Skip to content
NatureHQ

Reptiles and amphibiansspecies group

Poison frogs

Dendrobatidae

The poison is not theirs. They take it from mites and ants — which is why a frog bred in captivity is not poisonous at all.

Poison frogs do not make their poison. They take alkaloids from the mites and ants they eat and concentrate them in the skin — which is why frogs bred in captivity are not poisonous, and why wild populations of one species differ with what lives where they do.

The fact that reorganises everything about these animals is that the poison is not theirs. Frogs raised in captivity on crickets and fruit flies have essentially no skin alkaloids; give them the right arthropods and the alkaloids appear. In the wild, the principal sources are oribatid mites and certain ants, and the frog’s job is to take up these compounds without being harmed and store them in skin glands. This is sequestration rather than synthesis, and it makes the defence a property of a food web. Several consequences follow that a species-level description would miss. Toxicity varies geographically within a single species, because the arthropod fauna does. It varies between species enormously — the family contains many frogs with little or no chemical defence at all. And a captive-bred frog in a vivarium is genuinely not dangerous, which is not a diminished version of the wild animal so much as the same animal minus an ingredient it was never making. The name misleads twice over. Only a few species were ever used to poison darts, and these frogs are poisonous rather than venomous: the toxins act on contact or ingestion, and there is no fang, sting or delivery apparatus of any kind. The bright colouring is a warning signal in the technical sense, and it is worth being careful about what that means. There is a real association in this family between conspicuousness and toxicity — brighter species do tend to carry more — but the signal works through what local predators have learned, and the association is a tendency rather than a rule that can be applied to an unfamiliar frog.

Early coverage · 33% complete · reviewed 2026-09-03

What this page covers

Around three hundred species in Central and South America. Toxicity varies enormously across the family — many species carry little or none — and only a handful were ever used on darts.

Often confused with: All members of the family being dangerously toxic, when most are not; Being venomous, which they are not — there is no delivery apparatus; Making their own toxins, which they do not

Quick facts

Where the toxins come from
Diet — principally oribatid mites and ants
In captivity
Not poisonous, because the dietary source is absent
Poisonous, not venomous
No fang, sting or delivery apparatus of any kind
Not uniformly toxic
The family spans highly toxic species and many with little defence

Borrowed chemistry

The single fact that reorganises everything else about these animals.

Poison frogs do not make their poison. They take alkaloids from the mites and ants they eat and store them in the skin — so a frog raised on other food is not poisonous, and wild populations differ with their prey.

Established

Specialists would state this without hedging. Multiple independent lines of evidence agree.

Dendrobatid skin alkaloids are sequestered from dietary arthropods, principally oribatid mites and ants, rather than synthesised. Frogs raised on alkaloid-free diets lack skin alkaloids, and alkaloid profiles vary geographically with the composition of the local arthropod fauna.

Who this applies to
Established for dendrobatid poison frogs; comparable dietary sequestration occurs in several other groups.
Studied in
Dendrobatidae
Why we rate it this way, and what the caveats are
EstablishedHigh confidence

Demonstrated experimentally by manipulating diet and confirmed in the field by matching alkaloid profiles between frogs and their prey.

How far it can be extended

Dietary origin has been demonstrated across multiple dendrobatid species and confirmed by matching compounds between prey and frog in the field.

Caveats

  • Sequestration is not passive: the frogs have mechanisms for taking up and storing these compounds without harming themselves, and in some cases modifying them.
  • Species and populations differ substantially in which alkaloids they carry and in how much, so "poison frog" covers a wide range of actual toxicity.

Still unanswered

  • How the frogs tolerate compounds that are toxic to other animals, which is only partly understood.

Last reviewed 2026-09-03

The evidence (2 studies)

How we know

Feeding poison frogs the wrong food

Do poison frogs manufacture their alkaloids, or take them from something they eat?

Poison frogs were raised in captivity on diets lacking their natural arthropod prey, and their skin was analysed for alkaloids. Specific dietary items were then reintroduced and the skin analysed again.

What happened

Frogs raised without their natural prey had essentially no skin alkaloids. Alkaloids appeared when appropriate arthropods were included in the diet.

What it shows

That the toxins are sequestered from food rather than synthesised. This reframes the defence as a property of a food web rather than of a species, and explains both the loss of toxicity in captivity and the variation between wild populations.

What it does not show

Captive diets are simplified, so this identifies dietary origin without establishing the full set of natural sources — that came from later fieldwork matching compounds between wild frogs and their prey. It also does not explain how the frogs tolerate compounds toxic to other vertebrates.

The controls — what makes this evidence rather than a story
  • Diet as the only manipulated variable, with frogs otherwise raised identically.
  • Chemical analysis of skin rather than inference from behaviour or appearance.
  • Reintroduction of dietary items, which tests the mechanism in both directions rather than only removing it.

From Dietary source for skin alkaloids of poison frogs (Dendrobatidae)

Sequestration is not passive, which is worth saying because "they get it from their food" can sound like the frog is a container. These animals take up compounds that would harm most other vertebrates, transport them to skin glands, concentrate them there, and in some cases chemically modify them. The frog is doing real physiological work; what it is not doing is manufacturing the molecules from scratch.

Why toxicity varies between populations

Because the food does, and the defence is a property of a food web.

