DEODORANT EFFICACY OF TCC 77 transferred TCC were determined after use of the soaps A, C, D, E, and F. Twenty-four female volunteers always received random pieces of the above-mentioned soaps on Fridays. They were obliged to use these soap bars as they usually use a soap for --1 week on Monday and Wednesday morning, samples were taken from the lower forearms to determine the casual level of TCC. Subsequently both arms were washed by rubbing the forearm directly under running tap water of --30øC with the assigned soap bar for two minutes, and another series of samples was prepared to determine the content of TCC at the skin surface. The following week the soap pieces were substituted by one of the other soaps. INFLUENCE OF THE DURATION OF WASHING AND OF REFATTENING AGENTS Soap D was compared with soap G (containing wool wax alcohols) as described above, except that seven volunteers used the soaps for ten seconds, seven volunteers used them for 40 seconds, and nine volunteers for 120 seconds. INTERRELATION BETWEEN THE LEVEL OF SKIN SURFACE LIPIDS, SKIN SURFACE ROUGHNESS, AND TCC TRANSFER The skin surface lipids on the inside of one forearm of nine volunteers were determined by solvent extraction (2). The skin surface roughness of contralateral sites was deter- mined by profilometry (3). Afterwards both forearms were washed for two minutes under running tap water of 35øC using soaps I and K respectively. The content of TCC was determined as described above. INFLUENCE OF THE CONTENT OF SOLUBILIZER ON TCC TRANSFER Twenty volunteers washed their forearms for two minutes under running tap water of 35øC using soap G and soap K (containing an excess of 5.26% nonylphenolpolyglycol ether). The level of TCC on the skin surface was determined as described above. SNIFF TEST COMPARING SOAPS A AND D A panel of 50 volunteers of both sexes was given a control soap to use to wash the axillae in the evening and morning for three days. The axillary odor of the volunteers was then assessed by two independent assessors using a numerical scoring system for odor intensity. On the basis of this assessment, the 30 volunteers with the highest axillary odor were selected and split into two test groups having approximately the same mean and range of odor levels. For the next three days one group used soap A to wash the left axilla and soap B to wash the right axilla, and the other group washed according to the reverse pattern. After three days the odor levels were reassessed and both groups went back to using the control soap for two days before returning for a third assessment. In the final stage there was a product crossover, i.e., each group reversed its washing pattern for three more days, with a fourth assessment at the end of this period. The results were expressed as the ratios of odor intensity between axillae for each volunteer at each assessment. This test was performed by Toxicol Laboratories, Ledbury, En- gland.

78 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS SNIFF TEST COMPARING SOAPS C AND H The investigation was conducted as a double-blind study: a) On the first day the volunteers, who had refrained from using deodorant products for at least 24 hours and had used a deodorant-free soap, washed both axillae with a de- odorant-free placebo soap (= H). After this standardization procedure, the volunteers placed a ball of pure cotton under each axilla, wore a T-shirt made of cotton (washed with perfume-free detergents), and sat quietly for 30 minutes in a room with a tempera- ture of--30øC. After this period of slight sweating, the cotton balls were removed, sealed in a glass container free of noticeable smell and stored for 1-3 hours in a refriger- ator at 2øC. Thirty minutes before evaluation the glass jars were removed from the refrigerator and allowed to warm to room temperature (20øC). Two technicians experi- enced in this kind of investigation and with sensitive noses evaluated the atmosphere in every container according to quality and quantity of odor. The following scores were used to describe the odor: 0 = no odor 1 = odor just detectable 2 = odor of slight intensity 3 = odor of moderate intensity 4 = odor of strong intensity b) On the second day this procedure was repeated, except that soap C (containing 0.13% TCC) and soap H were used (random application sites). c) The scores (sum of two individual scores) characterizing each sample were compared by the paired t-test. RESULTS CONTENT OF TCC "SANDY" GRAINS The following procedure was used to determine whether the grains on the surface of the "sandy" soap cakes B had a high content of TCC and whether the inhomogeneity there- fore consisted mainly of deodorant: grains isolated from the surface of soap B contained aproximately 40 times the amount of TCC found in soap A. SPATIAL DISTRIBUTION OF CI AND Na IN SOAP CAKES B, C, AND F Figure 1 shows the distribution of chlorine--characterized by the little white spots-- in soap B, and Figure 2a presents the result of an analogous analysis of Soap F. The uneven distribution of chlorine in soap B is easily recognized when soap B is compared with soap F, i.e., the solubilization of TCC has the expected result. The low intensity of chlorine in Figure la compared with Figure 2a--both products contain 1.3% TCC --is probably a consequence of aggregation within soap B that inevitably will cause a decrease in dissolved or finely dissipated powdered compound. The presumably linear relationship between the number of white spots and the content of TCC can be esti- mated by comparing Figure 2a with Figure 3a, which represents the distribution and quanity of chlorine within soap C the reduction of solubilized TCC by a factor of 10 is expressed very well by the less intense, but even, distribution of chlorine. In contrast to chlorine, sodium is distributed very evenly in exactly the same area as used for the