A parasitoid wasp will find a potato aphid, climb onto it, and lay an egg inside it. The egg will not develop. The aphid carries on feeding and giving birth, and two weeks later there are no mummies on the leaf and the colony is bigger than it was.
Nothing went wrong with the shipment. That is what Aphidius colemani does to a potato aphid, and it is published: it stings them and “is unable to complete development in this host.” The wasp was the wrong one for the aphid, and everything else followed.
Aphids get sold to and written about as one pest. The organisms that eat them do not treat them as one pest. So this is the article about telling them apart — five species, what each one actually looks like, and which control matches it.
The short version
- Green peach
- Small, yellowish-green, antennal bumps that lean toward each other. Colemani or matricariae
- Melon / cotton
- Small, very variable color, cornicles black all the way. Colemani
- Potato
- Much bigger, elongate, pink or green, long cornicles. Ervi — colemani cannot develop in it
- Foxglove
- Pale green, dark spots at the cornicle bases, banded legs. Ervi
- Oleander
- Bright yellow with black legs and cornicles, on milkweed, oleander or Hoya. A different problem
- If you cannot tell
- Colemani has the widest recorded host range of the three — not matricariae
- What matricariae is really for
- Cooler rooms. Its optimum is about five degrees below colemani's
- The catch on oleander aphid
- Green lacewing was tested on it and could not develop
Why the species is a pass or fail
With a predator, getting the aphid species wrong is usually a matter of degree. A lacewing larva that meets an aphid it does not much like will often eat it anyway. With a parasitoid it is not a matter of degree, because the wasp is not eating the aphid — it is trying to raise an offspring inside it.
A review of Aphidius colemani in greenhouse crops gives the boundary exactly. The wasp is sold for “the economically important aphids Myzus persicae (green peach aphid) and Aphis gossypii (melon or cotton aphid),” and it has a recorded host range of over forty-one aphid species. And then: it “will sting the potato aphid, Macrosiphum euphorbiae, but it is unable to complete development in this host,” nor is it effective “for foxglove aphid … or chrysanthemum aphid.”
What the trade shorthand gets wrong
You will see this summarized as colemani for small aphids, ervi for large ones. That is a memorable rule and it is the wrong mechanism.
The published boundary is host suitability, not size. And the one study that actually varied host size found that within colemani’s suitable hosts, larger hosts produced larger and fitter wasps. Use the species list, not the rule of thumb.
The five you are most likely to have
These five cover almost everything a houseplant grower, a greenhouse and an ornamental garden will turn up. Color is the least reliable character on the list and the one people reach for first, so it is not the leading column.
| Species | What it looks like | Where you find it |
|---|---|---|
|
Green peach aphid Myzus persicae |
Small, about 1.7–2.0 mm, yellowish or greenish. Cornicles match the body color and swell unevenly along their length. The tell is at the front: the bumps the antennae sit on are “slanted toward each other,” leaving a notch between them | Almost anything — hundreds of hosts across more than forty plant families. The commonest aphid in a mixed indoor collection |
|
Melon or cotton aphid Aphis gossypii |
Small, 1–2 mm, and “quite variable in color” — mottled green, whitish, yellow, dark green, almost black. The reliable character is that the cornicles are black, not body-colored, and there is no notch between the antennal bases | Cucurbits above all, but also pepper, okra, hibiscus and a long list of ornamentals. The one still going in a hot August |
|
Potato aphid Macrosiphum euphorbiae |
Noticeably larger — around 3.5 mm against the green peach aphid’s 2 — and more elongate, with long cornicles, a long tail and antennae longer than its own body. Pink and green forms both occur. The antennal bumps diverge rather than lean together | Tomato, potato and other nightshades, roses, and scattered at all heights on a plant rather than massed at the tips |
|
Foxglove aphid Aulacorthum solani |
Pale shiny green with darker green spots at the base of each cornicle and black banding on the legs and antennae. It has a notch between the antennal bases like green peach aphid, so the cornicle spots are what separate them | One of the broadest host ranges of any aphid — recorded from over 82 plant families. Cool rooms and greenhouses especially |
|
Oleander aphid Aphis nerii |
Unmistakable: bright yellow to orange with black cornicles, antennae, legs and tail. 1.5–2.6 mm | Oleander, milkweeds — including tropical milkweed and butterfly weed — and wax plant, Hoya carnosa. Almost never anywhere else |
The one character worth learning
If you only take one thing to a hand lens, take the cornicles — the two tubes projecting backward off the rear. On green peach aphid they are the same color as the body. On melon aphid they are black, over their whole length. On foxglove aphid the body is pale but there is a dark patch where each one joins. On potato aphid they are conspicuously long.
