A large solitary-travelling cat that lives inside a structure it maintains entirely by scent. Its stripes are drawn in the skin before it has any hair, and its orange coat is close to invisible to the colour-blind animals it hunts.
Almost everything commonly said about tigers is nearly right, which makes them harder to write about than an animal surrounded by outright myths. Take the most repeated fact of all: that a tiger’s stripes are on its skin, not just its fur. Developmental work in cats shows the pattern really is established in the skin — as thickened and thinned bands in the embryo, before the animal has grown a single hair — and what the skin then does is instruct hair follicles which colour to grow. The pattern is a skin property, the pigment is a hair property, and the popular version compresses those into a claim about shaved tigers that nobody appears to have checked. The camouflage question has the same shape. People reasonably object that orange is a strange colour to hide in green, and the objection assumes the audience is human. Deer and wild pigs are dichromats: they lack the cone class that separates red from green, so a tiger’s orange and the vegetation behind it are close to the same colour. It is conspicuous to us; we are not the audience. Tigers do travel alone, and "solitary" is where most accounts stop. Radio-tracking from Nepal and from the Russian Far East shows a land-tenure system: a resident female holds largely exclusive ground, a male’s much larger range overlaps several females, and the arrangement is maintained by scent marks that stay informative for days. Two tigers who never meet in a year can each know a great deal about the other. When daughters grow up most settle beside their mothers, sometimes on ground their mother gives up, while sons leave properly and often never acquire any ground at all. What ties the ecology together is prey. A tigress in Chitwan holds a range a fraction the size of one in Sikhote-Alin; what differs is how much there is to eat. That is why conservationists talk about empty forest: a forest can look perfect from the air and hold no tigers, because its deer and pigs have been snared out of it.
Developed record · 92% complete · reviewed 2026-08-11
What this page covers
One species across Asia, in a fragmented remnant of a range that once ran from Turkey to the Russian Far East and south to Bali. How many subspecies it contains is genuinely unsettled: the twentieth-century scheme drawn from skulls and skins named nine, whole-genome work supports six living populations, and a two-subspecies arrangement is also in current use. All agree the divergence is shallow.
Often confused with: Ligers and tigons, which are captive hybrids with lions and have no wild population; Leopards and jaguars, which are spotted rather than striped and are separate species in the same genus; The Tasmanian tiger, which was a marsupial and not a cat at all
Quick facts
Where the stripes are
The pattern is in the skin; the pigment is in the hair
Why orange works
Deer are dichromats — orange and green look much the same
Solitary?
Travels alone; lives in a scent-maintained structure of neighbours
Territory size
No single figure — ten times larger in Russia than in Nepal
The famous version is close to right, and wrong in the one place that matters.
The pattern is laid down in the skin before there is any hair — but the colour is in the hair
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
In cats, coat pattern is established in two stages. A periodic pre-pattern of gene expression is set up in embryonic skin before hair follicles form, with Dkk4 marking the regions that will become dark and the skin itself visibly thickened and thinned across the pattern. Follicular melanocytes subsequently interpret that map, producing eumelanin-rich hair in the marked regions and pheomelanin-rich hair elsewhere. Taqpep determines whether the resulting pattern is narrow-striped or broadly blotched. The pigment that a viewer sees is deposited into the growing hair shaft; the dermis is not correspondingly pigmented.
Who this applies to
demonstrated in domestic cats and cheetahs, and understood to apply across the cat family
Studied in
Felis catus, Acinonyx jubatus, Panthera tigris
You may have heard
“Tigers have striped skin, not just striped fur — shave one and the stripes are still there”
This is one of those facts that is repeated everywhere and sourced nowhere, and the interesting thing is that the real answer is better than the myth. The pattern genuinely is a property of the skin: it is drawn there in the embryo, as thickened and thinned bands, before the animal has a single hair. But what the skin does with that map is instruct hair follicles which colour of hair to grow. The pigment you see is in the hair. So "the stripes are in the skin" is right about where the pattern lives and wrong about where the colour lives — and the reason everyone believes the strong version is that a tiger skin, meaning a pelt, is obviously striped. A pelt still has its fur on.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
The genetic mapping, the expression work in embryonic skin and the cross-genus confirmation in cheetahs are mutually independent lines converging on the same two-stage model.
How far it can be extended
The mechanism was mapped in domestic cats and then confirmed to explain the king cheetah phenotype in a different genus, and the same pattern-forming genes are present across Felidae. It has not been demonstrated directly in a tiger.
Caveats
The developmental work is in domestic cats; the extension to tigers rests on shared genetics rather than on tiger embryos.
Whether a shaved tiger shows visible dark stripes is a separate question about melanin in the dermis, and NatureHQ has not found a primary measurement of it either way.
The mechanism explains stripe geometry, not why the background colour is orange.
Still unanswered
How much visible pigment, if any, is present in tiger dermis beneath the dark stripes?
What sets the spacing of the stripes, given that spacing varies systematically between populations?
Matches photographs of confiscated skins to photographs of the living tigers they came from, which is why a pelt is unmistakably striped and why people take that as evidence about skin.
How we know
Looking for a stripe in an animal that has no hair yet
When does a cat’s pattern first exist — and does it exist before the fur does?
The obvious way to study a coat pattern is to look at the coat, and that can only ever tell you about the finished product. This went the other way: at embryonic cat skin, taken at successive stages before hair follicles have formed and before any pigment cell has made anything, looking for the earliest physical sign that a pattern is being laid out. Single-cell expression analysis identified which genes differed between regions, and the skin itself was examined for structure. The test is severe in a useful way — at these stages there is no hair to be striped and no pigment to be dark, so anything found cannot be the pattern being read off the fur.
What happened
The pattern is already present in embryonic skin as alternating thickened and thinned regions, with the gene Dkk4 expressed in what will become the dark areas. This is long before follicles, hair or pigment exist.
What it shows
That the pattern is a property of the skin, established there first, and that the colour is applied afterwards by follicles reading the map. It also explains individuality: a self-organising process of this kind reliably produces a cat-shaped pattern and never produces exactly the same one twice, which is what every tiger population estimate depends on.
What it does not show
It does not show that adult tiger skin is visibly striped. That is a question about melanin in the dermis of a grown animal, and this is a question about where a pattern is specified in an embryo — the two get run together constantly. Nor is it a tiger study: it is domestic cats, extended to tigers by shared genetics rather than by observation. Embryonic material is scarce, so the stage series has gaps.
The controls — what makes this evidence rather than a story
Stages sampled before follicle formation, so hair cannot be the source of any pattern found.
Expression measured cell by cell rather than in bulk tissue, so a difference between neighbouring regions cannot be averaged away.
Skin structure assessed independently of gene expression, giving two unrelated signals of the same pattern.
