Review Article

Blackwell Publishing,
REV I ELtd.
W ARTICLE

Is the human skin a pheromone-producing organ?

Bettina M Pause
Institute of Psychology, Christian-Albrechts-University Kiel, Germany

Summary It is controversial whether or not humans convey specific compounds within their body
odours which can potentially affect the physiology and behaviour of others. Such
compounds are called pheromones and have been discovered in many other species,
including mammals. It has been suggested that humans might have a special organ
within their nose that can transmit such chemosensory information. However, the
evidence for this organ is highly questionable. In any case, the main olfactory system is
a highly diverse system, capable of transmitting pheromonal information.
So far, no single substance has been found that acts as a chemical messenger for
erotic attraction. On the other hand, studies investigating the pheromonal properties
of natural complex body odour have proven that it does deliver information about the
sender and that it has an effect on the physiology and likely behaviour of other humans.
Its significance for human mating preferences probably lies not in driving them to choose
the right mate but rather in warning them not to choose the wrong one.
Keywords: androstenes, body odour, human leucocyte antigen (HLA), major histocom-
patibility complex (MHC), pheromone, vomeronasal organ (VNO)

has been proven that humans react physiologically or

Introduction
Recently published opinions about human pheromones
seem to be far apart from each other: Whereas a recent
study published in Nature suggested that it ‘provides
definitive evidence of human pheromones’,1 a review
in Science published at about the same time, indicated
that ‘it therefore seems implausible that humans might
experience significant behavioural or endocrine regulation
by pheromones’.2 The following review is intended to
give an overview on the present research on human
pheromones. In detail, it will be questioned whether
human skin products acting as potential pheromones could
be identified, whether the paths of sensory processing
of possible pheromones are known, and whether it
Correspondence: Bettina M Pause, Institute of Psychology, Christian-
Albrechts-University, Olshausenstr. 62, 24098 Kiel, Germany,
E-mail: bmpause@psychologie.uni-kiel.de
Accepted for publication 1 June 2004
behaviourally to specific body odour compounds.
What is a pheromone?
In 1959, Karlson and Lüscher3 introduced the first
pheromone definition: ‘Pheromones are defined as sub-
stances which are secreted to the outside by an individual
and received by a second individual of the same species,
in which they release a specific reaction, for example, a
definite behaviour or a developmental process’.3 Importantly,
their definition was decisive for a differentiation of releasing
from priming pheromones. Priming pheromones produce
a change of state in the receiver, usually a change in
hormonal secretion that primes the animal for a later
response. Releasing or signalling pheromones elicit an
immediate behavioural response from the individual
receiving the stimulus. The third class of pheromones are
called information pheromones – chemosignals indicating
the animal’s identity or territory.
Regarding the conflicting literature on human phe-
romones, Meredith4 suggested that it may be scientifically

© 2004 Blackwell Publishing Ltd • Journal of Cosmetic Dermatology, 3, 223–228 223

