226 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS tion level it is now interesting to see from Table HI that this level corresponds to a basic p of 0.5 in other words, this is the kind of level of detectability at which a panel member is as likely to say a difference exists as often as not whenever he judges it. This seems to be a very logical sort of level to take for the base line kind of differences below which one does not want to bother to know. Having built this known difference into the test set, then the operation calls for panel members to group the samples into as many different groups as they can detect. Any samples which are grouped together more frequently than the difference between the Bulk and the Contaminated Bulk (to use the terminology used in the test situation) are then indistinguishable the test samples should be indistinguishable from each other and a control sample. This same technique is combined with some multivariate statistical analyses employing a mathematical procedure known as the method of Principal Components to enable meaningful interpretation to be put on the way in which the samples are grouped together. It is not proposed to discuss this further in this paper except to indicate that the techniques we have described have considerable potential for further use and development. CONCLUSION This paper has been written to show that the close collaboration of the statistician and the cosmetic chemist in this difficult field of sensory assessment can be of considerable mutual benefit. In particular, I would lay stress on the following points:- (1) This is a field in which the concepts of probability are of vital import- ance - the probability of making choices regardless of any discrimina- ting power as compared with the probability to be found when making discrimination. (2) This is a field in which it is very easy to lose sight of the essential purposes of the tests unless great care is exercised by both statistician and the cosmetic chemist. In particular, it is too easy to assume that too much has been proved by a very limited test simply because it makes life complicated to believe otherwise. (3) This is a field in which the statistician can use the tools of his trade to get out a lot of hidden information, provided he can suggest the experimental designs before the work is done, not afterwards. (Received: 7th February 1970)

SENSORY TESTING -- A STATISTICIAN'S APPROACH 227 REFERENCES (1) Fisher, R. A. Design of experiments 6th Edition. 11 (1935) (Oliver & Boyd, Edinburgh & London). (2) Bengtsson, K. and Helm, E. Principles of taste testing. Wallerstein Lab. Comm. 9 171 (1946). (3) Lockhart, E. E. Binomial systems and organoleptic analysis. Food Technol. 5 428 (1951). (4) Green, M. W. A note on the use of triangle design in taste testing of pharmaceutical preparations. J. Am. Pharm. Assoc. Sci. Ed. $$ 380 (1955). (5) Fourman, V. G. Taste panels for pharmaceutical flavors. Dr,g Cosmetic D•d. 77 762 (1955). (6) Ostle, B. Statistics in research. 58 (1954) (Iowa State College Press, Iowa). (7) Hardes, J. M. Sensory tests and consumer acceptance. J. Sci. Food Agric. • 477 (1953). (8) Gridgeman, N. T. Sensory item sorting. Biometrics, 15 298 (1959). (9) Richardson, E.G. The science of orchestral instruments: Some recent •vork. Discovery 1• 87 (1953). DISCUSSION MR. A. ELLIOT•r: At this Symposium three distinguished speakers have recom- mended three different methods of selecting perfumes for new cosmetic products. Mr. Erni (10) suggested it should be the perfumer's responsibility to select the perfume, Mr. Landon {11) suggested the marketing manager should make the final choice, and finally you suggested that the statistician, with the aid of panel tests, should choose the perfume. The suppliers of perfume compounds are anxious to point out that a distinctive perfume will probably only be liked by about 1-o/ of the population and yet still be O/o a commercial success, whereas a perfume which is generally acceptable is not distinc- tive and hence will not sell the product. In view of this, can you tell me of a method of panel testing which will select a distinctive, rather than generally acceptable, perfume for new products? THE LECTURER: It is not my view that the statistician should have anything to do with the actual selection of the perfume. It should be the potential customer, when all is said and done--and all that the statistician can offer to do is to help both the cosmetician and the marketing man to find the best way of obtaining this information about the potential customer. The question of hoxv one goes about looking for the proportion of people that are going to find the perfume acceptable is very involved and it depends to what group you are aiming your product. It is not a statistician's function to answer (only to ask), and what he can do is simply to draw up rules for saying--if you are making decisions like this (which are, after all, commercial decisions or, possibly, perfumery decisions) what kind of samples you have to draw, what kind of numbers of people you are going to have to interview before you have some reasonable assurance of having met that target. Without talking about specific cases (and I do not think there is much point in doing that here), this is a question of laying down specifications. Ma. C. SEBLE¾: How much rechecking takes place witIx your panel, not so much from the point of view of sensory fatigue, but from a "preference" angie? (10) Erni, M. Evolution in perfumery. Presented at the Symposium on "Perfumery" at Eastbourne, on 21st April 1970. (11) Landon, M. F. The application of perfume to cosmetics and toiletries. Presented at the Symposium on "Perfumery" at Eastbourne, on 20th April 1970.