The Keystone in the Arch: How a Single Species Can Remake a Whole System; Your Garden Included

An earlier piece in this series built the living world from the bottom up - insects as the foundation everything else stands on. This one turns the picture upside down and looks from the top. Because just as a system can be starved from below, it can be held together, or brought down, from above, by a single species whose influence is wildly out of proportion to its size. Remove that one species and the whole edifice can shift, sag, or collapse.

Ecologists call these load-bearing species keystones, and the story of how we came to understand them is one of the most illuminating in all of ecology. It also, it turns out, has a great deal to say about your garden and, at the end, about any system built of interdependent parts, including the organisations we work in.

The stone that holds the arch

The idea began with someone messing about on a rocky shore. In the 1960s the zoologist Robert Paine ran a now-famous experiment in the tide pools of the Pacific Northwest: he simply removed all the ochre starfish, Pisaster ochraceus, from a stretch of coast and watched what happened. The starfish preyed mainly on mussels, keeping that aggressive, space-hogging competitor in check. With the starfish gone, the mussels ran riot, crowding out everything else, and within a year the number of species on that shore had fallen from fifteen to eight.

Paine had shown something genuinely new: that not all species are equal. One creature, never especially numerous, had been quietly holding the entire community open. At the suggestion of the conservationist Estella Leopold (daughter of Aldo, whom we met in the piece on edges), he gave it a name drawn from architecture - the keystone species, after the wedge-shaped stone at the top of an arch. That stone bears the least load of any in the arch, yet pull it out and the whole thing falls. A keystone species is exactly that: small in weight, decisive in effect.

When the ripple runs downhill: trophic cascades

The keystone's influence often travels through the food web in a chain reaction ecologists call a trophic cascade; the effect of a top predator rippling downward, tier by tier. The cleanest, best-documented example lives underwater. Studying the coast of Alaska in the 1970s, James Estes found that sea otters were the keystone of the kelp forest. Otters eat sea urchins; urchins, left unchecked, mow down kelp. Where otters were present, urchin numbers stayed low and vast, luxuriant kelp forests flourished, sheltering thousands of species of fish, invertebrates and birds. Where otters had been hunted out, urchins multiplied and grazed the kelp down to bare rock, a desolate moonscape ecologists call an "urchin barren." One furry predator, and the difference between a teeming forest and a wasteland. The pattern has since been confirmed again and again as otters have recovered along the Pacific coast.

An honest word about a famous story

You have very likely seen the other great cascade story, the wolves of Yellowstone. The version that went viral runs like poetry: grey wolves, reintroduced in 1995, thinned and spooked the elk; relieved of constant browsing, willow and aspen sprang back along the rivers; beavers and songbirds returned; the riverbanks stabilised; the wolves, it was said, had changed the very course of the rivers.

It's a wonderful story. It is also, the science now suggests, because a rewilding site that only tells the flattering version isn't worth much. A careful 2022 reanalysis by Elaine Brice, Eric Larsen and Dan MacNulty found that earlier studies had inadvertently cooked the books, sampling the tallest, healthiest young trees and so overestimating the recovery several-fold. Take a properly random sample and a trophic cascade is still there, but a weaker, patchier one, tangled up with drought, climate, bison and the beavers themselves, and far from the tidy domino-fall of the popular telling. The debate is still live in the journals today.

So the honest position is this: keystones and cascades are real and important, and predators genuinely shape the systems beneath them, but nature is not a row of dominoes that topples on cue. The effects are often subtler, slower and more context-dependent than a good story wants them to be. (The keystone concept itself has been stretched thin by over-use; hundreds of species now claim the title.) Keep both truths: the principle is sound; the reality is messy.

The builder: keystones that engineer the habitat itself

Not every keystone is a predator working from the top down. Some are ecosystem engineers species that physically rebuild the habitat, and in Britain the great returning example is the beaver.

A beaver doesn't eat its way through a food web; it builds. Its dams slow and pond the water, and out of that single behaviour spills an entire wetland; pools and marsh and wet woodland that host an explosion of insects, amphibians, fish and birds, while also storing water against drought, filtering pollution and blunting the flood peaks downstream (as we explored in the piece on holding water). This matters especially in a country where only around 14% of rivers are in good ecological condition. After centuries of extinction here, the beaver is coming home: in February 2025 the government set out England's first policy for licensed wild release, moving beyond fenced enclosures, and the first licensed wild releases followed in 2025 in Dorset and on Exmoor. One industrious rodent, remaking whole valleys.

Your garden has keystones too

You can't keep wolves, and a beaver would be a poor houseguest, but the principle scales all the way down to a suburban plot, and learning to see your garden's keystones is one of the most useful shifts a rewilder can make. A keystone is any element whose presence or absence ripples out of all proportion to its size.

