J. Soc. Cosmetic Chemists, 16, 515-596(1965) Success and Failure of Odor Classification as Applied to Reactions to Erogenous Odors ERNST PAUKNER, Ph.D.* Presented before the Third Congress of the I.F.S.C.C., June •1-•6, 19•, New York City Synopsis--One of the problems of successful perfuming is the difficulty of obtaining from consumers an indication of their ideas about the ideal fragrance, appropriate to the product under consideration. An interviewing method is described which reliably defines the con- sumer's ideal of the desired perfume type. This method (semantic differential) is illustrated with a description of the concept of an "erogenous fragrance." Several single odorants, at different concentrations, and some well-known perfumes are included in the test. The data are analyzed by factor analysis to determine which of the •naterials comes closest to meeting the expectation of the ideal. The Henning and the Crocker-Henderson methods, traditional systems of odor classification, are evaluated and shown to be less suited for the determination of odor qualities. INTRODUCTION Early this year, it was reported that work is in progress in India and England on a perfume which is to serve as an adjunct to birth con- trol measures (1). This perfume, when used by Indian women, is supposed to have the same kind of effect on Indian men as a repellent has on mosquitoes: it is, therefore, an antierogenous perfume. When he smells it, "Romeo no longe. r longs for Juliet," according to B. L. Raina, Director of the Indian Institute of Family Planning. On the other hand the concern of the perfumers in the Western world has been the * Laboratories of drom, Bertelestrasse 75, M/inchen-Solln, Germany. 515

516 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS creation of fragrances which will make Romeo long for Juliet. In this paper the results of an investigation of the psychology of erogenous fragrances will be presented. Consider, for example, the case of a manufacturer who is trying to attain broad public acceptance for a line of cosmetics. He is searching for an erogenous fragrance. This fragrance should give the woman who uses products with this fragrance the conviction that she is irresistible. With the help of his marketing research department or an outside consultant organization, the manufacturer will be able to determine the most appropriate advertising theme, the most enticing package and the most effective merchandising approach. He can also find out whether, for example, a product in powder form or a liquid would be more con- vincing. Modern interviewing techniques can assess the feelings of po- tential customers in questions of this kind quite accurately. However, up till now the cosmetics manufacturer has had no way of getting a de- scription of the most important thing in this case: the most effective fragrance for his product. The reason why market researchers find it so difficult to establish the character or message of a perfume is the lack of words to describe this message. It is easy for a respondent to state in an interview that there should be a little less blue in a shade of violet which is meant to harmo- nize with an erogenous product. But how could he describe how a per- fume should be changed in order to meet his expectations ? Experienced market research specialists indicate that they know of no reliable method to determine either the impressions and feelings evoked by a fragrance or the consumer's image of the ideal fragrance for a given product. As a result of the difficulty in finding out what the consumer wants, rather haphazard methods are still being used in fragrance selection. Instead of asking potential purchasers what the odor of cosmetic prod- ucts should be like, the procedure often consists in selecting among the many available perfume oils a few that are considered appropriate. Next, a few people with responsible positions in the company (or their wives) decide which of these the public will like best. This method is unsatisfactory, since too many purely accidental factors are involved. In order to find a better approach a research pro- gram on the psychology of odor was started in these laboratories several years ago (2, 3). Lately, the specific aim in these laboratories has been the search for a reliable classification of odor characters and for an exact way of expressing similarity or differences between odors. If one looks for an odor which comes as close as possible to an ideal image, one has to