Smell has no natural scale like the wavelength of colour or the frequency of sound. Each period found its own way to divide it: by medicinal use, by music, by type odours, by geometry, by numeric code, by molecular shape, then by the perfume market. This page outlines the main classifications by checked publication year.
| · Year | · Who | · Work | · Classes |
|---|---|---|---|
| 1752 · 1756 | Carl Linnaeus | Odores medicamentorum Amoenitates Academicae dissertation; reprinted in Amoenitates Academicae III | 7 classes |
| 1857 | G. W. Septimus Piesse | The Art of Perfumeryoctave of odours, later the odophone | ≈ 50 odours on a musical stave |
| 1865 | Eugène Rimmel | The Book of Perfumesclasses by type odour | 18 classes |
| 1895 | Hendrik Zwaardemaker | Die Physiologie des Geruchsphysiology of smell | 9 classes |
| 1916 | Hans Henning | Der Geruchsmell prism | 6 corners |
| 1927 | E. C. Crocker, L. F. Henderson | four qualities, scale 0 to 8 | 4 qualities |
| 1951 | Paul Jellinek | odour effects diagram | 2 axes |
| 1952 | John E. Amoore | stereochemical hypothesis | 7 primary odours (1963) |
| 1984 | Société Française des Parfumeurs | first classification, women's fragrances | 5 families |
| 1984 | Michael Edwards | The Fragrance Manualretail fragrance guide | |
| 1990 | Société Française des Parfumeurs | 2nd edition, adds citrus and woody | 7 families |
| 1991 | Linda Buck, Richard Axel | Cellodorant receptor gene family | about 400 receptors in humans |
| 1992 | Michael Edwards | fragrance wheel first printed | fragrance wheel |
| 2004 | Buck, Axel | Nobel Prize in Physiology or Medicine |
1752 · 1756
In 1752 Carl Linnaeus presided over the dissertation Odores medicamentorum at Uppsala, defended by Andreas Wåhlin. It was reprinted in 1756 in volume 3 of Amoenitates Academicae. Its aim was to infer the use of a remedy from its odour, in line with the medicine of the time.
Linnaeus set out seven classes: aromatic, fragrant, ambrosial, alliaceous, hircine, foul and nauseous. The first three are pleasant. It is among the earliest systematic odour classifications still cited.
1857
Septimus Piesse, an English chemist and perfumer, published The Art of Perfumery; the 1857 Philadelphia edition by Lindsay & Blakiston is held on Project Gutenberg. Piesse compares odours with sounds: some odours agree like keys within one octave. Almond, heliotrope, vanilla and orange blossom form one group; citron, lemon, orange peel and verbena a higher octave.
Later editions develop the idea into the "odophone": about fifty odours placed on a musical staff, sharp odours high, heavy odours low. It is a blending metaphor, not a measurement. The perfumer's word "note" belongs to the same line of thought.
1865
Eugène Rimmel published The Book of Perfumes in London in 1865; a French translation followed in 1870. Rimmel ranked only pleasant odours. He borrowed the idea of primary colours: each class has a type odour, and other odours are grouped by resemblance to it.
Rimmel's table has 18 classes. Examples of type odours: rose (with geranium, sweetbriar), vanilla (with balsam of Peru, Tolu, benzoin, tonka), clove, a citrus group (bergamot, orange, cedrat), lavender (spike, thyme, marjoram), mint, anise (star anise, caraway, dill, coriander), almond, musk, fruit. Rimmel notes that some odours, such as wintergreen, fit no class.
1895
Dutch physiologist Hendrik Zwaardemaker published Die Physiologie des Geruchs in 1895. He extended Linnaeus's framework to nine classes and added subclasses:
Dutch physiologist Hendrik Zwaardemaker published Die Physiologie des Geruchs in 1895. He extended Linnaeus' frame to nine classes with subclasses: ethereal, aromatic, balsamic (fragrant), ambrosial, alliaceous, empyreumatic (burnt), caprylic (hircine), repulsive, nauseous. Zwaardemaker also built an early olfactometer to measure odour thresholds.
1916
German psychologist Hans Henning published Der Geruch in 1916. He placed six primary odours at the corners of a triangular prism: flowery, fruity (ethereal), spicy, resinous, burnt, putrid. Every other odour lies on the prism's surface as a continuous transition between corners.
