Best companion plant combinations!

Hi permies,

I've compiled a list* on what are the best companion plants based on their relationships to each other. If a plant has a mutually good relation to every other plant in the group then it forms a group with the other plants!

Companion plant combination with 5 plants
(there is only one with 5 plants!)

Common bean (climbing/pole), Carrot, Marjoram, Cucumber, Lettuce

Companion plant combination with 4 plants

Squash, Common bean (climbing/pole), Marjoram, Corn

Squash, Broad Bean, Marjoram, Corn

Squash, Runner Bean, Marjoram, Corn

Common bean (climbing/pole), Potato, Marjoram, Corn

Common bean (climbing/pole), Carrot, Marjoram, Cucumber

Common bean (climbing/pole), Carrot, Marjoram, Lettuce

Companion plant combination with 3 plants
(can you spot the three sisters? :) )

Common bean (climbing/pole), Sage, Rosemary

Sweet cherry, Marigold, Lettuce

Potato, Broad Bean, Brussel Sprouts

Squash, Common bean (climbing/pole), Corn

Asparagus, Tomato, Parsley

Runner Bean, Cucumber, Lettuce

Beet, Lettuce, Broccoli

Carrot, Marjoram, Onion

I found way more combinations but I couldn't find a way to show it properly in the forum, so I've put them on a page here: https://permadb.com/companion-plant-combinations

Let me know if you want me to lookup anything else in the data!

ben

* These lists are all based on data, I've added from various resources to a database and then used this to compute these lists. As a few of you might already know, I am working on a permaculture plant database which is open for everyone to use and change. Please see this thread for more info!

Just wondering What criteria did you use to determine a “ relationship?”

For me it is usually, similar to crop rotation in that I plant things from different plant families, or similar to a pasture mix (legume+grass+squash+herbs+leafy green)
Legume family + grass/corn family + squash/melon family + mint/thyme family + carrot/celery family + cabbage family + spinach family/lettuce family

Janet Reed wrote:Just wondering What criteria did you use to determine a “ relationship?”

Hi Janet! It is based on the database behind. There you can add connections between plants. For example you can add that squash is a good companion to corn. And you can also say corn is a good companion to squash. See here for example.

I then queried the whole database and said: Give me groups of 3,4 or 5 plants which have only these mutual connection between themselves. There is no group (yet!) of 6 plants which has this by the way.

The source of the data is what people added to the database. I've added these companions by myself, mostly based on these sources.

S Bengi wrote:For me it is usually, similar to crop rotation in that I plant things from different plant families, or similar to a pasture mix (legume+grass+squash+herbs+leafy green)
Legume family + grass/corn family + squash/melon family + mint/thyme family + carrot/celery family + cabbage family + spinach family/lettuce family

Hey S Bengi,

can you please elaborate on that a bit. If I understand this right, there might be plants which are maybe not good companions but good follow up crops. Is this right? If yes, then I could imagine another type of relation which would be "good follow up plant for crop rotation" or something like that. What do you think about that?

Legume family and the Onion Garlic family don't do too well togather, due to allicin compounds that the garlic family produces, luckily the garlic family have tiny, shallow root systems. So higher order legumes are modtly okay with them.

Things in the same family are usually okay with each other, but they usually share pest, so not really a win.

Cauliflower and Broccoli, seems very picky about who they are grown with. As to why I am not too sure.

Everyone seems to love the mint/thyme family because they help confuse bad bugs and provide homes for the good bugs. Marjoram seems to be the best, but I wonder if Winger Savory is even better.  

Legume enrich the soil for everyone even for the onion family, but the allicin compound affects the nitrogren fixation routine. But for everyone else they love the fact that the allicin compound cuts down nematodes and other soil pest.

I was thinking more along the lines of S Bengi and wondering if there was any scientific basis for your data base

These are interesting thoughts! Currently there is no scientific research behind the database, beside the experience of users contributing to it or the external sources (websites/books/...) which are linked to a piece of information added.