That single feature separates four of the five, and it does not depend on the body color, which is the character that varies most and misleads most.
Which wasp matches which aphid
Three Aphidius species are sold for aphids. They are not three brands of the same wasp.
| Wasp | What it is actually for | The published basis |
|---|---|---|
| Aphidius colemani | Green peach aphid and melon aphid above all. The broadest recorded host range of the three — over forty-one aphid species — and the best of four aphidiine parasitoids tested against melon aphid | Stings potato aphid and cannot develop in it. Not effective on foxglove aphid |
| Aphidius ervi | Potato aphid and foxglove aphid — the two colemani cannot do. Also pea aphid and several cereal aphids | Described as oligophagous: legume, nightshade and cereal aphids |
| Aphidius matricariae | Green peach aphid, including the tobacco biotype. Its real advantage is temperature: its development optimum is about 25 °C against colemani’s 30 °C | Reared from both green peach and foxglove aphid in a field survey, where it was the most abundant primary parasitoid from both |
A correction we are making to our own advice
Aphidius matricariae has been described — including by us — as the general-purpose choice for when you cannot identify the aphid. That is not what the literature says. The wide host range belongs to A. colemani, and in a direct comparison colemani outperformed matricariae on melon aphid.
What matricariae has that colemani does not is cooler weather. If your plants live somewhere that sits in the mid-sixties Fahrenheit rather than the high seventies, that is the reason to choose it — and it is a good one.
There is a fourth match that we cannot supply, and it belongs here because the alternative is the reader guessing. Lysiphlebus testaceipes is the parasitoid named in the extension literature for oleander aphid. We do not carry it. That is the next section.
The yellow ones, which are a different problem
If the aphids are bright yellow-orange with black legs, and the plant is a milkweed, an oleander or a Hoya, you have oleander aphid — and most of what is sold for aphids has a documented problem with it.
The plant is the reason. Milkweeds and oleanders make cardiac glycosides, and this aphid does not merely tolerate them. In a study comparing it with the monarch caterpillar on the same plant, the aphid retained the complete range of all 25 cardenolides present in the plant, where the monarch retained eleven. It is drinking a defended sap and keeping the defense.
What that does to the predators was tested directly, and recently. Seven commercially available generalist predators were put on oleander aphid. Three could survive, develop and reproduce on it. Four could not — the authors’ words are that the aphid was “unsuitable prey … low foraging efficiency, survival and negligible impact.”
Coped with it
Brown lacewing Micromus angulatus, the fourteen-spot ladybird Propylea quatuordecimpunctata, and the hoverfly Sphaerophoria rueppellii. The brown lacewing was the best of the seven, with a higher attack rate and shorter handling time.
Could not
Green lacewing Chrysoperla carnea, two-spot ladybird Adalia bipunctata, and two others. Of green lacewing specifically: it “did not show a significant response to varying densities” of the aphid — it did not even hunt harder when there were more.
Read that precisely, because we sell one of them
Green lacewing is Chrysoperla carnea — the species that failed. That applies directly to our lacewing larvae and our lacewing eggs, and we would rather you knew before you bought them for this aphid.
Our ladybugs are Hippodamia convergens, which was not in that trial. Two ladybirds were tested and they split — one coped, one did not. So the honest statement is that ladybirds are divided on this aphid and ours has not been tested on it, not that ladybugs fail.
They also fight back, together
The chemistry is not the only defense. A separate study found that oleander aphid colonies respond to an attacker with a coordinated twitching and kicking that immobilised hoverfly, brown lacewing and ladybird larvae, and that “always caused a repelling effect” on Aphidius colemani. The same paper found solitary aphids were “unable to prevent oviposition.”
A crowd defends itself and an individual does not, which is an unusually direct argument for not letting this particular colony get large.
Where the sources disagree, and which one was a test
University of Florida’s factsheet on this aphid, written in 2017, says the glycosides “appear not to harm the parasitoids and generalist insect predators.” The 2025 trial above put seven of them on the aphid and four failed.
