Skip to content
NatureHQ

Mammalsspecies group

Zebra

Equus

Modelled through a lion’s eyes, zebra stripes blur into grey by fifty metres. Whatever they are for, it is not fooling lions.

The stripes are not camouflage and they do not confuse lions — modelled through lion vision they blur into grey by fifty metres. What they track, across every wild horse, ass and zebra, is where biting flies are. How they work on flies is still unresolved.

Few questions in natural history have attracted as many confident answers as this one, and for most of 150 years the answers were argued rather than tested. Five hypotheses were seriously proposed: that stripes conceal a zebra in tall grass or heat haze; that they confuse a predator; that they let zebras recognise one another; that they cool the animal; and that they deter biting flies. Each is plausible, each was defended vigorously, and almost none of it was ever put to a test that could distinguish them. What changed things was noticing that the five make different *geographic* predictions. If stripes conceal from predators, striped equids should occur where those predators are; if they cool, striping should track temperature; if they deter flies, it should track flies. So the striping of every living horse, ass and zebra was mapped and tested against range-wide data for all five at once. Only one predictor survived: the distribution of tabanid and tsetse biting flies. The predator explanations then took a more direct blow. Photographs of wild zebras processed through models of lion and hyena vision showed the stripes to be unresolvable beyond about fifty metres in daylight, and much less at dusk when these predators mostly hunt. A lion close enough to see stripes has already found the zebra. Then the fly result was tested directly, and the design is what convinces. Horseflies approached zebras and horses at similar rates and landed on zebras far less. The same horses were then dressed in striped, black and white coats: striped coats received fewer landings, while the uncovered head — identical in every condition — showed no difference, which rules out any whole-animal explanation such as smell or temperature. The flies were not deterred at range; they failed to decelerate properly on approach, colliding or veering away.

Developed coverage · 47% complete · reviewed 2026-09-03

What this page covers

Three zebra species, within a genus that also contains horses and asses. The stripe evidence is comparative across the whole genus, which is what makes it possible to test explanations against each other.

Often confused with: A horse with stripes, when zebras are separate species with different social systems; One explanation for stripes, when at least five were seriously proposed; Black stripes on white, or the reverse, which is not a meaningful question

Quick facts

What striping tracks
The distribution of biting flies — and nothing else tested
Not camouflage from lions
Unresolvable to lion and hyena vision beyond about fifty metres
The direct test
Horses in striped coats got fewer fly landings than the same horses in plain ones
Still unresolved
Why a striped surface disrupts a fly’s landing approach

Five explanations, tested against each other

The breakthrough was noticing that they make different geographic predictions.

Across every wild horse, ass and zebra, striping tracks one thing: where biting flies are. It shows no association with predators, woodland, temperature range or social living.

Well supported

Good evidence backs this, though some details remain open.

Comparative analysis of striping across all extant equid species and subspecies, tested against range-wide geographic predictors, found consistent association only with the distribution of tabanid and tsetse biting flies. No support was found for predictors corresponding to the camouflage, predator-confusion, thermoregulatory or social-recognition hypotheses.

Who this applies to
All living equid species and subspecies, compared across their geographic ranges.
Studied in
Equus
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

An unusually strong comparative design — five hypotheses tested against the same dataset, four failing on their own predictions — and independently supported by an experiment putting striped coats on horses. It remains a geographic association plus a proximate experiment, without an established mechanism.

How far it can be extended

The analysis covers the whole genus rather than a sample, which is what allows competing hypotheses to be tested against one another.

Caveats

  • A geographic association is not a mechanism, and how stripes disrupt flies is not established.
  • Within plains zebras specifically, stripe definition correlates with temperature — and temperature correlates with fly abundance, so the two are difficult to separate.
  • Biting flies transmit disease, so the benefit may be about infection rather than about the bites themselves; this has not been separated.

Where researchers disagree

  • An analysis of stripe variation within plains zebras across Africa found the strongest correlation with temperature rather than with fly abundance, which the authors read as consistent with a thermoregulatory contribution. Temperature and biting-fly abundance covary, so the two analyses may be detecting the same underlying gradient from different directions, and direct tests of stripe-driven cooling have generally failed to find an effect of useful size.

Still unanswered

  • What visual process in the fly is being disrupted, which the landing experiments demonstrate without explaining.

