A column about curious stories from the worlds of science, technology, and film. Today — a mushroom sold at the world’s largest wild-mushroom market, cooked at home and in restaurants, and feared so much that waiters start a timer before serving it: undercook it, and you’ll see little people.
Every summer — the height of the season — the emergency department in Kunming braces for a particular kind of patient. The person is conscious, their tests are normal, their stomach hurts a little, and they explain in detail that a dozen elves are climbing across the ceiling while tiny bathers dive into a bowl of soup after the spoon. Yunnan doctors see hundreds of such cases each season.
Until recently, this story carried three bad reputations at once: a clinical anecdote from rural China, a legend from the Papuan highlands of the thirties, and Albert Hofmann’s silent refusal to find even a single active molecule in the mushroom sent to him. In January, BBC Future brought all three threads together, and in June the University of Utah laid out the first-ever genomic atlas of the genus Lanmaoa — and it turned out that neither psilocybin nor ibotenic acid nor any other known hallucinogen lives in the mushroom. This is where things get interesting.
The first honest statistics on “little people” came independently from Chinese doctors. A retrospective study by the First People’s Hospital of Yunnan Province, published in the Hong Kong Journal of Emergency Medicine, examined 81 cases of Lanmaoa asiatica poisoning in 2023. The picture was surprisingly neat: two-thirds of those affected were women (72.8%), the average age was about forty-two, and the peak came in July (67.9% of cases). Four out of five bought the mushrooms at a market; 95% ate them at home. Not a single group poisoning occurred in the sample: everyone ate their own private misery — an individual batch of undercooked bolete.
Severity broke down as follows: mixed type (stomach plus mind) — 64.2%; purely neuropsychiatric — 24.7%; purely gastrointestinal — 11.1%. The main initial symptom was visual hallucinations, in 69% of patients; then came fatigue, dizziness, nausea, and vomiting. Everyone’s laboratory results were normal: liver, kidneys, heart, coagulation — steady numbers. Median hospitalization was three days; 92.6% were discharged home; deaths — zero. Six people (7.4%) had to be sent to psychiatric wards with prolonged symptoms. The protocol’s treatment was ordinary: gastric lavage, activated charcoal, restoration of fluid and electrolyte balance, and a short course of antipsychotics.
Comparing it with “familiar” mushroom intoxications reveals the first oddity. With psilocybin mushrooms, psychiatric symptoms begin within two hours at most; with fly agarics containing ibotenic acid, within 30–120 minutes. Here, the stomach kicks in after 1–6 hours, but the mind after 12–24 hours: a trip shaped by such a delay looks like nothing else. The study’s authors say it plainly: the nature of the toxin and its mechanism remain unknown, while the long delay nudges the diagnosis toward ordinary gastroenteritis. At the height of the season, a doctor has to ask every patient about mushrooms — otherwise nobody will spot the elves behind the symptoms of dehydration.
Accounts collected by Colin Domnauer, a University of Utah PhD student, form an unexpectedly stable scene. People see dozens or hundreds of little people, from three to thirty centimeters tall. Around 90% describe them as elves, clowns, or fairy-tale figures in bright clothes. They climb over the patient’s clothes while the patient is dressing; they sit on the plate while they are eating — a motif described verbatim in a 1991 publication by the Chinese Academy of Sciences. Close your eyes, and the images become more vivid.
The crucial detail — the one that makes the mushroom worth a clinical monograph — is that the little people obey physics. They do not pass through walls. They fall off tables, move around objects, jump from a spoon into soup. One patient told Domnauer that the Lilliputians jumped from her spoon into the bowl, swam around there, and one rode into her mouth along with a spoonful of soup. This is not some blurred symbolic vision — these are detailed scenes in the actual space of a room, with due respect for gravity, inertia, and the geometry of doorways.
By pharmacological logic, all this sounds implausible, and Domnauer says so outright: “I don’t know of anything else that produces such consistent hallucinations.” Ordinary psychedelics are a lottery: the same person can have two experiences with the same substance and get different plots. Here, everyone gets the same plot, across three countries and three generations. And the trip lasts one to three days, reaching week-long hospitalization in severe cases. Perhaps that is why neither Yunnan, nor Papua New Guinea, nor the Philippines has ever developed a reputation for using the mushroom recreationally: “It has always been eaten simply as food,” Domnauer says. Three days of Lilliputians are not seasonal fun, even for the most desperate.
