was overall is the subject of the article on Viking diet, which also covers isotope analyses that average over far longer spans than a stool.
This is where it gets most interesting, because a shift has crept into the popular tradition. Many accounts say the stool held “several hundred eggs”; Jennings’s retrospective also speaks of hundreds of eggs. Morgana Camacho and her colleagues report Jones’s conclusion differently in 2018: the person was parasitised by at least a small number of maw-worms and several hundred whipworms, meaning animals, not eggs. From the egg counts in the stool, then, Jones inferred the number of worms in the gut; I could not check this against his original text. Such an infestation would be classed today as a heavy one, they go on to quote him, but well within the limits of human tolerance.
That second half of the sentence tends to vanish from headlines. “Heavy but tolerable” turns into “sickly Viking” or “riddled with worms”. Both go beyond the evidence. The number itself also needs care. Camacho and her team stress that inferring worm numbers from egg counts is uncertain: as worms multiply, each lays fewer eggs, and eggs are not spread evenly through the faeces. In their comparison, by their account, the York piece reaches the highest whipworm value: their Table 4, citing Jones, lists it at 66,000 eggs per gram. If the table reports him correctly, “several hundred eggs” is far too few for the whole stool.
What such an infestation meant can only be inferred from modern medicine. According to the World Health Organization, people with few worms usually have no complaints. Heavy infections can bring diarrhoea, abdominal pain, malnutrition and, in children, impaired development. Whether this person suffered, we do not know. The often repeated constipation, supposedly following from the piece’s length and weight, goes back to popular texts, which credit it to a student who restored the piece in 2003. It is a vivid guess, not a finding.
“If worms trouble a person within, take the juice of plantain, pound and wring it, and give it to him to drink.”Old English Herbarium, London, British Library, Cotton Vitellius C.iii, fol. 22r; Old English wording (“Gif men innan wyrmas eglen…”) as given by Thijs Porck, my own translation
The recipe belongs to the Old English, not the Norse tradition, and proves nothing about the person from Pavement. It shows only that intestinal worms counted as a treatable complaint in medieval England. The text also advises binding the crushed plant to the navel. More on such remedies is in the article on Viking medicine.
The worms say less about the individual than about the surroundings. Roundworm and whipworm spread through eggs in human faeces, which must first mature for about three weeks in the soil. Infection therefore comes not directly from fresh faeces but through contaminated earth, unwashed vegetables, water or hands. An infestation of this type presupposes that faeces lay about in the neighbourhood, and over some time. The rest of the 1983 volume fits this well: plants and insects at Pavement paint a picture of moist, foul conditions outdoors with plenty of rotting matter and large quantities of leather waste. Whether people lived or worked in the buildings, or shared them with livestock, the authors could not decide.
None of this is distinctively Scandinavian. At the latrine of Viborg in Denmark, used between 1018 and 1030, Martin Søe and colleagues likewise identified the human whipworm by DNA, along with roundworm and liver fluke. A British study by Ryan and colleagues found high roundworm rates in archaeological samples not in the early Middle Ages (13.9 per cent) but in Roman times (38.3 per cent) and the high and late Middle Ages (31.5 per cent). Those are other places and sample types, not figures for Jórvík, and for York itself Hall and Kenward, according to a review, even report more worm eggs from the Anglian period than from the Viking Age. So the Pavement stool does not prove any special filthiness among the Norse. It shows what threatened townspeople everywhere before modern sewers.
The apparent contradiction with all those combs and tweezers dissolves the same way, as the article on Viking hygiene explains: grooming does not protect you from contaminated ground.
Very probably, but that certainty rests on the methods of 1983. Outer shape, according to Shillito and colleagues, is weak proof, because no solid reference data exist and dog or pig droppings can resemble human ones. Today one would test bile acids or ancient DNA. Cereal bran, the context and the worm evidence all favour a human origin. Yet the 2018 paper by Søe and colleagues also shows that eggs of different whipworm species are easily confused under the microscope; in their material they found pig and rodent species as well. I know of no more recent chemical or genetic confirmation for this piece. That is no reason to doubt the attribution, only a reminder of what supports it.
No, that cannot be shown. There is no body, no name, no sex and no statement of origin. “A Viking” is at first only shorthand for “someone living in Viking Age York”. Three things speak against the literal reading.
First, the date. Many texts say 9th century. I could not check whether and how Jones places the piece within the eleven phases. A laboratory report of 1988 that he co-wrote calls it a “Viking-age human coprolite”, and Jennings classes it the same way; that names an era, not a layer. The site runs into the late 11th century, and volume 17/3 puts the start of occupation before 850, that is, before the Great Army took the city in 866. Until the exact layer is established, the Viking Age remains a specialist classification without a checkable layer, and Norse rule does not necessarily follow.
