218 Proceedings of Royal Society of Edinburgh. [may 2, 
It may be urged as an objection to some of these conclusions, that 
the same odours are often produced by substances, chemically speak- 
ing, quite unlike each other. Thus benzoic aldehyde smells very 
much like nitrobenzene. In the case of taste, too, there are many 
examples of totally dissimilar bodies having indistinguishable acid 
or sweet tastes. In answer to this objection one has only to re- 
member that there are instances, equally numerous, of very different 
substances which produce the same colour sensations. One may 
produce exactly the same tint with either a chromate, a picrate, or 
an aniline dye. It would be strange, indeed, if among the com- 
plex vibrations of a compound, or even of an element, some tones 
were not of the same pitch, as some of the vibrations of substances 
quite dissimilar in general properties. When these tones fall within 
the scale of the visible spectrum, the scale of taste, or smell sensa- 
tions, we have, according to the vibration theory, a similar sensation 
produced. 
In this paper I have endeavoured to avoid all questions which 
are matters of speculation. I have dealt only with already ascer- 
tained facts, or those which can readily be verified. I do not 
attempt to offer any hypothesis to account for the action of vibrating 
matter on the olfactory end-organs. It may or may not be a 
mechanical or a chemical action. This question is not raised. We 
know next to nothing as to how it is that ether vibrations stimulate 
the cones of the retina, still less can we guess at the action of 
vibrating atoms and molecules of ordinary matter on the sensitive 
end-organs of the nose. My aim has been to establish the fact that, 
just as we have reason to connect differences in colour sensations 
with differences in the vibration of the ether, so, in like manner we 
have reason to connect differences in smells with differences in the 
vibrations which call them into existence. This analogy is estab- 
lished upon the following grounds: — 
(1) In passing from the lower to the higher members of one of 
Mendelejeff’s groups, such molecular vibrations as have been investi- 
gated tend to become lower in pitch. At the same time the colour, 
taste, and smell sensations alter in character when present. 
(2) In passing from the lower to the higher members of an 
organic series, such as the alcohols, such molecular vibrations as 
have been investigated tend to become lower in pitch. When pre- 