I think in general it is a problem in permaculture that there is very little "science" behind most information you have on permaculture. My assumption is, that no one really brought all the info together in one place, so many people can cross-check and add their thoughts to it. I think while "real science" might not be possible so easily, a crowd-sourced version of a wikipedia style platform would work. If you have a lot of people saying "Yes I grew this companion plant combination and it worked better then have them single" this would be almost as good as real science, wouldn't it? :)

Your website looks really nice. I like it. Big data and AI to find the patterns between them is real. I wonder if there is a way for people to vote up or vote down these combinations. Then we could give them a rating in addition to popularity. Think ebay/amazon. I would esp like to to see those up vote or down vote paired to soil type/climate.

I know that lavender and peach/almond does well in Arizona/Nevada/Utah, but not so well on the East Coast.

S Bengi wrote:Your website looks really nice. I like it. Big data and AI to find the patterns between them is real. I wonder if there is a way for people to vote up or vote down these combinations. Then we could give them a rating in addition to popularity. Think ebay/amazon. I would esp like to to see those up vote or down vote paired to soil type/climate.

I know that lavender and peach/almond does well in Arizona/Nevada/Utah, but not so well on the East Coast.

This is an awesome idea. I am thinking about how to "crowd validate" such info a lot. I was actually thinking about the common 1-5 star rating. But up/down might be more simple? This should also apply to basically all info in the database: If it can be verified by many people it has higher chances to be true. The soil type/climate per user is something I want to add to. This makes it also more easy to suggest plants or again verify info. If a user writes that something is growing or working well for them, that doesn't mean it does the same in a different climate or any other similar factor.

Ben Knofe wrote:These are interesting thoughts! Currently there is no scientific research behind the database, beside the experience of users contributing to it or the external sources (websites/books/...) which are linked to a piece of information added.

I think in general it is a problem in permaculture that there is very little "science" behind most information you have on permaculture. My assumption is, that no one really brought all the info together in one place, so many people can cross-check and add their thoughts to it. I think while "real science" might not be possible so easily, a crowd-sourced version of a wikipedia style platform would work. If you have a lot of people saying "Yes I grew this companion plant combination and it worked better then have them single" this would be almost as good as real science, wouldn't it? :)

No....I don’t think it would be as good as real science.

And I think there is a lot of s incense behind permaculture and a lot of science mentioned here.

I think people will respond because they like to input their experiences.

I think a more practical and scientific approach might be something like Jacke & Toensmeier's metrics for compatibility in their textbooks on forest gardening.

Here's my interpretation for how to pick companion plants for a closely planted garden (with less open space for weeds, and for healthier soil):

1. Start by considering the soil. Is your soil more carbon-heavy and/or fungal? Is  there heavy clay that you need roots to penetrate and aggregate? Is it pretty dead/sterile and in need of healthy microbes?  Do you need to inoculate legumes with bacteria and plan cover crops to sustain soil life in the off season? Is there a niche for a deep tap root among the shadow mat-forming roots of the veggie you want there? Do you need some brassica family plant to fumigate the soil after a former crop suffered disease there? How fertile is it?

2. Next consider the conditions of the plants you want to plant together. Are they all dry-loving, rock-dwelling Mediterranean plants? Are you trying to combine two plants that love moisture and rich organic matter with one plant that prefers sand and a chance to dry out? Do they all need full sun or cooler afternoon shelter? How do they tolerate crowding, humidity, wind, atmospheric pollution from a nearby road, or pests?

3. Next try listing your niches for each garden bed. Do you have one plant to cover the ground and provide weed suppression, one that occupies the middle height, and one that will climb a trellis and provide the other two a bit of summer shade? Will one feed heavily on nitrogen, one trade with bacteria for its own nitrogen, and one feed only lightly on soil nitrogen during the growing season when the others need it? Will one bloom earlier to attract pollinators and feed them until the critical period when another blooms? Will the scent of one drive away pests from another? Conversely, is one a known host species for a pest or disease that will critically impact another?

