Fungal mycelium really does produce electrical spikes. The vocabulary of fifty words came from choosing where to cut the recording.
Fungi coordinate a great deal — chemically, by fusing their filaments into one connected body, and with electrical activity that has been recorded since the 1990s. What has not been shown is that any of it constitutes language, and the widely reported fungal vocabulary comes from a threshold chosen by an analyst.
A mycelium has a real coordination problem. It may cover several square metres, it has no brain, no nerves and no blood, and its far edges need to behave as parts of one organism — abandoning exhausted patches, thickening cords towards new food, fruiting all at once. It solves this, and the mechanisms are worth knowing: chemical signals diffusing through hyphae, hyphal fusion that turns separate filaments into a single connected network, and electrical activity that propagates and changes when the fungus is stimulated. Then there is the claim that arrived in 2022 and has not left, that fungi have a vocabulary of about fifty words. That claim rests on real recordings and on a step that is not a measurement: the spikes were grouped into trains wherever the gaps exceeded a threshold the analyst chose, and the resulting groups were compared with word lengths in human languages. No fungus was shown to respond to a train — no receiver was tested at all, which is the minimum a communication claim needs. This page separates what has been measured, which is genuinely interesting, from what has been said about it.
Developed coverage · 92% complete · reviewed 2026-08-31
What this page covers
Signalling and coordination within and between fungi: chemical signals, hyphal fusion, electrical activity and the interpretation of all three. Electrophysiological work covers only a handful of cultured species.
Often confused with: Mycorrhizal networks between plants
Quick facts
Electrical spiking
Recorded since the 1990s, and responds to stimulation
The "50 words"
A grouping threshold chosen by the analyst, not a measurement
Receiver tested?
No — no fungus has been shown to respond to a spike train
How a mycelium becomes one body
Hyphal fusion — separate filaments joining into a connected network
The short answer, and the part the headlines removed.
The short answer
Do fungi have a language of about 50 words?
No — or at least, nothing has shown that they do. Fungi produce electrical spikes, and that is a real measurement. The words were produced by an analyst deciding how long a gap ends a word.
A 2022 study recorded voltage between electrode pairs in four fungal species, identified spikes, and grouped spikes into trains wherever the interval between them exceeded a chosen threshold. The resulting train lengths resembled the distribution of word lengths in several human languages, and that resemblance is where the vocabulary came from. Change the threshold and you get a different vocabulary. More decisively, nothing in the work tested a receiver: no fungus was shown to behave differently for having received a train, which is the minimum a communication claim requires. The paper itself says other explanations cannot be excluded; the coverage did not. What remains true and interesting is that fungal mycelium is electrically active, that the activity changes after a wound or a flame or contact with food, and that a mycelium does coordinate across metres without anything resembling a nervous system.
What a fungus is, and what a mycelium has to coordinate
Fungi have not been shown to have words. The spikes are real; the vocabulary is an interpretation.
Popular claim, unsupported
Widely repeated, with no good evidence behind it.
The identification of word-like units in fungal electrical recordings depends on an analyst-chosen inter-spike interval threshold for grouping spikes into trains, and on a statistical similarity between train-length and word-length distributions. No experiment has demonstrated that any fungus alters its behaviour in response to a received spike train, which is the minimum requirement for a communication claim.
Who this applies to
Claims that fungal electrical spiking constitutes a language or vocabulary, as reported for four species.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Basidiomycota, Ascomycota
You may have heard
“Fungi talk to each other using a vocabulary of about 50 words.”
The fifty comes from cutting a voltage recording into groups wherever the gaps exceed a threshold the analyst picked, then noting that the resulting group lengths resemble word lengths in English. Nothing in the work asks whether a fungus responds to a group, and until something does, the word "word" is doing all the work.
Why we rate it this way, and what the caveats are
Popular claim, unsupportedModerate confidence
The underlying recordings are sound and NatureHQ treats them as such. What is unsupported is the step to language, which rests on an analytical choice and an analogy rather than on a test — and the test that matters, whether a receiver responds, has not been run.
How far it can be extended
The analysis covers four species under laboratory conditions and tests no receiver.
Caveats
This is not a claim that fungi lack internal signalling. They plainly coordinate growth across a mycelium, by chemical and probably electrical means.
The original paper is more cautious than the coverage it generated.
Still unanswered
Whether a receiver experiment — delivering a recorded train to a fungus and measuring its response — would find anything.
