The Scent Atlas

Chapter I

Smelling is easy; saying is hard

Ask a friend to describe the colour of a ripe mango and they will reach for a ready word: yellow, orange, saffron. Ask them to describe how it smells and something different happens. They will usually name another thing — “like mango, obviously”, “a bit like turpentine?”, “sweet, sort of tropical” — or tell you whether they like it. They do not lack a nose. They lack a shelf of shared words.

That gap is the subject of this atlas. Evidence When researchers asked English speakers to name everyday smells, responses were long, hesitant and mostly built from sources (“like banana”) or judgements (“that’s disgusting”). Speakers of Jahai, a language of the Malay Peninsula with a set of abstract smell words, named the same smells about as easily and consistently as they named colours (Majid & Burenhult, 2014). The study is small and specific, but it quietly removes an old excuse: difficulty in naming smells is at least partly a matter of language and practice, not simply biology.

Interpretation If description is a skill, it can be practised — and that matters more than it sounds. Perfume is sold almost entirely in words: on boxes, in advertisements, by counter staff, in reviews. A person who can describe what they smell can test those words against their own experience. A person who cannot is left to borrow someone else’s story about a bottle, which is exactly what the story was written for.

So the atlas has three modest goals. First, give you a working vocabulary that does not require French or prestige. Second, explain the conventions perfumers and writers use — families, the top/heart/base pyramid, notes and accords — as tools, with their limits showing. Third, keep a clear line between three different things that marketing tends to blur: what you perceive, what someone says it evokes, and what is physically in the liquid.

Chapter II

Families are maps drawn by people

Established framework Perfumers, retailers and critics sort fragrances into families — citrus, floral, woody, amber (once commonly called “oriental”), aromatic, fougère, chypre, gourmand and so on. These classifications are genuinely useful: they help a shop assistant suggest something similar, and they help you say “I usually like the woody ones” without needing a single technical word.

But there is no single official family tree. Different houses, writers and retailers draw the borders differently, rename families over time and argue about where things belong. A lavender perfume can be “aromatic”, “fougère” or “floral” depending on who is filing it. The twelve exhibits above use only five plain-language groups, chosen for teaching, not because they are the true structure of smell.

Interpretation The best way to use families is as a first guess, never as a verdict. If you love one “woody” perfume and dislike another, the family name did its job of getting you to the right shelf — and then stopped being useful. The interesting description begins where the family label ends.

Chapter III

Top, heart, base: a useful simplification

Established framework Perfume marketing almost always describes a fragrance as a pyramid: top notes you notice first, heart notes that form the main character, base notes that linger longest. The physical idea behind it is real. Different aroma materials evaporate at different rates — this is volatility — so the balance of what reaches your nose changes over minutes and hours. Small, light citrus molecules tend to fade quickly; heavier woods, resins and musks tend to persist.

As a teaching device, the pyramid is excellent. It tells a beginner something true and important: a perfume is not one smell, it is a smell that changes. Smell a perfume on a paper strip at the counter, again after twenty minutes, again after two hours, and you are learning more than any advertisement can tell you.

As a description of what physically happens, it is too tidy. Materials do not queue politely in three tiers. They evaporate together from the start, at different rates, so a “base” material is present in the first second and a “top” material can linger as a trace. Many modern perfumes are deliberately built to feel “linear”, changing little. Skin, temperature, the amount applied and even fabric all shift the picture.

A worked example: reading a box honestly

Illustration Imagine a box that lists “Top: bergamot, pink pepper. Heart: rose, saffron. Base: oud, vanilla.” A careful reader can take three honest things from it: the maker wants the opening to feel bright and spicy; they want rose to be the main character; and they want it to finish warm and dark. What the reader cannot conclude is which materials are in the bottle, in what amounts, whether the “oud” comes from agarwood at all, or how long any stage will last on their skin. The pyramid on a box is closer to a menu description than a recipe.

Chapter IV

Note, ingredient, accord

Three words do most of the work in fragrance writing, and they are constantly confused. Keeping them apart is the single most useful habit this atlas can offer.

Note

A description of an impression: “a rain note”, “a fig note”. It says what something smells like to the person describing it. A note can exist without any matching ingredient.

