This is the longest guide in the set and the one we would keep if we had to throw the rest away. Every fragrance description you will ever read assumes you know what these words mean. Almost nobody does.
Descriptions of smell are opinion, ours mostly and sometimes the trade's, while prices, dates and processes are fact. We have tried to keep those two things visibly separate.
The citrus and the herbs
Bergamot
A small bitter citrus, mostly grown in Calabria, pressed from the peel rather than distilled. It smells like lemon with the sourness taken out and something faintly floral and tea-like put in. It is one of the most common top notes in perfumery, standard in nearly every eau de cologne and every classic chypre.
Natural bergamot is phototoxic. It contains furocoumarins, chiefly bergapten, which can burn skin in sunlight. The industry's answer was bergapten-free bergamot, processed to remove the dangerous fraction. Hold on to that when someone tells you natural means safe. The safe bergamot is the processed one.
Lavender
Steam distilled from the flowering tops. Clean, herbal, slightly camphorous, with a sweetness underneath that most people do not notice until it is pointed out.
Lavender plus oakmoss plus coumarin is the fougère accord, invented in 1882, and it is the backbone of the men's aisle to this day (the full story is in our short history of perfume). If you think you dislike lavender because of drawer sachets, try it in Le Male or Cool Water and reconsider.
The flowers
Rose
Two species matter. Rosa damascena, the damask rose, mostly from Bulgaria and Turkey, gives the classic rose otto: bright, slightly spicy, a little honeyed. Rosa centifolia, the May rose, grown around Grasse, is softer and greener and usually made as an absolute rather than distilled.
The economics are brutal and they explain everything about rose in fragrance. Distilling damascena yields around 0.02 to 0.03% oil. That means somewhere between 3,500 and 5,000 kilos of petals for one kilo of rose otto, as low as 2,600 in a very good year and as high as 8,000 in a bad one.
So when a modestly priced bottle lists rose, it is not lying, but the rose in it is mostly built from other materials. That is simple arithmetic, not a scandal.
Try: Portrait of a Lady, Rose 31, Lyric Woman.
Jasmine
The other great floral expense, and the one with the better story.
One kilo of jasmine flowers is roughly 8,000 individual blossoms, and a picker gathers about 350 grams an hour. Reported figures for flowers to absolute range from about 350 kilos per kilo of concrete up to 700 or 800 kilos of flowers per kilo of absolute, and the sources disagree.
It is picked before dawn, and that is chemistry rather than romance. Benzyl acetate, the compound behind the bright part of jasmine, rises about sevenfold between afternoon and pre-dawn, and flowers picked after sunrise have already lost volatiles to the heat. In Grasse the season runs August to October.
Jasmine smells creamy, slightly overripe, and faintly animal underneath. That last part comes from indole, and it is why some people find jasmine dirty.
Tuberose
The loudest white flower there is. Creamy, waxy, almost buttery, with a green sharpness at the edges and something close to menthol if you get very near it.
Tuberose divides people more sharply than almost anything else in perfumery. There is no moderate opinion about Fracas, the reference Cellier built in 1948, and nobody has bettered it since.
Orange blossom and neroli
Same tree, different products, and the distinction turns up on labels constantly.
Neroli is steam distilled from bitter orange blossom: bright, green, clean, slightly metallic. Orange blossom absolute is solvent extracted from the same flower: heavier, sweeter, more indolic and honeyed.
Neroli reads fresh, orange blossom reads sensual, and yet they are the same flower.
Ylang-ylang
Distilled from the flowers of Cananga odorata, mostly Madagascar and the Comoros. Banana-like, custardy, intensely sweet, with a rubbery edge.
It is distilled in successive fractions, so the first material off the still is different from the last, and they are graded and sold separately. That is unusual and it means "ylang-ylang" on a label covers a wide range of smells.
Iris, or orris
The most expensive material of the lot, and the one whose price makes the most sense once you understand it.
The scent is not in the flower. It is in the rhizome, the underground stem, and it is not there when you dig it up. The rhizomes are peeled, sliced, dried and then stored for at least three years, during which oxidation slowly converts odourless compounds in the root into irones, which are what you actually smell.
Then the yields. One tonne of dried powder gives roughly two kilos of orris butter. Global production is measured in tonnes a year, not thousands. Prices run tens of thousands of dollars a kilo depending on irone content, and orris absolute refined to high irone concentration can run several times higher again.
From planting to finished butter is six to eight years, the longest production cycle of any natural perfumery material.
It smells like cold earth, violets, face powder, and something faintly like raw carrot. People either find it the most beautiful thing in perfumery or they find it smells like a handbag. We are in the first group and we understand the second.
Violet, and the ionones
There is almost no violet in violet fragrances, because violet flowers yield essentially nothing usable.