Pheromones in humans • B M Pause
useful to add a mutual benefit criterion to the definition of
pheromones. This would require that the chemosensory
communication in question should contribute to the
evolutionary fitness for the sender and the receiver.
However, examples of pheromonal communications in
rodents have been well described: When groups of female
mice are housed together, their oestrous cycles slow
down and eventually stop (Lee-Boot effect), and if they are
exposed to the odour of a male, they begin cycling again
(Whitten effect). The observation that a high proportion
of pregnancies fail when a female mouse is exposed to the
smell of a male other than the stud male is called the
Bruce effect. The Vandenbergh effect describes the accel-
eration of the onset of puberty in a female mouse caused
by the odour of a male.
Expression of body odour signals
The apocrine and apoeccrine glands develop with puberty
and are discussed as the source of possible pheromones
in humans. Whereas initial studies indicate that the
apoeccrine contribute strongly to the intensity and
emotional valence of body odour 5 the secretions of
the apocrine glands have been studied in more detail.
Whereas freshly collected apocrine secretion is odourless,
incubation with the resident bacteria results in the
production of a characteristic odour. It has been shown
that the male axillary flora seems to be dominated by
coryneform bacteria, and that females show a dominance
of micrococci in their axillae.6 In vitro incubation of
sterile odourless apocrine sweat with micrococci results
in a sweaty type of odour, which has been identified with
isovaleric acid. Isovaleric acid has additionally been
found in the vaginal secretions of the rhesus monkey and of
human females.7 Coryneforms seem to be responsible
for the development of a pungent odour, delivered by
androgen steroids.8 Androgen steroids (androstenone,
androstadienone, androstadienol, and androstenol) can
also be found in other human body fluids (like urine,
plasma, saliva and semen). Recently, Zeng et al.9 detected
that the major contributors to the male sweaty axillary
odour are C6-C11 unsaturated acids, with the most
abundant being (E)-3-methyl-2-hexonic acid.
Perception of body odour signals (Table 1)
Biological findings
Three different systems should be considered when looking
for a chemosensory organ responsible for pheromonal
signal transduction. Whereas the olfactory system is
thought to be responsible for the perception of smelling
224 © 2004 Blackwell Publishing Ltd • Journal of Cosmetic Dermatology, 3, 223– 228
volatile molecules, the trigeminal nerve endings contribute
to odour perception by the processing of cold, pungent or
burning sensations. Particularly, the vomeronasal organ
(VNO) is considered to be responsible for the transduction
of pheromones, which do not necessarily have to have
a smell and can also be nonvolatile molecules.
The olfactory sensory neurones are bipolar cells with
the unmyelinated axons terminating in the main olfac-
tory bulb. Whereas each sensory neurone expresses one
olfactory receptor (OR) gene, it has been estimated that
there are about 500–750 different OR genes in humans.10
In search of pheromone-specific ORs, it has been found
that humans lack the high-affinity receptors for isovaleric
acid, which have been described in the mouse.10
It could be demonstrated that olfactory signals from a
particular OR are targeted to specific stereotyped clusters
of neurones in the olfactory cortex (anterior olfactory
nucleus, piriform cortex, olfactory tubercle, entorhinal
cortex) of the mouse.11 Similar brain areas belong to
the primary human olfactory cortex, which additionally
includes the amygdala. The secondary olfactory cortex is
characterized by the thalamus, hippocampus, hypotha-
lamus and neocortical structures.12 Neocortical process-
ing of olfactory information is unique among all other
modalities, because it does not require the thalamic relay
of sensory information.
During the last 10 years, the question whether a human
VNO exists or not has attracted much attention. Among
mammals it is well developed in a number species (e.g.
rodents, rabbits and hoofed animals) but cannot be found
in others (e.g. toothed whales and Old World monkeys).
In human embryos, a VNO can be detected, containing
bipolar cells similar to the developing VNO in other species.
However, endoscopic investigations in adult humans
show that the appearance of the VNO varies from one
inspection to another4 and moreover, that the shape, size
and orientation of the VNO between humans is highly
variable.13
In the mouse, VNO neurones pass the chemosensory
information via axonal transport to a structure called the
accessory olfactory bulb (AOB). However, cells within the
human VNO have not been shown to have axons leaving
the epithelium nor making synaptic contact with axons
in the epithelium; an AOB related structure has not been
found even after explicit research.4 A functional gene
closely related to the mammalian VNO receptor genes
has recently been detected in humans, but is expressed in
the main olfactory epithelium.14
It should be added that, in animals, the olfactory and
the vomeronasal systems seem to act in a highly interac-
tive manner and some behavioural consequences of
body odour perception do not require the VNO at all. For
 Pheromones in humans • B M Pause

Table 1 Pheromones: expression and perception

Pheromones are substances considered to affect the behaviour and physiology of other members of the same species.
Androgen steroids, isovaleric acid and long chain unsaturated acids are single substances released by humans which may have pheromonal
properties.
There is no convincing evidence that humans have a functional vomeronasal organ (VNO), responsible for sensing pheromones. However, in
mammals the main olfactory system is able to transmit pheromonal information.

Table 2 The search for pheromonal effects of complex biological mixtures has been more successful than the search for pheromonal effects
of single compounds

Androstenes, from human body fluids, can be smelled by most people. However, some people have a specific anosmia for these odours.
As the behavioural effects of androgenic steroids are contradictory, the role of learning processes and context need to be considered in future
research.
Complex human body odour contains several messages which are socially relevant.
The endocrine state of a woman can be altered by the chemical compounds in another person’s body fluids.

Table 3 The human leucocyte antigen (HLA) system seems to affect mate selection

The HLA system is responsible for self-non-self recognition within subjects (via the immune system).
The HLA system may also be responsible for self-non-self recognition between subjects (via the olfactory system).
In ethnically homogenous couples, HLA matches are observed less frequently than expected by chance.
Humans seem to show stronger brain responses to body odours from subjects with a similar HLA-type.
HLA associated body odours might provide cues that prevent evolutionarily harmful mating.

