Hans Henning (1885 to 1946) was a German psychologist. In 1916, as a Privatdozent at the University of Frankfurt, he published Der Geruch with Johann Ambrosius Barth in Leipzig. The front matter notes that parts had appeared earlier in the Zeitschrift für Psychologie, volumes 73 to 76. In the same year he qualified for habilitation at Frankfurt with a thesis on smell.
The main history page gives the six corners and the criticism. This page goes into the geometry of the prism, how Henning ran his experiments, and why later psychologists could not reproduce his results.
| Author | Hans Henning (1885–1946) |
| Work | Der Geruch, Leipzig: J. A. Barth, 1916 |
| Six primaries | würzig, blumig, fruchtig, harzig, brenzlich, faulig |
| Model | surface of a regular triangular prism |
| Odorants | 415 in the main series |
| Observers | about 15 named adults, a few children; group tests with 46 students |
Henning states that he reversed the method of earlier authors. During the experiments he did not open a chemistry book; he consulted only perception. Observers were given odours without knowing what they were and arranged them in series by similarity. From those series six points emerged, which he called primary odours: the points where the direction of similarity turns.
The series show continuous transitions. From clove, through laurel, thyme and lavender, then vanilla and vanillin, observers moved step by step from spice to flower. From rose, through grapefruit, orange blossom, neroli and petitgrain, the floral turns fruity. From turpentine, through pine and juniper, towards acetone at the end of the fruity series.
Examples in the table are substances Henning's observers placed in each corner's series.
| Corner | German | Examples in the series | In the library |
|---|---|---|---|
| Spicy | würzig | fennel, marjoram, cardamom, nutmeg, anise, pepper, ginger, cinnamon, cassia, clove, caraway | nutmeg, cardamom, ginger, clove |
| Hoa / Flowery | blumig | heliotropin, tonka, coumarin, geranium, jasmine, opopanax, ylang-ylang, mimosa | jasmine, ylang-ylang, mimosa, opopanax |
| Fruity | fruchtig | orange, bitter orange, lime, lemon, citronella, lemongrass, citral, bergamot; ethyl acetate, acetone | orange, lemon, lemongrass, bergamot |
| Resinous | harzig | incense, turpentine, pinene, mastic, copaiba, olibanum, cedarwood, eucalyptus, cajuput, juniper berry, rosemary, myrrh | olibanum, mastic, copaiba, juniper berry, myrrh |
| Putrid | faulig | carbon disulphide, hydrogen sulphide; towards spicy: dill, chervil, parsley, celery, asafoetida, garlic, onion, horseradish | asafoetida, garlic, onion |
| Burnt | brenzlich | tar, pyridine; roasted coffee between spicy and burnt; burning mastic between burnt and resinous | birch tar |
The prism has two triangular faces. One triangle joins flowery, fruity and putrid; the other joins spicy, resinous and burnt. The three side edges join flowery to spicy, fruity to resinous, putrid to burnt. That gives three rectangular faces: flowery, fruity, resinous, spicy; putrid, burnt, spicy, flowery; and putrid, burnt, resinous, fruity.
Henning chose the shape from the crossing of the two diagonals on each rectangle. Going from resinous to flowery, observers pass an odour they also meet going from spicy to fruity; that odour must resemble all four corners. He placed thujone here, with wormwood and tansy, and juniper berry nearby. On the second face, between putrid and spicy, he placed celery-like lactones and the apiole of parsley. On the third, some observers placed grapefruit that had begun to spoil, and a few synthetic musks.
Simple odours lie only on the edges and surface; the interior is kept for mixtures. Henning considered and rejected two other shapes. An octahedron, like the colour solid, would require burnt and putrid to be opposite poles like white and black, but his observers found many substances between them. A flattened pentahedron would set the crossing points too far off centre for observers to sense all four resemblances.
The opening chapter lists the observers by name. They include the philosopher Hans Cornelius, the psychologist Adhémar Gelb, Henning's wife, a former wine merchant and his two daughters, the institute's mechanic, and some children of about 11. Henning says outright that without his wife's acuity the results would hardly have been possible. Some disputed points were checked in group tests with 46 students.
The main odorant set held 415 substances; he judged that so many odours resemble one another that not every substance was needed in every series. To avoid missing an odour, the group also made four visits to the zoo, ten to the flower houses of the Palmengarten and one to a chemical laboratory. He notes that on the first day observers were lost enough to take roasted coffee for violet, but by the second day the series were usable.
After the psychology, Henning added a chemical part: he held that all substances in one odour class share a type of bond within the molecule. Later chemistry did not bear this out.
American psychologists tested the prism soon after. MacDonald published an experimental study in the American Journal of Psychology in 1922, and Findley a further study in the same journal in 1924. According to Doty's 2025 review, Findley found the differences between observers' placements closer to chance than to consistency. In 1927 Dimmick reviewed the methods for investigating olfactory qualities in the Psychological Review.
The same weaknesses come up again and again: few observers, several of them close to the author, no statistical testing, and, as Doty notes, pungent trigeminal sensations mixed into odour judgments. What survives of Henning is the idea that odour varies continuously rather than falling into separate boxes. Statistical odour-space maps from the late 20th century also work with continuous transitions, though they no longer use six corners.
The main history page dates the criticism from 1927; the first replication studies date from 1922.
| Thuja | |
| Wormwood | |
| Tansy | |
| Juniper berry | |
| Grapefruit | |
| Vanilla |
Updated 09/2026. Figures follow the year stated in each source.
↑ Top