Domestic cats with different adult patterns compared, so the embryonic signal can be checked against the animal it produces.
A pattern has to come from somewhere, and the surprising answer is that it comes from the skin, long before there is anything to be patterned. In cat embryos, before a single hair follicle has formed and before any pigment cell has made anything, the skin is already marked out into alternating thickened and thinned regions, with the gene Dkk4 switched on in what will become the dark areas. The map is drawn first. The colour arrives later.
What happens later is that pigment-producing cells in the hair follicles read that map and grow the appropriate hair — dark eumelanin-rich hair where the map says dark, orange pheomelanin-rich hair elsewhere. A second gene, Taqpep, sets whether the pattern comes out as narrow stripes or broad blotches, which is the same switch that produces the king cheetah’s run-together markings in an entirely different genus.
The pigment you see when you look at a tiger is inside the hair shafts. The pattern that determines where each hair’s colour goes is a property of the skin, established in the embryo.
So the popular claim is half right in an interesting way. "The stripes are in the skin" correctly describes where the pattern lives and incorrectly describes where the colour lives. It is worth being clear that NatureHQ could not find a primary measurement of how much visible pigment sits in the dermis of an adult tiger beneath a dark stripe — the claim is repeated constantly and sourced nowhere. What is easy to explain is why everyone believes it: a tiger skin, meaning a pelt, is unmistakably striped, and a pelt still has its fur on.
Words used here
Pre-pattern
A map of where a pattern will go, established before the thing being patterned exists. In cats it is set up in embryonic skin and later read by hair follicles.
Eumelanin and pheomelanin
The two pigments cats make. Eumelanin is the black-brown one and produces the stripes; pheomelanin is the red-yellow one and produces the orange ground colour.
The objection is fair. It assumes the animal looking is a person.
A tiger is not orange to a deer — deer cannot tell orange from green
Well supported
Good evidence backs this, though some details remain open.
Most ungulate prey of tigers are dichromats lacking the long-wavelength cone class that separates red and green in trichromatic primates. Computational analysis of natural scenes shows the colour minimising detectability against a background depends strongly on the observer’s visual system, and that colours conspicuous to a trichromat can approach optimal concealment against foliage for a dichromat. Comparative analysis across felids associates patterned coats with closed, complex habitats. Spatial-frequency analysis shows a tiger’s stripe pattern matches the frequency content of its background at distance, while remaining highly conspicuous close up.
Who this applies to
tigers hunting dichromatic mammalian prey
Studied in
Panthera tigris
You may have heard
“Tiger stripes are camouflage — which is strange, because orange stands out”
The objection is reasonable and it dissolves once you ask who is looking. Deer, wild pigs and cattle are dichromats: they have two cone types, not three, and the red–green channel that makes orange leap out at a human simply is not there. To the animal the tiger is hunting, its orange and the green of the vegetation are close to the same colour. Add the stripes, which match the spatial frequency of grass and scrub at the distance a stalk is decided, and the tiger is well hidden from everything it needs to be hidden from. It is conspicuous to us. We are not the audience.
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
The visual-system argument is computational rather than behavioural — no experiment has measured detection of a tiger by a deer — and the comparative felid analysis fits stripes less cleanly than spots. The convergence of three independent approaches is what carries it.
Caveats
No behavioural experiment has tested how well a deer detects a tiger; the visual argument is modelled.
The comparative felid work explains spots better than stripes, so the tiger is not the strongest case in that dataset.
Tigers occupy habitats from mangrove to snow forest, and one explanation is unlikely to fit all of them equally.
Still unanswered
Why stripes rather than the spots most closed-habitat cats carry?
Do the marked differences in stripe density between populations track anything about their habitats?
Fennell et al., 2019 · Journal of The Royal Society Interface
Shows optimal concealing colour depends on the observer’s visual system, and that a dichromat cannot make the red–green distinction that renders orange conspicuous.
Allen et al., 2011 · Proceedings of the Royal Society B: Biological Sciences
Associates patterned coats with closed habitats across the cat family, while fitting spotted patterns better than striped ones.
How we know
Asking what colour is invisible — to a deer
Is a colour that looks conspicuous to a human necessarily conspicuous to the animal it needs to hide from?
The question sounds unanswerable because you cannot ask a deer. The way round it is to stop asking about appearance and start computing detectability. Photographs of natural scenes were analysed to find, for each background, the colour that would be hardest to detect against it — and the computation was run twice, once for a trichromatic visual system like a human’s and once for a dichromatic one like that of most hoofed mammals, which lacks the cone class that separates red from green. The comparison is the experiment: the same scene, the same arithmetic, two different observers.
What happened
The best concealing colour depends heavily on the observer. Colours that are strikingly conspicuous to a trichromat can be close to optimally concealing against foliage for a dichromat, because the red–green distinction that makes them stand out is simply unavailable.
What it shows
That "but orange stands out" is an argument about human vision and not about camouflage. To deer, wild pigs and cattle — the animals a tiger hunts — orange and green are close to the same colour. It also demonstrates a general point worth carrying to other animals: camouflage has to be assessed against the eyes that matter, and ours almost never are.
What it does not show
No deer was tested and no tiger was involved. This is a computation about colour detectability using modelled visual systems, not a behavioural measurement of how long a real prey animal takes to spot a real predator. It also says nothing about the stripes themselves, which are a question about pattern rather than colour and rest on separate evidence.
The controls — what makes this evidence rather than a story
Real natural scenes rather than artificial backgrounds, so the result is not an artefact of a laboratory colour chart.
The two visual systems differ only in cone complement, so any difference in the answer is attributable to that alone.
Optimality computed rather than judged by eye, which removes the experimenters’ own trichromatic vision from the assessment.
The same procedure run for maximum visibility as well as minimum, checking the method finds the expected answer at both ends.
Humans are trichromats: we have three cone classes, and the extra one gives us a red-green channel that most mammals do not have. Deer, wild pigs and cattle are dichromats. The distinction that makes a tiger blaze out of a green background for us is a distinction they cannot make, and computationally, colours that are conspicuous to a trichromat can approach optimal concealment against foliage for a dichromat.
The stripes do separate work, and it is about distance. Their spatial frequency — how fast the pattern alternates across the animal — matches the frequency content of grass and scrub at the range where a stalk is decided. Close up the same stripes are highly conspicuous, which is why a tiger in a zoo enclosure looks like the least camouflaged animal alive.
Stripes do not make a tiger invisible, and no serious version of the argument claims they do. They delay detection at the specific range at which a stalk succeeds or fails.