Ingredient (material)

A physical substance in the formula: a natural extract such as an essential oil or absolute, or a single aroma chemical. Ingredients have names, safety data and regulatory status. Notes do not.

Accord

A combination of materials that together create one impression none of them gives alone — the way several colours of paint can mix into a new one. A “fig” or “rain” note is usually an accord.

Worked example: the “sea breeze” that contains no sea

Illustration A fictional perfume lists a “sea breeze note”. Ask the three questions in order. What is the note? An impression of cool, salty, open air. What are the ingredients? Unknown from the label — possibly a few synthetic materials that read as “marine”, plus citrus and musk. Is there an accord? Almost certainly; nothing from the sea is distilled for this. None of this makes the perfume dishonest. The dishonesty would begin only if the marketing claimed the bottle contained “real ocean minerals” or did something to your body because of them.

Interpretation Notice that “natural” and “synthetic” are about ingredients, while “fresh” and “warm” are about notes. Mixing the two vocabularies — “this smells natural” — is common and understandable, but it is a claim about a feeling, not about the formula.

Chapter V

Memory, colour and the stories we smell with

Everyone has a smell that ambushes them: a hair oil, a particular soap, the inside of an old school bus. Evidence Researchers have tested this “Proustian” experience. In one well-known study, memories brought back by an odour were rated as more emotional than the same memories cued by a picture, and people felt more “brought back” to the original moment (Herz & Schooler, 2002). That supports something many people feel — but it does not show that smell is always the strongest memory cue for everyone, or that a given perfume will reliably trigger a given emotion.

Evidence Smell description is also easily steered by the other senses. In a widely cited experiment, a white wine dyed red with a flavourless colouring was described by tasters in the vocabulary normally used for red wines (Morrot, Brochet & Dubourdieu, 2001). Later researchers have debated how far that finding generalises, but the lesson for perfume is direct: a pink bottle, a word like “velvet” or a photograph of a beach starts shaping your description before the liquid reaches your nose.

Worked example: “It smells like my grandmother”

Illustration Suppose someone smells a powdery rose perfume and says it smells like their grandmother. That is a real and valuable description — it is a source description, the most natural kind. But it carries little information for anyone else, because everyone’s grandmother smelled different. A second sentence turns it into shared language: “powdery, soft, a little soapy, like face powder and rose water”. The memory stays yours; the qualities become something a friend can check.

Evidence Biology adds another layer of difference. A single gene variant in one human odour receptor was found to change whether people perceived a steroid odour called androstenone as unpleasant, mildly pleasant, or barely at all (Keller et al., 2007). That study concerns one compound, not perfume in general, but it is a clear demonstration that two honest noses can disagree about the same molecule.

Hypothesis This atlas suspects that practising quality words (warm, dry, green, powdery) alongside source words (like pencil, like rain) makes people better at noticing when a description is coming from the label rather than the liquid. That is plausible, but it has not been tested here.

Chapter VI

Saying it plainly

Fragrance writing has a prestige vocabulary — sillage, dry-down, chypre, opulent, skin scent — and some of it is genuinely useful shorthand. But you do not need it to say something precise. Here is a ladder that works with any smell, from coffee to cologne. Each rung adds information.

1. Reaction. Do I like it? How much? (Honest, but tells others least.)
2. Source. What does it remind me of? “Like pencil shavings.” “Like a new raincoat.”
3. Qualities. Temperature (warm/cool), weight (light/heavy), texture (dry, creamy, powdery, sharp), brightness, sweetness, intensity.
4. Change. What happens after ten minutes? An hour? Did something disappear or appear?
5. Context. Where was I, what did I just eat or see, what did the label tell me before I smelled it?

Before and after

The plain version is longer, but every phrase is checkable. “Mysterious” and “sensual” describe a mood the writer wants you to feel; “bitter”, “leathery” and “from across the room” describe what they actually noticed.

Chapter VII

Honest bottles: marketing and safety

Evocative stories are part of the pleasure of perfume, and nothing in this atlas asks for them to disappear. A name like “Monsoon Library” is a poem, not a contract. The ethical line is crossed when a story starts making factual claims the product cannot support.

