Catnip is a plant, not a drug, and it is worth starting there because the way people talk about it suggests otherwise. It is a herb in the mint family, it contains a volatile compound that a proportion of cats can detect, and the behaviour that follows is a reflex rather than an experience anyone can describe from the inside.
The short answer to what is catnip: Nepeta cataria, a perennial herb native to Europe and Asia and now naturalised almost everywhere, whose leaves and stems carry an iridoid monoterpene called nepetalactone. About half to two thirds of domestic cats respond to it. The rest cannot, and never will, because the ability is inherited rather than learned.
Nepeta cataria is a coarse, grey-green herb that grows to about a metre and produces spikes of small white or pale purple flowers. It is easy to grow and difficult to contain, which is why it turns up in gardens it was not invited into and why it is listed as invasive in parts of North America.
The active compound is concentrated in the leaves and stems, in the essential oil. It is volatile, which is the single most practical fact about catnip: the compound leaves the plant continuously, which is why a bag of dried catnip that has been open on a shelf for a year smells of dust and produces no response. Freshness is not a marketing point, it is the difference between working and not working.
Nepetalactone is an iridoid, a class of compounds that plants use defensively. It is also, incidentally, an effective insect repellent β more on that later, because the recent research on it is the most interesting thing in this field.
One naming confusion is worth clearing. Catnip and catmint are not the same plant in practice, even though the terms get used interchangeably. True catnip is Nepeta cataria. The ornamental catmint sold in garden centres is usually Nepeta Γ faassenii, a hybrid bred for flowers and tidy habit, and it carries far less nepetalactone. A cat that ignores a garden catmint is not necessarily a non-responder; it may simply have been given the wrong plant.
The response begins with smell, not with taste, and this matters for how catnip should be offered.
When a cat sniffs catnip, nepetalactone reaches the olfactory epithelium at the back of the nasal cavity. From there the signal travels along the olfactory nerve to the olfactory bulb, and the bulb projects to the amygdala and the hypothalamus. Those two structures handle emotional response and the hormonal regulation that sits underneath it, which is why the behaviour that follows looks like a reproductive display rather than like play.
The behaviour itself has a characteristic sequence. The cat sniffs, then licks or chews at the source, then shakes its head, then rubs its chin and cheek against it, then rolls. Some cats vocalise. Some drool. Some become briefly hyperactive and tear around; others flop and go quiet. Which of those a given cat does is individual, and the intensity does not tell you anything about the cat's health or its bond with you.
Two details in that sequence are worth noticing. The chewing and the rubbing are not incidental β they are how the compound gets transferred onto the cat's own face and coat, which is the mechanism behind the mosquito finding described below. And the response is not mediated by the vomeronasal organ, the structure cats use for pheromones, which is why a cat that has lost its sense of smell through nasal disease may stop responding.
Roughly a third to a half of domestic cats show no reaction at all, and there is nothing wrong with them.
The trait is inherited as an autosomal dominant, which means a cat needs only one copy of the variant to respond. A cat without it will never respond, at any age, to any quality of catnip. This is not a matter of exposure, preference or temperament. It is a receptor question.
The practical consequence for an owner is that a cat which ignores catnip is not broken, bored or depressed. Roughly one in three cats is a non-responder, and the number is high enough that it is the first thing to rule out before looking for a behavioural explanation. Siamese and Abyssinian cats are frequently cited as showing lower response rates, though the sample sizes behind that claim are small and the dominance pattern predicts exactly the rates that are observed across breeds anyway.
An owner who wants to know whether their cat carries the trait cannot find out by trying harder. They can find out by offering silver vine, which reaches a larger proportion of cats including many catnip non-responders.
A kitten under about six months usually does not respond, and this is normal rather than diagnostic.
The response develops alongside sexual and neurological maturity, and the commonly quoted threshold is six months. A kitten that ignores catnip at ten weeks may respond fully at eight months. A kitten that ignores it at eight months is more likely to be a genuine non-responder, though the only way to be certain is to try again later.
What this means practically is that catnip is not a kitten toy. There is no harm in a small amount, but there is no point either, and a kitten's enrichment budget is better spent on play that develops coordination and social skills. Most of the guidance suggesting an introduction at eight weeks is about safety rather than about response β a very young kitten eating a large quantity of dried herb is a stomach ache waiting to happen, and a kitten is exactly the animal likely to eat rather than sniff.
Older cats are a separate question, and the answer is less clean. Some geriatric cats respond less, some respond the same, and some develop a stronger response because they are less mobile and the rolling is one of the few vigorous things they still do. A change in response in an old cat is not in itself a sign of illness, but a cat that suddenly stops responding to something it used to love is worth watching for other changes.