4. Finally consider labor and timing. Can you stagger your plantings to get the ground cover established first? Are there perennials you can incorporate to help support your more needy and fragile annuals? Are there crucial beneficial insects or microbes you want to provide habitat and food throughout the growing season to prevent problems during the brief window of your annual crop? Will one plant grow too tall during the wet season and cause mildew problems on another, or shade it out when it's little? Can you transplant older seedlings to make the timing work better? Will one plant have traits like thorns or irritating chemicals that make it harder for you to tend the others?

There's no perfect list for companion plants because the variety of conditions, soils, timing windows in your climate, pests and beneficials are too numerous and complicated. But you can list the niches that each plant can fill and thereby provide a guide for possible combinations.

For example, the famous 3 Sisters example includes one sprawling ground cover that suppresses weeds far away from its own root center, one tall plant that feeds heavily on nitrogen, and one climbing vine that helps anchor the tall plant against lodging in wind, and that has its own source of nitrogen thanks to bacterial alliances. All three enjoy rich organic matter in moist soil, but the squash and the beans can tolerate some summer shade (or even require it) while the corn welcomes full sun (and provides the others that shelter). All three can be planted roughly at the same time and harvested at the same time (depending on varieties), or all three can provide a mixed season harvest. So that's why they make good companions;  they occupy compatible niches, minimize competition, and provide beneficial services to each other while requiring little intervention from the farmer.

In my garden, black alium aphids attacked my Egyptian walking onions in such swarms that the onions were dying, too early in the season for many aphid predators to be active.  So I transplanted all of them into a bed of rosemary (with a ground cover of golden sedum). Rosemary is known as one of the resinous, aromatic plants that help get rid of onion aphids. It worked. So for me, those are now good companion crops.

Likewise, the caterpillars that have been attacking my madder plants stop their browse line just above the height of the French marigolds that I have growing in the same bed. So to me, those are good companions.

Likewise, my scarlet runner beans were stunted and dying this spring thanks to aphid-farming black ants. I interplanted luffa, and the luffa plants started recruiting 'bouncers' (protective ants) as soon as they had established a few true leaves. The red ants that were being 'hired' by the luffa (the luffa pays them with sugar at EFNs, 'extra-floral nectaries') included the poor stunted bean vines in their protection plan and drove away the black ants and the aphids. The beans and luffa are now sharing a trellis. Nasturtiums are covering the soil below them as an extra insurance, since aphids like nasturtiums more than bean leaves and should attack those first if they return.

Conversely, one year I interplanted trellised cucumbers and turnips and parsley. However, I didn't know that cucumber vines have tiny hairs that severely irritate my skin. With gloves, I was able to handle them to trellis them and harvest the cucumbers, but I couldn't reach between the vines to harvest any turnips or parsley once the cucumbers were growing well. The hairs got to me even through long-sleeved shirts. So that was not a good companion planting for me.

So, that's how I'd measure what defines a good companion plant. Maybe the database could include fields for such niches (root type, height, nutritional needs, conditions and timing) and beneficial alliances.

Janet Reed wrote:No....I don’t think it would be as good as real science.

And I think there is a lot of s incense behind permaculture and a lot of science mentioned here.

Janet Reed wrote:I think people will respond because they like to input their experiences.

You are right and I was half joking. I think the problem is, while there is a lot of info existing, it is not easily accessible. I am looking for ways to make existing info, may it be scientific, a personal experience or another source of information easily accessible, so people can focus on growing things instead of searching through fragmented info. I'd also say that while there is a lot of science behind some things, you don't need all of it to grow a successful garden that feeds you and the family. I think someone can grow the three sisters successfully without understanding 100% the science behind it. They probably could not design their own guild/companion from scratch but would still have a great harvest of corn, beans and squash :)

Sadb O'Conner wrote:I think a more practical and scientific approach might be something like Jacke & Toensmeier's metrics for compatibility in their textbooks on forest gardening.

