TESTING THE EFFICIENCY OF DEODORANTS 117 many possible loopholes, however, that I can spend only a few brief mo- ments quickly summarizing them: 1. It is possible that fresh perspiration does have an odor, either when it comes from all people or from some, and that the work determining that it is odorless cannot be verified experimentally. Killian defines sweat as "the mixture of excretory products, including water, of the coil glands" and goes on to point out that it "contains practically the same metabolites as does the urine but in significantly lower concentrations the principal solid components are sodium chloride, sugar, urea, ammonia, amino acids, lactic acid, uric acid, phosphates and sulfates." One must be very skep- tical as to whether a product answering such a description would be odor- ] eSS. 2. It is possible that during the sterilization of perspiration, other com- ponents thereof may have been removed or altered, thus causing changes that would not take place on the skin because of the use of a germicidal soap. As a result, the sterile perspiration may have been odorless, but for reasons other than the removal of the bacteria. 3. It is possible that sterile perspiration is odor]ess, but that perspira- tion having small quantities of bacteria therein does not have diminished odor, because of the fact that the bacterial decomposition may take place as a result of a catalytic action, with a small number of bacteria causing change just as obnoxious as a larger number of bacteria. 4. It is possible that the continued and continuous use of a soap con- taining hexachlorophene may result in the growth of bacterial strains resist- ant to this material, and that such strains may be responsible for bacterial decomposition causing odors as strong as or even stronger than those caused by bacterial decomposition when no bactericidal agent has been used.* 5. It is possible that the odor of the perfume in the soap or deodorant is sufficiently strong so that no group of olfactory-sensitive judges can deter- mine that there is less perspiration odor. Does this mean that we can draw a conclusion that the hexachlorophene claims are false? Not at all. It merely means that these claims are un- proved. Until such time as there is a body of acceptable evidence to show that hexachlorophene works as a deodorant, the claims based on the theory cannot be given serious consideration. Parenthetically, I would like to mention that there is, to my knowledge, one--and only one--published piece of literature that seems at first glance to verify experimentally the theory of deodorant action expounded by the advocates of germicidal deodorants. This is the work of Killian (3). * I am informed in a private communication from M. J. Shnitzler, Gillette Safety Razor Co., Boston, Mass., that unpublished studies have been conducted by that company on the skin flora of people who have been using a shaving cream containing hexachlorophene daily over a period of four years, and that no such strains have been encountered.

118 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Ironically enough, this information appears in an article devoted to chloro- phyll and its derivatives. This is peculiar because certain of the claims made for chlorophyll are contradictory to certain claims made for hexa- chlorophene. The theory that perspiration odor can be reduced by inges- tion of an internal deodorant is mutually exclusive with the theory that perspiration odor is caused by bacterial decomposition. The information in the Killian paper on germicidal soap is rather meager, it being only incidental to a much more complete study of chlorophyll. For that reason, the KillJan publication can hardly be considered a serious work on the study of the deodorant effectiveness of a bactericidal soap. The only factual data on germicidal deodorants appearing in the work of KillJan consists of information about an experiment in which it was demonstrated that when plain soap was used, it took 12 out of 15 subjects from 8 to 11 consecutive days of bathing to acquire "satisfactory protection against the development of perspiration odors," and that this same pro- tection was attained within a period of four days by all 15 subjects using a germicidal soap. Without going into a complete analysis of this small-scale experiment, it is interesting to note that it indicates that the germicidal soap has no prac- tical value as a deodorant. According to this study, daily washing with ordinary soap is as effective in protecting against the development of per- spiration odors as daily washing with germicidal soap, except that the point of greatest effectiveness is reached after fewer days with the germicidal than with the ordinary soap. Therefore, a soap containing hexachloro- phene, we are forced to conclude, would have deodorant value (over and above ordinary soap) only for people who have not been bathing in the past but who are going to start bathing from this point on and then such a soap containing the germicide would give better deodorant protection than the ordinary soap only for the first 8 to 11 days. A more devastating refu- tation of the pro-hexachlorophene argument is difficult to imagine. Why is the work on the testing of deodorants of such low caliber ? The reasons are twofold. First, some companies seem to have taken a short- sighted viewpoint. They feel that the claims, even if untrue and un- founded, will not do anyone a•y harm and may even do some good in promoting more hygienic habits. These companies do not understand that the public cynicism that must inevitably arise as a result of false claims and exaggerated advertising will cost them far more in dollars and cents than can be gained by ephemeral successes. A second source for the poor work is found in the inherent difficulty of verifying results. It is virtually impossible to repeat an experiment in the manner utilized in other scientific disciplines. As a result of confusion over method and disagreement over results, prod- ucts are now being marketed about which even the most enthusiastic