The honest caveat is that nobody has run the behavioural experiment. No deer has been tested on how long it takes to spot a tiger. What exists is a computation about detectability using modelled visual systems, a spatial-frequency analysis of photographs, and a comparative survey across the cat family that associates patterned coats with closed habitats — and which, awkwardly, fits spots better than stripes. Three lines pointing the same way, none of them a direct test.
Words used here
Dichromat
An animal with two cone types in its eye rather than three. Most mammals are dichromats; primates including humans are the unusual ones. A dichromat cannot separate red and green.
Spatial frequency
How rapidly a pattern alternates across space. Camouflage works better when the animal’s spatial frequency matches its background’s, and how well it matches depends on viewing distance.
No two tigers are marked the same, and this is how tigers are counted
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
Tiger flank patterns differ between individuals to a degree sufficient for reliable individual identification from photographs, and are stable through adult life. Photographic capture–recapture using paired camera traps, with individuals identified by stripe pattern and non-detection modelled explicitly, is the standard method for estimating tiger density. Automated matching that fits a three-dimensional body model to each image, so that pattern is sampled from a consistent body region regardless of camera angle or posture, achieves reliable identification and can also match a skin to photographs of the living animal it came from.
Who this applies to
all tigers
Studied in
Panthera tigris
You may have heard
“Every tiger has a unique stripe pattern, like a fingerprint”
This one is essentially right, and it is worth knowing that it is not a piece of trivia — it is load-bearing. Every published estimate of how many wild tigers exist depends on it. Before pattern identification, tigers were counted from pugmarks on the assumption that each track was a different animal, which has no way of knowing what it missed. The fingerprint comparison is also more apt than people intend: like a fingerprint, the pattern is not designed to be unique, it is unique as a side effect of how it forms.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
Individual identification from stripe pattern underpins two decades of published density estimation, and the automated method was validated against sets of animals whose identities were independently known.
Caveats
Uniqueness is demonstrated as sufficient-for-identification within study populations, not proved absolutely across every tiger alive.
Both flanks differ from each other, so a survey photographing only one side counts a different set of individuals from one photographing the other.
Cubs change as they grow; identification applies to established adult patterns.
Still unanswered
Do close relatives share more pattern similarity than unrelated tigers, and enough to be detectable?
Explains why the pattern is reliably tiger-shaped yet never twice identical: a self-organising developmental process rather than a template.
How we know
The count that starts by admitting what it did not see
How many tigers are in a forest, when you cannot possibly see all of them?
The method it replaced counted pugmarks and treated each distinguishable track as a different animal, which has no way of knowing what it missed and no way of stating its own uncertainty. This design starts from the opposite assumption. Paired camera traps were placed facing each other along routes tigers actually use, so that any animal passing is photographed on both flanks — necessary because a tiger’s two sides carry different patterns. Individuals are then identified from those patterns, producing a record of which animals were photographed on which occasions. That record is analysed with capture–recapture models, whose whole purpose is to estimate how many animals were present and never photographed at all, from the pattern of how often the photographed ones reappeared.
What happened
Tiger density can be estimated with quantified uncertainty from photographs alone, with no animal captured or handled, because individuals are identifiable and the number missed can be modelled rather than guessed.
What it shows
That a wildlife count can be a statistical estimate rather than a tally, and that the honest version of the question is "how many were there, given what we saw and how likely we were to see it". It is the methodological foundation under every subsequent statement about tiger numbers.
What it does not show
It does not solve sparse populations, where too few photographs of too few individuals leave the models underdetermined — which is exactly where the conservation questions are hardest. The area a trap array effectively samples has to be estimated, and that choice moves the density figure substantially. And it assumes patterns are matched correctly by the human doing the matching, an assumption later work automated in order to remove.
The controls — what makes this evidence rather than a story
Paired traps, so an animal is not counted as two individuals by being photographed on different sides.
Trap placement on known travel routes rather than at random, with the sampled area then estimated explicitly rather than assumed.
Capture histories analysed with models that estimate detection probability, so the answer carries a stated uncertainty instead of a confident single number.
Applied at four sites of differing density, so the method is tested where it is easy and where it is hard.
This is usually offered as trivia and it is load-bearing. Every published estimate of how many wild tigers exist depends on individuals being identifiable from photographs. Before that, tigers were counted from pugmarks on the assumption that each distinguishable track was a different animal — a procedure with no way of knowing what it had missed and no way of stating its own uncertainty.
What replaced it works by admitting the gap. Paired camera traps photograph both flanks of any animal passing, which matters because a tiger’s two sides carry different patterns. Individuals are identified, and the resulting record of who was photographed when is analysed with capture–recapture models whose entire purpose is to estimate how many animals were present and never photographed at all.
A method that models what it missed usually produces smaller numbers than one that assumes it saw everything. Photographic capture–recapture frequently did, which made it politically as well as statistically inconvenient.
The pattern is stable enough through adult life for this to work, and distinctive enough that automated matching — fitting a three-dimensional body model to each photograph so that pattern is sampled from the same body region regardless of camera angle — can identify individuals reliably. The same software can match a confiscated skin to photographs of the living tiger it came from, which turns a monitoring tool into a forensic one.
Words used here
Capture–recapture
A way of estimating how many animals are in a population from how often the ones you do detect turn up again. The repeats tell you how likely you were to detect anything, which tells you how many you missed.
They do make a friendly noise. It is not a purr, and the reason is not the bone.
Tigers cannot purr — the friendly noise they make is a chuff, and it stops when they breathe in
Well supported
Good evidence backs this, though some details remain open.
Purring, defined acoustically as a call produced continuously through both inspiration and expiration without interruption at the turn, is not recorded in the roaring felids. Tigers instead produce prusten or chuffing, a low-intensity friendly greeting call made on the outbreath, together with a range of rumbles, moans and the roar. In the lion, tiger and jaguar the epihyoid element of the hyoid chain is an elastic ligament rather than a bone, whereas in cheetahs and domestic cats it is fully ossified; the elastic condition tracks roaring across species. Vocal-fold measurements show the roaring cats have unusually large folds with a thick flat surface whose mechanical properties predict very low fundamental frequencies at low lung pressure.
Who this applies to
tigers, and the other roaring cats
Studied in
Panthera tigris, Panthera leo, Panthera onca
You may have heard
“Big cats cannot purr because of a bone in the throat”
Two things are muddled here. The bone part is real but is not the mechanism: one link in the chain that suspends a tiger’s larynx is a ligament where a house cat has bone, and that difference does track which cats roar. What actually does the work is the vocal folds, which in a tiger are large and heavy with a thick flat surface — built to vibrate slowly and loudly, not to sustain the fast cycle a purr needs. And the reason people argue about this is that tigers plainly do make a friendly, throaty, cat-ish noise: the chuff. It is real, it means roughly what a purr means, and it is not a purr, because it stops when the animal breathes in.
