82 SMELL, TASTE, ALLIED SENSES 



the hairs, a process that must precede the initiation of the 

 olfactory nerve-impulse. The solute may be any one 

 of an immense variety of substances whose primary char- 

 acteristics are that they are not only soluble in water but 

 also in oil. The amount of these substances necessary 

 for olfaction even in the case of the least odorous of 

 them is very small and in that of the most odorous in- 

 credibly small. The amounts that are usually estimated 

 for olfaction are those contained in what is believed to 

 be the minimum volume of water or of air necessaiy 

 for stimulation, but of the very minute amount of odor- 

 ous substance contained in this volume only a veiy small 

 fraction of it can reach the olfactory hairs. Much must 

 be carried away in the general current or left stranded 

 on non-olfactory portions of the nasal surfaces. Whether 

 the olfactory hairs can concentrate this material or not 

 remains to be ascertained, but even assuming that they 

 can, the effective concentration must be of an extremely 

 low order. 



The substances thus brought in solution into the ol- 

 factory hairs must initiate those nervous changes that 

 eventually produce the olfactory sensations. There 

 ought, therefore, to be some relation between these sub- 

 stances and the resulting sensations. It is generally 

 assumed that the substances that act as olfactory stimuli 

 fall into classes associated with corresponding classes 

 of sensations. As already indicated this conception may 

 be called the component theory of olfaction, and if we 

 assume, for instance, that the six classes of odors dis- 

 tinguished by Henning are separate classes, a view that 

 Kenning, however, opposes, then these classes would 



