THE SCENT OF FLOWERS AS A MEANS OF ATl'BACTING ANIMALS. 205 



soDie kind of swinging movement, and that this movement is transferred to the 

 nerve-endings as soon as the molecules come into contact with the olfactory organ. 

 Since the nerve-endings are not exposed the transference must take place through 

 the portion overlying the nerve-ends, and it must depend upon the structure of this 

 superficial layer which is exposed to the air whether the transference takes place 

 quickly or slowly, completely or only partially. It is necessary that the stimulation 

 of the nerve-endings, wliich we imagine to be a form of movement, should be con- 

 ducted to the central organ if it is to be perceived as smell. But now arises this 

 difficult question: Do the various sensations of smell depend upon the fact that 

 different nerve-endings are stimulated by different odorous substances, and that 

 a particular scent, e.g. that of Oil of Lavender, is only perceived when those 

 particular nerve-ends are stimulated which are sensitive to the kind of vibration 

 undergone by the molecules of the Oil of Lavender? Or, are they caused directly 

 by the movement of the molecules of any odorous substance being transmitted by 

 any olfactory nerve-fibre to the central organ, and there producing a definite sense 

 of smell? In this case the same nerve-fibre which had just transmitted the vibra- 

 tions of the lavender oil would be capable in the next moment of transferring to the 

 central organ those belonging to the molecules of chloroform. 



The one hypothesis assumes that certain parts of the central organ, as well 

 as the nerve-fibres leading to them, diflfer essentially from one another in their 

 capability of being stimulated, although they seem to our senses to be of exactly 

 the same structure. One part can only be stimulated by Oil of Lavender and is 

 not affected by chloroform molecules, another part is only set into a corresponding 

 movement by the swinging of chloroform molecules, but is not in sympathy with 

 the particular movement of those of lavender oil. But to favour this hypothesis is 

 to assume an enormously large number of different nerve-endings in the olfactory 

 organ considering the innumerable quantity of different odoriferous substances that 

 exist — even if it be granted that there is a place only for groups of similar sub- 

 stances in the olfactory organ and not for each singly, the individual scents of each 

 group being only produced by the different degree of the stimulation. The other 

 hypothesis assumes that each olfactory nerve-fibre according to its structure is 

 enabled to transmit the different forms of movement which occur at its peripheral 

 end to the central organ. The particular movements of the molecules of lavender 

 oil would not only afiect the nerve-ends, but would continue as a specific form of 

 movement through the whole nerve-fibre to the central organ, and would be there 

 perceived as the scent of lavender oil. This same nerve-fibre which had just 

 transmitted the scent of lavender might in the next moment transmit the vibrations 

 of chloroform and produce the chloroform smell. Such conduction resembles that 

 of a telephone at least in this that different words spoken at one end through the 

 same telephone can be heard unaltered at the other end. The assumption of a 

 conduction of specific forms of movement set up by odorous substances from the 

 periphery through the whole nerve-fibre up to the central organ, also renders it 

 necessary to assume that the olfactory nerve-fibres are not stimulated at all by 