What function, if any, the spike trains serve within the mycelium.
The source of the claim. Its measurements are of voltage; its lexicon is produced by a grouping threshold, and the paper itself states that other explanations cannot be excluded.
Supplies the duller competing reading. Recorded in a forest, the potentials rose after rainfall — that is, they tracked the water status of the fungus, which is a considerably better-supported explanation of structured electrical activity than a vocabulary.
Words used here
Spike train
A cluster of voltage spikes close together in time. Where one train ends and the next begins is a decision made when analysing the recording.
Fungal mycelium produces electrical spikes, and the pattern changes when the fungus is disturbed.
Well supported
Good evidence backs this, though some details remain open.
Electrode recordings from fungal mycelium and fruiting bodies show spontaneous voltage fluctuations resembling action potentials, arriving in irregular trains, whose rate and pattern alter following localised stimulation such as wounding, heat or contact with a food source.
Who this applies to
A small number of cultured species recorded under laboratory conditions with implanted electrodes.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Basidiomycota, Ascomycota
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
The phenomenon has been recorded independently since the 1990s and is not in doubt. Its scope is narrow — few species, laboratory conditions — and separating genuine biological activity from electrode artefact over long recordings is a persistent difficulty.
How far it can be extended
Recordings exist for only a handful of species, and electrode implantation is itself a wound, which is one of the stimuli that changes the signal.
Caveats
Activity within one mycelium is coordination within a single body, which is a different phenomenon from communication between individuals.
What the spikes do, if anything, is not established by their existence.
Still unanswered
Whether the spikes carry information, and if so what kind.
Whether any fungal behaviour changes in response to a spike train arriving from elsewhere.
Recorded spike trains in four species and found the patterns differ between them. The recordings support this claim; the linguistic interpretation is a separate matter.
The descriptive baseline the later analytical work is built on: two distinct spike populations, fast and slow, arriving in irregular trains. It establishes that the activity has structure without proposing what the structure means.
The only recording in the ground rather than in culture. It answers the standing objection that the phenomenon is an artefact of laboratory substrate, and shows the activity rising after rain.
Diagram
The measurement, and the two columns it belongs in
Everything on the left is data. Everything on the right is what a language claim would require and does not have.
The same explanation in words
A two-part figure. On the left, a schematic of the experiment: a strand of mycelium or a fruiting body with two electrodes inserted at different points along it, wired to a recorder. To the right of that, a voltage trace against time, running over minutes to hours, showing a flat baseline interrupted by irregular clusters of sharp spikes a few millivolts high — the spike trains. Beneath, two boxes. The first, headed Demonstrated: voltage fluctuations occur without any stimulus; they cluster into trains; the trains differ between species; and some change after a flame, a wound, or contact with a piece of wood. The second, headed Not shown: that a train carries meaning; that any fungus responds to one; that the groupings are units of anything at all; and that no receiver has been shown to behave differently for having received one. A closing line states that cutting a signal into word-like units is a choice made by the analyst and is not a finding about the fungus.
How we know
Counting the gaps between fungal voltage spikes
Do the electrical spikes in fungal mycelium group into anything like words?
Pairs of electrodes were inserted into the mycelium or fruiting bodies of four fungal species and the voltage between them recorded for long periods. Spikes were identified in the traces. Spikes closer together than a chosen interval were treated as belonging to one train; a longer gap ended the train. The resulting train lengths were then compared statistically with the distribution of word lengths in several human languages.
What happened
Spikes did cluster into trains, and the distribution of train lengths resembled the distribution of word lengths in the languages compared. The author reported a median of around fifteen distinguishable groupings, and a possible repertoire of up to about fifty.
What it shows
Fungal mycelium produces voltage fluctuations that are not uniformly spaced but arrive in bursts, and the pattern of bursts differs between species. That much is a measurement.
What it does not show
It does not show that the trains mean anything. The word-like units exist because an analyst chose how long a gap ends a train, and a different threshold yields a different vocabulary; the resemblance between two length distributions is a statistical similarity, not evidence of shared function. Most importantly, no receiver was tested: nothing in this design asks whether any fungus, or anything else, behaves differently for having received a train. That test is the minimum a claim about language requires, and it has not been done.
The controls — what makes this evidence rather than a story
Four species were recorded rather than one, and their spike patterns differed from one another.
Reference recordings from the substrate were used to check that the signal came from the fungus.
