Most trees you walk past in a forest are not standing alone. Beneath the soil, their roots are wrapped in — or penetrated by — microscopic fungal threads called hyphae, which connect individual trees into a shared underground network. The relationship between tree and fungus is called a mycorrhiza, from the Greek for "fungus" + "root". The network as a whole is often nicknamed the Wood Wide Web.
About 80% of land plants live with mycorrhizal partners, and the relationship is genuinely ancient: it dates back at least 450 million years, to when the first plants colonised bare rock and needed help reaching nutrients. Life on land as we know it would arguably not have been possible without these fungi.
The two main kinds of mycorrhiza
- Ectomycorrhizal fungi wrap around the outside of root cells, forming a dense sheath (the Hartig net). Common in temperate forests — oaks, beeches, pines, spruces, birches, Douglas fir. The fungal partner is typically a Basidiomycete or Ascomycete, and the association is visible to the naked eye as a fuzzy fungal mantle on a fresh root tip.
- Arbuscular (endomycorrhizal) fungi grow into root cells, forming branching tree-like structures called arbuscules. More common in tropical forests, grasses, and most crop plants. The fungus is from the Glomeromycota, a phylum that has lived symbiotically with plants for so long it has effectively lost the ability to live independently.
A handful of plants associate with both types, and a few (e.g. some heathers, some orchids) form yet further specialised associations. The vast majority of the world's biomass is supported by one of the two main types above.
The deal: sugars for minerals
Mycorrhizae are a two-way mutualism. The tree gives the fungus sugarsmade by photosynthesis — in some forests, up to 30% of the tree's photosynthate is sent underground. In return, the fungus gives the tree access to water and minerals(especially phosphorus and nitrogen) that the tree's own roots are too thick and too few to reach.
Why phosphorus? Because in most soils, plant-available phosphate is present at concentrations of just a few micromolar — tightly bound to iron and aluminium oxides and effectively immobile. A single teaspoon of healthy forest soil can contain several kilometres of hyphae. Where a tree root is millimetres wide, a hypha is only micrometres — thin enough to slip into soil pores inaccessible to roots and bring back nutrients from a volume hundreds of times larger.
What flows through the network
- Carbon — Suzanne Simard's landmark 1997 Nature paper used radioactive tracers (¹³C and ¹⁴C) to show that carbon labelled in birch trees appears in neighbouring Douglas firs through the fungal network — and vice versa, depending on which species had the photosynthetic advantage that season. Net flow can reverse with the seasons: birches drop their leaves in winter and become net receivers from evergreen Douglas firs.
- Nutrients — phosphorus, nitrogen and minerals move from hub trees to shaded seedlings, dramatically improving their survival. In experimental setups, seedlings deprived of network access show 3–10× lower survival rates than connected ones.
- Water — hyphae conduct water along potential gradients, helping plants survive dry patches. The hyphal "hydraulic continuum" reduces the rate-limiting resistance of soil micro-pores.
- Chemical warning signals — when a tree is attacked by insects, defence hormones (especially jasmonic acid and its volatile ester methyl jasmonate) can travel through the network and switch on defences in neighbours before the attacker arrives. This has been demonstrated in pine forests under bark beetle attack and in tomato/bean systems in greenhouse studies.
Mother trees and hubs
Some trees in the network are vastly more connected than others. These are the mother trees— usually old, large individuals at the centre of dozens or hundreds of fungal connections. Simard showed that a single old Douglas fir can be connected to hundreds of other trees of several species. They are the hubs of what mathematicians call a scale-free network — the same kind of structure as the internet, airline routes, or your brain's neuronal connectivity. We will analyse the spectral properties of such networks rigorously in Module 4.
A genuine scientific debate
The poetic "Wood Wide Web" framing has been criticised by some scientists as overly romanticised. The cooperative view (associated with Simard) emphasises that trees support kin, seedlings and dying neighbours. The sceptical view (Toby Kiers, Justine Karst and others) argues that fungi are primarily acting in their own interest — redistributing nutrients to keep productive partners alive, not out of forest-level altruism. The 2023 review by Karst, Jones & Hoeksema in Nature Ecology & Evolution argued that many popular claims (trees "talking", forests "caring" for seedlings) are not adequately supported by current evidence, and that positive-citation bias has amplified weakly substantiated findings.
The truth is probably in between: cooperative outcomes can emerge from individually self-interested actors, without anyone having to "intend" cooperation. This is the same theme as biological markets in economics, swarm intelligence in ants, and the emergent behaviour of brain neurons. Kiers's own 2011 Science paper showed that arbuscular fungi and roots engage in mutual reward: roots send more carbon to fungal strains delivering more phosphate, and fungi deliver more phosphate to roots paying with more carbon. Cooperation appears as the equilibrium of a local market — without anyone calculating market prices. It is what the rest of this course tries to capture mathematically.
What disrupts the network
- Soil compaction (heavy logging machinery, construction) physically tears hyphae and reduces porosity.
- Fungicides and some pesticides directly poison fungal partners.
- Excessive nitrogen pollution from agriculture and combustion: when nitrogen is abundant, trees invest less in fungi, weakening the network. This is a major issue in European forests downwind of intensive farming.
- Clear-cutting removes the hub trees that anchor connectivity. Even when replanted, young plantations have impoverished mycorrhizal networks for decades.
- Fire regime change: too-frequent fires destroy soil organic matter and hyphal networks faster than they can regenerate.
Old-growth forests have far richer mycorrhizal networks than young or managed plantations — a scientific argument for preserving large old trees rather than replacing them with fast-growing monocultures.
Why it matters beyond ecology
- Forest management: planting seedlings with access to an established network increases their survival 3–10×.
- Climate: mycorrhizal networks store a significant fraction of soil carbon. Damaging them releases carbon to the atmosphere.
- Agriculture: commercial mycorrhizal inoculants can reduce fertiliser use in maize, wheat and many vegetables. The technology is still maturing but is a serious item on the regenerative-agriculture agenda.
- Philosophy and cognition: the network forces us to reconsider what "intelligence", "communication" and even "individuality" mean — without resorting to anthropomorphism.