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
The anatomy is solid and the acoustic review is careful, but the conclusion depends on a definition of purring that not every author adopts — which is precisely why the question keeps being argued.
How far it can be extended
The hyoid and vocal-fold features are shared across Panthera, and the absence of purring on an acoustic definition holds across the roaring species.
Caveats
The conclusion rests on defining purring as continuous through both breath phases; on a looser definition some authors count tiger rumbles.
The hyoid explanation is correlational and has been partly superseded by the vocal-fold work.
Almost all of this comes from captive animals and post-mortem material.
Still unanswered
Is the chuff homologous to the domestic-cat purr, or an independently evolved friendly signal?
Measures the vocal-fold properties that make a tiger larynx suited to slow, low, loud vibration rather than to the fast sustained cycle purring requires.
The roar is built into the vocal folds — the animal does not have to strain to make it
Well supported
Good evidence backs this, though some details remain open.
Excised tiger and lion larynges tested mechanically show vocal folds that are unusually large, with a thick flat medial surface and a distinctive loosely organised superficial layer. Measured stress–strain and shear properties predict a very low fundamental frequency and a low phonation threshold pressure, meaning the folds begin to vibrate at modest subglottal pressure and produce high acoustic power for the effort expended.
Who this applies to
tigers and lions
Studied in
Panthera tigris, Panthera leo
You may have heard
“A tiger roars from sheer power”
The opposite, more or less. The roar is loud and low because the tissue is shaped to be loud and low: large, heavy folds with a flat square surface that vibrate readily and cheaply. Measured in the laboratory, they start vibrating at surprisingly modest lung pressure. The animal is not straining to produce that sound — the instrument is built so it does not have to.
Why we rate it this way, and what the caveats are
Well supportedHigh confidence
Direct mechanical measurement of the tissue, with acoustic predictions that match what the animals actually produce.
How far it can be extended
Both species tested show the same fold morphology and mechanical properties, and the roaring cats share the associated hyoid condition.
Caveats
Excised tissue in a rig is not a vocalising animal; predictions were not checked against the same individuals alive.
Very few larynges, obtained opportunistically.
Explains the capacity to roar, not what tigers use roaring for, which is much less well studied.
Still unanswered
How far does a tiger roar carry in the forests tigers actually occupy, and what does a hearer learn from it?
Describes the surrounding hyoid and pharyngeal anatomy in which those folds sit.
How we know
Why a roar is easier than it sounds
Does a tiger roar because it is powerful, or because its vocal folds are built that way?
You cannot instrument a roaring tiger, so the question was moved out of the animal. Vocal folds from tiger and lion larynges, obtained after death, were mounted and tested mechanically — stretched lengthways and sheared sideways — to measure how stiff the tissue is and how it resists deformation. Those measurements feed a physical model of vibrating tissue that predicts two things independently of any recording: the range of frequencies the folds can produce, and the minimum lung pressure needed to start them vibrating at all. The predictions can then be compared with what the living animals are known to do.
What happened
The folds are unusually large, with a thick flat surface and a loosely organised superficial layer, and their measured properties predict a very low fundamental frequency at a low phonation threshold pressure. The tissue begins vibrating at modest lung pressure and yields high acoustic power for the effort.
What it shows
That the roar is a property of the instrument rather than of the exertion. It also gives the better answer to why tigers do not purr: purring needs a small stiff fold cycling fast and continuously, and this is the opposite piece of equipment. That explanation does not require anyone to believe a bone in the throat decides what noise a cat can make.
What it does not show
Excised tissue in a rig is not a living animal, and no individual was recorded alive and then tested after death, so the chain from tissue to sound is completed by modelling rather than by observation. The larynges were few and obtained opportunistically. And the study explains the capacity to roar, not what tigers roar *for* — which is much less well understood than the mechanism.
The controls — what makes this evidence rather than a story
Two species with the same vocal behaviour tested, so a result peculiar to one animal would show up as disagreement.
Tissue properties measured directly rather than inferred from the sounds produced, which is what makes the prediction independent.
Both stretch and shear measured, since a fold that is stiff in one direction and soft in the other behaves differently from one that is uniformly either.
Predicted frequency range checked against known vocalisations rather than the model being fitted to them.
The usual explanation is anatomical: one link in the chain of small bones suspending the larynx, the epihyoid, is an elastic ligament in lions, tigers and jaguars where it is solid bone in cheetahs and domestic cats. That difference is real and it does track which cats roar. It is not, however, the mechanism, and stating it as one produces the odd impression that a ligament decides what noise an animal makes.
The mechanism is in the vocal folds. Tiger folds are large and heavy with a thick flat surface, and when the tissue is measured mechanically its stiffness predicts a very low fundamental frequency at a low phonation threshold — meaning it starts vibrating at modest lung pressure and yields enormous acoustic power for the effort. A roar is not the sound of straining. It is the sound of an instrument built to do exactly that cheaply.
Purring needs the opposite equipment: a small stiff fold cycling rapidly and continuously through both the in-breath and the out-breath. That last part is the definition that settles the argument. Tigers make a warm, throaty greeting called a chuff or prusten, which anyone would call a purr if they had not been told otherwise — and which stops when the animal breathes in.
On the standard acoustic definition, a purr continues without interruption through inhalation and exhalation. A tiger’s chuff is made on the out-breath only.
A friendly, breathy greeting call made by tigers and some other big cats, produced on the out-breath. It means roughly what a purr means and is not one.
Phonation threshold pressure
The minimum lung pressure needed to set vocal folds vibrating. A low threshold means loud sound for little effort.
Tigers walk alone inside a structure of known neighbours, inherited ground and constant correspondence.
Tigers travel alone but live inside a structure, and they maintain it by leaving messages
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
Radio-tracking shows tigers occupy structured individual ranges rather than moving at random: a resident female holds a largely exclusive area, a resident male’s much larger range overlaps those of several females, and neighbours avoid one another temporally while sharing ground. Scent marking is heavy and spatially non-random, concentrating along boundaries and habitual travel routes, with rates highest for resident males and in areas of range overlap. Marks remain informative for days after deposition, so information passes between individuals who do not meet.
Who this applies to
tigers, across populations from Nepal to the Russian Far East
Studied in
Panthera tigris
You may have heard
“Tigers are solitary animals”
True about the travelling arrangements and misleading about everything else. A tiger walks alone, and a tiger also knows exactly which neighbours it has, where their ground begins, and roughly what each of them has been doing lately — because they are all continuously leaving each other messages on the same trees and trails. The communication is dense and constant; it is just asynchronous. Calling that solitary is like calling someone antisocial because they write letters instead of visiting.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
The same land-tenure structure is documented independently at opposite ends of the species range, in habitats differing by an order of magnitude in prey density.