It is worth saying what the honest interpretation of this activity probably is. A mycelium is one organism spread over a large area, and it has to act like one: cords thicken towards food, unproductive regions are abandoned and their material recovered, whole networks fruit at once. Electrical activity that propagates and responds to local events is exactly the sort of thing coordination inside a single body would use. That is a different phenomenon from communication between individuals, and the difference is not a technicality — it is the entire question.
Words used here
Action potential
A brief, self-propagating change in voltage across a membrane. In animals it is the basis of nerve signalling; fungal spikes resemble one without implying the rest.
Two wires, a logger, and a long argument about which of the wiggles belong to the fungus.
The method is simpler than the conclusions drawn from it. Two fine electrodes are pushed into a piece of mycelium or a fruiting body a short distance apart, wired to an amplifier and a logger, and the difference in electrical potential between them is recorded for hours or days. What comes out is a wandering baseline with sharp excursions in it. Those excursions are the spikes.
Three things about that description do most of the work in the arguments that follow. The recording is differential — it measures the difference between two points, so it says nothing about either point on its own. Inserting an electrode is a wound, and wounding is one of the stimuli known to change the signal, so the act of measuring is also one of the things being measured. And a spike is not an object the fungus produces: it is whatever crosses a detection threshold in the analysis. Move the threshold and the number of spikes changes.
How we know
Wiring up a mushroom in a wood, and waiting for rain
Does the electrical activity recorded from fungi in the laboratory happen in the ground, in weather?
Electrodes were inserted into fruiting bodies of an ectomycorrhizal fungus growing in soil in a forest, wired to a logger, and the potentials recorded continuously across a period that included a rainfall event. Several fruiting bodies at different separations were recorded at once, so signals could be compared between neighbours and between distant individuals.
What happened
Potentials were small and largely unstructured before the rain and rose afterwards. Signals from fruiting bodies close together were more strongly related to each other than to those further away.
What it shows
Fungal electrical activity is not confined to laboratory cultures. It occurs in the ground, and it changes with an ordinary environmental event.
What it does not show
It does not show that the activity carries information. The most economical reading is the opposite of the interesting one: that the potentials track the water status of the fungus, which is exactly what changes when it rains. Nearby fruiting bodies also share a microenvironment, so their signals being alike is not evidence that they are connected to each other rather than exposed to the same rain. One species, one site, one rainfall event.
The controls — what makes this evidence rather than a story
Field conditions rather than culture, which removes the objection that the phenomenon is an artefact of laboratory substrate.
Multiple fruiting bodies at known separations, so the relationship between signals could be tested against distance.
A recording window spanning before and after the rain, making the fungus its own control.
The obvious objection to the whole literature — that it describes cultures in dishes, where an implanted wire and an artificial substrate could account for everything — has now been partly answered. Electrodes placed in fruiting bodies growing in soil in a forest recorded much the same kind of activity, and it rose after rain. That is a real advance, and it also supplies the least exciting explanation available: potentials that track the water status of the fungus. Rain changes what is in the tissue, and what is in the tissue is what an electrode measures.
A spike is a threshold crossing in an analysis, not a discrete event the fungus emits. The count depends on where the threshold is set.
Where one train of spikes ends and the next begins is a second, separate choice, and it is the one the vocabulary claim rests on.
Electrode insertion wounds the tissue, and wounding alters the signal being recorded.
Long recordings drift — electrode chemistry, temperature, moisture — and drift can look like slow biological activity.
Recordings from soil and weather are recent, and they suggest the activity tracks water at least in part.
None of this makes the phenomenon an artefact. Olsson and Hansson recorded stimulus-sensitive activity in growing mycelium in 1995, three decades before anyone called it language, and the activity changed when a distant part of the same mycelium was offered a piece of wood. Something real is being measured. What the methodology explains is why the step from that measurement to a lexicon is so much larger than it looks.
Words used here
Differential recording
Measuring the voltage between two electrodes rather than against a fixed reference. It reveals differences along the tissue and hides anything the two points share.
Artefact
A feature of a recording produced by the equipment or the act of measuring rather than by the thing being measured.
Chemistry, fusion and growth — how a mycelium behaves as one organism.
The best-established fungal signalling is chemical, and some of it is unusually well characterised. Fungi with mating types recognise a compatible partner by peptide pheromones and matching receptors, and will grow towards one; the recognition is specific enough that a mismatch produces no response at all. Fungi also sense their own density and their neighbours through diffusible compounds, and adjust growth, sporulation and enzyme production accordingly.
