The mycorrhizal network

Almost every land plant on Earth lives in symbiosis with one or more species of soil fungus. The plant gives the fungus sugars produced by photosynthesis. The fungus gives the plant water and mineral nutrients — phosphorus, nitrogen, micronutrients — that it can extract from a much larger volume of soil than the plant’s own roots can reach^[1]. This relationship is called mycorrhiza, literally “fungus-root.”

It is not a fringe arrangement. Roughly 90% of vascular plant species form mycorrhizal associations, and most major forest tree species are obligate — they cannot grow normally without their fungal partners^[1][4]. The above-ground forest you see is the visible third of a much larger living system. The other two-thirds is the soil column, the fine root mass, the fungal hyphae, and the decay community.

What the network actually does

The two big trades are well documented and uncontroversial:

Trees give fungi sugars. Around 5–30% of a tree’s annual photosynthate ends up moving through its mycorrhizal partner^[1]. This is a substantial energy subsidy from the plant kingdom to the fungal kingdom, and it underwrites the entire forest soil food web.

Fungi give trees water and minerals. Mycorrhizal hyphae are typically a hundred times thinner than fine roots and reach into pore spaces roots cannot enter. They scavenge phosphorus from soil minerals, mobilise nitrogen from leaf litter, and deliver water across the dry season. Forest trees grown without their fungal partners are stunted, water-stressed, and more susceptible to root pathogens^[4].

A third trade is more recent and is where the science gets careful. In 1997, Suzanne Simard, then at UBC, published the first field demonstration that carbon labelled with isotopes moved between paper birch and Douglas-fir saplings through a shared ectomycorrhizal network^[2]. That paper made the cover of Nature, kicked off two decades of research into the so-called “wood-wide web,” and inspired a popular book in 2021^[3]. The honest summary in 2026 is: yes, carbon and signalling molecules do move between trees through fungal networks under some conditions, in some forest systems. The strongest claims — that mature trees deliberately allocate resources to specific kin saplings, that defence signals propagate routinely between trees — are still being evaluated and replicated, and at least some of the popular framing has run ahead of the field-measurement evidence^[4]. The conservative reading is the safe one: fungal networks are real, ecologically important, and not yet fully characterised.

Why the network is also old growth

The mycorrhizal community in a forest is itself successional. Pioneering tree species, on disturbed soil, partner with a small number of fast-spreading fungal species. As the forest matures and the soil organic-matter horizon thickens, the community shifts toward longer-lived, slower-growing, and more diverse fungi. By the time a stand carries the structural indicators of old growth, its fungal community is also distinct from younger stands — richer in species, more partitioned by depth and substrate, with persistent rhizomorphs that have lived through the same centuries the trees have^[4].

This is the part that “just plant more trees” fails on. The above-ground trees can be replanted. The mycorrhizal community cannot be replanted — it has to recolonise from inoculum sources, regrow through compacted post-clearfell soil, and re-establish the depth-stratified, species-stratified architecture that built up over centuries. Recent work from groups like the Society for Protection of Underground Networks (SPUN) is mapping fungal diversity globally and finding that the highest-diversity hotspots correlate strongly with old, undisturbed forest^[5]. They do not correlate with stands that were clearfelled and replanted.

Implication for the things humans actually do

Three practical consequences fall out of the network biology, each well established in the soil-science literature^[6]:

1. Soil disturbance is the main thing to avoid. Compaction by heavy machinery, burning of slash, removal of the duff layer — these damage the fungal community far more than the cutting itself does. Some forestry methods (selective harvest with low-impact equipment, helicopter logging in sensitive terrain, indigenous-led harvest) preserve most of the fungal community. Conventional clearfell with whole-tree removal does not.

2. Reforestation outcomes depend on inoculum. Successful reforestation, especially of slow-growing native species, often requires either retaining old-growth refugia within reach of the planting site (so fungi recolonise naturally) or actively inoculating seedlings with appropriate fungal partners in the nursery.

3. The carbon cycle has a fungal floor. Most of the long-lived organic carbon in forest soils is fungal — cell walls, secondary metabolites, dead hyphae — not plant tissue. Damaging the fungal community during “forest restoration” can release more carbon than was sequestered by the new trees for their first several decades.

How to think about it

The most useful mental shift is to stop thinking of a forest as trees plus background and start thinking of it as trees plus fungal community plus soil column, with each piece building or eroding the others. From this angle, “protect old growth” and “keep the soil intact” are the same instruction. From this angle, “tree-planting offset” programmes that plant on degraded fungal soil are doing visibly less work than the same number of trees in a stand that already has its fungal partners.

The next time someone says we’ll just regrow it, ask: regrow the trees, or the fungal community? They are not the same time-scale.