Here's my interpretation for how to pick companion plants for a closely planted garden (with less open space for weeds, and for healthier soil):

1. Start by considering the soil. Is your soil more carbon-heavy and/or fungal? Is  there heavy clay that you need roots to penetrate and aggregate? Is it pretty dead/sterile and in need of healthy microbes?  Do you need to inoculate legumes with bacteria and plan cover crops to sustain soil life in the off season? Is there a niche for a deep tap root among the shadow mat-forming roots of the veggie you want there? Do you need some brassica family plant to fumigate the soil after a former crop suffered disease there? How fertile is it?

2. Next consider the conditions of the plants you want to plant together. Are they all dry-loving, rock-dwelling Mediterranean plants? Are you trying to combine two plants that love moisture and rich organic matter with one plant that prefers sand and a chance to dry out? Do they all need full sun or cooler afternoon shelter? How do they tolerate crowding, humidity, wind, atmospheric pollution from a nearby road, or pests?

3. Next try listing your niches for each garden bed. Do you have one plant to cover the ground and provide weed suppression, one that occupies the middle height, and one that will climb a trellis and provide the other two a bit of summer shade? Will one feed heavily on nitrogen, one trade with bacteria for its own nitrogen, and one feed only lightly on soil nitrogen during the growing season when the others need it? Will one bloom earlier to attract pollinators and feed them until the critical period when another blooms? Will the scent of one drive away pests from another? Conversely, is one a known host species for a pest or disease that will critically impact another?

4. Finally consider labor and timing. Can you stagger your plantings to get the ground cover established first? Are there perennials you can incorporate to help support your more needy and fragile annuals? Are there crucial beneficial insects or microbes you want to provide habitat and food throughout the growing season to prevent problems during the brief window of your annual crop? Will one plant grow too tall during the wet season and cause mildew problems on another, or shade it out when it's little? Can you transplant older seedlings to make the timing work better? Will one plant have traits like thorns or irritating chemicals that make it harder for you to tend the others?

There's no perfect list for companion plants because the variety of conditions, soils, timing windows in your climate, pests and beneficials are too numerous and complicated. But you can list the niches that each plant can fill and thereby provide a guide for possible combinations.

For example, the famous 3 Sisters example includes one sprawling ground cover that suppresses weeds far away from its own root center, one tall plant that feeds heavily on nitrogen, and one climbing vine that helps anchor the tall plant against lodging in wind, and that has its own source of nitrogen thanks to bacterial alliances. All three enjoy rich organic matter in moist soil, but the squash and the beans can tolerate some summer shade (or even require it) while the corn welcomes full sun (and provides the others that shelter). All three can be planted roughly at the same time and harvested at the same time (depending on varieties), or all three can provide a mixed season harvest. So that's why they make good companions;  they occupy compatible niches, minimize competition, and provide beneficial services to each other while requiring little intervention from the farmer.

In my garden, black alium aphids attacked my Egyptian walking onions in such swarms that the onions were dying, too early in the season for many aphid predators to be active.  So I transplanted all of them into a bed of rosemary (with a ground cover of golden sedum). Rosemary is known as one of the resinous, aromatic plants that help get rid of onion aphids. It worked. So for me, those are now good companion crops.

Likewise, the caterpillars that have been attacking my madder plants stop their browse line just above the height of the French marigolds that I have growing in the same bed. So to me, those are good companions.

Likewise, my scarlet runner beans were stunted and dying this spring thanks to aphid-farming black ants. I interplanted luffa, and the luffa plants started recruiting 'bouncers' (protective ants) as soon as they had established a few true leaves. The red ants that were being 'hired' by the luffa (the luffa pays them with sugar at EFNs, 'extra-floral nectaries') included the poor stunted bean vines in their protection plan and drove away the black ants and the aphids. The beans and luffa are now sharing a trellis. Nasturtiums are covering the soil below them as an extra insurance, since aphids like nasturtiums more than bean leaves and should attack those first if they return.