In 1893 Ferdinand Tiemann and Paul Krüger, working out what made orris smell the way it does, found a related family of compounds that were simpler and far cheaper to make. They called them ionones, and they smell of violets. Violet became affordable overnight.
Ionones also have a quirk: they blank out fast. Many people find violet notes seem to vanish and return, which is olfactory adaptation rather than the fragrance fading.
The woods and roots
Oud, or agarwood
The most expensive wood in the world, and it is only valuable when the tree is sick.
Oud comes from trees of the genus Aquilaria. Healthy Aquilaria wood is pale and worthless. When the tree is wounded and infected by fungus, it defends itself by producing a dark aromatic resin, and that resin, laid down over years, is agarwood.
Aquilaria malaccensis has been listed on CITES Appendix II since 1995, and the rest of the agarwood-producing genera followed in 2004, requiring export permits and proof that trade will not threaten survival.
Price depends entirely on grade and runs from around a hundred dollars a kilo for low grades into six figures for the rarest. That range is the answer to the common claim that oud is the most expensive material in the world. Top-grade agarwood is in that conversation. Most oud sold is not top grade, and a very large share of what is labelled oud in Western designer fragrances contains no agarwood at all, only synthetics built to suggest it.
Real oud smells medicinal, barnyard, sweet and woody at once, and it is challenging. Most people's first oud is not oud.
Sandalwood
Warm, creamy, milky, soft. The most comforting wood there is, and one of the clearest cases of a material nearly destroyed by demand.
Santalum album, Indian or Mysore sandalwood, is the historic standard, with santalol content around 70 to 90%. Legal production in Karnataka fell from 3,000 tonnes in 1978 to about 20 tonnes by 2002, a drop of over 99% in under twenty-five years. The Indian state owns sandalwood trees even on private land, though Karnataka loosened trade rules in 2022.
The supply moved. Australia grows Santalum album on a large scale in Western Australia, with millions of trees planted, and Santalum spicatum, Australian sandalwood, is a different species with santalol around 20 to 40% and a drier, sharper smell. Prices reflect it: Indian sandalwood typically runs many times the cost of Australian, kilo for kilo.
If a mass-market fragrance lists sandalwood, it is almost certainly a synthetic sandalwood material. Those have become very good.
Cedar
Two unrelated things share the name. Virginian cedar is a juniper and smells like pencil shavings, dry and clean. Atlas cedar is a true cedar, softer and more resinous.
Cedar is structural. It props up other things and rarely leads.
Vetiver
A grass, and the scent is in the roots. Haiti is the largest producer, around half of world output, with Java and India the other main sources. Givaudan has run a certified fair-trade sourcing programme with a cooperative of around 250 Haitian producers since 2012.
Haitian vetiver is smoky and earthy. Javanese is harsher. Bourbon, from Réunion, is the one perfumers speak about with reverence.
It smells like damp soil, grapefruit rind and smoke, and it is the backbone of Guerlain Vétiver, and, alongside cedar, patchouli and benzoin, of Terre d'Hermès too.
Patchouli
The most misunderstood material in perfumery, because a generation met it as a head-shop oil rather than as a perfumery material.
The leaves are dried and fermented before distillation, which is unusual, and good patchouli improves with age in a way almost nothing else does. Properly used it is dark, sweet, earthy and slightly chocolatey, and it is doing structural work in a huge share of modern fragrances, including Angel. If you think you hate patchouli, you hate cheap patchouli.
The mosses and resins
Oakmoss
A lichen, not a moss, harvested from oak trees. It smells like a forest floor: damp, bitter, green, inky. It is the material that makes a chypre a chypre.
It is also the clearest case of regulation with a real reason behind it. Oakmoss contains atranol and chloroatranol, potent contact allergens implicated in sensitisation in roughly 1 to 3% of the general population. It is restricted, not banned, and it is now commonly replaced with synthetics such as Evernyl.
Our guide to why fragrances change handles what that did to the classics. The short version: the chypres in the old books are not the chypres on the shelf.
Labdanum
A sticky resin from the Mediterranean rockrose. Warm, leathery, sweet, faintly animalic. It is the backbone of the amber accord and it also does a lot of the work in leather compositions.
Benzoin
A balsamic resin from the Styrax tree. Sweet, vanilla-like, powdery, comforting. It is the soft edge of most amber accords.
Amber
Read this one carefully, because the word is doing three jobs and confusing all of them.
Amber the gemstone is fossilised tree resin and it has no scent. It is not a perfume material and never has been.
Amber in perfumery is an accord, not an ingredient: typically labdanum plus benzoin plus vanillin and often coumarin and patchouli. Coty's Ambre Antique in 1905 is an early landmark of it.
And the name probably comes from ambergris, which is a different substance again, from a whale, covered below.
Three meanings, one word. When a bottle says amber, it means the accord.
