Skip to content
NatureHQ

Claim check

Do ants treat each other’s wounds?

Mostly supported

The core is right; the usual phrasing overstates part of it.

The behaviour is real, precise and measurably effective — injured ants are carried home and their wounds are treated with antimicrobial secretions that substantially cut mortality. What the coverage adds is intention. Nothing in the research shows any ant deciding to help, and the ants that are left behind are left because they stop signalling, not because anyone judged them beyond saving.

The claim as it circulates

“Ants are paramedics: they rescue injured comrades from the battlefield and perform medical treatment on their wounds.”

Where you may have met it: Widely shared coverage of the Megaponera rescue and wound-treatment studies; Social media posts framing the behaviour as ant medicine; Headlines describing ants as surgeons or doctors

Taking it apart

What was claimed
That ants act as medics — rescuing the wounded, triaging who can be saved, and applying medical treatment.
What was actually observed
In the termite-hunting ant Megaponera analis, workers injured during raids release a pheromone that causes nestmates to carry them back to the nest. Most recover and rejoin later raids. Nestmates then apply secretions from the metapleural gland directly to the wound; chemical analysis found compounds with antimicrobial and wound-healing properties, and ants prevented from receiving treatment died of infection far more often.
What the evidence supports
That a specific chemical signal triggers carrying, that wounds receive a targeted application of antimicrobial compounds, and that this produces a large and measurable survival benefit at the colony level.
What it does not support
That any ant assesses, decides or intends. The apparent triage — fatally injured ants being left — is not a judgement made by anyone: those ants simply do not produce the signal. The work covers one species with an unusually injury-prone lifestyle, and does not generalise to ants as a group.

The detail that decides how this should be read is the one most often omitted. Ants too badly injured to survive do not call for help, and are not carried. Read as compassion, that looks like a heartbreaking decision. Read as a mechanism, it means nothing is deciding anything: the colony rescues whoever signals, and the signal stops when the injury is severe enough.

That does not make the finding less impressive. A social insect with a chemically triggered rescue response and a specific antimicrobial wound treatment is a genuinely remarkable piece of colony economics, and the survival data are unusually clear. It is simply a different kind of remarkable from the one the headlines describe.

Where the evidence lives

The claims underneath

Each one carries its own evidence, scope and caveats. Expand any of them to reach the studies.

Some ants carry injured nestmates home — but only the ones that call for help

Well supported

Good evidence backs this, though some details remain open.

Injured Megaponera analis workers release a pheromone that triggers nestmates to carry them back to the nest, where most recover and rejoin later raids. Fatally injured individuals do not produce the signal and are left behind.

Who this applies to
a termite-hunting ant of sub-Saharan AfricaDo not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Megaponera analis
Why we rate it this way, and what the caveats are
Well supportedHigh confidence

Field observation combined with experimental manipulation of the chemical signal, with survival outcomes measured.

How far it can be extended

This species raids termites at unusually high injury cost, which is what makes rescue pay. Most ants do nothing of the kind.

Caveats

  • Nothing here decides who is worth saving: fatally injured ants simply do not signal.
  • Reported widely as ants performing medicine, which imports intention the studies do not test.
  • One specialised species with an unusually injury-prone lifestyle.

Still unanswered

  • Why do severely injured workers stop signalling?
  • Do other ants with costly raids show anything similar?

Last reviewed 2026-08-09

The evidence (2 studies)

Go deeper

Ants

The colony solves problems no ant is solving. That is not a metaphor; it is the mechanism.

Last reviewed 2026-08-09