The response lasts somewhere between five and fifteen minutes, most commonly around ten, and then it stops.
What follows is a refractory period of roughly half an hour to two hours during which the cat will not respond to catnip at all, however much is offered. This is not tolerance and it is not satiation in the ordinary sense. The receptor system has desensitised and needs time to reset.
The refractory period is the built-in dose limit on the whole thing. It is also the reason catnip cannot be used as a continuous distraction: a cat left with a catnip toy will have ten minutes of interest and then ignore it for an hour, which is not what an owner hoping to occupy a bored cat is after.
Building a routine around it works better than leaving it out. Offer catnip two or three times a week rather than daily. Repeated daily exposure produces habituation in some cats, where the response weakens over weeks β the compound is still working, and the cat has simply stopped finding it worth reacting to. Rest, and the response usually comes back.
Every article about catnip says it is non-addictive, and almost none of them says why. The answer came out of a study published in 2021, and it is more interesting than the reassurance.
The research, by Uenoyama and colleagues in Science Advances, worked mainly with silver vine rather than catnip, and with its principal compound nepetalactol, which is closely related to nepetalactone. Three findings came out of it.
The first was that exposure raised plasma beta-endorphin in the cats. Beta-endorphin is one of the body's own opioid peptides, and an increase in it produces euphoria and analgesia.
The second was that blocking mu-opioid receptors with naloxone suppressed the rubbing response. Naloxone is the drug given in opioid overdose; it occupies the receptor without activating it. If naloxone stops the behaviour, the behaviour is running through the opioid system. That is the strongest possible evidence that the response is endorphin-mediated rather than merely endorphin-associated.
The third, and the one that explains the non-addictive claim mechanically rather than as a promise: the compound triggers the release of the cat's own endorphins rather than binding the receptor itself the way morphine does. There is no exogenous molecule sitting on the receptor, so there is nothing to build tolerance against in the pharmacological sense, and nothing to withdraw from. The cat's own system does the work and then stops.
That distinction is worth holding onto, because it separates catnip from a whole category of things that are described as "natural" and are not harmless. A compound that binds a receptor directly is a drug regardless of where it came from. A compound that prompts the body to release its own signalling molecules, briefly and within the body's own limits, is a different proposition.
What catnip is not is a treatment for anything. It does not reduce anxiety in a clinical sense, it does not treat pain, and it does not modify a behaviour problem. The enrichment value is real β a cat that plays vigorously for ten minutes is a cat that has used energy and rehearsed a hunt sequence β and that is the extent of the claim.
The third result of that study is the one that tends to get left out of catnip articles, and it is the most surprising.
Cats that rubbed and rolled in the compound had significantly fewer Aedes albopictus mosquitoes land on them than cats that had not. The rolling, in other words, is not only pleasure. It is a cat applying an insect repellent to its own face and head.
The finding connects to a much older observation: nepetalactone had already been shown to repel mosquitoes when applied to human skin, which is why the researchers tested the hypothesis in cats in the first place. The demonstration in a living animal was done with silver vine's compound, but the two are close relatives and the effect is very likely general.
Two cautions before anyone replaces their mosquito strategy with a catnip plant. The concentration and formulation matter enormously, and the effect measured in a study is not the effect of a cat rolling in a herb for ten minutes in a garden. And repellent effects on the cat's own face do not protect a human sitting next to it.
Four plants produce this kind of response, and knowing the alternatives matters because a third of cats are non-responders to the most common one.
Silver vine β Actinidia polygama, a climbing shrub from East Asia, also called matatabi. It contains nepetalactol and actinidine, and it reaches a higher proportion of cats than catnip does, including many that ignore catnip entirely. The 2021 study was largely a silver vine study for that reason. The usual form is a dried powder or a dried gall, and the response tends to be more intense than catnip's in cats that respond to both.
Valerian root β Valeriana officinalis. A smaller share of cats respond, and the compound responsible is different again, an actinidine-type molecule rather than an iridoid. It is also the one most likely to be raided by rats in a shed, which tells you something about its appeal across species.
Tatarian honeysuckle β Lonicera tatarica, sold as a wood chew. Reaches a decent proportion of cats, and the solid form makes it a better option for a cat that eats rather than sniffs.
Catmint β Nepeta Γ faassenii, the ornamental one. Weak by comparison, and usually the reason an owner concludes their cat is a non-responder when it is not.
The practical sequence for a cat that ignores catnip is: check the catnip is fresh, then try silver vine. Do not buy four products at once. One variable at a time, or you will not know which one worked.
Most of what this site covers is about dogs, so the dog question deserves a direct answer: catnip is not toxic to dogs.
