E. D. Adrian
I. - Olfactory discrimination
In: L'année psychologique. 1949 vol. 50. pp. 107-113.
Citer ce document / Cite this document :
Adrian E. D. I. - Olfactory discrimination. In: L'année psychologique. 1949 vol. 50. pp. 107-113.
doi : 10.3406/psy.1949.8434
\. — Fonctions sensorielles.
by E. D. Adrian
Physiological Laboratory, Cambridge.
The study of sensation may be said to begin with the stimulus,
the material change in the environment, and to end with an
event on the mental plane which cannot be weighed and mea
sured in physical units. It must be undertaken as a co-operative
effort to which the physicists, physiologists and psychologists
must each contribute an important share. And those who are
best fitted to traverse the whole area must be at home in all
three districts, must understand the physical events which
make the stimulus, the physiology of the sense organs and the
operation of the mind. No one in these days has shown himself
so well equipped for such a task as our friend Professor Piéron.
He can speak with authority on the material events as well as
the mental, and his writings commend themselves to workers
in every part of the field.
Since our own activity is guided so much by what we see
and hear and since our thoughts are largely formulated in words pictures, it is natural that the eye and the ear should be the
two sense organs of greatest interest to mankind. They have
been intensively studied for many years for they have the great
advantage of giving rise to mental experiences which we can
analyze. The olfactory organ plays a minor role in guiding human
behaviour; its loss would not cripple us like the loss of sight or
hearing, and although we can distinguish innumerable smells,
we seem to lack the mental equipment needed for analyzing PSYCHO-PHYSIOLOGIE 108
what we experience. For this reason we understand very little
about the mechanism of the olfactory organ, although in many
animals it must give information even more important than
that from the ears or the eyes.
In recent years electro-physiological methods have greatly
assisted the study of the eye and the ear. Records of the messages
in the auditory and optic nerve fibres have become a fresh source
of information and have resolved many points which were in
dispute. The method does not involve the analysis of our sensory
experience and so might well be applied to the olfactory organ.
This has now been attempted and though the results are incomp
lete, they are already encouraging. They indicate at least one
of the ways in which smells may be distinguished and recogni
The experiments have been made on rabbits and cats. The
olfactory bulb is exposed and one or two fine wire electrodes
are thrust into it until the exposed tip of the wire is in contact
with the axons of the mitral cells which form the olfactory tract.
The nerve fibres from the receptors are non-medullated and
difficult to reach, but the signals from them are relayed through
the synapses in the glomeruli to the mitral cells. Since the recep
tors outnumber the mitral cells in the ratio of several hundred
to one the impulse in one mitral axon must be regarded as the
product of a large number of receptors distributed over a con
siderable area in the olfactory epithelium.
The discharge in the mitral axons represents the receptor
signals modified by convergence and synaptic conduction, but
there is a further complication in that the nerve cells in the
olfactory bulb, like those in the cerebral cortex, tend to remain
in constant activity in the absence of stimulation. As in the
cortex their must be abolished by deep anaesthesia
before the effects of olfactory stimuli can be clearly distinguished.
Fortunately the receptors seem to be very little affec
ted by urethane or the barbiturates but it would be more sati
sfactory if the anaesthetic could be avoided. One disadvantage
is that with an anaesthesic the force of inspiration is usually
weakened so that the stimulation of the olfactory organ is not
as effective as it would be in the normal state. D. ADRIAN. OLFACTORY DISCRIMINATION 109 E.
The nature of the olfactory stimulus.
In some experiments made 10 years ago in the hedgehog
(Adrian, 1942) it appeared that the olfactory receptors might
be stimulated by the mechanical effect of the air current through
the nose though the effect was greatly increased when the air
contained oderous molecules. This view was probably mistaken.
It arose from the fact that an olfactory discharge could be detec
ted at each inspiration although the air seemed to be odourless,
but it has now been found that air filtered through absorbent char
coal will not stimulate unless the force of inspiration is abnor
mally high. Gross mechanical stimulation of the epithelium is
certainly effective and discharges can be set up by osmotic and
ionic changes in the fluid bathing it, but the normal stimulus
is the arrival of a sufficient quantity of oderous material within
a given time.