Caveats
Radio-tracking yields locations rather than behaviour, so encounters and avoidance are inferred from proximity in space and time.
What information a scent mark carries is inferred from placement and timing, not read directly.
Collar samples are small everywhere, because collaring tigers is difficult and expensive.
Still unanswered
What does a mark actually encode — identity, sex, reproductive state, time since deposition, or all of these?
How much of the observed overlap between neighbouring females reflects tolerance rather than failure to exclude?
Finds the same structure in the Russian Far East at a completely different scale, and more female range overlap than the exclusive-territory model predicted.
Daughters settle next door to their mothers, sometimes on ground she hands over
Well supported
Good evidence backs this, though some details remain open.
Dispersal in tigers is strongly sex-biased. Females typically settle close to their natal range, frequently adjacent to their mother, and in some cases occupy a portion of the maternal range that the mother relinquishes. Males disperse considerably further, pass through an extended transient period without a range, and may never acquire one. The resulting population structure is a set of local female kin clusters with males circulating between them.
Who this applies to
documented in detail in Chitwan, with consistent patterns elsewhere
Studied in
Panthera tigris
You may have heard
“Tiger cubs leave their mother and are on their own”
Half of them are. Young males leave properly — long distances, years without ground of their own, and many never get any. Daughters mostly do not. A tigress typically settles beside her mother, often on ground her mother gives up to make room. That means much of a tiger population is a patchwork of related females living next to each other, which is not what "solitary" brings to mind and is closer to how lionesses are usually described.
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
The detailed evidence comes from one long-term population with individually known animals followed from birth; other sites agree in pattern but with less complete life histories.
Caveats
Chitwan is unusually productive floodplain habitat where a daughter can be accommodated nearby; philopatry may be less affordable where prey is sparse.
Dispersing males routinely leave the study area and their outcomes are unknown, so male figures are lower bounds.
A single long-term population carries most of the weight.
Still unanswered
Does maternal range transfer occur in low-density populations such as the Amur, where ranges are ten times larger?
Sunquist, 1981 · Smithsonian Contributions to Zoology
Independently records daughters establishing ranges adjacent to their mothers within the same study population.
Radio-tracking in Chitwan through the 1970s established the shape of it. A resident female holds a largely exclusive area. A resident male’s range is much larger and overlaps those of several females. Neighbours do share ground, and avoid one another in time rather than by never using the same places. This is a land-tenure system, and the same structure turns up decades later in the Russian Far East at a completely different scale.
It runs on scent. Tigers spray-mark heavily and non-randomly, concentrating on boundaries and habitual travel routes, with resident males marking far more than anyone else and rates rising where ranges meet. A mark stays informative for days. That is how a species whose members rarely meet maintains a continuous social structure: the communication is dense and constant, and asynchronous.
What each sex does with the ground it grew up on
Aspect
Females
Males
Where they settle
Close to their natal range, often adjacent to their mother
Far away, after an extended period with no range at all
Ground acquisition
Sometimes a portion of the maternal range, given up to make room
Must displace an existing resident, and many never do
Range size
Smaller; overlaps neighbouring females somewhat
Much larger; overlaps several females
Marking rate
Lower
Highest of any class, especially at boundaries
Population effect
Local clusters of related females
Circulation of unrelated males between clusters
Most of a tiger population is a patchwork of related females living next to one another, because daughters mostly do not leave. Sons do.
None of this makes tigers social in the way lions are — there is no group that hunts, feeds or raises young together. What it does is make "solitary" a description of the travelling arrangements rather than of the social biology. A tigress is very often living beside her mother, on ground her mother arranged for her, in continuous chemical conversation with neighbours she has never stood next to.
Words used here
Land tenure
A system in which individuals hold and defend particular ground over time, with recognised boundaries and rules about who may use what — as distinct from animals simply wandering where they like.
Philopatry
Settling near where you were born. In tigers it is strongly sex-biased: females are philopatric, males are not.
The same animal needs ten times more ground in Russia than in Nepal, and the difference is dinner
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
Tiger range size varies by roughly an order of magnitude across the species’ distribution, tracking prey density. Female ranges in the productive floodplain grassland of Chitwan are small, while female ranges in the Russian Far East, where ungulate density is a fraction of that, are far larger, with male ranges larger again in both cases. Across sites, tiger density is predicted by prey biomass.
Who this applies to
tigers across their range
Studied in
Panthera tigris
You may have heard
“A tiger’s territory is about X square kilometres”
There is no such number, and quoting one is the giveaway that a source has copied another source. A tigress in the grassland of Chitwan can hold a range a fraction the size of one in the Russian Far East, and it is the same species behaving the same way. What changes is how much there is to eat. That is not a footnote about tigers — it is the central fact about them, and it is why a forest can look intact from the air and hold no tigers at all.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
Range sizes measured at both extremes of the species range by comparable telemetry, and the density–prey relationship measured independently across eleven sites with detection probability accounted for.
Caveats
Range estimates depend on the statistical method used to draw a boundary, which differs between studies and eras.
Both comparison populations are in protected areas; ranges in surrounding managed land differ.
Prey density is one driver among several, including human disturbance and terrain.
Still unanswered
How much of the remaining variation between sites is explained by human disturbance rather than by prey alone?
Sunquist, 1981 · Smithsonian Contributions to Zoology
Provides the small-range end of the comparison from high-density floodplain habitat.
Tigers are not opportunists — they select large ungulates, and the size window is narrow
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
Analyses comparing tiger diet against independently measured prey availability show consistent selection for ungulates of roughly 60–250 kg, with most-preferred prey mass near 100 kg, and avoidance of small prey relative to availability. Sambar, wild pig and similar large ungulates are taken disproportionately. Very large prey such as adult gaur are killed but not preferred. Where tiger, leopard and dhole occur together, they partition prey by body size rather than by area.
Who this applies to
tigers across the range, from Indian forest to the Russian Far East
Studied in
Panthera tigris
You may have heard
“Tigers eat anything they can catch”
Measured against what is actually available, they do not. Tigers select a fairly narrow band of large ungulates and pass over smaller animals more often than chance would predict — the same window turns up in India and in Russia despite the species on offer being completely different. That is not a detail. It means a tiger needs a forest with large deer and pigs in it, and a forest that has lost those cannot hold tigers however green it looks.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
The same size preference emerges from a range-wide synthesis and from single-site work using measured availability, despite entirely different prey species being present at different sites.
Caveats
Scat-based and kill-based diet estimates disagree systematically, and the synthesis mixes both.
Small prey digests completely and is under-represented in every method.
Preference cannot be fully separated from differences in how catchable each prey species is.
Still unanswered
How much do individual tigers specialise within the population-level preference?