The clearest case of fungal chemical signalling, and it is a recognition system rather than a message: a hypha grows towards a partner whose pheromone its own receptors can read, and ignores one whose it cannot.
Many fungi have thousands of mating types rather than two sexes
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
Sexual compatibility in many basidiomycetes is governed by two unlinked mating-type loci, each with many alleles. Two mycelia can mate if they differ at both loci, which in species such as Coprinopsis cinerea and Schizophyllum commune produces thousands of mutually compatible mating types rather than two sexes. Compatible hyphae fuse, and the resulting mycelium carries two genetically distinct nuclei side by side without fusing them.
Who this applies to
tetrapolar basidiomycetes; other fungal groups differ
Studied in
Basidiomycota, Coprinopsis cinerea
You may have heard
“Fungi have thousands of sexes”
A good headline resting on the wrong word. Sexes imply different gametes — large eggs and small sperm — and these fungi have neither. What they have is a compatibility system with many alleles, so almost any two individuals can pair. Calling them sexes makes a genuinely unfamiliar arrangement sound like a familiar one multiplied.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
The loci have been characterised genetically and the compatibility predictions tested by crossing isolates.
How far it can be extended
Multi-allelic mating-type systems are widespread among basidiomycetes. Many ascomycetes and other fungi have simpler bipolar systems, and some are asexual.
Caveats
Not all fungi work this way: bipolar systems with two mating types are common, and many fungi reproduce asexually.
Mating type is not sex — there are no distinguishable male and female structures in these species.
The genetics is best characterised in a small number of laboratory species.
Still unanswered
What maintains such high allele diversity at the mating-type loci?
Kües, 2000 · Microbiology and Molecular Biology Reviews
The two-locus mating system, hyphal fusion and the persistence of the dikaryon.
The reason coordination is a problem worth solving in the first place. The organism is the network, and the network is large.
A mushroom is a fruiting body — the fungus itself is a network in the soil or wood
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
The vegetative body of most fungi is a mycelium of hyphae permeating its substrate. Fruiting bodies are transient reproductive structures. Individual mycelial genets can occupy many hectares and persist for centuries, as established by genetic identification of isolates across a forest.
Who this applies to
fungi generally; the size record established in Armillaria
Studied in
Fungi, Armillaria gallica
You may have heard
“The biggest living thing in the world is a fungus in Oregon”
Broadly right and worth qualifying. "Biggest" depends on whether you count area, mass or connectedness; the mass is estimated rather than weighed; and whether one genetically uniform mycelium is one organism is a choice about definitions rather than a discovery about nature.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
Basic mycology, supported by genetic delimitation of individual genets across landscapes. The extent of a single individual is the part that involves estimation.
How far it can be extended
Mycelial organisation is fundamental to the kingdom outside the single-celled yeasts, and large genets have been identified in several genera independently.
Caveats
Mass and age figures for large genets are extrapolations, not measurements.
Whether a spatially continuous genet counts as one individual is a definitional question.
Yeasts are fungi and are not mycelial, so the picture does not cover the whole kingdom.
Still unanswered
How old can a fungal individual actually get?
How much of a large genet remains physically connected rather than fragmented?
How mycelial networks extend, find resources and remodel themselves — the behaviour of the actual organism.
How the network gets built is the best-evidenced act of fungal signalling there is, and it goes almost entirely unmentioned in the coverage. Two filaments approaching one another do not simply collide. Each detects the other at a distance, and they take turns: one tip adjusts its direction, then the other, back and forth until they meet, at which point the walls between them are dissolved and the two cytoplasms become one. Taking turns is the informative part — it means both are emitting and both are responding, which is the structure of an exchange rather than of one filament following a gradient.
Two fungal filaments detect each other at a distance, turn towards each other and merge into one.
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
Fungal germlings and hyphae of a compatible individual reorient their growth towards one another before contact, in a mutual and alternating fashion consistent with a reciprocal chemical exchange, then make contact and dissolve the intervening cell walls to form a continuous cytoplasmic connection.
Who this applies to
Demonstrated in detail in a small number of laboratory model fungi, chiefly Neurospora crassa, where the mutants that break the process have been identified.
Studied in
Ascomycota, Basidiomycota
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
Directly imaged in living cells, with genetic mutants in which the homing and the fusion fail separately. This is the best-evidenced case of one fungal cell acting on another.