With the more volatile substances the list of those which
will set up a discharge of impulses in the olfactory tract agrees
in general with those which have a perceptible smell to the
human nose. Thus the inert gases, water vapour, CO and CO2
do not stimulate but C2H2, H2S, C6H6 etc. stimulate strongly. Air
in the laboratory which is without odour to the human nose
usually stimulates unless it has been specially purified and the
effect has been traced to contamination with oil. On the other
hand it may be difficult or impossible to detect a discharge with
a substance of low volatility, partly no doubt because the action
of sniffing cannot be duplicated in an anaesthetized animal.
As might be expected from our own olfactory experiences
the discharge is usually limited to each period of inspiration
unless the smell is very strong. It follows that the molecules
soon become ineffective and that the rate at which they are
brought to the olfactory epithelium will be an important factor
in stimulation. What makes them lose their effectiveness is
unknown. Many of them are chemically active and may form
inert compounds, but the hydrocarbons (e. g. pentane, ben
zene etc.) are relatively stable and unless their effect is due
to impurities we must suppose that it depends on physical-
changes in the surface membranes of the receptors. 110 PSYCHO-PHYSIOLOGIE
a) Specific Receptors. — A single electrode in the olfactory
bulb or tract records the axon potentials in a number of units
though one series of impulses may stand out above the rest.
The receptors which give rise to the discharge are confined to
a particular region of the olfactory epithelium but are widely
distributed within that region. Now it is found that when seve
ral trains of impulses can be distinguished there is seldom any
sign that different smells produce a different ratio of activity
in the various conducting units. This does not prove that all
the receptors in the area are equally sensitive to different smells,
for small differences in threshold might well be missed. But as
far as the majority of the receptors are concerned the differences,
if there are any, must be quite small. Occasionally a series of
impulses can be associated with a particular class of smell, as
though there were a few receptors with marked specific sensi
tivity. This should be noted, although it has happened too
rarely to be significant except as a hint to further work, and in
no case has the specificity been absolute.
b) Time-relations and Spatial Distribution of Discharge. — But
when we compare the development and time relations of the
discharges set up by different smells and still more when we
compare the discharges from different regions of the olfactory
organ, it becomes clear that all smells do not act alike. In the
rabbit the distinction in time relations is most marked as be
tween substances with a fruity or ethereal smell (Amyl Acetate,
Ethyl Acetate, Ether etc.) and substances with an oily smell
(Pentane, Benzene, paraffin oil, petrol etc.). The esters give an
abrupt discharge, starting early in each inspiration and ending
as soon as the air movement ceases and the hydrocarbons give
one which rises and declines more gradually and usually outlasts
the period of inspiration.
When simultaneous records are made from two electrodes in
different parts of the olfactory bulb, the differential effect is
more obvious. We can then compare the intensity as well as the
time relations of excitation in two different parts of the olfac
tory organ and it is found that there is the same grouping of
smells into the ethereal and the oily class. In the rabbit's olfac
tory bulb an electrode near the anterior pole will record the E. D. ADRIAN. OLFACTORY DISCRIMINATION HI
excitation of the forward and dorsal parts of the organ and an
electrode placed deeply in the posterior half of the bulb will
record from the ventral and aboral parts which are further from
the main airway through the nose. Amyl Acetate and the ethe
real smell in general give more effect at the anterior electrode
Fig. 1. — Oscillograph records made from two points in the rabbit's olfac
tory bulb to show the characteristic differences in the localization of acti
vity set up by different smells. In each record the white line signals inspi
ration, the upper oscillograph tracing (thick line) records activity in the
posterior part of the bulb, the lower tracing records activity in the anter
ior part. The rabbit was anaesthetized with urethane and was breathing
purified air to which the stimulating substance was added.
In A the stimulant is Amyl Acetate
In B — Pentane
In C — Coal Gas
With Amyl Acetate (A) the activity is mainly in the anterior part of the
olfactory bulb and the latency is short.
With Pentane (B) both regions are active but the discharge begins earlier
in the posterior part.