Prevention, meanwhile, has been refined into folk wisdom. At Yunnan mushroom hotpot restaurants, the waiter sets a timer for fifteen minutes and gives instructions: don’t eat until it rings — or you’ll see little people. Heat treatment completely removes the psychoactivity — scientific confirmation of what local culture already knew without us: it must be boiled.
The term now used to describe the syndrome was a gift not from mycology, but from psychiatry. In 1909, the French psychiatrist Raoul Leroy described hallucinations of crowds of miniature, lively, friendly little people inhabiting the patient’s real space — and called them lilliputian hallucinations, after the island in Swift’s Gulliver’s Travels. Leroy had seen them himself. He also noticed links to withdrawal, dementia, and retinal degeneration — conditions in which the brain fills in the world with smaller inhabitants.
A hundred years later, the body of solid evidence fits into a single line. A systematic review by Jan Blom (Neuroscience & Biobehavioral Reviews, 2021) found only 226 non-mushroom-related documented cases of Lilliputian hallucinations since 1909. A third of those patients did not recover completely. Yet the image itself — a little people living in the cracks of everyday life — exists in the folklore of virtually every culture, from dwarfs to house spirits. The result is a paradox: a syndrome known to humanity for thousands of years is tiny in the clinic, and science has no mechanism for it.
That is precisely why the mushroom is not a curiosity but a potential key. Lilliputians in dementia or delirium tremens arrive randomly and amid accompanying chaos; here, there is a deterministic trigger that can be isolated and studied. “We may now be able to understand where in the brain they originate,” says Dennis McKenna, an ethnopharmacologist commenting on the research for the BBC. If a molecule is found that reliably switches on a recalculation of scale, neuroscience will gain a tool for a fundamental question: why does the brain keep a ready-made “little person” template at all, and what does it think of itself when that template activates?
The story had been dragging on since the thirties. In 1934, missionaries in the Papuan highlands reported to Europe on “mushroom madness”: locals would sometimes “go mad” after eating wild mushrooms. In the sixties, Gordon Wasson and Roger Heim went there — the very people who had introduced the West to the existence of psilocybin mushrooms in the first place. They narrowed the culprits down to the bolete family, collected samples, and sent them to Albert Hofmann — the chemist who discovered LSD and isolated psilocybin. Hofmann found nothing. The team closed the case with a verdict of “cultural folklore,” and the investigation slept for sixty years.
It woke up in fragments. In 1991, researchers at the Chinese Academy of Sciences described the Yunnan cases — with clothing, dishes, and intensified visions when the eyes were closed. In 2015, taxonomists finally formally described the genus Lanmaoa (a paper in Fungal Diversity established four new genera of boletes at once). And only in 2023 did Domnauer arrive in Kunming at the height of the season and do what nobody else had done: walk through the market aisles asking one direct question — “Which one makes you see little people?” The sellers giggled and pointed. Genomes from the purchased mushrooms confirmed the suspicion.
The final act added geography. In 2024, Domnauer heard the same stories from an Indigenous community in the high Northern Cordillera of the Philippines: the local mushroom sedesdem — and its descriptions matched the Chinese and Papuan ones down to the details. Outwardly, the Philippine specimens differed — smaller and pinker than the Chinese ones — but the DNA said: the same species, L. asiatica, previously thought to be endemic to Yunnan. In December 2025, Domnauer’s supervisor Bryn Dentinger reached Papua New Guinea too — and came away empty-handed: an unusually dry year, no mushrooms to be found. So the legendary nonda remains a question mark. If it is the same species, that is a biogeographical sensation; if it is another, then the same effect evolved independently in different corners of the planet. There is precedent: a year earlier, geneticists showed that psilocybin arose twice in fungi, in the unrelated lineages Inocybe and Psilocybe.