  • Between species: the family spans highly toxic frogs and many with little or no chemical defence, so "poison frog" is a family name rather than a description.
  • Between populations: the same species can differ in alkaloid profile and quantity, tracking the arthropods available locally.
  • Between seasons and individuals: intake varies, and so does what is stored.
  • In captivity: essentially absent, unless the animal is deliberately fed alkaloid-bearing prey.

The practical reading of this is that toxicity cannot be inferred from identification alone. Knowing the species tells you what range to expect; it does not tell you what a particular animal is carrying, and the honest general statement is that these are wild animals best not handled rather than that any given frog is dangerous.

The colours, and what they can and cannot tell you

A real association, and not a rule you can apply to an unfamiliar animal.

Within the poison frog family there is a real tendency for more conspicuous species to carry more toxin. It is a tendency within one family, not a rule that transfers to animals in general.

Emerging evidence

Real findings exist, but too few or too recent to be settled.

Comparative analyses within Dendrobatidae report a positive association between signal conspicuousness and alkaloid quantity or toxicity, consistent with honest aposematic signalling in this clade. The association is clade-specific and does not license inference from conspicuousness to toxicity across animals generally.

Who this applies to
Poison frogs specifically; the association is not claimed beyond the family.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Dendrobatidae
Why we rate it this way, and what the caveats are
Emerging evidenceModerate confidence

The association is reported in comparative work within the family and fits the aposematic account. It is a correlation across species measured in different ways, and how tightly signal and defence track each other remains argued.

How far it can be extended

The point of stating it is that it holds within this family and fails as a general rule, so generalising it is precisely the error the claim guards against.

Caveats

  • Toxicity varies within a species by population and by season, so a species-level association does not predict what an individual frog carries.
  • Conspicuousness is usually scored against human or modelled avian vision, and the relevant predators differ.

Still unanswered

  • How signal honesty is maintained when the defence is sequestered from food rather than produced, since a frog cannot adjust its brightness to match a bad year for mites.

Last reviewed 2026-09-03

The evidence (2 studies)

Bright colouring is not a reliable guide to whether an animal is dangerous. Conspicuous animals may be signalling to mates, mimicking something defended, or simply not conspicuous to the eyes that matter.

Well supported

Good evidence backs this, though some details remain open.

Conspicuous coloration arises from multiple selective sources — aposematic signalling, sexual signalling, Batesian mimicry of defended models, thermoregulatory and structural constraints — and appears conspicuous only relative to a given visual system. Colour therefore does not function as a reliable indicator of chemical defence across taxa.

Who this applies to
Applies across animals; the counterexamples are numerous in every direction.
Studied in
Animalia
Why we rate it this way, and what the caveats are
Well supportedHigh confidence

The association between conspicuousness and defence is real but far too leaky to support the popular inference, and the exceptions run in both directions and are well documented.

How far it can be extended

Conspicuous undefended species, cryptic highly toxic species, and mimics of both kinds are documented across many groups.

Caveats

  • There is a genuine statistical association in some groups — among poison frogs, brighter species do tend to be more toxic — and this claim is that the inference is unreliable, not that it is baseless.
  • "Bright" is judged by human vision. An animal cryptic to us may be conspicuous to a bird that sees ultraviolet, and vice versa.

Still unanswered

  • How honest aposematic signals remain over evolutionary time, given that a mimic benefits from the signal without paying for the defence.

Last reviewed 2026-09-03

The evidence (3 studies)

This family is where the popular rule comes closest to working. Among dendrobatids there is a genuine tendency for brighter species to be more toxic, and the warning-signal interpretation is well supported here. What does not follow is the general inference. Across animals as a whole, conspicuousness has several unrelated causes, harmless mimics borrow the appearance without the chemistry, and plenty of extremely toxic animals are drab. A tendency within one family is not a field guide.

Related

The research behind this page

3 studies, newest first. Each one has a page explaining what it found and what it could not show.

2007Proceedings of the National Academy of Sciences

Oribatid mites as a major dietary source for alkaloids in poison frogs

Oribatid mites contained a wide range of the alkaloids found in poison frog skin, identifying them as a major dietary source alongside ants.

2005Trends in Ecology & Evolution

The complex business of survival by aposematism

Aposematism depends on predator psychology rather than on the signal alone.

1994Journal of Chemical Ecology

Dietary source for skin alkaloids of poison frogs (Dendrobatidae)

Frogs raised without their natural prey lacked skin alkaloids almost entirely.

This page is a stop on a longer route

A guided journey reads several subjects in a deliberate order, with an argument for why one follows another. You can join in the middle.

Where to go from here

Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.

How complete this page is, and what it is still missing

NatureHQ publishes its own gaps. This page is at 33% completeness against what we would call a finished subject, and was last reviewed on 2026-09-03. It carries 2 claims and answers 3 mapped search questions.

  • 4 high-priority search intent(s) not yet covered
  • no research from the last few years is attached — check for newer work
  • more experiments could be explained in plain English
  • no popular claim about this subject has been checked yet
  • How the frogs tolerate compounds toxic to other vertebrates is only partly understood.
  • Parental care, in which several species carry tadpoles individually to water, is notable and outside this page.
  • Conservation status varies widely across the family and is not covered here.