SKIN PERMEABILITY AND WATER LOSS 353 300 I•l) with an ethanol-water mixture (95:5), followed by two rinses (2 x 300 I•l) with distilled water and gentle drying with a cotton swab. Measurement conditions. Utilizing literature data relating to the kinetics of urinary excre- tion of benzoic acid administered intravenously and orally (16, 17) or percutaneously (18, 19,20) in different species, the total quantities absorbed during the four days fol- lowing application were calculated, after counting in scintillation liquid the quantities found in the urine in 24 hours, which contains 75% of the total quantity absorbed. TRANSEPIDERMAL WATER LOSS After completion of the benzoic acid treatment, TEWL was measured with an Evapo- rimeter EPIC (Servo Med, Sweden) from a contralateral site (same anatomic region) in each subject. The handheld probe was fitted with a 1-cm tail chimney extension to reduce air turbu- lence around the hydrosensors and metallic shield (supplied by Servo Med), eliminating the possibility of sensor contamination. Measurements (gm' m -2' h -1) stabilized within 30-45 seconds. Immediately following TEWL measurements, skin temperature was taken with a surface probe (Tele-Thermometer, Yellow Springs Instruments Co., USA). All TEWL were corrected to a common skin temperature of 30øC, reducing this thermal variable (21). Since the room environment was comfortable (room temperature 20øC, relative humidity 60%) and the subjects were physically inactive, the TEWL should closely reflect stratum corneum water flux without sweating interference. RESULTS Table I shows, for each site studied, the quantities of benzoic acid present in the urine at 24 hours, as well as the calculated total quantities absorbed over a period of four days. The table also gives the values of the TEWL corresponding to each site. As shown in Figure 2, for all the sites taken together a linear relation exists between the values of the total penetration of benzoic acid and the values of the TEWL (Figure 2) (r = 0.97, p 0.001). DISCUSSION Although it is widely accepted that anatomic site plays an important role in the degree of absorption of molecules and in TEWL, the literature contains no data defining their relationship. Our results (Table I, Figure 2) show that for the anatomic site studied and within the range of TEWL and penetration values determined, there exists a direct relation be- tween the permeability of the skin to water and the percutaneous absorption of a non- water-soluble compound, such as benzoic acid. Thus it can be seen that the values for both TE¾7L and benzoic acid penetration differed significantly for the arm, the ab- domen, and the forehead, and are linearly correlated. However, even if it is true that no statistically significant differences have been found between the values reported for the

354 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Table I Percutaneous Absorption and TEWL Values According to Anatomic Site Percutaneous Absorption of Benzoic Acid Urinary Excretion Total Penetration 0-24 h In 4 Days (nmoles ß cm- 2) (nmoles ß cm- 2) TEWL Anatomic Site (a) (b) (gm ' m -2 ' h-•) Back 6.41 8.55 4.51 (c) (0.99) (1.32) (0.57) Arm 7.06 9.41 4.34 (0.77) (1.02) (0.07) Chest 8.78 11.70 4.73 (0.98) (1.30) (0.26) Thigh 9.38 12.50 4.39 (1.07) (1.43) (0.32) Abdomen 11.44 15.26 4.99 (1.44) (1.93) (0.38) Forehead 20.74 27.65 8.12 (2.71) (3.61) (0.64) (a) (b) calculated values from (a): -- O.75 (c) S.D. thigh and the chest, it is worth noting that these values do fall on the correlation plot obtained and thus confirm it. The cutaneous permeability of benzoic acid varied according to the site in the following manner: back arm chest thigh abdomen forehead. These results are in excellent agreement with previous studies (13,18) involving various molecules of different chemical structures, and confirm that the forehead displays the greatest permeability. The laws describing diffusion through a membrane accord a major role to membrane thickness. These general laws have been applied to several biophysical phenomena, including TEWL and percutaneous absorption. However, examination of the literature shows that it is not unusual for these laws, predictated by pure mathematical logic, to be found wanting when applied to a discontinuous membrane of great physicochemical complexity such as the stratum corneum. Similarly, our results show that the forehead is about two to three times more perme- able to water and to benzoic acid than the other sites studied here, despite the fact that the SC thicknesses are comparable in all cases (12 [x on average for 18 layers) (22,23). This example seems, therefore, to run counter to the inverse relation which should exist between permeability and membrane thickness. Hence, simple consideration of the thickness of the SC cannot of itself explain the differences in TEWL and penetration observed between anatomic sites. Other criteria must be considered. By virtue of its density of active sebaceous glands, the forehead is the richest of all those studied in terms of sebum, which forms an uneven film of thickness varying form 0.4 to