Both are in front of you. The difference worth weighing is that the first is a summary statement and the second is a controlled experiment that post-dates it, and that the 2025 work tested the specific commercial species you would actually buy.
If it is on milkweed, there is a second question
Monarch caterpillars and oleander aphids turn up on the same plants, and the aphids are far more visible. Nothing in this article is a reason to spray a milkweed — an insecticide that reaches the aphids reaches the caterpillars, and the caterpillars are usually the reason the milkweed is there.
A strong jet of water knocks oleander aphids off without touching the chemistry, and it is the intervention with the fewest side effects on a plant you are growing for something else.
When you genuinely cannot tell
Sometimes the aphids are too small, the lens is not strong enough, or there are two species on the same plant. Three honest options, in order of how much they cost:
- Use the plant as the clue. Nightshades and roses with big elongate aphids scattered up the stems point to potato aphid. Cucurbits with very variable small ones point to melon aphid. A cool greenhouse with curled, spotted leaves and surprisingly few aphids points to foxglove aphid, which does disproportionate damage at low numbers. Milkweed, oleander or Hoya with bright yellow ones is oleander aphid and nothing else.
- Choose the wasp with the widest recorded range, which is colemani. Over forty-one aphid species, and the best of four parasitoids tested on melon aphid. Its two documented blanks are potato and foxglove aphid, and both of those are identifiable by size and by the cornicle spots respectively — so if you have ruled those two out by eye, colemani is the wider bet.
- Use a predator instead of a parasitoid. A predator does not need the host to be physiologically suitable, only catchable. Aphidoletes is the one usually reached for here, because the adult midge finds the colony by scent and lays into it, so you are not required to know what is in the colony. Oleander aphid remains the exception to this and to most things.
Identifying before you buy
Look at the cornicles first, not the color. Body-colored, black, dark-spotted at the base, or conspicuously long — that one character separates four of the five species here, and color separates none of them.
Size is a real character between potato aphid and the rest. About 3.5 mm against about 2. Put them beside each other on the same leaf and it is obvious; in isolation it is not, so compare rather than judge.
Photograph them against something with a known size. A phone macro shot next to the edge of a fingernail or a grain of rice gives you a scale you can come back to, and it gives us something to work from if you ask.
On milkweed, check what else is on the plant before doing anything. Monarch eggs and caterpillars use the same plants. That rules out sprays and it changes what a reasonable outcome looks like.
Take matricariae for a cool room, not for uncertainty. Its published development optimum is about 25 °C against colemani's 30 °C. That is what it is genuinely better at.
Common questions
Does it really matter which aphid I have?
For a predator, not much — it will eat most of what it catches. For a parasitoid wasp it decides the outcome. Aphidius colemani will sting a potato aphid and, in the words of the published review, “is unable to complete development in this host.” The aphid survives, and from the outside that looks the same as a release that failed.
What is the difference between Aphidius colemani and Aphidius matricariae?
Host range and temperature. Colemani has the broader recorded range — over forty-one aphid species — and beat matricariae in a direct comparison on melon aphid. Matricariae's advantage is that it works cooler: development optima are about 25 °C for matricariae and 30 °C for colemani. Choose matricariae for a cool room, not for uncertainty.
I was told colemani is for small aphids and ervi for large ones. Is that wrong?
It is a useful shorthand for the wrong reason. The published boundary is host suitability — colemani cannot complete development in potato aphid or control foxglove aphid, and ervi is the agent for those two. But the one study that actually manipulated host size found that within colemani's suitable hosts, larger hosts produced larger and fitter wasps. So use the species list rather than judging by size.
My milkweed is covered in bright yellow aphids. What do I use?
That is oleander aphid, Aphis nerii, and it is the hardest aphid in this article to buy a solution for. It sequesters the plant's cardiac glycosides, and in a 2025 trial of seven commercial predators, four could not develop on it — including green lacewing. The parasitoid named for it in the extension literature is Lysiphlebus testaceipes, which we do not stock. On a plant grown for monarchs, a strong jet of water is the intervention with the fewest side effects.
Will ladybugs eat oleander aphids?
We do not know, for the ladybug we sell. The 2025 trial tested two ladybird species and they split — the fourteen-spot ladybird coped, the two-spot did not. Neither is Hippodamia convergens, which is what we ship. Anyone telling you either way with confidence is going beyond the evidence.
