232 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Microbiology The modified Williamson and Kligman technique (the double-scrub method) (13) was used to identify and quantify organisms from the mid-forehead and cheeks. An area of 4 cm 2 was scrubbed twice using a sterile cotton swab moistened with 0.1% sterile Polysorbate 80 (Tween © 80, VWR Scientific, Philadelphia, PA) in 0.075 M phosphate buffer (pH 7.9) for a period of 20-30 seconds. The swab was immediately placed in a test tube containing 2 ml of the above detergent solution. Tenfold dilutions of the pooled scrub fluid were prepared using 0.05% Polysorbate 80 in 0.0375 M phosphate buffer as the diluent. Microorganisms were identified and counted by plating 50 Ixl of appropriately diluted pooled scrub fluid on Schaedler's agar medium. For aerobes, the media were incubated at 37øC for 48 hours. Incubation under anaerobic conditions was used to quantify Propionibacterium acnes. Assessment of dryness/scaling D-squames. D-squames (CuDerm Co., Dallas, TX) are adhesive-coated, clear discs de- signed to collect individual corneocytes or cell aggregates from the desquamating por- tion of the outer horny layer. These are now widely used in the quantitative assessment of dry, scaly, xerotic skin (14). Test sites on the forehead, cheeks, and legs were first delipidized by wiping with gauze soaked in ether-acetone (1:1). D-squames were then firmly applied for 5-10 seconds, followed by careful removal and placement on storage cards. D-squames were assessed by means of image analysis. The degree of scaling was quantified by integrating two parameters: a) the percentage area covered by scales and b) the distribution of scales according to five thickness levels to yield the desquamation index (14). Visual assessment. The lower legs, cheeks, and forehead were visually assessed for dryness/ scaling by using a 0-4 scale: 0: None 1: Mild = slight scaling 2: Moderate = small scales, minimal flaking 3: Moderately severe = larger scales and flakes 4: Severe = many large, loose scales with cracking and fissuring STATISTICAL ANALYSIS Data analysis was carried out using the two-tailed Student's t-test. A p-value 0.05 was considered significant. RESULTS TRANSEPIDERMAL WATER LOSS The TEWL data are presented in Figure 2. TEWL was lower in blacks, reaching statistical significance at p 0.05 on the legs and cheeks. TEWL in blacks was also lower on the forearm, but the difference was not statistically significant.

COMPARISON OF BLACK AND WHITE SKIN 233 22 16.5 11 [' 5.5 0 Cheeks Blacks Whites Forearms Legs Figure 2. Transepidermal water loss (mean + SD). Asterisk indicates significant difference (p 0.05). HYDRATION The DPM data that reflect skin hydration are summarized in Table I. The initial level of hydration was found to be higher for blacks on the cheeks (p 0.05). Differences between the races were not statistically significant on the forearms and legs. The level of water buildup under occlusion by the probe was significantly higher for white skin on the forearms (p 0.05) and directionally higher on the legs (p 0.06). Racial differences were not detected on the cheeks. Regression analysis of the slope revealed lower rates of moisture buildup on the forearms (p 0.05) and legs (p 0.05). The cheeks did not show statistically significant racial variations (Figure 3). Table I Differences in Capacitance Readings of Black and White Skin Initial reading (mean + SD) a Reading at 60 seconds (mean + SD) b Site Blacks Whites Blacks Whites Cheeks 141.37 + 25.55 119.03 + 23.72* 454.20 + 148.58 413.87 + 153.81 Forearms 107.60 + 15.62 115.53 + 21.64 144.33 + 46.86 175.13 + 72.55* Legs 96.43 + 6.59 97.80 + 7.27 113.4 + 21.49 130.33 + 35.19+ Initial level of hydration. Moisture buildup in 60 seconds. *,+ Significant difference between black and white skin at p 0.05 and p 0.06, respectively.