The idea of continuous transition proved lasting. The method was criticised from 1927 for relying on few subjects and lacking statistical testing.
1927
Chemical engineer Ernest C. Crocker and chemist Lloyd F. Henderson proposed in 1927 that any odour can be described by four qualities: fragrant, acid, burnt, caprylic. Each is scored from 0 to 8, giving a four-digit code. Vanillin is 7122; acetic acid is 3803.
The system was easy to use in industry and was taught for decades, though agreement between assessors was limited.
1951
Austrian perfumer Paul Jellinek published Die psychologischen Grundlagen der Parfümerie in 1951. His son, J. Stephan Jellinek, edited the English edition The Psychological Basis of Perfumery in 1997.
Jellinek's diagram sorts odours by perceived effect rather than source, on two opposed axes: animalic and erogenous against fresh; sweet and heavy (he called it narcotic) against bitter and sharp (stimulating). It served to balance a formula. A 2009 study of perfumers' odour space found dimensions close to this diagram.
1952 · 1963
British biochemist John E. Amoore published "The stereochemical specificities of human olfactory receptors" in Perfumery & Essential Oil Record in 1952. The hypothesis: odour follows molecular shape and size, which fit receptor sites like a key in a lock.
His 1963 Nature paper "Stereochemical theory of olfaction" named seven primary odours: camphoraceous, musky, floral, pepperminty, ethereal, pungent, putrid. Amoore later studied specific anosmias to look for more primaries. The receptor idea was partly borne out by the 1991 discovery, though the number of receptors is far larger than seven.
1984 · 1990
The Technical Commission of the Société Française des Parfumeurs (then the Société Technique des Parfumeurs de France) drew up its first fragrance classification in 1983, published in 1984. The team was Raymond Chaillan, Jean Kerléo, Raymond Pouliquen and Guy Robert. The 1984 edition covered women's fragrances only, in five families: floral, chypre, fougère, amber, leather.
The 1990 edition added hesperidic (citrus) and woody, making seven families, to cover the growing eau de toilette market. Further revisions followed in 1998 and 2001. According to the SFP, the current version has seven families and 47 subfamilies across women's, men's and unisex fragrances. The library's family pages follow this frame: see Fragrance families.
1984 · 1992
In 1984 Michael Edwards, then working in Sydney, self-published The Fragrance Manual, a guide to help retailers suggest perfumes. The 1992 edition was the first to include a fragrance wheel: families arranged in a circle, each blending into its neighbours. The guide was renamed Fragrances of the World in 2000, and the wheel has been revised several times since.
The wheel suits retail because it shows a path from a liked scent to its neighbours. The library has its own version at Scent wheel and the Families & Facets table.
1991 · 2004
In 1991 Linda Buck and Richard Axel reported in Cell a large gene family encoding odorant receptors in the rat olfactory epithelium. Each receptor responds to a group of molecules; each molecule activates a combination of receptors. Odour is coded combinatorially, not by a few primaries. They shared the 2004 Nobel Prize in Physiology or Medicine.
Humans carry about 400 functional odorant receptor genes. No table of primaries maps neatly onto the biology. A 2025 review in Chemical Senses concludes that no universal relation between odour quality and physical or chemical parameters has been found. Classifications keep their value as a shared vocabulary for people who work with scent.
These systems fall into two lines. The scientific line (Linnaeus, Zwaardemaker, Henning, Crocker and Henderson, Amoore, Buck and Axel) looks for the structure of smell itself. The perfumery line (Piesse, Rimmel, Jellinek, SFP, Edwards) arranges materials and perfumes for work and sale. The library's fragrance families belong to the second line.
Years in the table are publication years checked against the printed work, a library catalogue or the issuing body. Where sources disagree, the sources section says so.
Fragrance families on the SFP frame. Families & Facets. Scent wheel. Volatility chart.
Linnaeus: many texts give 1756, the reprint year in Amoenitates Academicae; the dissertation dates from 1752. Rimmel: Doty's review gives 1867; the London edition and press reviews carry 1865. Edwards: many pages date the wheel to 1983; checked sources show the manual in 1984 and the wheel first printed in 1992. Jellinek: one source ties the diagram to a 1949 book; this page uses 1951, the checked German edition.