46 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS SMELL--THE PHYSICAL SENSE A. W. MIDDLETON, B.Sc., PH.D., F.R.I.C.* The sense of smell is shown to be analogous to that of hearing and to be a "physical sense." The fundamental physiological or physical basis o• the different oltactory phenomena is explained, and the explana[ions are a sui[able criteria •or the assessment o• experimental work. A l•lea is made •or the establishment of a co-ordina[ing cen[re for research on olfaction. IN THIS article it is proposed to explain the behaviour of the sense of smell in a qualitative manner and to work out a basis on which quantitative studies can be made. The appropriate inductive approach for the development of a ½atisfactory theory regarding the sense of smell, is: 1. To collect the properly established observations. 2. To consider any information available concerning other senses, and also of the relevant properties of gases. 3. To develop from the whole, consistent explanations of the mechanism of response of the olfactory organs to odorous stimuli. 4. To plan future experimental work in a co-ordinated manner, and to modify the theory as may be required by its results. PART I OLFACTORY OBSERVATIONS TO BE EXPLAINED Basic Odour Research • collects together the work of many authors on olfaction from which the following accepted smell behaviour can be gleaned. (a) Adaptation and Fatigue. Continued exposure to an odour can more or less rapidly lead to inability to smell that odour, without necessarily causing inability to smell other odours: on continued exposure to an odour such as that in the cream room of a cosmetic laboratory, the smell of the perfume compounds in use rapidly disappears, but other odours can still be detected in that room. (b) Anosmia. Persons exist who are unable to smell (or to differentiate between members of) certain groups of odorants whilst smelling others normally. (c) Paralysis. A few substances, of which formaldehyde is outstanding, paralyse the sense of smell to all odours to the same degree. * Chesebrough-Pond's Ltd., London, N.W. 10.

SMELL--THE PHYSICAL SENSE 47 (d) Smell in Insects. Many insects' smell-sensilla are all peg-like hair derivatives. Stimulation of the sense of smell of some of them does not require moisture (Dethier2). (e) Smell in Animals. Animals have a "smell response," some having a lower threshold than man. (f) Molecular Structure of Odorants. (i) There is a considerable body of information regarding the change in odour as serial changes in constitution, or changes in atomic arrangement in molecules, are made (Moncrieff-'). Thus stereoiso- mers can have different smells, e.g., the position of groups in the molecule among isomers can alter the smell, and unsaturated substances generally have stronger odours than saturated sub- stances. Moreover, substances of different chemical composition can have similar odours. (ii) When homologous series of organic chemicals are smelled, it is seen that the more evanescent odours are the more volatile members with the lower molecular weights. As the series is ascended, the intensity of the smell increases to a maximum and then decreases as the substance loses volatility. (g) Odour Analysis. This is the ability to dissect a smell into several components whilst being subjected to the stimulus of the whole, the extent to which this can be done being dependent on the past experience of the observer. If two odours of about equal strength are smelled, then either a blend or both individuals can be recognised, but if one is much stronger than the other then the weaker is not recognised. (h) Perception. (i) The intensity of perception. According to the Weber-Fechner law this appears to be a log function of the external chemical concen- tration. (ii) Threshold. There is a limiting concentration below which an odour cannot be smelt: this can be as low as 10 '•-• Molar for some substances. (iii) Differences between odours can be detected much more sensitively when the odours are near to their respective threshold concentra- tions. (iv) Change of quality. The odour of many substances changes in quality as the dilution of the substance is altered. Thus, j asmin absolute has a fatty odour to it which completely disappears when diluted. (v) Effect of vdodty of air stream. The olfactory threshold depends on the air pressure and when determined by air blast methods is different from that determined by other methods (Jones'). (i) Masking. It is sometimes possible to hide one odour by the use of