How far it can be extended
Fusion is observed across both major fungal phyla and is required for the networked mycelium that fungi generally form, but the signalling molecules are characterised in only a few species.
Caveats
This is signalling that builds a single organism, not a message passing between individuals — the distinction the rest of this subject turns on.
Two hyphae of incompatible individuals meeting usually produces a barrier rather than a fusion.
Still unanswered
What the signal exchanged during the approach actually is; the alternating pattern is imaged, the molecule is not identified in most species.
How often fusion happens in soil and wood, as opposed to on agar.
Synthesises the live-imaging and genetic work showing mutual reorientation before contact, and the wall dissolution that follows it.
How we know
Filming two fungal filaments deciding to become one
When two fungal filaments join, is one following the other, or are both signalling?
Germinating fungal spores were imaged live at high magnification as they grew towards one another, in wild-type fungi and in mutants selected for failing at particular stages. The geometry of the approach was recorded: which tip changed direction, when, and in what order, up to the moment of contact and the dissolution of the wall between them.
What happened
Both tips reorient, and they do so in turn — one adjusts, then the other — in a back-and-forth that continues until they meet. The walls between them are then dissolved and the cytoplasm becomes continuous.
What it shows
This is a two-way exchange, not one filament tracking a gradient from a passive partner. Taking turns implies each is both emitting and responding, which is the structure of a signal exchange, and the mutants show that the homing and the fusion are separately encoded rather than incidental.
What it does not show
In most species the molecule being exchanged is not identified — the alternation is imaged, the chemistry is inferred. The work is also concentrated on model fungi growing on agar, and how often hyphae find each other this way inside soil or wood is not established by it. And it remains signalling that assembles one organism, which is a different thing from a message sent between two.
The controls — what makes this evidence rather than a story
Mutants in which homing fails but growth does not, separating the attraction from the growing.
Mutants in which the tips meet but do not fuse, separating the contact from the merger.
Live imaging of the whole approach rather than fixed end-points, which is what makes the alternation visible at all.
Hyphal fusion — anastomosis — joins separate filaments so that a mycelium becomes one continuous, interconnected network rather than a set of parallel threads. It is the physical basis of everything else here.
Nutrients and signals then move through that network, so a discovery at one edge can change behaviour at another.
Cord-forming fungi thicken the connections towards a resource and withdraw material from unproductive regions — foraging, by an organism with no ability to move.
Where two different individuals meet, the response is usually combat rather than cooperation: visible dark barrier lines in decaying wood are the boundary between two fungi neither of which is winning.
Fungi respond to damage, to light, to gravity and to the chemistry of the substrate, and adjust where they grow and when they fruit.
The channels that are demonstrated: airborne chemistry going out, and light coming in.
Walk into a wood after rain and the smell is fungal. That smell is a signal channel, and unlike the electrical one it has receivers that have been tested. Fungi release small airborne compounds — the alcohols and ketones behind the smell of mushrooms, the terpenes behind damp and mould — and in experiments where a fungus and another organism share air but never touch, those compounds measurably change what the other organism does.
The one fungal channel where a sender, a receiver and a changed behaviour have all been shown in the same experiment — with the caveat that the concentrations used are often higher than anything in a forest.
Fungi release compounds into the air that change how insects, plants and other fungi behave — without any contact.
Well supported
Good evidence backs this, though some details remain open.
Fungi emit volatile organic compounds, chiefly alcohols, ketones, aldehydes and terpenes, which in bioassays where organisms share an atmosphere but never touch measurably alter insect attraction and repulsion, plant growth and root architecture, and the growth of competing fungi and bacteria.
Who this applies to
Emission is general across fungi. The demonstrated effects come from chamber bioassays at concentrations that may exceed those in soil or open air.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Fungi
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
That fungi emit these compounds is not in question — it is what mushrooms and mould smell of. That the compounds change the behaviour of exposed organisms is repeatedly demonstrated in chambers. Whether those effects occur at field concentrations is much less certain.
How far it can be extended
Which compounds a fungus emits varies enormously by species and substrate, and effects measured in a sealed chamber cannot be assumed to occur in a forest.
Caveats
Emitting a compound is not the same as signalling with it. Much of what a fungus releases is metabolic by-product that other organisms have learned to read.
Chamber concentrations are often far above what an organism would meet in soil or air.
Effects run in both directions — attraction and repulsion, promotion and inhibition — depending on compound, dose and receiver.