Conversely, one year I interplanted trellised cucumbers and turnips and parsley. However, I didn't know that cucumber vines have tiny hairs that severely irritate my skin. With gloves, I was able to handle them to trellis them and harvest the cucumbers, but I couldn't reach between the vines to harvest any turnips or parsley once the cucumbers were growing well. The hairs got to me even through long-sleeved shirts. So that was not a good companion planting for me.

So, that's how I'd measure what defines a good companion plant. Maybe the database could include fields for such niches (root type, height, nutritional needs, conditions and timing) and beneficial alliances.

Wow this is a great post! Thank you. That's exactly the kind of information I am looking for. Mostly this is locked in blogs and/or boards and is hard for people to find but it is so immensely valuable!  

In general: Obviously the way I formed combinations here is just one of many possible ways to do that. Currently this is the only way possible with the given info entered. I was thinking of not only looking for companions but also how much different attributes the group would have. As an example: a group of three plants where the sum of all plant's attributes is Nitrogen-fixer, insect attracting, mulch producing, soil improving, pest repelling is probably better then a group of three plants where all the plants only do mulch and attract insects. But currently the data is very limited and mostly added by me so I am hoping to get people join in and add all kinds of data like Sadb described (example: root type, height, nutritional needs, conditions and timing) and then we can refine the approach!

This is such a big deal!! Thanks for making a system for this! We need to get this everywhere ASAP

Can I share this to groups on facebook

Josh Golden wrote:Can I share this to groups on facebook

Hi Josh! Sure, share it wherever it might help people. If you can, let people know that this is a user generated projected and any help is appreciated! Thanks!

I like this method of "fake science". It is actually how I learned almost everything I know. Just ask people that seem to be good at something and take their word for it.

Benefits of this crowdsourced (right? Right???) method:

Way faster than independant research! All the info is right there (not all the info is there YET)

"Good enough" accuracy. Not thesis reference quality but like wikipedia, at least 93% accurate!

Self adjusting. If someone gives bad info, the thumbs down will expose it as such.


Good work Ben. You are on to something. Keep it up.

I was all excited because there was actually a reason why alliums and legumes are anti-companions, but you already have this information.  

I am curious if this holds up with older beans.  So far my only really productive and dependable perennial vegetable is a bean.  After a few years does the pest control of the onions start to justify a decrease in nitrogen fixing?  

Casie Becker wrote:I was all excited because there was actually a reason why alliums and legumes are anti-companions, but you already have this information.  

I am curious if this holds up with older beans.  So far my only really productive and dependable perennial vegetable is a bean.  After a few years does the pest control of the onions start to justify a decrease in nitrogen fixing?  

What do you think about doing and experiment? Do a small bed of your beans interplanted with Alliums and one without and see if there are any differences. I'd love to see the results of that. I guess that's the whole idea here: We all need to experiment and share the results more :)

Right now my only perennial onions are Canadian onions.  They are summer dormant so I have been patiently waiting for them to fill in this flower bed before harvesting heavily.  As you can see they are forming flower heads before going to sleep.  Thinking about it, they are probably the ideal variety for this kind of experiment.   They sleep through most of the beans growing season and would get full benefit of the nitrogen fixed during their growing season.

When those buds open there are both traditional flowers and bulbs in the head.  It drops both viable seeds and those bulbs.  I happily used those flower heads to locate the ones escaped into the lawn and harvested them earlier this week.  I could move some to one side of the bean bed and then pay attention for the next couple of years.  That seems very doable and could be a great way to extend the harvest in that bed.  

I should mention that I got my first onions from someone on this forum but I can't remember who to credit.   They're a native onion that grows wild down the Great Plains from Canada to the gulf coast.