example, in the domestic pig, the release of androstenone

by the male elicits a receptive mating stance in the oestrous
female. However, blocking the access of chemosensory
information to the female VNO does not alter the sensitiv-
ity and behavioural response to androstenone.15
Psychophysical and neuropsychological findings
The sensitivity of humans to body odour compounds has
been best documented for androstenone and described
to depend on genetic, developmental and environmental
factors. Whereas probably, all very young children are
able to detect this odour, about 40% of them lose their
sensitivity during puberty and become anosmic to this
substance.16 However, it has been documented that
sensitivity to androstenone can be acquired in anosmic
subjects as a result of repeated exposure.17 This effect
seems to be based, at least in part, on a stimulus-specific
enlargement of receptor sensitivity.18
A recent study investigated human brain activation
through putative pheromones (androstadienone, estra-
tetraenol) by positron emission tomography:19 A gender-
specific activation of the hypothalamus was observed in
males only when they smelled estratetraenol and only
in females, when they smelled androstadienone.
Response to body odour signals (Table 2)
Information pheromones
Body odour contains information about the gender20 and
the relatedness of individuals:21 Breast-fed babies as
young as 6 days old are able to olfactorily identify their
mother, and mothers are able to olfactorily recognize
their newborn. Even nonfamily members can successfully
match the t-shirts of mothers and their children. Moreover,
siblings are able to recognize each other by olfactory
cues. Additionally, a pronounced brain response to the
olfactory self has been reported.22
Whereas the former studies are supportive for a strong
genetic impact on the expression of human body odour,
effects of the endocrine system have been indicated by
studies examining female body odour during the course
of the menstrual cycle.23,24
A new area of research on human information,
pheromones have been recently initiated by Chen and
Haviland-Jones.25 As experiments in rodents show that
olfactory perception of stressed conspecifics results in
alterations of the physiology and behaviour of the per-
ceiving animal,26 they asked their subjects to identify the
emotion of odour donors from the smell of underarm-pads.

© 2004 Blackwell Publishing Ltd • Journal of Cosmetic Dermatology, 3, 223–228 225

Pheromones in humans • B M Pause
It appeared that the smell of frightened men could be best
identified by either sex.
Priming pheromones
More than 30 years ago, it was first reported that
the menstrual cycles of females living together tend to
synchronize over time.27 Just recently, it could be proven
to be a chemosensory mediated effect: Stern and McClintock1
showed that the presentation of odourless compounds
from the axillary region of females differentially can alter
the length of the menstrual cycle of recipient women.
However, it should be noted that a number of studies
failed to find effects of menstrual synchrony28 and
secondly, that the mutual benefit of this effect seems to
be questionable.4 Schank28 pointed to the fact that the
biological costs of synchrony were enormous, in particular
regarding the loss of female mating choice.
Signalling pheromones
The question whether body odour or its substances have
the capacity to directly alter human mood or behaviour
has mainly been investigated using androgen steroids.29
Investigating the behavioural consequences of andros-
tenol exposure, some studies reported that females
describe themselves as more submissive whereas others
found that females seem to have contact with men more
often. The treatment of restroom stalls with androstenol
did not show to have any effect on females, but men
avoided using the sprayed stalls.
When women were exposed to androstenone, they
rated themselves as less sexy, and they rated males on
photographs to be less sexually attractive. However, in
a dentist waiting room, they seemed to use seats sprayed
with androstenone more frequently than expected by
chance.
At present, the literature on signalling effects of
human pheromones seems to be conflicting, possible
confounding effects should be considered, e.g. the influ-
ence of the individual learning history and the role of the
context.
In mammals, it is evident that some of the relatively
fixed behavioural response patterns elicited by pherom-
ones are learned during a sensitive period. For example,
the Bruce effect in mice implies that the female recognizes
the pheromone of the mating male. In humans, an example
of the learnt nature of responses to body odour is the find-
ing that newborn bottle-fed babies seem not to olfactorily
recognize their mothers (see above). It is in line with these
considerations, that Jacob et al.30 recently observed that
the positive mood of females in response to androstadi-
226 © 2004 Blackwell Publishing Ltd • Journal of Cosmetic Dermatology, 3, 223– 228
enone improves in combination with a male tester, but
not with a female tester.
Another new line of research on immunogenetic
markers of olfactory individuality and its consequences
in the receiver might also overcome some of the problems
faced in the traditional pheromone research.
The major histocompatibility complex (MHC)
(Table 3)
Cell-surface molecules belonging to the MHC present
peptides derived from pathogenic organisms to the T-cell
receptor. In most vertebrates the MHC shows an extreme
polymorphism, which is assumed to have an evolutionary
advantage. E.g., an individual with many different MHC
alleles can respond to a wide array of pathogens and
is thus more likely to survive. It is proposed that sexual
selection maintains this MHC variation by influencing
the chemosensory communication between individuals.
Several experiments have demonstrated that the indi-
vidual body odour in mice and rats is associated with their
MHC type and that male and female mice prefer to mate
with a conspecific that differs at the MHC.31 Additionally,
it has been found that the Bruce effect can be elicited,
when the only difference between the stud and the
strange males are in the alleles of the MHC. However,
as mate choice can be manipulated by foster nursing,
the behavioural response to MHC-related body odours
seems to be prone to early learning experiences.
In the last years, strong evidence has been accumu-
lated that human body odour is also partly determined by
the MHC (in humans referred to as Human Leucocyte
Antigen, HLA). Rats can be trained to differentiate
human urine samples with respect to the HLA type of the
odour donor.32 Subjective rating studies show that males
and females prefer t-shirts of subjects with a dissimilar
HLA-type.33,34 Investigating mate choice in a natural,
inbred and ethnically homogenous population (Hutter-
ite), it could be found that, in couples, HLA haplotype
matches occur less frequently than expected by chance.35
Additionally, preliminary studies indicate that the electrical
brain response to the axillary odour of HLA-similar odour
donors is more pronounced than the brain response to
the odour of HLA-different odour donors.36 In female
odour perceiving subjects, however, this effect seems to
be modulated by their menstrual cycle.37 These results
indicate that odours from HLA-similar donors may act as
warning signals, indicating that caution is warranted to
avoid inbreeding. Therefore, a negative selection mecha-
nism in humans is proposed, preventing evolutionary
harmful mating, but leaving positive selection to other
cues (visual, social, etc.).