Claims worth doubting

• Health and mood cures. “Relieves anxiety”, “heals insomnia”, “boosts immunity”. Enjoying a smell is not the same as being treated by it. Treat therapeutic promises from fragrance sellers with strong scepticism.
• “Pheromone” attraction. Whether humans have pheromones in the animal sense is still contested, and the androstenone research above shows that people do not even agree on how some of these compounds smell.
• “Chemical-free”. Everything in a bottle, including water and rose oil, is made of chemicals. The phrase signals a mood, not a fact.
• “Natural means safe.” Some natural materials are among the most restricted in perfumery because of allergy or skin-sensitisation concerns.
• Invented precision. Exact longevity hours, “scientifically proven” attraction, or compliance badges without a named, checkable source.

Safety, briefly and firmly

Established framework The International Fragrance Association (IFRA) publishes the IFRA Standards, an industry self-regulatory programme that bans, limits or sets purity criteria for certain fragrance materials, based on safety assessments by the Research Institute for Fragrance Materials (RIFM) reviewed by an independent expert panel. Two cautions: IFRA Standards sit alongside national laws rather than replacing them, and a seller’s mention of “IFRA” is not by itself proof that a particular product has been assessed. This atlas makes no compliance claims about anything.

Interpretation A historical, illustrative note: in many Indian bazaars, attars and perfume oils have long been sold with a dab on the wrist and a story about their origin. The storytelling is part of a living craft worth respecting. The same respect means asking, kindly, what a material actually is — and being suspicious of any seller who cannot say.

Chapter VIII

Objections, limits and misconceptions

The strongest objection: “Words ruin smell.”

A serious critic might say: smell is pre-verbal and private. Teaching people to label it trains them to experience the label instead of the smell — to sniff a perfume, recognise “bergamot” from the box, and stop paying attention. Worse, a shared vocabulary quietly turns one culture’s categories (European perfume families, English adjectives) into the “correct” way to perceive.

That objection is partly right, and this atlas cannot fully answer it. Labels can replace attention; the colour-and-wine study shows how easily a cue can redirect a description. Two responses, though. First, people already use words for smell — mostly borrowed from advertising. The choice is not between language and pure experience but between someone else’s language and your own. Second, the Jahai findings suggest that a richer vocabulary can sit alongside fluent, attentive smelling rather than replacing it. The practical safeguard is order: smell first, then read the label.

Limits of this atlas

• The families, tags and vignettes are one writer’s choices, shaped largely by English and by Indian everyday life. Another culture would draw the drawers differently.
• The exhibits describe notes, not real materials, and the instrument manipulates words, not molecules.
• The research cited is a small, curated sample. Each study has its own population, method and limits, described on the source shelf below.
• People with reduced or altered smell — temporarily or permanently — may find many descriptions here do not match their experience. Their experience is not a lesser version of anyone else’s.

Common misconceptions

“Humans have a weak sense of smell.”

This old belief has been challenged as more cultural assumption than measurement (McGann, 2017). Humans differ from dogs, but “weak” is not a safe summary.

“Humans can smell exactly one trillion odours.”

A 2014 study proposed at least a trillion; a 2015 reanalysis argued the method could not support that number. The honest answer is that nobody knows the count (Gerkin & Castro, 2015).

“A top note disappears after fifteen minutes.”

Evaporation is gradual and overlapping, and depends on the material, the amount and the conditions. Fixed timings are rules of thumb at best.

“If it says rose, there is rose in it.”

A note is a description. It may or may not correspond to a natural rose material.

“Experts smell the truth; beginners guess.”

Experts are better trained at noticing and naming, but they are also exposed to expectations and labels. Training sharpens attention; it does not remove context.

Chapter IX

An open question

If smell vocabulary is partly learned, then the words we inherit decide which differences we notice and which ones blur together. English speakers inherit a vocabulary built heavily by perfume marketing. Jahai speakers inherit one built from forests, foods and animals. Many Indian households inherit a vocabulary of attars, spices, rain on earth and temple flowers that rarely appears on a perfume box at all.

So here is the question this library leaves on the table: whose words should we learn to smell with — and what would we start noticing if the vocabulary on the box came from our own kitchens, streets and weather instead?

There is no answer key. The journal is a place to start collecting yours.