More photosynthesis per square meter
In its first years, a tree uses only a fraction of the light that falls on its patch. Everything green below and beside it catches the rest, and in the end every leaf feeds the soil.
Plant guilds
A plant guild is a group of plants around a central tree in which every species contributes something, and in which some members are meant from the start to stay only for the first few years. This page explains what that means in practice, when the effort pays off and when it doesn't, and walks through three real-world examples: an apple tree in a home garden, a meadow orchard and a walnut.
A group of plants around a central tree that get along well and where every species chips in: fixing nitrogen, covering the soil, supplying mulch, feeding beneficial insects, blocking wind, keeping voles away.
What makes it special is the time frame. The tree in the middle (let's call it the star) gets bigger every year and changes everything around it: light, water, root space. So a guild is never finished. What stands in the sun today will stand in the shade in ten years, and some plants are meant only for the early years from the start: the nurse, which helps the young tree along and later makes way, or the placenta (herbs and annuals that cover the tree circle for the first few summers and then disappear by themselves).
That is why a guild always comes with a timeline. This is the real difference from a simple "what grows well with what?" chart.
Companion planting is about one season: which vegetable grows next to which in the same bed, this year. Carrots and onions, cabbage and celery: there are proven charts for this, and for annual crops they are exactly the right tool.
A plant guild revolves around a woody plant and thinks in decades. The questions are different: How wide will the canopy get? How much light will be left underneath in year X? What will still be standing then, and what should have been removed by that point?
The two don't exclude each other. Under a young fruit tree, vegetables in polyculture can do perfectly well for the first few years, until the shade arrives. You just shouldn't plan them as one thing.
Because a tree on its own in a mown lawn gets almost none of what a group can give it. Here are five things a well-stocked guild does:
In its first years, a tree uses only a fraction of the light that falls on its patch. Everything green below and beside it catches the rest, and in the end every leaf feeds the soil.
Roots at different depths, mulch grown on the spot, covered ground that soaks up rain instead of letting it run off. Grass does almost none of this; a guild does all of it.
An apple tree takes years to produce its first fruit. Strawberries, herbs and berry bushes yield from the first summer on, so the patch is productive while the tree is still growing.
Flowers all season, cover in the shrubs, food in winter. If you want less aphid and scab trouble on your tree, the guild brings their natural enemies up into the canopy.
Holding water, cooling the soil, slowing the wind, storing carbon. These effects are measurable, and they are the reason a forest garden is more than a collection of fruit trees.
The team around the star
A guild works like the team around a star player: one carries the main harvest, everyone else makes that possible. A single plant often fills several roles: comfrey is a deep-rooter, ground cover, mulch supplier, medicinal herb and bee forage all in one. And the other way round, several species often share one role. So treat the list as a checklist, not a shopping list.
To be honest
I have been practicing permaculture for more than fifteen years, and in all that time I have never thought in guilds. I think in layers and functions. So I have certainly planted an evergreen elaeagnus next to a fruit tree, or worked out suitable ground covers and planted them under trees. But I have never set out with the intention of building a textbook plant guild. There is a reason for that.
The usual depictions of plant guilds give the impression that you have to create something perfect: tree X needs exactly this nitrogen-fixing companion, plus a windbreak of these small trees, these berry bushes with these ground covers. When I transfer an example like that to my own garden and think it through, I notice that it doesn't fit at all. I already have a windbreak here in the form of a hedge. I might want my star right next to a path, and that's where the trampoline or the patio is. I also want to keep that one sightline. And so on.
So I prefer to look at the big picture first. From the vision to the planning of individual spaces; there, the question of which stars absolutely have to go into the garden and where they fit best. Then: which small trees and shrubs are wanted, which ones suit these spots particularly well, and which companion plants in which places support and strengthen the whole concept.
People constantly ask me which trees and shrubs can be planted together and how well they get along, much as if they were companion plants in a vegetable bed. So I understand the enthusiasm for guilds. They are good ideas and inspiration. Then they have to be adapted to the specific site, and that's why I don't follow a guild recipe. I follow the guide further down.