The sweet materials
Tonka and coumarin
Tonka beans are the seeds of a South American tree and they are full of coumarin, which smells of hay, almond, vanilla and warm skin.
Coumarin is the molecule that started modern perfumery. William Henry Perkin synthesised it in 1868, the first synthesis of a naturally occurring aromatic compound, and it reached the market by 1876. Six years later Paul Parquet built Fougère Royale around it and invented a family.
Vanilla
The second most expensive spice in the world, and the story is better than the smell.
Vanilla orchids outside Mexico produce no fruit, because their pollinator does not live there. The problem was solved in 1841 on Réunion by Edmond Albius, an enslaved boy of about twelve, who worked out that a thin membrane separated the male and female parts of the flower and that lifting it with a sliver of bamboo let them meet. Every vanilla bean in the world is still pollinated by hand using his method. He died poor.
The price is violent. Madagascar grows about 80% of the world's vanilla, and between 2012 and 2016 a kilo rose roughly eightfold. Then Cyclone Enawo hit in March 2017 and pushed it higher again.
Which is why almost none of the vanilla you smell is vanilla. Around 85 to 88% of the world's vanilla flavour is synthetic vanillin, much of it from petrochemical guaiacol, and less than 1% comes from actual vanilla beans.
There is not enough vanilla on earth for it to be anything but arithmetic, not a fragrance industry scandal.
Ethyl maltol
Smells like candy floss and burnt sugar, and that is not a metaphor: it is the described character of the molecule, first patented by Pfizer in 1969. It is the material that makes a gourmand a gourmand, and Angel in 1992 is where it went from a flavour material to a perfume idea. Used heavily it is unmistakable and slightly cheap-smelling, which is part of why Angel remains divisive more than three decades on.
The animal materials
Musk
Every musk you can buy is synthetic. That is the fact to lead with, because the folklore says otherwise.
Natural musk came from the gland of the male musk deer, and the deer had to be killed. It took the glands of roughly 160 animals to produce one kilo of musk. In 1979 CITES listed all musk deer species, the most threatened on Appendix I with a complete commercial ban and the rest on Appendix II under strict regulation.
The replacement arrived long before the ban, and by accident. Albert Baur was working on explosives when he produced a strongly musky compound in 1888. The nitromusks came out of TNT chemistry. Modern musks fall into four groups: nitromusks, polycyclic, macrocyclic and linear or alicyclic. The macrocyclics, which are closest to the natural material and much better tolerated, became commercially viable in the late 1990s.
The Body Shop's White Musk in 1981 sold synthetic musk as a virtue rather than a compromise, and changed what the public thought the word meant.
Musk is also the family most affected by specific anosmia. If a musky fragrance smells like nothing to you, your nose may simply not have the receptor. That is covered in our guide to how smell actually works, and it is not a defect.
Ambergris and ambroxan
The best story in perfumery, and almost everything commonly said about it is slightly wrong.
Ambergris is not whale vomit. It forms in the intestine of sperm whales, in only about 1 to 5% of them, when the indigestible beaks of squid bind together into a mass that the whale cannot pass. It is closer to a gallstone than to sick, and closer to faeces than to either. It is expelled or released at death, floats for years, and cures in seawater and sunlight into something that smells of salt, tobacco, skin and warm paper. Fresh, it smells appalling.
The legal position is complicated and you should not take a beach find home without checking. Here is the structure of it.
CITES excludes ambergris. Under Resolution Conf. 9.6, urine, faeces and white ambergris naturally excreted by a sperm whale are treated as waste products rather than parts or derivatives of a protected species, and so fall outside the Convention, even though the sperm whale itself is on Appendix I.
National law then diverges, and that is where people get caught. The United States bans possession and trade under the Endangered Species Act. Australia bans it too.
Canada we could not settle, and we would rather tell you that than guess. Canada implements CITES through the Wild Animal and Plant Protection and Regulation of International and Interprovincial Trade Act, which turns on whether something is a CITES specimen, and by the Resolution above ambergris is not one. That suggests the CITES route does not prohibit it here. But the American and Australian bans do not come from CITES either; they come from separate domestic endangered-species law, and Canada has its own in the Species at Risk Act. We found no Canadian source stating a position on ambergris specifically.
So: not established, and if you are ever holding a lump of it on a Canadian beach, get advice before you sell it rather than after. Prices vary enormously by quality, commonly reported in the tens of thousands of dollars per kilo, with some claims running far higher.
Almost every "ambergris" note you smell is ambroxan. Ambroxan is made from sclareol, which comes from clary sage, and it has largely replaced ambergris since the 1990s on both cost and supply grounds. The molecule itself is older than that: Ambrox was discovered by Stoll at Firmenich in 1949, found by breaking down an expensive natural material to identify the trace compound responsible. It is also the loud, mineral, slightly metallic material doing much of the work in Dior Sauvage and a thousand things like it.