Karanth and Sunquist, 1995 · Journal of Animal Ecology
Compares tiger, leopard and dhole diets against measured prey abundance in one forest, showing size-based partitioning.
A number quoted for a tiger’s territory is a sign that a source has copied another source. Female ranges in the productive floodplain grassland of Chitwan are a fraction of the size of female ranges in Sikhote-Alin, where ungulate density is far lower — the same species, behaving the same way, needing an order of magnitude more ground. Range size is a readout of prey density.
Tigers are also fussier than they are given credit for. Measured against what is actually available rather than against a kill list, they select ungulates in a fairly narrow band — roughly 60 to 250 kilograms, with the strongest preference near 100 — and pass over smaller prey more often than chance predicts. Very large animals such as adult gaur are killed but not preferred. Where tiger, leopard and dhole share a forest, they partition prey by body size rather than by area, which is how they coexist at all.
The same preferred prey-size window appears in India and in the Russian Far East, despite the available species being entirely different. That consistency is what makes it a preference rather than a menu.
Put those together and the ecology has one governing variable. How many tigers a forest holds is set by how many large ungulates it holds — which is a fact about deer and pigs, and turns out to be the fact that matters most for the species’ survival.
Words used here
Prey biomass
The total weight of prey animals available per unit area, rather than the number of them. It predicts predator density better than counts do, because a predator needs kilograms rather than individuals.
A tigress is committed for two years per litter, and most cubs do not make it
Well supported
Good evidence backs this, though some details remain open.
Wild female tigers produce litters averaging around two to three cubs. Cubs remain dependent on their mother for approximately eighteen months to two years, so inter-litter intervals are long. Mortality before independence is high, and a female’s lifetime output of surviving offspring is correspondingly small. Population growth rate is most sensitive to cub and subadult survival and to adult female survival.
Who this applies to
measured in detail in the Amur population, with broadly similar patterns elsewhere
Studied in
Panthera tigris
You may have heard
“Tigers breed well in captivity, so wild numbers should recover quickly”
Captive breeding rates say nothing useful about wild recovery, and the arithmetic is why. A wild tigress is occupied with one litter for the better part of two years, and most of the cubs in that litter will not reach independence. So losing a single breeding female removes far more than one tiger from the population — it removes years of future output. It also explains why populations recover slowly even where protection is working, and why the number that matters in a recovery programme is not how many tigers there are but how many adult females are breeding successfully.
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
Detailed reproductive histories exist for one population in the harshest part of the range; tropical populations breed somewhat faster, so the specific figures should not be read as species-wide.
Caveats
The Amur population is the most northerly and least productive; tropical rates are higher.
Cub deaths are usually inferred from a cub ceasing to be detected rather than observed directly.
Poaching removed collared animals mid-study, truncating some reproductive histories.
Still unanswered
How much of the difference in reproductive rate between Amur and tropical populations is climate rather than prey?
Provides independence ages and post-independence survival for a tropical population.
A wild tigress has a litter of around two or three cubs, and those cubs depend on her for something like eighteen months to two years. She cannot begin again until they are gone. Mortality before independence is high, so a female’s lifetime output of surviving offspring is small even when everything goes well.
The arithmetic is the point. Losing one breeding female does not remove one tiger from a population, it removes years of future output, and that is why populations recover slowly even where protection works. It is also why the meaningful number in a recovery programme is not how many tigers there are but how many adult females are successfully raising young.
Captive breeding rates tell you nothing useful about wild recovery. In captivity a female can produce litters far more frequently, because she is not the one feeding them.
The nine-subspecies figure is a scheme the genetics has revised, and the revisions disagree
Contested
Researchers actively disagree, and the disagreement is substantive.
The traditional subspecies arrangement, based on skulls and pelage, has been repeatedly revised by molecular work. Marker-based analysis of verified-origin tigers identified groups only partly matching the classical scheme and recognised a previously undescribed Malayan population. Whole-genome analysis of surviving populations supports six extant genetically distinct units with low diversity, evidence of a severe historical bottleneck, and population-specific selection signals. Divergence between all living groups is shallow, consistent with recent common ancestry. A further arrangement recognising only two subspecies — a mainland and a Sunda island group — is also in current use, including by the IUCN; NatureHQ has not built the evidence record behind that proposal and does not weigh it here.
Who this applies to
the classification of living and recently extinct tiger populations
Studied in
Panthera tigris
You may have heard
“There are nine subspecies of tiger, three of them extinct”
That is a scheme drawn from skulls and skins in the twentieth century, and the genetics has been revising it since 2004 without yet settling. Current genome work supports six living units; the earlier marker study drew them differently and added a Malayan tiger nobody had described; and a two-subspecies arrangement is also in current use. What they agree on is that the differences between tigers are shallow — the whole living radiation is recent — so the argument is partly about how much difference deserves a name. Anyone quoting a confident number is quoting one of several.
Why we rate it this way, and what the caveats are
ContestedModerate confidence
The underlying genetic data are good and the disagreement is real: it concerns how much divergence justifies a named subspecies, which is a question about taxonomic convention as much as about tigers.
Caveats
Extinct populations are represented by museum material or not at all, so any count of historical subspecies is partly unrecoverable.
The two-subspecies arrangement is named but not assessed: NatureHQ has not built the evidence record behind it, so this claim weighs only the marker and genome studies it cites.
Sample sizes per population are small, unavoidably.
Subspecies boundaries have management consequences, which means the argument is not purely academic.
Where researchers disagree
Marker-based and genome-based analyses draw the population boundaries differently, and a two-subspecies arrangement is also in current use. The disagreement is about how much divergence a subspecies name should require rather than about the underlying data, which is why more sequencing has not settled it.
Still unanswered
Should conservation units be defined by genetic distinctiveness, local adaptation, or the practical geography of the remaining populations?
Earlier marker-based structure that agrees on shallow divergence and few units while drawing the boundaries differently, and which first recognised the Malayan tiger.
Represents the morphological subspecies scheme that the molecular work revised, and is the source of the counts still in general circulation.
Tigers average larger, but a Sumatran tiger is smaller than any lion
Well supported
Good evidence backs this, though some details remain open.
Body size in tigers varies substantially between populations, with the northern Amur and mainland Bengal populations at the large end and the island populations, notably Sumatran, markedly smaller. Genome-wide analysis identifies selection on body-size and metabolic genes in the Amur population. Museum-derived maximum records are biased upward by selective collection of exceptional trophy animals, so upper figures quoted for both tigers and lions are not comparable to population means.
