Companion planting with herbs: what extension services actually support

The charts are not science, and an extension service says so in print

Linda Chalker-Scott, extension horticulturist at Washington State University, wrote a paper titled "The Myth of Companion Plantings". Her words, not a paraphrase:

"There is no scientific basis, however, for any of the several lists that exist describing 'traditional companion plants'. Like horoscopes, these lists may be fun to use, but they should not be perceived or promoted as scientifically valid any more than astrology."

She goes further, noting that some popular sources claim companion pairs can be discovered "by sensitive crystallization of their extracts" or through a plant's "rhythm, its vibration, its music, and its note". Her bottom line is that traditional companion plant charts have entertainment value, not scientific value.

Illinois Extension reached the same place more recently and more gently: there has not been much research at a garden scale, and many recommendations rest on questionable methods and anecdote. Even the Three Sisters, corn with beans and squash, which is genuinely ancient and genuinely sensible, has "no compelling, published evidence that it benefits plant productivity or soil quality" according to that piece.

Washington State University Extension, "The Myth of Companion Plantings", no printed date; the PDF metadata says 2008. Illinois Extension, Good Growing, 7 March 2025.

What is supported: diversity, confusion and predators

The same WSU paper is clear about what the research does show, using the terms scientists prefer, intercropping and plant associations rather than companions:

• Diverse plantings attract and keep beneficial insects, including the predators and parasitoids that do the actual pest control.
• Several species in one area disrupt a herbivorous insect's ability to find its host plant, apparently through both visual and smell related mis-cues.
• Roots are connected below ground, largely through mycorrhizal fungi, which can move nutrients such as nitrogen between species.

Read that middle point twice, because it reframes the whole subject. The mechanism is not that a plant repels a pest. It is that a mixed planting is harder to read.

The specific pairs an extension service is willing to put its name to

Penn State Extension published a page distinguishing the anecdotal from the research based, and it names pairs:

• Basil with tomato, masking the tomato from thrips and discouraging egg laying by the tomato hornworm moth.
• Thyme under tomatoes, discouraging egg laying by the yellow-striped armyworm moth.
• Calendula near brassicas, attracting the beneficial insects that eat aphids.
• Coriander, dill and fennel near leafy greens, attracting lacewings and parasitoid wasps.
• Flowering rosemary, oregano, basil and thyme, bringing in native sweat bees for pollination.

And the sentence that ties it back to the mechanism: these herbs "don't repel insect pests but rather confuse them".

Penn State Extension, herbs make good plant partners and companions, updated 22 April 2026.

The finding that argues against our own niche

We are going to print this one because leaving it out would be dishonest. The WSU paper notes a study reporting that traditional companion plants, including aromatics such as mint, had little or no disruptive effect on insect behaviour, which suggests that aroma alone is not a good reason to choose a plant association.

So: plant the herbs because you want the herbs, and because a mixed bed with flowers in it supports the insects that do pest control for you. Do not plant them because a chart promised that the smell keeps something away.

What to do with all that, practically

1. Mix species instead of planting blocks of one thing. That is the effect with the most support behind it.
2. Let some herbs flower. The flowers, not the leaves, are what feed the beneficial insects.
3. Put the aggressive spreaders in containers before they make the decision for you: see the herbs that take over.
4. Match the plant to your zone first, because none of this matters if it does not survive February: herbs by USDA zone.

Sources, with the date of each

• Washington State University Extension, Linda Chalker-Scott, "The Myth of Companion Plantings". No date printed on the document; PDF metadata indicates 2008.
• Illinois Extension, Good Growing, companion planting, 7 March 2025.
• Penn State Extension, herbs make good plant partners and companions, updated 22 April 2026.

Not used, because we could not open them: a UMass Extension fact sheet on companion planting, which returned a server error, and a University of Arizona Extension PDF we could not read. If either contradicts the above, we would rather find out and correct this page than quote what we did not read.