336 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS t o t o . •' t• t•. •' t Figure 3. Recording of the phenomenon. At time t o bottom of the "sampling chamber" is pulled up. Three minutes after, the system has reached equilibrium. At time t• the bottom is pulled down AF• -- AF0 (frequency difference variation of the cavities) is proportional to the TEWL value. Experimental protocol The measurements were carried out at room temperature (22-25øC) and humidity (35-45%). These physical parameters (humidity and temperature) were recorded continuously during the measurements. Each volunteer was taken into the laboratory one hour before the measurement was made and encouraged to relax in order to be in equilibrium with his new environment. The measurements were carried out four times successively on the dorsal side of the forearm, but the first recording was systemati- cally rejected because stress can alter this result. After obtaining these first values, which characterize the TEWL of the patient, we applied the product under examina- tion to the skin. In the case of measurements on emulsions, it is essential that they are carried out 1 hr after application in order that the water in the emulsion has had time to evaporate and is consequently not included in the TEWL value. Previous measure- ments of water loss of the emulsion applied on a plastic plate have shown that 1 hr was sufficient for allowing water to evaporate. TRANSEPIDERMAL WATER LOSS VALUES With the previously described device, we have measured the TEWL on 21 volunteers from our laboratories. The results appear in Table I. The third column shows the averaged value of the three measurements the last column represents a simplified clinical statement on the state of the skin. For all 21 volunteers the mean value of TEWL is 0.61 mg/cm2/hr, and the standard deviation is 0.3 mg/cm2/hr. The data already reported in the literature do not allow a precise and quantitative description of the phenomenon (7,8,10). Very different experimental conditions under which the various results were obtained could be an explanation. Two remarks can be made: ß the published values range from 0.3 to 1.2 mg/cm2/hr.

TRANSEPIDERMAL WATER LOSS 337 Table I TEWL Values on the Forearm of 21 Volunteers N ø TEWL Sex Age mg/cm2/H Type of Skin 1 F 41 0.63 D 2 F 44 1.17 D 3 M 33 0.44 N 4 M 35 0.52 N 5 M 26 0.56 D 6 M 39 0.63 N 7 F 32 0.56 N 8 F 25 0.59 N 9 M 32 0.42 N 10 F 31 0.66 N 11 M 29 0.63 N 12 F 27 0.53 N 13 F 27 0.36 N 14 F 21 0.36 N 15 F 40 0.63 N 16 M 28 0.63 D 17 M 28 0.50 N 18 F 21 0.80 N 19 F 35 0.88 D 20 F 47 0.49 N 21 F 54 0.87 D Mean value ................................ 0.61 mg/cm2/H Mean value DS ............................. 0.75 mg/cm2/H Mean value NS ............................. 0.54 mg/cm2/H ß some authors find that the given values represent a rather wide average range, while for others it is a reproducible characteristic. Our own values of TEWL are in good agreement with those published by Grice et al. (4) and Spruit (3). The wide range shows the great diversity in the physical and chemical properties of S.C. Though the sampling was not enough to be statistically significant, we could note, as a trend, that "dry skin" had the highest TEWL values (0.75 mg/cm2/hr). It is worth noting that the low values published for the TEWL have been obtained by measurements at 0% RH. In these conditions the S.C. desorbs very slowly [see for example Scheuplein (9)] and it is difficult to know when its equilibrium state is reached. Moreover for the diffusion process there is a balance effect between the decrease of the diffusion coefficient due to the S.C. drying and the increase of the water vapor pressure gradient. These difficulties explain the conflicting results obtained for TEWL at 0% RH and the large range of values for TEWL at 0% RH given in the literature. PRODUCT INFLUENCE ON THE TEWL ANTIPERSPIRANT EFFECT Baker and Kligmann explained that their own results were due to the anticholinergic treatment applied to inhibit the perspiration (7). They believe that above 20øC