The paper in Mycologia (5 June 2026) is not about hallucinations. It is about foundations. Domnauer and Dentinger sequenced 53 genomes from members of the genus, including 21 type specimens, and built a phylogeny from 1515 orthologous genes — with full statistical support at every major node. The result: 17 species are recognized in the genus instead of the previous blurred list; four are new to science, two described in the same paper (L. fallax and L. carbonilivor); the other two new species had been hiding in old American collections under someone else’s labels. “How much biological novelty is sitting in plain sight — just waiting to be sequenced,” Domnauer puts it. “Without DNA sequencing, we would never have known.”
The key finding concerns chemistry. A scan of the L. asiatica genome for canonical biosynthesis genes found neither close traces of psilocybin genes nor genes for the ibotenic-acid pathway. If the mushroom causes the visions, then the visions are caused by a molecule science does not yet have — no name, no class. The familiar psychoactive toolkit is ruled out not by indirect symptoms, but by genetics: the relevant biosynthetic pathways simply are not there. At the same time, L. asiatica is the only psychoactive species in the entire genus of 17 known species.
What comes next is a technical route with no pretty shortcuts. The classic bioprospecting sequence: grind up fruiting bodies, extract the mixture, test the extract in vivo and in cell cultures (Domnauer’s work is already underway — mice given the extract went through a phase of hyperactivity and a long stupor), then separate the extract into fractions and look for the one where the activity lives. Slow, routine work, with the risk that several molecules may be working together in synergy. Judging by an interview with Nautilus, the researchers have no ready guess about the class of substance — and that is rare honesty: “I really don’t know. We don’t understand how this hallucination works in the brain, so it is impossible to say what types of compounds might be responsible.”
The story’s industrial backdrop is serious too. Lanmaoa is an ectomycorrhizal mushroom, a partner of pine; you cannot grow it on farms, only find it in forests and markets. As the popular bolete spreads, so does confusion with lookalikes: the authors state outright that frequent misidentification of edible boletes as poisonous lookalikes is a documented public-health problem for doctors and traders. The sequenced atlas of the genus is, among other things, an attempt to bring order to what may land on a stranger’s dinner table tomorrow.
The name hints at the mechanism. The mushroom’s local name, jian shou qing, translates as “turns blue in the hand”: the cut surface blooms blue within seconds — an oxidation reaction, the calling card of many boletes. Beneath the color lies a beloved umami flavor; beneath the flavor, an economy: Kunming’s markets trade in two hundred edible species, and the largest is considered the biggest wild-mushroom market in the world. Yunnan is China’s leading mushroom province — and also its leader in mushroom poisonings; every summer, its news is flooded with entire families who ate slippery jacks that turned out to be doubles of death caps.
The elf story belongs to this landscape without a trace of esotericism: the mushroom is eaten because it is delicious, stewed locally in chicken hotpot, boiled by the timer, and the hallucinations are an unpleasant side effect of ignoring that timer. None of the three cultures familiar with the mushroom built a special psychedelic practice around it — unlike mescaline-bearing Lophophora or Psilocybe. A two-day trip is incompatible with daily life. The Lilliputians come to those who did not keep an eye on the pan.
And against this backdrop sits the larger parenthesis — the reason all of this is being written, really. Mycologists estimate that science has described less than 5% of the planet’s fungal species. “Fungi are a vast biochemical and pharmacological library that we have only just begun to leaf through,” says Giuliana Furci, founder of the Fungi Foundation. Penicillin was the same kind of case a century ago: sediment in a Petri dish, behind which stood the whole antibiotic era. If an undercooked Yunnan bolete has its own psychoactive class, the question is not whether more such mushrooms exist, but how many there are — and what other modules of our brains are waiting for their key.
The honest conclusion is this: the molecule exists phenomenologically — three cultures independently describe the same syndrome; genetically, it has vanished without leaving a single known trace; chemically, it has not yet been caught. The mystery is unresolved on all three planes, but for the first time in sixty years it is in an active-search phase: there are clinical statistics, there is a genomic atlas, there is fractional bioprospecting in a Salt Lake City laboratory.
The most curious thing in this story is not even the mushroom. Lilliputians who respect gravity suggest that the architecture of perception contains a separate, long-refined module for recalculating scale — and that some undiscovered class of substances can switch it on at full volume. The mushroom is merely the pretext: the real instrument is the brain, whose button has been found. The story may end most beautifully when the new molecule receives a name. Whoever names it should remember that it all began with a fifteen-minute timer in a hotpot restaurant.