Leather
Not a material, an effect. Leather notes are constructed, historically from birch tar and now more often from synthetics.
The defining one is isobutyl quinoline, synthesised by Darzens in 1908, which smells bitter, tarry and green. It sat around for thirty-six years until Germaine Cellier used it at around 1% in Bandit in 1944 and terrified everybody.
The spices
Saffron
Expensive, and in perfumery usually a synthetic reconstruction rather than the spice. It smells leathery, slightly medicinal, warm and a bit like hay. It is the signature of a large share of modern Middle Eastern and oud-adjacent fragrances.
Pink pepper
Not pepper, but the berry of a South American tree: bright, rosy, slightly fizzy. It became fashionable in the 2000s and is now in an enormous number of openings because it adds sparkle without the sourness of citrus.
Cardamom
Green, cool, slightly soapy, faintly eucalyptus. One of the few spices that reads fresh rather than warm: that is why it turns up in modern masculines that want to be spicy without being heavy.
Tobacco
Rarely actual tobacco. The note is usually built to suggest dried leaf, honey and hay. Warm, sweet and slightly dusty. It sits very close to coumarin and tonka in the way it behaves.
The modern synthetics
These four have shaped what fragrance smells like more than any natural material of the last century.
Hedione
Faint on its own: a thin jasmine, some air. What it does is make everything around it brighter and more transparent. The molecule is a hydrogenated version of methyl jasmonate, which occurs in real jasmine, and it was made in 1960.
Eau Sauvage in 1966 was the first fragrance to use it, and once Roudnitska demonstrated what it was for, it went into an enormous share of everything since.
Hydroxycitronellal
Synthesised in 1905, and the material that makes lily of the valley possible. The flower yields nothing when you try to extract it, so muguet is always built, and this is the molecule it is built on.
It is now IFRA-restricted by product category, at around 2% in fine fragrance and less in leave-on skin products, where older formulas used far more. That is the specific reason Diorissimo today is not Diorissimo of 1956.
Calone
Synthesised at Pfizer in 1966 by J. J. Beereboom, D. P. Cameron and C. R. Stephens. The compound numbered 1951 was the one that smelled of watermelon and sea air.
And nobody was looking for it. The Pfizer team were preparing oxygenated derivatives of anxiolytic benzodiazepines, which is to say they were doing tranquilliser chemistry, and the smell of the sea fell out of it. The nineties smelled the way they did because of a side reaction in a drug programme.
It sat unused commercially until the patent lapsed, and Aramis New West for Him in 1988 used it at a level nobody had used before and started the aquatic decade, the same shift that later carried Cool Water, Acqua di Gio and Bvlgari Aqva through the years that followed.
Iso E super
Made in two steps from myrcene in 1973, and it carries cedar and vetiver with mineral and ambery facets. Velvety, woody, transparent, and hard to describe because it behaves more like a lighting effect than a smell. Geza Schoen, who built a whole brand on it, calls it "less an aroma than an effect which lends an indefinable radiance to the wearer."
In 2006 Schoen released Molecule 01, which is Iso E Super and nothing else, and it sold extremely well.
Roughly 20% of people are mildly anosmic to certain isomers of Iso E Super, which explains a great many arguments online. One in five people cannot properly smell one of the most widely used materials in modern perfumery. When somebody says a fragrance projects beautifully and somebody else says it disappeared in ten minutes, they may both be telling the truth. Terre d'Hermès carries it, among a great deal else.
A note on this guide's limits
Prices move, harvests fail, and regulations change. The material facts, what a thing is and where it comes from, will hold. Treat the numbers here as a snapshot from around mid-2026.
How to actually use this guide
Nobody reads an encyclopedia front to back, and you should not start.
Use it as a lookup while you smell. A fragrance you own says orris in the note list. Come here, read the entry, then go back and try to find the thing the entry describes. That loop, word to nose and back, is the entire method of our guide to training your nose, and this guide is its dictionary.
Pick five materials and learn them properly before touching the rest. We would take bergamot (Fougère Royale carries it, and so does Eau Sauvage), coumarin (Fougère Royale again, since it is the reason that fragrance exists), orris (Guerlain's L'Heure Bleue), oakmoss (Mitsouko, then read our guide to why fragrances change straight afterwards) and Iso E Super (Terre d'Hermès). Five materials, and between them you can account for a large share of what is in front of you in a shop.
Then use it as a bluff detector. When a description promises real Mysore sandalwood, or ambergris, or a rose you are told was picked at dawn, the entries here tell you what those things cost and whether they still exist at that price. Most of the time the answer is no, and the answer is more interesting than the claim.
One rule we would hold to. If you cannot name three materials in a fragrance you have worn for a year, you are not smelling it, you are recognising it. Those are different, and only one of them gets better with practice.
Part of Nose to Bottle, our complete fragrance manual. Read it in order here.