Last reviewed 2026-09-03

The evidence (4 studies)
What each hypothesis predicts, and what was found
HypothesisPredicts striping where…Outcome
CamouflagePredators hunt, or cover is tallNo association; and stripes are unresolvable to lions at range
Predator confusionCoursing predators are commonNo association; and unresolvable at the distances required
Social recognitionHerds are large and social living mattersNo association; zebras cannot resolve them at the needed distance either
ThermoregulationIt is hotNo association across the genus — though see the within-species result below
Biting-fly deterrenceTabanid and tsetse flies are abundantConsistent and strong association

It is worth appreciating why this design settles something that a century and a half of argument did not. Any single hypothesis can be supported by selective examples — there are certainly zebras in grass, and zebras in heat, and zebras in herds. Testing all five against the same complete dataset means they compete, and four of them failed against predictions they themselves generate.

Diagram

Five hypotheses, tested against the same data

Four failed against predictions they generate themselves.

Five hypotheses, tested against the same data at onceHypothesisMatched the geography?CamouflageNoPredator confusionNoSocial recognitionNoThermoregulationNo*Biting-fly deterrenceYes* Not across the genus. Within plains zebras, stripe definition does track temperature.
The same explanation in words

A five-row table listing each proposed function of zebra striping and whether it matched the geographic pattern across all living horses, asses and zebras. Camouflage: no. Predator confusion: no. Social recognition: no. Thermoregulation: no across the genus, with a footnote that within plains zebras stripe definition does track temperature. Biting-fly deterrence: yes, highlighted as the only surviving predictor. The design works because each hypothesis predicts a different geography, so testing them against one dataset makes them compete rather than each being supportable by selective examples.

Whatever the stripes are for, lions cannot see them

The receiver principle, applied to the most confidently repeated explanation.

Modelled through lion and hyena vision, zebra stripes blur into grey beyond about fifty metres in daylight — and far closer at dusk. By the time a predator can resolve them it has already found the zebra.

Well supported

Good evidence backs this, though some details remain open.

Visual modelling incorporating the spatial acuity and spectral sensitivity of Panthera leo and Crocuta crocuta indicates that zebra striping is unresolvable beyond approximately fifty metres in photopic conditions and at substantially shorter distances under crepuscular and scotopic light, distances at which detection by other cues has typically already occurred.

Who this applies to
Lion and spotted hyena vision, modelled against photographs of wild zebras.
Studied in
Panthera leo, Crocuta crocuta, Equus quagga
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

Built on measured visual parameters and standard modelling methods, and it makes a clear prediction that the geographic evidence independently matches. It is modelling rather than a behavioural test of predators, and acuity estimates carry uncertainty.

Caveats

  • This addresses whether stripes can be resolved, not whether a striped animal differs from a plain one in some way that survives blurring.
  • Acuity estimates for large carnivores carry uncertainty which propagates into the distance thresholds.

Still unanswered

  • Whether stripes affect a predator’s perception of motion at speed, which blurring at distance does not directly address.

Last reviewed 2026-09-03

The evidence (3 studies)

This is the argument this site makes on every camouflage page, arriving at a specific and consequential answer. The camouflage and confusion hypotheses both require that a predator can see stripes as stripes. Model the actual acuity of a lion or a hyena, in the light levels at which they actually hunt, and the pattern merges into uniform grey well before the distances at which those explanations would operate. The hypotheses are not merely unsupported; their precondition fails.

The same modelling disposes of social recognition on the same grounds. A zebra cannot resolve another zebra’s stripes far enough away for the pattern to serve as a long-range identity signal — which does not exclude close-range individual recognition, but does remove the version of the hypothesis that was being proposed.

Dressing horses in zebra coats

The experiment that turned a geographic correlation into a demonstration.

How we know

Putting zebra coats on horses

Striping tracks biting-fly distribution across all equids. Is the pattern itself doing something to flies, or is it something else about zebras?

Horseflies were filmed approaching and landing on captive zebras and domestic horses kept in the same field. The same individual horses were then dressed in striped, black and white coats, and fly behaviour was recorded again — so the animal was held constant and only the pattern varied.