Second, the population. The excavation reports speak of “Anglo-Scandinavian” for good reason. Julian Richards and Dave Haldenby describe a hybrid culture in northern England after the raiding phase, and mention scholars who see this identity as socially created rather than a matter of descent. York had existed long before 866, and descendants of the locals probably lived on there, though this cannot be shown in detail. Living in Jórvík did not automatically make anyone Scandinavian. The Danelaw, likewise, names a zone of rule and law, not an ancestry.
Third, the word itself. Strictly speaking, “Viking” meant a participant in raiding voyages. A craftsperson on Pavement would hardly qualify, even with Scandinavian roots. The correct wording, then: a human stool from Anglo-Scandinavian York.
English Wikipedia cautiously says it “may be” the largest example. A press note of 2003 calls it the largest complete specimen of its kind. In the tabloids, that “may be” hardened into a world record. I found no worldwide register of human coprolites with measurements against which the claim could be tested, and no recognised record authority. Besides, the human origin of a coprolite often has to be tested separately. Broken pieces and finds dissolved in pits cannot enter any ranking. The superlative is possible but unproven.
The money is no better. The figure of 39,000 dollars circulates everywhere. Its origin is usually traced to a Wall Street Journal piece of 1991 by Tony Horwitz about Andrew Jones. I could not read that piece, so the quotable remarks are second-hand: that this was the most exciting piece of excrement he had ever seen, and in its own way as irreplaceable as the Crown Jewels (an older rendering has “as valuable”). Even Jones’s role is unclear: Wikipedia has him valuing the piece for insurance, while the Bathroom Reader text has him complaining that a 39,000-dollar appraisal was far too low. Those who name the sum rarely say when it arose or in what currency. If the valuation exists, it would be an insurance figure, not a market price. The piece is not for sale.
What follows is pub-quiz knowledge, and the sources are correspondingly modest. A diary note in Times Higher Education of 13 June 2003 reports that a school visit broke the piece and that the Trust meant to show it again after gluing as a star exhibit in its Archaeological Resource Centre. Later versions speak of three pieces and of a teacher who dropped it. The contemporary Guardian report was not available to me, so the teacher remains second-hand. The student restorer is said to have been called Gill Snape.
Since 2008, according to Wikipedia, the piece has been at the Jorvik Viking Centre. A report on the reopening in April 2017, after the flood of December 2015, mentions a coprolite in the new artefact gallery; Atlas Obscura places it in the “Diet and Health” area. The museum itself mentions “our fossilised poo” on its page about Viking life, undated and without naming the find; its page on the artefact gallery does not mention the piece. I found no dated statement from the museum on the present position. Anyone hoping to see it should ask first. The story that a smell profile for the museum’s reconstructed latrine was derived from it also appears only in popular texts.
It gives neither the age nor the sex nor the occupation of its maker. It says nothing about rank, language or faith, nothing about war or government. It does not show whether this person died healthy or suffered from the worms, nor how typical they were of the neighbourhood. A single find cannot establish frequency. All these questions remain the business of skeletons, household finds, coins and the few written sources. Its value lies elsewhere: it ties the abstract layers of Pavement to the body of one person who ate cereal and carried worms.
Where was the Lloyds Bank coprolite found? In York, during York Archaeological Trust’s excavations at 6–8 Pavement, which began in 1972. The name comes from the Lloyds Bank branch at that address.
How old is it? The 9th century is usually quoted, and specialists call it Viking Age. All I could check is the frame of the site, running from before 850 into the 11th century. I could not check the exact layer.
Which parasites did the person have? Whipworm and roundworm, two nematode species. Jones, as reported today, inferred several hundred whipworms and a heavy but tolerable infestation.
What did the person eat? Certainly cereal, as the bran shows. The familiar “meat and bread” appears in retrospectives; I could not verify the meat content of the stool itself. Meat played a large part at the site overall.
Did it certainly come from a Viking? No. The label is a simplification. At most it is secure that a person in Anglo-Scandinavian York left it.
Jórvík: the history of Viking York · Viking hygiene · Viking diet · Viking medicine · Viking daily life · Leatherworking · Comb makers · The Great Army · The Danelaw
The Norse of Jórvík · Viking Body Care · Viking food & drink · The Vikings in England – from Lindisfarne to 1066 · Viking medicine.
All History & Finds articles →
The Lučans (Czech Lučané, also Lutschanen or Lucanians in older German and English) are a tribe that may never have …
Read more →The Lugii were not a tribe but a name that, in the words of Tacitus, was "spread across many communities": a league in …
Read more →Lusatia owes its name to a people of whom we have little more than a number in a list of peoples, two chronicle …
Read more →Between the Elbe and the Oder, a league of Slavic tribes got by without rulers in the early Middle Ages: the Lutici …
Read more →