The pond is the clearest of them. We've made the case before that water is the single fastest way to bring a garden alive, and this is why: a pond is a keystone. Add one and you don't just gain frogs - you gain the frogs and toads that patrol your borders for slugs, the dragonflies that hunt the air, the birds that come to drink, and the clouds of emergent insects that feed the bats at dusk. One feature, a whole cascade.

A native tree is your keystone plant. As the insect food-web piece described, an oak can support something like 2,300 species; even a birch or a hawthorn is an engine of caterpillars, and caterpillars are what turn into birds. Plant the tree and you seed the web.

Your predators are your wolves and otters. Hedgehogs, frogs, ground beetles, ladybirds and insect-eating birds are the top-down control that keeps aphids and slugs from overwhelming everything, the living alternative to a bottle of spray. Rewild for them and they do the pest control for free.

And here's the sobering mirror image. If keystones can build a system, removing one can unravel it. Fell the single mature tree, fill in the pond, pave the plot, or spray out the insect life, and you create the garden's own version of an urchin barren, a place where, with the checks and balances gone, a few opportunists take over and the richness drains away. Worse, remove a top predator and you can get what ecologists call mesopredator release: pull one thread and the wrong things multiply. Gardens, like coastlines, have load-bearing parts.

Which leads to the most empowering thought of all: in a space this small, you are the keystone. No other single agent has anything like your influence over whether this patch of ground teems or empties. The pond you dig, the tree you plant, the spray you put down, the predators you welcome, these are the decisions the whole system pivots on.

From the tide pool to the boardroom

It would be a shame to leave the idea in the garden, because the keystone is really a truth about any system built of interdependent parts, which is to say, the organisations we work in as much as the ecosystems we walk through.

Every team and company has its keystones: the quiet individual who holds a web of relationships and knowledge that no org chart records; the modest ritual or norm that turns out to be structuring the whole culture; the small practice that bears far more load than its prominence suggests. The systems thinker Donella Meadows spent her career mapping what she called leverage points - the places in a system where a small, well-chosen intervention produces outsized change. It is the keystone concept in another language: not everything matters equally, and the highest-leverage elements are rarely the biggest or the loudest.

The lesson for anyone ''leading'' a system, whether in a garden or a business, is twofold, and it runs in both directions. First, find your keystones and protect them, because they're holding up more of the arch than anyone notices. And second, beware the efficiency drive that quietly removes something small and load-bearing - the "redundant" role, the "unnecessary" ritual can lead to culture sliding into an urchin barren of its own, as the checks that held things in balance vanish and the opportunists move in.

Be the keystone

From Paine's starfish to the returning beaver to the pond at the bottom of your garden, the thread is the same: systems are not sums of equal parts. A few elements hold everything else up, and finding them, protecting them and restoring them is where the real leverage lies, whether you're mending a coastline, a catchment, a company or a back garden.

In the garden, the happiest part is that the keystone is within your gift. Dig the pond. Plant the tree. Put down the spray and let the predators back. Do those few load-bearing things, and the rest of the living web will rise to meet them, because you will have become the keystone in your own little arch.

References & further reading

• Paine, R.T. (1966). "Food Web Complexity and Species Diversity," The American Naturalist; and Paine, R.T. (1969), which introduced the "keystone species" concept. (The Pisaster starfish experiment and the arch metaphor.)

• Estes, J.A. & Palmisano, J.F. (1974). "Sea otters: their role in structuring nearshore communities," Science, 185, 1058–1060; and later otter–urchin–kelp studies (Estes & Duggins, 1995; and 30-year recovery data, 2025).

• Brice, E.M., Larsen, E.J. & MacNulty, D.R. (2022). "Sampling bias exaggerates a textbook example of a trophic cascade," Ecology Letters, 25, 177–188. (The honest reappraisal of the Yellowstone wolf story; the debate continued in Global Ecology and Conservation, 2025.)

• Rewilding Britain — keystone species and trophic cascades; and Carroll, S.B. (2016) The Serengeti Rules. https://www.rewildingbritain.org.uk/

• Defra / Natural England (Feb 2025) — Wild release and management of beavers in England; first licensed wild releases (Dorset; Holnicote, Exmoor). https://www.gov.uk/government/publications/wild-release-and-management-of-beavers-in-england

• Tallamy, D.W. — keystone plants and the oak (see also the companion piece in this series on the insect food web).

• Meadows, D. (1999). Leverage Points: Places to Intervene in a System. The Sustainability Institute. (The systems-thinking cousin of the keystone concept.)

• Companion pieces in this series — on the insect food web, holding water, ten ways to rewild your garden and living soil.