The Smell Report

Smell Report - download full report in pdf format Click the image above to download and read the full document, The Smell Report, using Adobe's Acrobat Reader.


The Smell Report

The human sense of smell

Although the human sense of smell is feeble compared to that of many animals, it is still very acute. We can recognise thousands of different smells, and we are able to detect odours even in infinitesimal quantities.

Our smelling function is carried out by two small odour-detecting patches – made up of about five or six million yellowish cells – high up in the nasal passages.

For comparison, a rabbit has 100 million of these olfactory receptors, and a dog 220 million. Humans are nonetheless capable of detecting certain substances in dilutions of less than one part in several billion parts of air. We may not be able to match the olfactory feats of bloodhounds, but we can, for example, ‘track’ a trail of invisible human footprints across clean blotting paper.

The human nose is in fact the main organ of taste as well as smell. The so-called taste-buds on our tongues can only distinguish four qualities – sweet, sour, bitter and salt -all other ‘tastes’ are detected by the olfactory receptors high up in our nasal passages.


Our smelling ability increases to reach a plateau at about the age of eight, and declines in old age. Some researchers claim that our smell-sensitivity begins to deteriorate long before old age, perhaps even from the early 20s. One experiment claims to indicate a decline in sensitivity to specific odours from the age of 15! But other scientists report that smelling ability depends on the person’s state of mental and physical health, with some very healthy 80-year-olds having the same olfactory prowess as young adults. Women consistently out-perform men on all tests of smelling ability (see Sex differences).

Schizophrenics, depressives, migraine sufferers and very-low-weight anorexics often experience olfactory deficits or dysfunctions. One group of researchers claims that certain psychiatric disorders are so closely linked to specific olfactory deficits that smell-tests should be part of diagnostic procedures. Zinc supplements have been shown to be successful in treating some smell and taste disorders.

Although smoking does not always affect scores on smell-tests, it is widely believed to reduce sensitivity.

A recent study at the University of Pennsylvania suggests that, contrary to popular belief, blind people do not necessarily have a keener sense of smell than sighted people. In their experiments on blind and sighted people, the top performers on most tests were (sighted) employees of the Philadelphia Water Department who had been trained to serve on the Department’s water quality evaluation panel. The researchers conclude that training is the factor most likely to enhance performance on smell tests. (University of Pennsylvania researchers are probably fairly clued-up on this subject – they designed the University of Pennsylvania Smell Identification Test (UPSIT) which is the standard test used in almost all experiments.)

The importance of ‘training’ in the development of smell-sensitivity is confirmed by many other studies. Indeed, this factor can sometimes be a problem for researchers, as subjects in repetitive experiments become increasingly skilled at detecting the odours involved.

Smell-sensitivity researchers have to be very careful about the odours they use in experiments, because a smell is not always a smell. Many odorous substances activate not only the olfactory system but also the ‘somatosensory’ system -the nerve endings in our noses which are sensitive to temperature, pain etc. This is why ‘anosmics’ – patients who have completely lost their sense of smell – can still detect menthol, phenylethyl alcohol and many other substances. In a study testing anosmics’ ability to perceive odorous substances, it was found that many so-called odours are in fact affecting the pain- and temperature-sensitive nerve-endings, rather than the olfactory receptors. Out of 47 ‘odorous’ substances, anosmics could detect 45. (Only two substances could not be detected by the anosmic patients: these were decanoic acid and vanillin, which affect only the olfactory receptors, and can thus safely be classified as ‘pure’ odours.) Some unpleasant ‘smells’ do more than just annoy or disgust us, they actually cause us pain.


Although smell-identification ability increases during childhood, even newborn infants are highly sensitive to some important smells: recent research shows that newborn babies locate their mothers’ nipples by smell. In experiments, one breast of each participating mother was washed immediately after the birth. The newborn baby was then placed between the breasts. Of 30 infants, 22 spontaneously selected the unwashed breast.

Other experiments have also shown that babies are responsive to very faint differences in body odour, but it is believed that infants are highly sensitive only to specific smells, rather than a wide range of odours.

In terms of odour preference, however, one significant study showed that 3-year-olds have essentially the same likes and dislikes as adults. Experiments conducted in the early 70s and replicated in 1994 revealed that children do not develop sensitivity to certain odours until they reach puberty. In these studies, 9-year-olds showed a pronounced insensitivity to two musk odours, although their ability to detect other odours was the same as that of postpubescents and adults.