Take three spores from a green or black mould, from one spore-bearer, siblings that sat shoulder to shoulder. Put them on a dish of agar or potato starch and wait. You get three colonies, three wedges, and between them a thin line of bare substrate where nothing grows.
Mikhail Vishnevsky, a mycologist, military doctor and author of more than twenty books on fungi, reads that line as a front. It is where the colonies meet and close up, and he calls it a "zone of military action". Each colony pours out substances that stop the neighbour from growing and spreading. The antimycotics and fungicides made there, he says, are the kind you can only marvel at. The three are family, and none of it matters. "No friendship, no merging": that is how it goes, every time.
This is the opposite of a picture many of us carry. Paul Stamets, the American mushroom entrepreneur, describes mycelia running through the whole planet, joined to one another, like the world-wide plug-in of the film Avatar, where you take yourself by the braid and connect to everything. Vishnevsky says it doesn't work that way. The forest you walk through is a patchwork of territories, and the fungi under your feet are neighbours with borders.
In every case, the relationship between mycelia is either ignoring or war. They never merge.
Во всех случаях отношения между мицелиями это или игнор, или война. Они никогда не сливаются.
— Mikhail Vishnevsky, Gleb Solomin Podcast, 86:59
There is one exception, and it is short. When two haploid mycelia meet, mother and father, they join to make the two-nucleus mycelium that will later grow mushrooms. That is reproduction. Outside it, the choice is to look past each other or to fight.
Size is the other problem. A mycelium is not even an anthill. It can have few cells, sometimes a single one, because many mycelia have no walls between their cells and are one continuous structure. Ants have hundreds of thousands of neurons and people have 86 billion. A fungal body has nothing like that to pool.
Cooperation does exist, and Vishnevsky grants it. A big old mycelium of a porcini or a birch bolete can spread over several square metres and weave the roots of one birch, then a second, then a third. If one of the birches runs short of water, the fungus can pump water through itself to that tree. It does not do this out of pity. The birches don't care about each other, he says, and the fungus doesn't care about everything around it either. It sits on those roots, and they have to be well fed. Its partners do well, so it does well: "our little feeding grove". He draws the line there: a positive arrangement you can imagine, "but no wider".
A theoretical model published in PNAS in January 2025 by Thomas W Scott arrives somewhere close. In the model, warning signals that plants pass along fungal networks are rarely evolutionarily stable. They help competing neighbours at a cost to the plant that sends them. A fungus, though, can be selected to watch its host plants and pass warnings on to protect its own trading partners.
The line between colonies now has a gene attached. A bioRxiv preprint in June 2026 reported the first allorecognition locus in a mushroom-forming fungus. The stretch of DNA, named somA, sits on chromosome 5 of Coprinopsis cinerea. It is linked to the barrage zone, the antagonistic line that keeps two different individuals apart.
So the planet-wide fungal mind has a plain obstacle: the fungi won't join. Vishnevsky does leave one door ajar. If every mycelium on Earth did merge and pull the trees in as well, then, in the Marxist phrase, quantity might turn into quality. "Who knows?" he says. They never merge, so nobody gets to find out.