Who this applies to
comparisons between tiger and lion populations
Studied in
Panthera tigris, Panthera leo
You may have heard
“Tigers are bigger than lions”
On average, yes — and the average conceals the interesting part. Tiger body size varies far more between populations than the tiger–lion difference does. A big Amur or Bengal tiger outweighs a lion comfortably; a Sumatran tiger is smaller than essentially any lion. So "which is bigger" has no single answer without naming which tiger. It is also worth distrusting the record weights on both sides: most come from trophy hunting, where the incentive ran towards the largest specimen and, occasionally, towards weighing it after a large meal.
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
The within-species variation and its genetic basis are well established; direct like-for-like comparison between tiger and lion body mass is hampered by inconsistent measurement standards across the older literature.
Caveats
Historical body masses were often recorded from trophy animals, sometimes with the stomach contents included, which inflates them.
Few modern studies weigh wild individuals of either species systematically.
Captive animals of both species are heavier than wild ones and frequently supply the numbers people quote.
Still unanswered
How much of the Amur–Bengal size difference is genetic and how much is prey availability during growth?
Finds selection on body-size and metabolism genes in the Amur population, giving the size difference a mechanism.
The nine-subspecies figure still in general circulation comes from a twentieth-century scheme drawn from skulls and skins. Molecular work has been revising it since 2004 without settling: marker-based analysis found groups only partly matching the old scheme and recognised a Malayan tiger nobody had described, whole-genome work supports six living populations, and a two-subspecies arrangement — one mainland group and one Sunda island group — is also in current use, including by the IUCN. What every analysis agrees on is that the differences between living tigers are shallow, because the whole surviving radiation is recent.
The disagreement is partly about tigers and partly about taxonomy — how much divergence a subspecies name ought to require. It is not academic, because subspecies boundaries determine which animals may be bred with which in captive programmes and which populations count as separate things to save.
Three answers to the same question
Basis
Living units
Note
Skulls and skins, 20th century
Six living, three extinct
Still the version in general circulation
Mitochondrial and nuclear markers, 2004
Groups only partly matching the above
First recognised the Malayan tiger
Whole genomes, 2018
Six, with population-specific selection
Amur shows selection on body size and metabolism
The size question runs into the same variation. Tigers average larger than lions, and tiger body size varies more between tiger populations than the tiger-lion difference does: a large Amur or Bengal tiger outweighs a lion comfortably, while a Sumatran tiger is smaller than essentially any lion. Record weights on both sides deserve suspicion, since most come from trophy hunting, where the incentive ran towards the largest specimen and occasionally towards weighing it after a large meal.
Words used here
Subspecies
A named population within a species, distinct enough to be recognised but not separate enough to be its own species. Where the line falls is a convention rather than a discovery, which is why counts differ.
A white tiger is an ordinary Bengal tiger with one letter changed in one gene
Established
Specialists would state this without hedging. Multiple independent lines of evidence agree.
The white phenotype results from a single amino-acid substitution in SLC45A2, inherited as a recessive. It removes pheomelanin, the red-yellow pigment of the background coat, while leaving eumelanin — and therefore the black stripes — intact. It is not albinism: eye and stripe pigmentation are unaffected. The variant occurred in wild Bengal tigers and is a natural polymorphism of that population. Its frequency in captivity results from repeated breeding within a very small founder group, and the health problems associated with captive white tigers are attributable to that inbreeding rather than to the pigment allele.
Who this applies to
white tigers, all of which are Bengal tigers or Bengal hybrids
Studied in
Panthera tigris
You may have heard
“White tigers are a rare species, or a snow-adapted subspecies”
Neither. Every white tiger is a Bengal tiger carrying two copies of one recessive change in a single pigment gene, which switches off the yellow-red pigment and leaves the black alone — which is why they still have stripes and are not albinos. No white tiger comes from anywhere cold; the animals they descend from are from India. The variant is genuinely natural and genuinely was found in wild tigers. What is not natural is how many white tigers exist in captivity, which is the result of breeding a very small number of related animals to each other over and over. The health problems that follow come from that, not from the colour.
Why we rate it this way, and what the caveats are
EstablishedHigh confidence
The causal variant was mapped in a pedigreed family and confirmed by sequencing; the gene is a well-characterised pigmentation locus across vertebrates.
Caveats
The mapping family is captive and highly related, which is what made the mapping tractable.
The claim that the variant is natural rests on historical records of white tigers shot or captured in the wild, not on a living wild population.
No white tiger has been recorded in the wild for decades.
Still unanswered
Would the variant persist in a wild population, or is it removed by selection quickly enough to explain its rarity?
Establishes the population structure against which a white tiger is placed as a Bengal variant rather than a distinct lineage.
A single amino-acid substitution in the pigment gene SLC45A2, inherited recessively, removes the red-yellow pigment from the background coat and leaves the black stripes untouched. That is the whole of it. A white tiger is a Bengal tiger with two copies of that variant — not a subspecies, not a snow adaptation, and not an albino, which is why the stripes and the eyes are still pigmented.
The variant is genuinely natural and was recorded in wild Bengal tigers. What is not natural is how many white tigers exist in captivity, which is the result of breeding a very small founder group to each other repeatedly. The health problems associated with white tigers come from that inbreeding, not from the colour gene — and running the two together gets both the biology and the welfare argument wrong.
Animal welfare
Why the white-tiger argument is not about the colour
The pigment variant itself is a single natural recessive and is not known to cause any health problem on its own. The defects associated with captive white tigers — the crossed eyes, the spinal and immune problems, the stillbirths — come from the fact that every white tiger alive descends from a very small number of related animals bred repeatedly to each other. Attributing them to the colour gene gets the biology wrong; attributing them to nothing gets the welfare argument wrong. It is also worth being clear that no white tiger belongs to any wild conservation programme, so presenting one as rare or endangered misdescribes what a visitor is looking at.
Where this applies: Global. Captive breeding regulation is national and differs widely; the biology below does not.
Words used here
Recessive
A variant that only shows when both copies of a gene carry it. Two orange tigers can therefore produce a white cub if both carry one copy.
Tigers vanish from forests that are still standing.
Empty forest is the problem — a perfect-looking forest with no deer holds no tigers
Well supported
Good evidence backs this, though some details remain open.
Across sites with paired estimates obtained by methods that model detection probability explicitly, tiger density is predicted by prey biomass. Depletion of large ungulates by hunting reduces tiger density independently of forest cover. Almost the entire remaining breeding population is concentrated in a small number of identified source sites making up a few per cent of the species’ remaining range.
Who this applies to
wild tiger populations
Studied in
Panthera tigris
You may have heard
“Tigers are threatened by habitat loss and poaching”
Both true, and the phrasing hides the more immediate mechanism. Tigers disappear from forests that are still standing, because the deer and pigs have been snared out of them. Conservationists have a term for the result — empty forest — and it is why satellite maps of remaining habitat overstate how much tiger country is left. Protecting prey turns out to be the operative lever, and it is a much less photogenic campaign than protecting tigers.