With Coal Gas (C) there is a prolonged discharge in the posterior part of
the bulb with very little activity in the anterior.
A similar distribution of has been found in all experiments.
and pentane and the oily smells at the posterior. Examples of
records from two points in the bulb are given in Fig. 1. The
most striking regional difference is found when the stimulus
is the ordinary coal gas used for illumination (Fig. 1 C). This
always gives an intense and persistent discharge at the posterior
electrode and little or none at the anterior.
The time relations of the discharge suggest that the esters
are concentrated and removed from the receptors more rapidly
than the oils. It is possible therefore that the differential effects PSYCHO-PHYSIOLOGIE 112
are due to the varying rates of air flow in different regions,
those where the rate is slow favouring the substances which can
build up the excitation more gradually. And the difference in
their speed of action may depend on nothing more recondite
than solubility in water, for the esters are slightly soluble
whereas the hydrocarbons are not. But whatever the cause may
be the effect is to make the two classes of substances give diffe
rent patterns of excitation in the olfactory organ, different in
spatial distribution as well as in time relations; and although in
these records the pattern is best seen with very weak stimuli,
its characteristics will remain over a wide range of concentrations
and with varying rates of air flow through the nose.
c) Spatial differences in the Cat. — If the rabbit distinguishes
smells by such large scale differences in the pattern of excitation
we must suppose that there are other classes of smell which give
other patterns than those which have been described. No other
differential effects have been found though this may be because
the rabbit's olfactory bulb is small and because it has not been
possible to record from more than two points simultaneously.
But in the cat there is a further differentiation. The bulb in the
cat is twice as large as in the rabbit and the olfactory organ is
more elaborate. There is the same tendency for hydrocarbons to
give a greater effect in the posterior and ventral region of the
bulb and the esters in the anterior, dorsal region, but there is
also a posterior dorsal region where the greatest effect is pro
duced by the smell of decayed meat or fish. Trimethylamine and
other amines with a fishy smell give larger discharges here than
elsewhere : they -are readily soluble in water but the discharge is
less abrupt than that given by amyl acetate. It seems to arise
in the dorsal clefts of the olfactory organ which lie in the cat
and dog above the main air passage and extend back over the bulb.
But the connections between the different parts of the olfactory
epithelium and the different parts of the bulb have not been
mapped out in the cat as they have been in the rabbit, and all
that can be said is that the amines, like the esters and hydro
carbons, produce a pattern of excitation which is sufficiently
characteristic to be seen in oscillograph records.
These results show that there are large scale differences in the
spatial and temporal pattern of excitation produced by different D. ADRIAN. OLFACTORY DISCRIMINATION 113 E.
smells : they do not show that the smells are distinguished in this
way by the animal, though they can be by the
electro-physiologist who has access to the olfactory bulb. But
as there has been little evidence for the existence of a mosaic of
specific receptors it is at least probable that olfactory discr
imination depends mainly on such differences in general pattern.
It would have seemed more probable if more different patterns
had been found, but the experimental procedure is not easy
and it may well turn out that there are other regions where
other classes of smell are specially effective.
It is a striking fact, not always appreciated, that the olfactory
organ has a very large sensitive surface, larger than that of the
eye in most animals. It has also a complex structure which
seems designed to. ensure that the stimulus shall not be the same
in every part. If smells are in fact analyzed on the basis of their
effectiveness in different regions it is clear that the more complex
the structure of the organ the greater will be the variety of olfac
tory patterns which can be formed. We can see therefore why the
dog should have a much larger and more elaborate organ than
the rabbit. There is no reason why a large organ should be more
sensitive to smells than a small one, but it will certainly provide
a screen on which the pattern of excitation can be mapped in
greater detail. Presumably the brain can learn to discriminate
these patterns as it discriminates those which are formed by
sounds on the basilar membrane of the ear and by visual stimuli
on the retina. And to the animals which rely on smell slight
differences in the temporal and spatial pattern of the olfactory
excitation may be just as significant as slight differences in the
visual pattern can be to man.
.Adhian (E. D.). — ,/. Physiol., 1942, 100, 459.
A. P VOL .1UIÎ.