One bottle of perfume, two wrists, the same single spray. Two hours later, the two trails of scent are no longer alike. This essay separates what has been measured about that from what is only retold.
Two wrists, one bottle
A single spray of perfume puts a thin film of liquid on skin. Most of it is alcohol. The rest is many kinds of odorous molecule, each with its own rate of evaporation. As soon as it leaves the nozzle, that mixture starts to come apart over time: light molecules go first, heavy ones stay.
On a smelling strip the process is fairly even. A strip is made of cellulose; it is not warm, it does not secrete oil, it does not breathe. Two strips of the same kind dipped in the same bottle give two nearly identical evaporation curves. Skin is different. Each area of skin has its own temperature, moisture, oil and roughness, and these change from person to person, from place to place, and over the hours of a day.
So the familiar saying that a perfume smells different on everyone is partly true. But the reason usually given, something loosely called skin chemistry, is the part that has been measured least. Clearer differences come from two places: the rate of evaporation from one particular surface, and the nose that is doing the smelling.
This essay takes each variable in that order: heat, then oil and water on skin, then the person smelling. The last part is kept for things that are often said but come with no figures.
Temperature and rate of evaporation
Each odorous molecule has a vapour pressure, a measure of how readily it evaporates at a given temperature. High vapour pressure molecules reach the nose first and fade early; low ones linger. Vapour pressure rises with temperature. A warmer surface sends the same molecule into the air faster. Skin temperature is therefore the first variable.
The library's volatility chart offers one way to picture this. By blotter substantivity figures for undiluted material, expressed lemon oil stays recognisable for about 4 hours, lavender about 8 hours, spearmint about 36 hours, patchouli 400 hours or more. The chart states its own limits: hours on a blotter are not wear time on skin, and temperature, dilution, lot and the nose all move the figure.
That means the same perfume on a warmer skin passes through its stages faster. The top notes project more strongly for a shorter time; the base shows itself earlier. On a cooler skin the path slows down. Two people standing side by side at hour two may be at two different points on the same curve.
Position on the body is a thermal variable too. Wrist, neck, inner elbow and the back of the hand are not equally warm or equally exposed to air. A trace of perfume behind the ear and one on the forearm of the same person can already differ, before a second person is even considered.
Oil and moisture on skin
The surface of skin is covered by a thin layer of sebum, sweat and water passing out through the epidermis. Many odorous molecules dissolve better in oil than in water. A thicker oily layer can hold such molecules longer before they leave; a drier surface holds less. That is the plausible mechanism behind the saying that oily skin holds scent longer than dry skin, and it is also what laboratories have started to measure.
In 1995 Vuilleumier, Flament and Sauvegrain published in the International Journal of Cosmetic Science a method for collecting scent vapour directly above the inner forearm. They used a mixture of eleven synthetic odorants in alcohol and measured the diffusion rate of each over time. The results showed each substance leaving the skin at its own rate, and the same substances leaving more slowly when applied from a bar of soap than from an alcohol solution.
The same study describes a panel of 80 people, 40 women and 40 men, whose sebum, hydration and pH were measured on the face, the neck, and the inner and outer forearm, and who were then sorted into skin groups. The authors note that the evaporation results they present were obtained on an average skin type. In other words, the measurement of skin had begun; linking each skin type to each evaporation curve had not.
In 2025 Hadjiefstathiou and colleagues returned to the question in the same journal. They measured the evaporation of odorous molecules on the skin of several volunteers, recorded differences between individuals, then measured skin properties to explain those differences. Statistical analysis indicated that both skin type and the molecules' own properties play a part. The authors call the results promising, a careful choice of word.
A reanalysis of older data by Saiyasombati and Kasting in 2003 adds one detail: on skin, odorous molecules both evaporate and are absorbed. And the substances in a mixture affect each other. The same twelve component mixture, with and without a musk fixative, gave two different kinds of evaporation curve. Scent on skin is the product of both the skin and the formula.
The person smelling is a variable too
The human nose has about 400 kinds of working odorant receptor, and their genes vary from one person to the next. A smell is not recorded in the same way by two genomes. This is where differences have been measured most clearly.
In 2007 Keller and colleagues published a concrete example in Nature. Androstenone, an odorous steroid, is found offensive and sweaty by some people, sweet and floral by others, and odourless by others again. The team found that the receptor OR7D4 responds selectively to androstenone, and that people carrying a common variant of the gene were, as a group, less sensitive to it and found it less unpleasant.
Being unable to smell one particular compound while the sense of smell is otherwise normal is called specific anosmia. It is fairly common with some synthetic musks. A perfume with a musk base may project clearly to someone standing nearby yet stay almost silent for the wearer, if the wearer belongs to the group that cannot smell that molecule. The essay A Scent Half the Room Cannot Smell goes further into this case.
There is one more variable: getting used to a smell. Smelling one odour continuously reduces its perception; stepping away from it for a while brings perception back. The wearer smells their own perfume for hours, while a passer by meets it for a few seconds. The same trail of scent, two people in two different states of the nose.
The trade has two words for these two sides. Tenacity is how long a scent stays recognisable on the surface itself, strip or skin. Sillage is the trail of scent left in the air as the wearer moves. The wearer usually judges tenacity; those around usually meet the sillage.
What has not been measured
A large share of explanations about perfume on skin come without figures. Skin pH, diet, hormones, medicines in use: each is often named as the reason a perfume changes its smell. In the studies the library has read, pH was measured on test panels, but no result links a particular pH value to a particular change in smell.
That does not mean these factors play no part. It only means their part has not been measured in a way that allows a firm statement. What has been measured more clearly are simpler things: surface temperature, the amount of oil and water, how the substance is put on skin, the ingredients of a formula affecting each other, and the genes of the person smelling.
A few familiar molecules also have lives of their own on skin. Linalool, the soft floral note found in lavender and coriander, and limonene, the smell of orange peel, are both small molecules that evaporate fairly quickly. In a perfume they usually belong to the part that opens first. Differences between two skins, where they exist, often show most clearly in these first minutes, while the light molecules are leaving at a rate set by the warmth of the surface.
Nor is there yet a way of classifying skin that allows a perfume to be chosen by skin type with much confidence. The only practical advice with a basis is the oldest advice in the trade: try it on the skin that will wear it, and wait long enough to go through the whole curve.
Skin is a surface
A surface has a temperature, a roughness, a layer of oil, a layer of water. Smelling strips, fabric, hair and skin are all surfaces, and each lets molecules leave at a pace of its own. In 2025 a new measuring device was tested on four surfaces: inert glass, a skin simulating membrane, a perfume test strip and human skin. That one phenomenon needed four surfaces to be measured says that the surface is not a minor detail.
Seen this way, the saying that everyone smells different loses some of its mystery. There is no private chemical code on each skin that turns a perfume into something else. There is a mixture of molecules, leaving at different rates from different surfaces, and recorded by different noses.
The fragrance family of a perfume does not change when it changes skin. A chypre stays a chypre, a citrus stays a citrus. What changes is the balance between the stages at a given moment, and the moment at which each stage shows itself.
The essay How Long Before a Scent Leaves continues with the question of time: why some scents stay for hours and others for days, and how the figure printed on a box should be read.
The bottle does not change.
Two skins only let it leave at two different paces.
The rest belongs to the nose that is smelling.