And in the end, an elaborate combined planting like this costs time and money, when the whole point is usually to save exactly that. Putting dozens of ground covers under a tree, only for them to be overgrown by the meadow or shaded out after a few years, is a lot of effort if it doesn't produce a satisfying yield. Yield can also simply be the joy of the flowers and insects that are there during that time. In a small home garden, where every square meter counts, guilds are a lovely inspiration for creating a colorful density in a tight space that is ecologically valuable and healthy for the tree.
That way a home garden quickly turns into a forest garden, and a forest garden is basically a network of guilds. Then you have to look at which guild goes with which. That's exactly why I work the other way round: instead of mentally placing ready-made guild sets everywhere, I first place the stars, after the hedge, the privacy screens and windbreaks and the important functional plantings (against erosion, for example). Then I look at how their shadows move through the day and across the year, how they interact, and I think about the spaces in between: Which guild partners do two neighboring stars have in common? Which of those really serve me and the concept, and which would rather get in the way in my case?
In order, that means: plan water and paths, place the stars, then decide at which points smaller trees and shrubs of which height should go in between, and which guild roles are still unfilled. Ground covers come last (for the first five years and for afterwards), and only if the area is within my daily attention.
A meadow orchard lies outside of that. It is visited rarely, basically only for harvesting and pruning. In a home garden it makes sense to plant strawberries under the young apple tree, but in a meadow orchard I sow beneficial wildflowers and perennials at planting time, and that's where my work there ends. Maybe I'll add elaeagnus that ripens at the same time as the apple variety, or other berry bushes, if that fits my harvesting system and I know I'll have the time and space to cut the elaeagnus down and chop it up later.
On arable land, where rows of trees are meant to enrich the field, you can build a syntropic agroforestry system: in the tree row, much like in a guild, there are lots of plants in between that later have to be cut back and eventually removed entirely. If you have the time and money, and still benefit from the interplantings during maintenance and removal (or from privacy and wind protection, food for insects and birds), you can implement guilds there too, designed as a row. On some sandy sites this is a valuable soil improvement, or even a prerequisite. But everyone has to factor in the extra work in planting and maintenance.
Worth it. Little space, lots of attention, and the harvest should start early. Every layer pays off here: strawberries under the young apple tree, herbs at the edge, ground covers later on.
Worth it. A forest garden is a network of guilds. You just don't plan it guild by guild: first the stars, then the spaces in between.
Probably not. Sow wildflowers at planting time, maybe add an elaeagnus with the same ripening time, and that's it. Anything more will disappear into the meadow anyway.
It depends. As syntropic agroforestry, definitely, if there is time and money for maintenance and removal. On poor sand it can be what decides whether the row establishes at all.
The most useful rule of thumb
Big picture first, then the details. What protects the whole site? Where does the wind go, where does the water flow, where are the paths? Whatever you can solve at that level, you don't have to repeat at every single tree.
And then plant only what really serves a purpose, not whatever is still unchecked on a list of roles.
Method
As described above, you don't plan an overall design guild by guild but in layers: first all the stars, then the next layer down of medium and small trees, then shrubs, then ground covers. But every time we get close to a star along the way, we go through the following:
A star is a large fruit tree, a nut tree or another canopy crop tree. What matters is how big it will be in twenty years, not how small it looks in its pot today.
Tip: If you have time, don't plant the star first. Start with the nurse or sacrificial trees: nitrogen fixers, soil preparers, pioneers such as black locust, alder, poplar or sycamore maple, depending on the site and the concept. The meadow orchard example below shows what that looks like.
After soil and climate, there are two things that leave no room for maneuver. Juglone: a growth-inhibiting compound that walnut trees release through their roots and leaves. Whatever refuses to grow next to a walnut won't grow there with a good timeline either. And shared diseases, especially among members of the rose family (fire blight, brown rot). Everything else is a question of time.
The big ones cast the most shade and so decide what can still grow underneath later. The emphasis is on later, which brings us to:
You can plant closer than any chart allows if you know you will step in. And you can put light-hungry species under a tree that will cast shade later, as long as they have been harvested or have disappeared by themselves by then. So a guild is not a picture but a process: a succession.
| Time window | What's there | What happens |
|---|---|---|
| Years 1-5 | Young star, nurse, placenta of annuals and herbs | Cover the soil, hold back the grass, let the tree establish. The harvest comes from the herb layer. |
| Years 5-10 | Shrubs bear fruit; the nurse is cut back or removed | The canopy casts shade. Whatever needs sun moves outward or goes. First harvest from the star. |
| From year 10 | Star, small tree, shade-tolerant ground covers, mulch plants | The understory is reshaped. The guild produces its own mulch. |
The roles from above are a checklist to tick off, not a shopping list. One plant often covers several of them, and many are long since filled on your site: if you have a hedge, you don't need another windbreak.