It is listed as non-toxic to dogs, cats and horses by the ASPCA, and there is no toxin in it. A dog that eats a catnip toy's contents will not be poisoned.
What a dog will not get is the response. The behaviour depends on a receptor system that dogs do not appear to share in the same form, so the rolling and the rubbing do not happen. What does happen, in a dog that eats a quantity of dried herb, is mild gastrointestinal upset β a bit of vomiting or soft stool, which resolves without treatment.
The real hazard in a dog household is not the herb, it is the toy. Catnip toys are small, often stuffed, frequently stitched with a seam that a determined chewer can open, and sometimes strung on elastic or ribbon. A dog that treats a catnip mouse as a chew toy is at risk from the mouse, not from the catnip β from the stuffing, the seam, the bell, or the string. If a household has both species, catnip toys should be play-supervised items kept off the floor rather than permanent fixtures.
There is a second-order point worth making for anyone with a dog that raids the litter tray or the cat's belongings. Catnip is sometimes described as a mild sedative in herbal medicine for humans, and that reputation occasionally leads people to offer it to an anxious dog. There is no evidence base for that in dogs, the response mechanism is not present, and an anxious dog needs a behavioural plan rather than a herb.
Catnip is safe for cats. It is non-toxic, it is not addictive in the mechanism described above, and there is no withdrawal. The behaviours look dramatic and are not distress.
The one genuine caution is quantity, and it applies to cats that eat the herb rather than sniffing it. A large amount can cause mild vomiting or loose stool. It is uncommon, it resolves on its own, and it is the reason to offer a pinch rather than leave a pile in a bowl.
A sensible offer is roughly half a teaspoon of dried herb for an adult cat, or a small stuffed toy, or a pinch rubbed into a cardboard scratcher. Fresh catnip leaves can be offered whole. The point is the scent, not the volume, and doubling the amount does not double the effect.
Supervision matters more for kittens and for the first exposure than it does long-term. Watch for a cat that chews and swallows rather than sniffs, and remove the source if it does.
Two things to avoid. Do not buy catnip products that list added oils, "catnip spray" concentrates of unknown composition, or herbal blends without an ingredient list β the herb is safe and the additives are the unknown. And do not use catnip as a way to manage a behaviour problem. A cat that is over-grooming, spraying or hiding needs a veterinary and behavioural assessment; a ten-minute rolling session is not an intervention.
The safety story that applies to dried leaf and fresh plant does not transfer automatically to catnip essential oil, and the two are sold side by side.
Dried herb is a few per cent essential oil by weight, bound in plant tissue and released slowly. A bottle of essential oil is that same compound at close to full strength, often diluted in a carrier that is itself unlisted. The dose per gram differs by orders of magnitude.
That matters for three routes. On skin, concentrated nepetalactone is an irritant, and cats groom themselves, so anything applied to the coat is swallowed. By mouth, a few drops of essential oil is a very different exposure from a pinch of leaf. And in the air, a diffuser running in a small closed room exposes a cat continuously for hours, which is nothing like a ten-minute sniff.
None of this makes catnip essential oil a poison in the way that, say, tea tree oil is to cats. It makes it a concentrated product with no established feline dose, sold for a different purpose. The herb is what you offer a cat. The oil is an insect repellent ingredient and a fragrance component, and it belongs out of reach of the animal it is named after.
A catnip toy in a house with four cats does something that a single-cat household never sees: it becomes a resource.
The response raises arousal. A cat in the middle of a rolling session is not in a state to share, and a second cat approaching the same toy at the same moment is inviting a swat. The result is that catnip sessions can produce the only genuine conflict some multi-cat households have.
The fix is structural rather than behavioural. Offer catnip in more than one place at once, spaced far enough apart that a cat can respond without another cat entering its space β separate rooms rather than opposite ends of one. Offer it to cats that respond, and do not leave a catnip toy on the floor as a permanent fixture. And supervise the first few sessions, because the cats that will compete are not always the ones you would predict.
For a household with one responder and several non-responders, the more useful observation is that the non-responders are not missing out on anything. They are not being deprived of a pleasure the others get. Their receptor system does not produce the response at all, and their enrichment needs are met by play, height, scratching surfaces and hunting games β the same list, catnip or no catnip.
Catnip is one of the few products where the storage method matters more than which one you buy, and this is a measurable rather than a theoretical point. Side-by-side testing of commercial catnip products has found that the difference between an airtight container and an open bag is larger than the difference between brands.
The reason is volatility. Nepetalactone leaves the plant material continuously at room temperature, and the rate depends on exposed surface area, air movement and heat. Powdered catnip loses potency faster than whole leaf for exactly that reason: more surface, more escape.