Why we rate it this way, and what the caveats are
Well supportedHigh confidence
The density–prey relationship rests on estimates that account for animals present but not detected, which is what distinguishes it from earlier work; the source-site concentration is separately documented, though its policy conclusion is contested.
Caveats
Eleven sites, all Indian and all with some protection, carry the quantitative relationship.
The source-site prioritisation argument is contested; other authors argue it underweights connectivity and recovery outside those sites.
The site assessment is now over a decade old and several populations have changed materially.
Still unanswered
How well does the density–prey relationship hold in South-East Asian forests, where prey depletion is most severe and data are sparsest?
Supplies the density estimation method without which the relationship could not be measured reliably.
Habitat loss and poaching are the two words attached to tiger decline, and between them they hide the more immediate mechanism. Across sites where both tiger and prey densities have been estimated with methods that account for animals present but not detected, tiger density tracks prey biomass. Snaring removes deer and pigs from forests that remain intact on any satellite image, and the tigers go with them.
The remaining population is also far more concentrated than a range map suggests. Nearly all breeding tigers are in a small number of identified source sites making up a few per cent of the species’ remaining range — most of them places with names, several of them the same places the long-term studies were done in.
Safety
What NatureHQ will not tell you about tigers and people
Where attacks on people have been studied carefully they trace to a small number of identifiable individual animals, usually old, injured or displaced, in places where forest meets settlement directly. That is a finding about attribution and it is not safety advice. NatureHQ has no basis on which to advise anyone who lives or works alongside tigers, and does not attempt to: what reduces risk depends on the landscape, the season, the local population and the particular animals in it. Anyone in that position should follow the guidance of the national park or forest department responsible for the area, which is the body that knows those things.
Where this applies: Global statement of scope. Every practical measure that matters here is national or landscape-specific, which is the reason for the limit.
When to get help: The relevant national park or forest department, and local emergency services. Not a website.
Attacks trace to a small number of identifiable animals, usually old, injured or without ground
Well supported
Good evidence backs this, though some details remain open.
Analysis of three decades of tiger attacks on people around Chitwan National Park attributes incidents to a small number of individual tigers rather than to the population at large. Those individuals were disproportionately old, injured, or displaced from established ranges. Incidence rose as tigers occupied buffer-zone forest adjoining settlement, and fell where the responsible individuals were identified and removed.
Who this applies to
documented for one landscape in Nepal over three decades
Studied in
Panthera tigris
You may have heard
“Tigers are man-eaters”
The risk is real and it is not a property of tigers. Where it has been studied carefully, attacks trace back to a handful of specific animals — typically old, injured, or pushed out of their ground and unable to hunt what they should be hunting — in places where forest meets settlement directly. Most tigers, in the same landscape, in the same years, never come near a person. NatureHQ is not in a position to give safety guidance to people who live alongside tigers, and does not attempt to; what it can say is that the sentence "tigers are man-eaters" describes the wrong subject.
Why we rate it this way, and what the caveats are
Well supportedModerate confidence
Careful long-term attribution in one well-studied landscape, with the important limitation that the responsible animal could not always be identified.
Caveats
One landscape, with a particular configuration of park, buffer zone and settlement.
Attribution to an individual was not always possible, so the individual-attribution rate is a lower bound.
The Sundarbans has a long and different history of attacks that this analysis does not address.
Still unanswered
Why do attack rates differ so much between landscapes with similar tiger and human densities?
Compiles three decades of attacks with individual attribution and examination of the tigers responsible where they could be located.
Words used here
Empty forest
Forest that is structurally intact but has lost its large animals, usually to snaring. It looks like habitat and does not function as any.
Source site
A place that produces more animals than it loses, so that it can supply surrounding areas. Almost all remaining wild tiger breeding happens in a small number of these.
Is adult tiger dermis actually pigmented beneath the dark stripes?
Why it matters: It is the most repeated fact about tigers and NatureHQ could not find a primary measurement of it. The pre-pattern work establishes where the pattern is specified, which is a different question from whether a shaved tiger looks striped.
What would settle it: Histology of skin biopsies from beneath light and dark regions of the same animal, which would be a by-product of routine veterinary work rather than a study anyone need commission.
Why stripes, when most closed-habitat cats have spots?
Why it matters: The comparative analysis that links patterned coats to closed habitats fits spotted cats better than striped ones, so the tiger is the awkward case in the dataset that supports the camouflage explanation.
What would settle it: Detection experiments with dichromatic prey against real vegetation, comparing striped and spotted patterns at realistic stalking distances.
What does a scent mark actually encode?
Why it matters: The entire social structure runs on marking, and what information a mark carries — identity, sex, reproductive state, how long ago — is inferred from where and when tigers place them rather than read directly.
What would settle it: Chemical characterisation paired with presentation experiments to captive tigers, testing which components change with status and which are discriminated.
Does maternal range inheritance happen where ranges are ten times larger?
Why it matters: Female philopatry is documented in detail in high-density floodplain habitat, where a daughter can be accommodated nearby. Whether an Amur tigress can afford the same is unknown, and it determines how those populations rebuild after loss.
What would settle it: Long-term genetic sampling of neighbouring females in the Russian Far East, establishing relatedness against spatial position.
What are tigers doing when they roar?
Why it matters: The mechanism is well characterised down to the tissue mechanics, and the function is not. For an animal that communicates mostly by scent, a very loud long-distance call is a conspicuous exception nobody has explained.
What would settle it: Acoustic arrays across a known population, relating roars to the identity, position and subsequent movement of individuals.
NatureHQ publishes its own gaps. This record is at 92% completeness against what we would call a finished subject.
14 high-priority search intent(s) not yet covered
Whether adult tiger dermis is pigmented beneath the stripes is left open rather than answered, because no primary measurement was found. It is the single most repeated claim about the animal.
Hunting behaviour itself — stalk distances, kill rates, success rates, how a tiger actually takes a sambar — is thin here relative to the prey-selection ecology.
The Sundarbans, which has the longest and strangest history of tiger–human conflict anywhere, is not treated; doing it responsibly needs the social science alongside the ecology.
Sensory biology beyond vision is essentially absent: tiger hearing, olfaction and whisker use are all real literatures and none of them is here.
Almost all of the field biology comes from two populations, Chitwan and Sikhote-Alin, which is a genuine limit on how confidently anything here generalises to South-East Asia.
The two-subspecies arrangement is named but not assessed. It is in current use, including by the IUCN, and NatureHQ has not built the evidence record behind it — so the taxonomy section weighs the marker and genome studies it cites and is explicit that it is not weighing that one.
Last reviewed 2026-08-11 · 15 claims · 42 search questions answered on this page