Still unanswered
Which of these effects persist at realistic field concentrations.
Whether any fungal volatile is emitted at a cost that only makes sense if a receiver responds — the test that would separate a signal from a smell.
Hung et al., 2015 · Applied Microbiology and Biotechnology
Assembles the emission chemistry and the bioassays across insects, plants and microbes, including the contactless exposure designs that make the effect attributable to the volatiles.
It is worth resisting the obvious upgrade. That a beetle is attracted by a compound a fungus emits does not make the compound a message to the beetle; a great deal of what fungi release is metabolic exhaust that other organisms have learned to read, which is eavesdropping rather than conversation. The test that would separate the two is whether the fungus pays a cost to emit something that only makes sense if a receiver responds, and for most fungal volatiles nobody has run it.
Coming the other way: an organism with no eyes, carrying proteins tuned to particular wavelengths, rebuilding its gene expression within minutes of the light changing.
Fungi have proteins that detect light, and use them to time spore release and orient their growth.
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
Fungi encode photoreceptor proteins responsive to blue, green and red wavelengths. Light acting through these receptors entrains circadian rhythms, induces sporulation, orients the growth of reproductive structures and alters the transcription of large gene sets within minutes of exposure.
Who this applies to
Characterised genetically in a handful of tractable species — Neurospora, Aspergillus, Phycomyces — under laboratory light regimes.
Studied in
Fungi
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
The receptors are identified molecules, and the mutants that lack them lose the response. This is a sense with a named mechanism rather than an inference from behaviour.
How far it can be extended
Photoreceptor genes of the same families are found across the fungal kingdom, though what each species does with the input differs.
Caveats
A sense, not a form of communication. The fungus is reading its environment, not being addressed by it.
Laboratory light regimes are far simpler than the light reaching a forest floor.
Still unanswered
What most fungi actually do with light information in the wild, where they spend most of their lives in the dark inside wood or soil.
Reviews the photoreceptor genetics and the developmental and transcriptional responses they drive, including the loss-of-function mutants.
Light is the clearest case of a fungal sense with a named molecule behind it, which is why it is here rather than filed under curiosities. A fungus that fruits at the right time of day, orients a spore-bearing structure towards the sky, or resets a daily rhythm is doing so through receptor proteins whose loss, in mutants, removes the response. That is a considerably better-evidenced piece of fungal perception than anything in the electrical literature, and it is a sense rather than a conversation — the fungus is reading its world, not being addressed by it.
Words used here
Volatile organic compound
A small carbon-based molecule light enough to evaporate at ordinary temperatures, so it travels through air. Most smells are these.
Photoreceptor
A protein that changes shape when it absorbs light of a particular wavelength, allowing a cell to detect and respond to it.
The fungal interaction people are most likely to have seen, and it is not cooperation.
Break open a piece of decaying wood and there is a fair chance of finding thin dark lines wandering through the pale interior, dividing it into irregular territories. They are not growth rings, decay stages or staining. Each one is a boundary between two fungal individuals that met inside the wood, and could not displace each other.
The thin black lines wandering through a rotten log are front lines: boundaries where two fungi met and neither could displace the other.
Well supported
Good evidence backs this, though some details remain open.
Encounters between genetically distinct wood-decay basidiomycete individuals are predominantly antagonistic, resolving as replacement of one by the other or as deadlock. Deadlock is marked by pigmented interaction zone lines in the wood, and outcomes reorder with temperature, water content and prior residency rather than following a fixed hierarchy.
Who this applies to
Wood-decay fungi, studied in paired encounters in agar and in wood blocks, and observed in naturally decaying timber.
Studied in
Basidiomycota
Why we rate it this way, and what the caveats are
Well supportedHigh confidence
Repeatedly demonstrated in paired cultures and in wood, and the lines are visible in any decaying log. What is conditional is the outcome of any given pairing, not the phenomenon.
How far it can be extended
Zone lines and combative displacement are observed widely across wood-decay basidiomycetes, though which species wins which encounter is local to the conditions.
Caveats
Paired encounters are a simplification; a log holds many individuals at once.
Zone lines are also produced against the wood itself and against damage, so not every dark line in timber is a fungal boundary.
Combat is the usual outcome between individuals. It is not the whole of fungal ecology, and says nothing about the mycorrhizal relationships the same species may hold with plants.