What to do. Buy whole dried leaf rather than powder where the option exists. Store it in an airtight container, in the dark, away from a warm surface. The freezer works well and is the single best storage location if you have space, because the low temperature slows the loss considerably. Let the container come to room temperature before opening it, or condensation will damp the herb.
Test the catnip before blaming the cat. A simple freshness check: rub a pinch between your fingers and smell it. Fresh catnip has a distinct, slightly lemony, slightly medicinal scent that is obvious once you know it. Catnip that smells of hay or dust is spent.
One more thing about the exchange. Because the response has a refractory period, a cat offered stale catnip twice in an hour may show nothing on either attempt for two different reasons. Change one variable at a time.
Catnip is easy to grow and hard to keep where you put it. It self-seeds prolifically, it spreads by runners, and it tolerates poor soil. In parts of North America it is classed as invasive, and a plant put in a border will be in the lawn within two seasons.
If you want to grow it anyway β and for a multi-cat household it is worth it β the practical approach is a large pot on a hard surface rather than a bed. That contains the runners and stops the self-seeding from becoming a project.
Harvest before the plant flowers for the strongest concentration, cut the stems in the morning after the dew has dried, and dry them in a dark, well-ventilated space rather than in sunlight, which drives off the compound you are trying to keep. Once dry, strip the leaves and store them as above.
A fresh plant in the house is also the best catnip toy there is, for the obvious reason that it is alive and therefore maximally fresh. A cat that ignores a bag of dried herb will often destroy a growing plant. That is not a contradiction; it is the freshness point demonstrated.
The response is not limited to house cats. Lions, tigers, leopards and jaguars all show it, which is a useful piece of evidence about what the trait actually is: an inherited receptor characteristic shared across the felid lineage rather than a quirk of domestication.
Not every felid responds. The pattern across species is patchy, and the differences track the same kind of genetic variation that separates responding domestic cats from non-responders. It is a good reminder that "cat" is not a single thing, and that a behaviour observed in one species of the family cannot be assumed in another without evidence.
It is not a drug in the intoxicant sense. The behaviour is a reflex triggered through the olfactory system, and it runs on the cat's own endorphins. There is no intoxication, no impairment, and no hangover.
It is not a sedative. It is often described as one because some cats flop and go quiet, but the same compound makes other cats tear around the room. The direction of the response varies by individual, and in the same cat it can vary by exposure. If you have been told catnip will calm an anxious cat, you have been told something that is true for a subset of cats and false for others.
It is not a treat. It is not food and it has no nutritional role. A cat that eats a lot of it gets a stomach ache, not a benefit.
It is not addictive. Covered above, with a mechanism.
It is not a substitute for play. The most useful way to use it is as an adjunct to a play session: rub it into a toy immediately before using the toy, so the sensory trigger is paired with the physical work of a hunt sequence. Catnip on its own is ten minutes; catnip plus a wand toy is a proper session.
It does not make cats aggressive, but it can raise arousal. The aggression reports usually trace back to a human hand in the middle of an intense rolling session. The cat is not being aggressive; it is over-aroused and the hand is in range. Use a toy, not a hand.
Safe, and pointless below about six months, because the receptor response has not developed. A small pinch is harmless; a large amount eaten will cause a stomach upset, and kittens are more likely to eat than sniff. Wait until closer to six months, or until the cat shows interest.
The refractory period. After five to fifteen minutes of response the system desensitises for half an hour to two hours, and during that window catnip produces nothing at all. It is not a sign of boredom or of a weak product.
No. There is no toxic dose in any practical sense, and the receptor system limits the response on its own. Very large amounts eaten can cause vomiting or loose stool, which is uncomfortable rather than dangerous.
Not in the same way. Dogs do not show the rolling and rubbing response, and the herb has no euphoric effect on them. It is non-toxic, so an accidental mouthful is not an emergency. The risk in a dog household is the toy rather than the herb.
Yes, considerably, provided you are comparing fresh leaf with dried herb of the same age. A growing plant is the freshest source there is. Between two dried products, storage history matters more than brand.
You can, but the response weakens over weeks in some cats. Two or three times a week with a sealed container in between produces a stronger and more consistent response than a permanent open bowl.
No, and the ones it does not work on are not unwell. Roughly a third to a half of cats carry no responsive variant, and no product will change that. Try silver vine before concluding anything about the cat.
Catnip is a small, well-understood plant with one interesting compound, one clear genetic rule and one genuine limit built into its own mechanism. The useful things to know about it are that freshness decides whether it works at all, that a third of cats will never respond, that the response lasts about ten minutes and then stops, and that the safest way to give a cat something to do with it is to pair it with a toy rather than leave it in a bowl. PawAtlas keeps the same position here as it does on enrichment for dogs: the compound is not the point, the session is.