Is coffee a legume? No. Coffee beans are the seeds of Coffea plants, which belong to the madder family (Rubiaceae). Legumes belong to a different family, the pea family (Fabaceae), and their seeds grow in pods. A coffee bean grows inside a fleshy cherry instead, so “bean” describes its look, not its botany.
Table of Contents
Are coffee beans legumes?
No, a coffee bean is not a legume. World Flora Online records Coffea arabica L. as an accepted name “in the genus Coffea (family Rubiaceae)” and the common bean, Phaseolus vulgaris L., as an accepted name “in the genus Phaseolus (family Fabaceae)”. The classification on those same two records puts coffee in the order Gentianales and the common bean in the order Fabales, so the two differ in family and in order.
Three differences do the work, and only one of them is about taxonomy.
- The fruit. A legume in the strict botanical sense is a dry fruit that usually splits along two seams to release a row of seeds: the pea pod is the textbook example. A coffee cherry is a drupe (Unigarro et al., Plants, 2025), a fleshy fruit that does not split; it is picked whole, and the seeds are separated from the fruit afterwards.
- Nitrogen. This is the difference with real agronomic consequences. USDA’s Agricultural Research Service describes Rhizobium as “bacteria that form symbiotic (mutually beneficial) relationships with soybeans, alfalfa, peanuts, beans, and other legumes to convert nitrogen gas from the air into fertilizer for the plants”, a process it names as nitrogen fixation. Coffee does not do this. A 2019 paper in Frontiers in Plant Science calls root-nodule symbiosis “restricted to four plant orders: Fabales (legumes), Fagales, Cucurbitales and Rosales”, and coffee is in none of them (World Flora Online puts it in Gentianales). Legume trees are also planted as coffee shade; a 2022 field study in Ecuador, which used three Fabaceae shade species, says residues of N2-fixing trees “might fulfil coffee nutrient requirements”.
- The family. Rubiaceae (the madder family, which also holds the gardenia) versus Fabaceae (the pea family), as the two records linked above confirm. Fabaceae has about 22,500 species and, in a 2025 review in Current Biology, is “the third largest family of flowering plants”. Coffea is a genus of tropical trees and shrubs; the 2025 revision that split Liberica into three species put the count at 133 known Coffea species (Davis et al., Nature Plants, 2025). Four of them are grown commercially.
| Trait | Coffee | A legume (the common bean) |
|---|---|---|
| Family | Rubiaceae (madder family) | Fabaceae (pea family) |
| Order | Gentianales | Fabales |
| Fruit | A fleshy cherry (a drupe) that does not split | A dry pod that usually splits along two seams |
| Seeds | Usually two per cherry | A row in the pod |
| Root-nodule nitrogen fixation | No | Yes, with Rhizobium bacteria |
What is inside a coffee cherry?
From the outside in, the National Coffee Association’s cross-section names five layers:
| Layer | Botanical name | What it is |
|---|---|---|
| Outer skin | Exocarp | The outermost layer; it turns from green to yellow to red as the cherry ripens |
| Pulp | Mesocarp | The juicy flesh around the beans |
| Parchment | Endocarp | A thin layer that covers and protects the beans as they develop |
| Silver skin | Testa (epidermis) | The outer layer of the bean itself |
| Bean | Endosperm | The core of the coffee bean |
Usually two seeds develop inside one cherry, one in each of its two chambers. Each seed is convex on one side and flattened on the other, with a groove along the flat side (Unigarro et al.). When only one seed develops it fills the cavity by itself and comes out rounded. The NCA puts that at roughly 5% of the world’s coffee and gives it its trade name: “This is called a peaberry (or a caracol, which is ‘snail’ in Spanish).”
Is a coffee bean a bean?
Not in the botanical sense. Dictionary.com defines a bean first as “the edible nutritious seed of various plants of the legume family, especially of the genus Phaseolus“. A coffee “bean” is the seed of a different plant: the NCA says coffee beans are “actually seeds found inside the cherries (also called the fruit) of coffee trees”.
The word was stretched later: the Online Etymology Dictionary gives Old English bean as “bean, pea, legume” and dates “coffee-bean” to the 1680s. The same Dictionary.com entry also lists “any of various other beanlike seeds or plants, as the coffee bean.” Vanilla and cocoa follow the pattern. A vanilla bean is the cured pod of an orchid (Vanilla planifolia, family Orchidaceae), a cocoa bean is the seed of Theobroma cacao (family Malvaceae), and neither plant is a legume. “Bean” describes how these things look in the hand, not what family they belong to.
Is a coffee bean a vegetable or a fruit? Neither. The bean is the seed; the cherry around it is the fruit, a drupe; and the bean is not a legume either.
Coffee species and their Latin names
All four commercial coffees are Coffea, and all four are Rubiaceae. Each has its own page here:
- Arabica: Coffea arabica, the species behind most of the world’s coffee.
- Robusta: Coffea canephora, lower-grown, higher in caffeine, rounder seed.
- Liberica: Coffea liberica, a big-seeded species that makes up a minor share of world production.
- Excelsa: Coffea dewevrei, separated from Liberica as its own species in 2025.
A fifth, Coffea stenophylla, is a West African species that, according to a 2021 study in Nature Plants, grows and crops “at a mean annual temperature 6.2–6.8 °C higher than Arabica coffee”. And if the question behind “are coffee beans legumes” was really a dietary one, see what food group coffee belongs to; the short answer there is none.
Corrections to earlier versions of this page
Earlier versions of this page got the headline answer right (Rubiaceae versus Fabaceae) and several supporting details wrong. Each is corrected above and named here.
- It said coffee and legumes are “not even distantly related”. That was wrong. NCBI Taxonomy places Rubiaceae and Fabaceae in the same large group of flowering plants (the eudicots) but on separate branches: asterids for coffee, rosids for legumes. They are distant relatives, in different families and different orders.
- It offered the black-eyed Susan vine as an example of a legume. It is not one. World Flora Online records Thunbergia alata Bojer ex Sims as an accepted name “in the genus Thunbergia (family Acanthaceae)”, a different family from Fabaceae. The plant that is a legume and has a similar-sounding name is the black-eyed pea.
- It described the parchment as “the outer layer” that “encases the bean”. The parchment is the endocarp, the innermost fruit layer; the outer layer of the cherry is the exocarp, the skin you can see. The NCA cross-section above is the corrected order.
- It claimed “over 20,000 species” of legumes, with no source. The figure was about right: a 2025 review in Current Biology puts Fabaceae at about 22,500 species, “the third largest family of flowering plants” (Hughes, Ringelberg and Bruneau, 2025). The number is now in the family point above.
- It said English “has always used ‘bean’ for anything seed-shaped”. That was wrong: the word began as the name of a legume seed, and “coffee-bean” is recorded from the 1680s (sources above).
- It said a coffee farm gets its nitrogen from “fertiliser, mulch or shade trees”, “which is often why” leguminous shade trees are planted. We could not source that motive, so the page now says only what the 2022 field study says.
- It made three more claims we could not source: that a coffee seed’s flat, grooved face is “exactly what makes” it look like a bean; that the “bean” label “stuck long before anyone was classifying plants by shared descent”; and that Liberica is “much the smallest of the four crops”. The first two are cut. For the third, the source we cite says only that Liberica and excelsa together are “a minor share of global production”.