Conclusions
Chemical signals of human natural complex body odour
deliver information about the sender that may have an 16
effect on the physiology and behaviour of the receiver.
17
Acknowledgements
The author is indebted to R. Ferstl and M. Heckmann
for their comments on a draft of the manuscript and would 18
like to thank M. Ribeiro-Nelson for the proofreading.
19
References
1 Stern K, McClintock MK. Regulation of ovulation by human 20
pheromones. Nature 1998; 392: 177– 9.
2 Keverne EB. The vomeronasal organ. Science 1999; 286:
716 –20. 21
3 Karlson P, Lüscher M. Pheromones: a new term for a class
of biologically active substances. Nature 1959; 183: 55 – 6. 22
4 Meredith M. Human vomeronasal organ function: a critical
review of best and worst cases. Chem Senses 2001; 26:
433 –45.
5 Heckmann M, Teichmann B, Pause BM, Plewig G. 23
Amelioration of body odour after intracutaneous axillary
injection of botulinum toxin-A. Arch Dermatol. 2003; 139:
57–59. 24
6 Leyden JJ, McGinley K, Hölzle E, Labows JN, Kligman AM.
The microbiology of the human axilla and its relationship to 25
axillary odour. J Invest Dermatol 1981; 77: 413 – 6.
7 Michael RP, Bonsall RW, Kutner M. Volatile fatty
acids, ‘copulins’, in human vaginal secretions. 26
Psychoneuroendocrinol 1975; 1: 153 – 63.
8 Gower DB, Ruparelia BA. Olfaction in humans with special
reference to odorous 16-androstenes: their occurrence, 27
perception and possible social, psychological and sexual
impact. J Endocrinol 1993; 137: 167– 87. 28
9 Zeng X-N, Leyden JJ, Lawley HJ et al. Analysis of
characteristic odours from human male axillae. J Chem
Ecol 1991; 17: 1469 – 92. 29
10 Zhang X, Firestein S. The olfactory receptor gene superfamily
of the mouse. Nature Neurosci 2002; 5: 124 – 33.
11 Zou Z, Horowitz LF, Montmayeur J-P et al. Genetic tracing 30
reveals a stereotyped sensory map in the olfactory cortex.
Nature 2001; 414: 173 – 9.
12 Zatorre RJ, Jones-Gotman M. Functional imaging of the 31
chemical senses. In: AW Toga, JC Maziotta, eds. Brain
Mapping: the Systems. San Diego: Academic Press;
2000: pp. 403 –24. 32
13 Abolmaali ND, Kühnau D, Knecht M et al. Imaging of the
human vomeronasal duct. Chem Senses 2001; 26: 35 – 9.
14 Rodriquez I, Greer CA, Mok MY, Mombaerts P. A putative
pheromone receptor gene expressed in human olfactory
mucosa. Nature Genet 2000; 26: 18 – 9. 33
15 Dorries KM, Adkins-Regan E, Halpern BP. Sensitivity and
© 2004 Blackwell Publishing Ltd • Journal of Cosmetic Dermatology, 3, 223–228
Pheromones in humans • B M Pause
behavioural responses to the pheromone androstenone are
not mediated by the vomeronasal organ in domestic pigs.
Brain Behav Evol 1997; 49: 53 – 62.
Dorries KM, Schmidt HJ, Beauchamp GK, Wysocki CJ.
Changes in sensitivity to the odour of androstenone during
adolescence. Dev Psychobiol 1989; 22: 423 – 35.
Pause BM, Rogalski KP, Sojka B, Ferstl R. Sensitivity to
androstenone in female subjects is associated with an
altered brain response to male body odour. Physiol Behav
1999; 68: 129 –37.
Wang H-W, Wysocki CJ, Gold GH. Induction of olfactory
receptor sensitivity in mice. Science 1993; 260: 998 –1000.