Can I have more than one species at once?
Yes, and on a mixed collection it is common. It is one of the better arguments for a predator rather than a parasitoid, or for pairing a wasp with something that does not care what species it catches.
Is the foxglove aphid really worse than the others?
For a given number of aphids, yes, on ornamentals. Its saliva distorts tissue, and NC State Extension describes it as causing irreversible cosmetic damage at much lower numbers than other aphids. The practical consequence is that “only a few of them” is not the reassurance it would be with green peach aphid.
References
- Prado, S.G., Jandricic, S.E. & Frank, S.D. (2015). Ecological interactions affecting the efficacy of Aphidius colemani in greenhouse crops. Insects 6(2): 538–575. The source for colemani’s host range of over 41 aphid species, for its use against green peach and melon aphid, and for the two blanks this article turns on: it stings potato aphid and “is unable to complete development in this host,” and it is not effective for foxglove aphid. Open access. doi.org
- Villegas, C.M., Ortiz-Martínez, S.A., Zepeda-Paulo, F. et al. (2017). Morphological variation of Aphidius ervi Haliday associated with different aphid hosts. PeerJ 5: e3559. The host list for A. ervi, including potato aphid and foxglove aphid. Open access. doi.org
- Zamani, A.A., Talebi, A., Fathipour, Y. & Baniameri, V. (2007). Effect of temperature on life history of Aphidius colemani and Aphidius matricariae. Environmental Entomology 36(2): 263–271. The development optima of about 30 °C and 25 °C respectively, which is the real difference between the two wasps. doi.org
- Sampaio, M.V., Bueno, V.H.P. & De Conti, B.F. (2008). The effect of the quality and size of host aphid species on the biological characteristics of Aphidius colemani. European Journal of Entomology 105(3): 489–494. The study that manipulated host size directly, and found larger hosts better within colemani’s suitable range — which is why this article does not repeat the small-aphid rule of thumb. Open access. doi.org
- Valentini, T., Castanier, J., Santoiemma, G., Tirello, P. & Pozzebon, A. (2025). Potential biocontrol services by seven aphid predators for the control of Aphis nerii infesting Mandevilla. Biological Control 208: 105847. The trial behind this article’s oleander aphid section: three of seven commercial predators could complete development on it and four could not, green lacewing among the failures. doi.org
- Malcolm, S.B. (1990). Chemical defense in chewing and sucking insect herbivores: plant-derived cardenolides in the monarch butterfly and oleander aphid. Chemoecology 1(1): 12–21. The sequestration measurement — the aphid retained the complete polarity range of 25 cardenolides against the monarch’s eleven. doi.org
- Hartbauer, M. (2010). Collective defense of Aphis nerii and Uroleucon hypochoeridis against natural enemies. PLoS ONE 5(4): e10417. The coordinated twitching response, its effect on Aphidius colemani, and the finding that solitary aphids could not prevent oviposition. Open access. doi.org
- McAuslane, H.J. Oleander Aphid, Aphis nerii Boyer de Fonscolombe. University of Florida IFAS Extension EENY-247. Appearance, the 1.5–2.6 mm size range, the host list including Hoya carnosa, and Lysiphlebus testaceipes as the named parasitoid. Also the 2017 statement that the glycosides appear not to harm generalist predators, which the 2025 trial above tested and partly contradicts. Free to read. ifas.ufl.edu
- Capinera, J.L. Green Peach Aphid (EENY-222) and Melon Aphid or Cotton Aphid (EENY-173). University of Florida IFAS Extension. Sizes, color ranges, and the black cornicles that separate melon aphid from green peach aphid. Free to read. ifas.ufl.edu
- University of California Statewide IPM Program. Green Peach and Potato Aphids, Pest Management Guidelines: Spinach, UC ANR Publication 3467. The antennal tubercle character — green peach aphid’s “slanted toward each other” — and the size and cornicle comparison with potato aphid. Free to read. ipm.ucanr.edu
- North Carolina State Extension. Foxglove Aphid. The dark spots at the cornicle bases, the banded legs and antennae, and the statement that it causes irreversible cosmetic damage at much lower numbers than other aphids. Free to read. ces.ncsu.edu
The wasp has to match the aphid. The predators are more forgiving, and slower to compound.