Care, harvest, use: these matter more than any compatibility chart. I gather nuts from the ground, so it has to stay walkable there. I pick cherries from the tree, so dense shrubs can grow underneath. With apples, plums and pears it depends: Does a ladder still need to reach the trunk? Do I want to have picnics underneath and keep the area clear? Should the sightline stay open? Ask these questions beforehand and you won't have to grub things out later.
Example 1 · Home garden
The most common case of all, and the one where the most goes wrong, because hardly anyone thinks about the grass.
For a young fruit tree, grass is the strongest competition there is. Its dense mat of roots takes exactly the water the tree needs to establish in its first years, and it does so in the same top layer of soil where the tree's fine roots are.
If you have a year, clear the area beforehand, and the most pleasant way to do it is with hay potatoes: lay seed potatoes on the turf (ideally on a first thin layer of hay), cover them thickly with hay or straw, and top it up once more in summer. In autumn you lift the potatoes out of the mulch, and underneath is an open, grass-free, well-rooted patch of soil. That is exactly where the tree goes in winter.
If you don't have that much time, hoe the tree circle clear and mulch it. Important: the graft union must never disappear under mulch or soil. Otherwise the scion variety grows its own roots, the rootstock loses its effect, and your small tree turns into a big one.
The tree circle stays open and mulched, but it doesn't have to stay empty. As long as the canopy casts hardly any shade, quite a lot fits in there:
What doesn't belong there is potatoes, at least not anymore. When harvesting, you dig exactly where the tree's fine roots now are. Before planting they are ideal; afterwards they are not.
And a nitrogen fixer? On established garden soil it is usually not needed. More nitrogen means more shoot growth, more pruning, softer cell walls (meaning branches that break more easily under the weight of fruit) and more aphids. If you are planting on poor sand, though, add an elaeagnus.
Once the tree bears fruit and the canopy casts enough shade that the first crew slowly dies back, we sow ground covers such as clover into the gaps and gradually plant small pieces of comfrey root underneath. A comfrey root can be split into dozens of small pieces, and each one grows into a new plant. It is also called a dynamic accumulator: it roots deep and produces masses of leaves, which you scythe several times a year and leave lying as food for the soil, ideally just before the fruit drops. Then the fruit lands softly on the comfrey leaves and doesn't get lost in the grass. Comfrey is also bee forage and good fodder for grazing animals, especially varieties like Bocking 4.
Beyond that, I look at how much harvesting work I still want from the patch. Further from the kitchen I use easygoing ground covers suited to the site that keep the grass out, and kitchen herbs for partial shade, as long as they don't get in the way of tree care or harvesting. And at some point an apple tree also makes a lovely climbing or swing tree for children. Once it has reached that size, the grass can happily take over again.
Example 2 · Meadow orchard
Out in the meadow, different rules apply: little time per tree, wide spacing, no watering. Whatever helps here has to be cheap and work on its own.
One year before the fruit trees arrive, push a hybrid poplar cutting into the ground at every future planting spot: a shoot two to five centimeters thick and one to two meters long, stuck in the soil, nothing more. Poplars and willows root from these by themselves. Cost: with a bit of luck, nothing, if you know someone with a poplar.
We then plant the fruit tree just west of it and girdle the established poplar near the base, meaning we remove a strip of bark all the way around. From now on it takes on several jobs at once:
Girdling is the key: with a ring of bark removed around the trunk, the poplar slows down noticeably and doesn't become a threat to the fruit tree. Later it is sawn off at ground level, and the spot is simply mown along with the rest until it gives up.
And then comes the real payoff: its roots stay in the soil and now serve as both pathfinder and food. The fruit tree can simply take over a root system that has drilled through the soil for years, deeper than any turf.
A poplar doesn't fix nitrogen. It is still one of the better nurse trees you can have, which shows that "nurse" means more than "nitrogen fixer".