Still unanswered
How much of the variation in real forest decay rates is explained by which individual won, as against which species is present.
Synthesises the paired-encounter experiments and the field observations of zone lines, including the reordering of combative hierarchies with temperature and moisture.
Supplies the consequence: if which individual holds a piece of wood is decided by combat, then decay rate depends on who arrived and won, which is what this experiment found.
This is the correction the popular picture of fungi most needs. A mycelium fuses readily with itself, and the network it builds with plant roots is a real and important structure — but between two genetically distinct individuals of the same kind, the usual outcome of meeting is combat. One replaces the other and takes its territory, or neither can advance and both invest in a barrier. Which happens is not fixed: warm the wood or dry it and hierarchies reorder, so a fungus that wins in one log loses in the next.
That conditionality is what connects this section to the decay page. Identical blocks of wood, left in identical weather, rot at very different speeds, and which fungi got there first predicts the outcome better than the climate does. This is the mechanism behind that finding. A log is not a passive substrate being processed at a rate set by temperature; it is contested ground, and its fate depends on who arrived and who won.
The fungus that makes an ant climb and bite does not invade its brain. It fills the muscles and leaves the brain alone.
Well supported
Good evidence backs this, though some details remain open.
Three-dimensional reconstruction of Camponotus castaneus infected with Ophiocordyceps unilateralis found fungal cells forming an interconnected network throughout the body cavity and surrounding and penetrating muscle fibres, including mandibular muscle, with the brain conspicuously free of fungal cells.
Who this applies to
One fungus and one ant host, in laboratory infections.
Studied in
Ophiocordyceps unilateralis, Camponotus castaneus
You may have heard
“The zombie-ant fungus takes over the ant’s brain.”
It leaves the brain untouched and takes the muscles instead — which is stranger. The ant’s nervous system is intact while its body is carried somewhere it did not choose.
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
A careful anatomical reconstruction with a clear and unexpected result, but a single host–parasite pair and a description of location rather than of mechanism.
Caveats
Other manipulating fungi work differently; this establishes nothing about fungal manipulation in general.
Where the fungus is does not establish the chemistry by which the behaviour changes.
Still unanswered
Which compounds produced by the fungus act on the host, and where they act.
Fredericksen et al., 2017 · Proceedings of the National Academy of Sciences
Serial sectioning and three-dimensional reconstruction locating fungal cells throughout the body and absent from the brain.
The most dramatic fungal effects on other organisms are chemical rather than electrical, and the best-studied case turns out to work differently from its reputation. Ophiocordyceps infects an ant, changes where it goes and makes it bite down on vegetation before it dies — and a three-dimensional reconstruction of an infected ant found fungal cells filling the body and wrapped around the muscles, with the brain conspicuously untouched. The nervous system is intact; the body is no longer taking instructions from it.
This is worth scoping carefully, because it is the fungal fact most likely to be over-extended. It is a specialised relationship between particular fungi and particular insects, refined over a long evolutionary history, and it establishes nothing whatever about fungi influencing anything larger.
Does any fungus respond to a spike train arriving from elsewhere?
Why it matters: This is the missing experiment. Until a receiver is shown to behave differently, electrical activity remains a property of the organism rather than a message between organisms.
What would settle it: Delivering a recorded or synthesised train to a mycelium and measuring growth, enzyme production or fruiting against a control receiving a scrambled version.
What function, if any, does the spiking serve inside the mycelium?
Why it matters: Coordination across a large body is a real problem that fungi demonstrably solve. Whether electrical activity is part of the solution, a by-product of transport, or something else is unknown.
How much of the recorded activity survives better electrode methods?
Why it matters: Inserting an electrode is a wound, and wounding changes the signal. Long recordings also drift. Separating biology from artefact is the first problem in this field.
How complete this page is, and what it is still missing
NatureHQ publishes its own gaps. This page is at 92% completeness against what we would call a finished subject, and was last reviewed on 2026-08-31. It carries 10 claims and answers 4 mapped search questions.
Quorum sensing is well characterised in bacteria and much less so in fungi; the page says little about how far the analogy holds.
Fungal responses to gravity, and to touch, are still only mentioned in passing. Light now has a section; the others would support one.
Nutrient redistribution within a mycelium — how a discovery at one edge changes allocation at another — is asserted rather than evidenced here, because the measurements are sparse.
The receiver experiment this subject repeatedly asks for has not been run by anyone, so the central open question stays open by default rather than by choice.