76 SMELL, TASTE, ALLIED SENSES 



their intermediates and mixtures. So far as an arrange- 

 ment of odors is concerned the clarity of Henning's 

 scheme is at once its most attractive and most suspi- 

 cious feature. 



8. Chemical Relations of Odors. The scientific 

 value of any classification of odors will depend upon the 

 success with which such a classification brings the odor- 

 ous substances as stimuli into relation with the receptor. 

 A satisfactory classification ought to make evident the 

 number of elements or components concerned in olfac- 

 tion. That olf action is made up of a number of compo- 

 nents is far from established, but what may be called 

 the component theory of olfacton is generally assumed 

 by the majority of writers on this subject Zwaardemaker 

 (1895). That the classification outlined by Zwaarde- 

 maker shows very little of this feature is readily admit- 

 ted even by this author himself. Quite aside from the 

 fact that it may include irritants as well as true odorous 

 substances, its classes do not stand up well under experi- 

 mental test. Nagel (1897) tested this question in an 

 investigation of the odors of vanillin and cumarin. 

 These two substances, according to Zwaardemaker 's 

 classification, belong not only to the same class of bal- 

 samic odors but to the same subdivision, the vanilla odor. 

 They ought, therefore, to show considerable olfactory 

 similarity. Nagel attempted to test this relationship by 

 ascertaining whether the temporary exhaustion of the ol- 

 factory organ by one of these substances would influence 

 its receptive capacity for the other. To carry out this 

 he prepared an aqueous solution of the two substances 

 in such proportions that the smell of only vanillin could 

 be recognized. He then exhausted the olfactory organ 