Example 3 · The special case
The only case where a species is truly ruled out rather than just a point in time. Walnuts release juglone, a compound that inhibits the growth of many neighbors. It is strongest under the canopy, and present in smaller amounts wherever walnut roots reach, which is far beyond the canopy. No timeline and no pruning will help against it.
| Tolerates juglone | Doesn't tolerate it | |
|---|---|---|
| Trees | Hazel, oak, maple (most), quince, hawthorn, black locust, honey locust, white mulberry, pecan | Apple, pear, sweet chestnut, linden, alder, hackberry |
| Shrubs | Elder, wild roses, most viburnums, rose of Sharon, black raspberry, wild grapes | Blueberry, blackberry, aronia, grapevine, some viburnums, some ornamental cherries |
| Perennials and herbs | Jerusalem artichoke, onion and leek, saffron crocus, sweet cicely, daylily, coneflower, bee balm, hollyhock, sedum, white clover, dandelion, chickweed | Rhubarb, asparagus, alfalfa, lilies, perennial kale, bush clover, and from the vegetable garden tomato, potato, pepper |
After Jacke & Toensmeier, Edible Forest Gardens Vol. 2. It is a North American list, so some European species are missing. Cherries and plums vary: some ornamental cherries are sensitive, some sweet cherries tolerant.
Here is how it plays out over time, using the example on the right: for the first five years, a sycamore maple or a poplar stands right next to the star as a mulch tree; after that it goes entirely. Elder and elaeagnus are allowed to come into bearing, get shaded out by the canopy after five to ten years and are then removed, partly so there is room under the walnut again to gather the nuts from the ground. The hazels stay and provide the second nut harvest, and the raspberries stay too; other berries would work just as well instead. And in rows rather than arcs: that is more practical, especially for tying in. If you need to keep out wind and cold from the north, put a row of arborvitae there.
About the nurse: if you want to prune little later on, choose alder. The walnut will slow it down anyway, and that takes care of the removal by itself. If you want to give the processes a stronger push, choose an elaeagnus or a black locust; with black locust, keep an eye on height, growth rate and suckers.
Because the walnut is a late leafer, it stays bright underneath until May. A bed of daylilies fits on the sunny side, though you will have to scythe it if you want to gather the nuts properly from the ground. Or limb the walnut up high and grow onions underneath.
Elder would also work under the walnut, because like elaeagnus it can be limbed up. If you plant dense shrubs underneath instead (roses, currants or gooseberries), you lose the chance to gather the nuts properly from the ground. The ground under a walnut should stay walkable. A cherry is different: we harvest it from the tree, not from the ground, so dense shrubs can grow underneath.
In the app
The tedious part of a guild isn't the idea, it's the bookkeeping: how wide will each plant get, when is pruning due, what comes out when. That is exactly what the app takes off your hands.
Your site in year 1, 5, 10 and 20. Nothing makes it clearer, faster, why dense planting is fine, and when the nurse has to come out.
Climate zone, heat sum, water balance, soil fertility, pH and stone content for your coordinates. The filter that comes before any guild question.
With mature height, mature width, growth rate, light requirements, flowering and harvest times. Look them up in the plant encyclopedia or find ones that suit your site with the plant finder.
What is due in which year: coppice the nurse, replace the starter plants, replant ground covers. The part that otherwise gets lost in your head.
Suggestions role by role, each with a reason and with the date that goes with it. Plus a check for existing gardens: what's missing, what clashes.
Stored for all 16 German states, Austria, Switzerland and neighboring countries, and applied automatically while you draw. All the rules.
Next steps
Everything on this page revolves around a group of plants around one tree. The more important decisions usually lie one level up: Where does the water flow, where does the wind come from, where do the paths run? Which pattern repeats across the site? What should this land be in twenty years?
In Permaplans, vision planning helps with that. And if you want a proper design for your land, with someone who looks at the site and thinks it through with you: that's what I do professionally, as a permaculture designer.
Taro, permaculture designer and forest gardener. More about me
Why a forest garden? The framework in which guilds make sense in the first place: layers, yield, soil, and how to get started.
Mycorrhiza: which plant goes with which fungus. Why woody plants of the same fungal type belong next to each other, and what is left of the "Wood Wide Web".
Harvest calendar: what ripens when, for 188 hardy food plants.