What happened

Flies approached zebras and horses at similar rates but landed on zebras far less often. Striped coats received fewer landings than plain coats on the same horses, while the uncovered head showed no difference between conditions. Flies failed to decelerate properly on approach to striped surfaces, frequently colliding or veering away.

What it shows

That the striped pattern itself disrupts a fly’s final approach, and that the effect belongs to the pattern rather than to anything else about a zebra. The uncovered head is what makes this hard to argue with — the difference is confined to the part that was striped.

What it does not show

It does not explain what visual process is being disrupted, which remains open. One fly family in one region, using captive zebras and domestic horses, and it does not address tsetse or other biting flies.

The controls — what makes this evidence rather than a story
  • The same horses in every coat condition, which removes any difference between individuals or species.
  • The uncovered head as an internal control: identical in all conditions, so any whole-animal explanation such as smell, temperature or behaviour would show up there too.
  • Zebras and horses in the same field on the same days, so conditions and fly populations are shared.
  • Approach rate measured separately from landing rate, distinguishing deterrence at range from disruption on arrival.

From Benefits of zebra stripes: behaviour of tabanid flies around zebras and horses

The uncovered head is the detail that makes this design hard to argue with. If striped horses received fewer landings because of something about the animal — its smell, its temperature, its behaviour — the head would show it too, since the head was uncovered in every condition. It did not. The difference was confined to the covered body, which is where the pattern was.

The flies were not repelled from a distance. They approached striped and plain animals at similar rates and then failed to decelerate properly on final approach, colliding with the surface or veering away.

Based on Across every wild horse, ass and zebra, striping tracks one thing: where biting flies are. It shows no association with predators, woodland, temperature range or social living.

What is still open

Including one result that does not fit.

The fly account is much the best supported and it is not the end of the matter. An analysis of stripe variation within plains zebras across Africa found stripe definition correlating most strongly with temperature — more strongly striped animals in warmer places — which is what the thermoregulatory hypothesis predicts and the fly hypothesis does not. Temperature and biting-fly abundance covary strongly, so the two analyses may be detecting the same underlying gradient from opposite directions. Direct tests of stripe-driven cooling have generally failed to find an effect of useful size, which is why NatureHQ treats this as an unresolved complication rather than a rival answer.

  • Why does a striped surface disrupt a fly’s landing?

    Why it matters: The effect is demonstrated and the mechanism is not. Candidate explanations involve how stripes interfere with the optic flow a fly uses to judge its approach speed, but none has been established — and until one is, the best-supported answer to "why do zebras have stripes" rests on an unexplained visual effect in an insect.

    What would settle it: Controlled work on tabanid flight control against striped patterns, varying stripe width, contrast and orientation to find which parameters the disruption depends on.

  • Is the benefit about bites, or about the diseases flies carry?

    Why it matters: Tabanids and tsetse transmit serious disease. If the selective pressure is infection rather than blood loss or irritation, the strength and geography of the benefit look quite different, and the two have not been separated.

Related

Claims about this, checked

Things people have heard, and what the evidence actually supports.

The research behind this page

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

2019PLOS ONE

Benefits of zebra stripes: behaviour of tabanid flies around zebras and horses

Flies approached zebras and horses at similar rates but landed on zebras far less often.

2016PLOS ONE

Zebra stripes through the eyes of their predators, zebra and humans

Stripes are indistinguishable to lions and hyenas beyond roughly fifty metres in daylight and at much shorter distances at twilight or night, by which point the predator has almost certainly detected the zebra by other means.

2015Royal Society Open Science

How the zebra got its stripes: a problem with too many solutions

Within plains zebras, stripe definition on the torso correlated most strongly with temperature, with more strongly striped animals in warmer regions.

2014Nature Communications

The function of zebra stripes

Striping was consistently and strongly associated only with the distribution of biting flies.

2009Philosophical Transactions of the Royal Society B

Animal camouflage: current issues and new perspectives

Camouflage mechanisms differ in which stage of visual processing they defeat.

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 47% completeness against what we would call a finished subject, and was last reviewed on 2026-09-03. It carries 2 claims and answers 5 mapped search questions.

  • 6 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
  • Zebra social systems, ecology and conservation are outside this page, which is about the stripes.
  • The mechanism by which stripes disrupt fly landing is unresolved and is stated as such.
  • Whether the benefit concerns bites or fly-borne disease has not been separated.