Savic I, Berglund H, Gulyas B, Roland P. Smelling of odorous
sex hormone-like compounds causes sex-differentiated
hypothalamic activations in humans. Neuron 2001; 31:
661– 8.
Doty RL, Green PA, Ram C, Yankell SL. Communication of
gender from human breath odours: relationship to perceived
intensity and pleasantness. Horm Behav 1982; 16: 13 –22.
Porter RH. Olfaction and human kin recognition. Genetica
1999; 104: 259 – 63.
Pause BM, Krauel K, Sojka B, Ferstl R. Body odour-evoked
potentials: a new method to study the chemosensory
perception of self and non self in humans. Genetica 1999;
104: 285 – 94.
Doty RL, Ford M, Prety G, Huggins GR. Changes in the
intensity and pleasantness of human vaginal odours during
the menstrual cycle. Science 1975; 190: 1316 –7.
Singh D, Bronstad PM. Female body odour is a potential cue
to ovulation. Proc R Soc Lond B 2001; 268: 797– 801.
Chen D, Haviland-Jones J. Human olfactory
communication of emotion. Percept Motor Skills 2000;
91: 771– 81.
Cocke R, Moynihan JA, Cohen N et al. Exposure to
conspecific alarm chemosignals alters immune in BALB/c
mice. Brain Behav Immun 1993; 7: 36 – 46.
McClintock MK. Menstrual synchrony and suppression.
Nature 1971; 229: 244 – 5.
Schank JC. Measurement and cycle variability:
re-examining the case for ovarian-cycle synchrony in
primates. Behav Processes 2001; 56: 131– 46.
Kirk-Smith M. Culture and olfactory communication. In:
A Gardner, ed. Ethological Roots of Culture. Dordrecht,
Holland: Kluwer Academic Pubs; 1995: pp. 385 – 406.
Jacob S, Hayreth DJS, McClintock MK. Context-dependent
effects of steroid chemosignals on human physiology and
mood. Physiol Behav 2001; 74: 15 –27.
Yamazaki K, Singer A, Beauchamp GK. Origin, functions
and chemistry of H-2 regulated odourants. Genetica 1999;
104: 235 – 40.
Ferstl R, Pause B, Schüler M et al. Immune system signalling
to the brain: MHC-specific odours in humans. In:
CN Stefanis, CR Soldatos, AD Rabavilas, eds. Psychiatry:
A World Perspective, Vol. 2. Amsterdam: Elsevier;
1990: pp. 751– 5.
Wedekind C, Füri S. Body odour preferences in men and
women: do they aim for specific MHC combinations or
227
Pheromones in humans • B M Pause
simply heterozygosity? Proc R Soc Lond B 1997; 264:
1471– 9.
34 Jacob S, McClintock MK, Zelano B, Ober C. Paternally
inherited HLA alleles are associated with women’s choice
of male odour. Nature Genet 2002; 30: 175 – 9.
35 Ober C, Weitkamp LR, Cox N et al. HLA and mate choices
in humans. Am J Hum Genet 1997; 61: 497– 504.
36 Krauel K, Pause BM, Mueller C et al. Central nervous
correlates of chemical communication in humans.
228
In: C Murphy, ed. Olfaction and Taste XII. New York:
Annals of the New York Academy of Sciences; 1998:
pp. 628 –31.
37 Pause BM, Krauel K, Eggert F et al. Perception of
HLA-related body odours during the course of the
menstrual cycle. In: RE Johnston, D Müller-Schwarze,
PW Sorensen, eds. Advances in Chemical Signals in
Vertebrates. New York: Kluwer Academic/Plenum
Publishers; 1999: pp. 201–7.
© 2004 Blackwell Publishing Ltd • Journal of Cosmetic Dermatology, 3, 223– 228