MODIFIED FCAT 155 a 21-day cumulative irritation test (6). In a four-day patch test (internal company protocol using semi-occluded patches on four consecutive days), these products are comparable to the negative irritant controls (water and isotonic saline). The products are non-irritating and are negative for sensitization and in human repeat insult patch tests (7). Forearm washing protocol. In tests involving comparisons of baby wipe products, baby wipe application consisted of repeated washing cycles of the defined test sites with the test products. Washing sessions were 1 • hours apart, and were conducted four times each day, with the exception of the final day of the test, when one or two washing sessions were conducted. For each washing cycle, a baby wipe was folded twice, and wiped in a back and forth motion with light pressure on the test site for either 15 or 30 seconds. In some studies, the baby wipe was refolded to expose an unused portion, and the back and forth wiping was repeated. In some studies, the duration of the washing procedure for the last washing session of each day was doubled by using a second baby wipe and repeating the two-step washing procedure. The technicians conducting the washing procedure were instructed to take care to apply equal pressure to all test sites. A study of specific differences in the washing protocol are given below in the Results section. BOTH PHASES Subjects. For each study, the protocol was approved by the test facility's Institutional Review Board. Participants in all studies were healthy adult volunteers, ages ! 8-65, who had signed an informed consent form. Subjects were excluded from participation if: (a) they had allergies to soaps or fragrances, (b) they were taking anti-inflammatory corti- costeriods or other medications that may interfere with test results, (c) they had active dermatitis, (d) they were pregnant or lactating, or (e) they had sunburn, acne, scar tissue, or other skin abnormality at the test site. In addition, for experiments during the FCAT modification phase, subjects were excluded if they had participated in a forearm patch test within four weeks prior to the study. Initial studies used both male and female subjects. However, the relatively high amount of body hair on the forearms of the male subjects made visual scoring for erythema and dryness difficult. Therefore, the final two studies used female panelists only. For each panelist, the forearms were divided into upper and lower test sites. These sites were at least 2 cm apart, and 2 cm from either the elbow or the wrist. Test patch systems or products were assigned to test sites using a randomization schedule. Panelists in each study were given the same brand of bar soap to use for personal cleansing and instructed not to use lotions, cremes, or moisturizers on the test sites, nor to wash the test sites during bathing throughout the duration of the test. In those experiments where modi- fied diaper patches were used, subjects were instructed to put a plastic bag on the arm to cover the modified diaper patch while bathing or showering. Evaluation of skin condition. In these experiments, transepidermal water loss (TEWL) was measured using a ServoMed Evaporimeter EP ! © employing standardized procedures (8). This instrument concurrently measures evaporation rate (range 0-300 g/m2hr), relative air humidity, and water vapor partial pressure. Readings were taken in triplicate. All measurements were taken after the subjects had spent a 30-minute acclimation period in the test facility's controlled temperature and humidity room. Specific room settings

156 JOURNAL OF COSMETIC SCIENCE varied slightly, but all were between 66 ø and 72øF, and 30% and 50% relative humid- ity. In all studies, individuals with a TEWL reading of 7.5 g/m2hr at baseline (i.e., prior to the first patch application) were excluded from the study. A Minolta Chromameter CR-200 © was used to score redness. A complete description of the use of this instrument for similar applications has been given previously (9). Briefly, this instrument reads color in a three-dimensional format along coordinates L*, a*, and b*. Coordinate L* indicates the level of brightness between black and white. Coordinate a* indicates the red-green balance (with 100 being red only and - 100 being green only). Coordinate b* indicates the yellow-blue balance. For this program, the reading along the a* coordinate was of greatest interest since higher a* values indicate greater irritation. Measurements were obtained by gently resting the aperture of the measuring probe on the skin surface. The L*a*b* values were directly inputted to a personal computer. Triplicate a* readings were taken at each test site. Visual scoring was conducted by expert graders under a 100-watt incandescent daylight bulb. For each study, a single expert grader was used throughout the study. Persons scoring the test sites were unaware of the specific treatment at each test site. In the first study in the FCAT modification phase, dryness and erythema were graded on scales of 0-4, with "0" indicating perfect skin and "4" indicating a severe reaction. In all other studies, these scales were expanded to the seven-point scales given below. Erythema grading scale 0 = none 1 = barely detectable redness 2 = slight redness 3 = moderate redness 4 = heavy or substantial redness 5 = severe redness 6 = extreme redness Dryness grading scale 0 = none 1 -- patchy, slight powderiness, small scales 2 = general, slight powderiness, small lifting scales 3 = general, moderate powderiness, cracking and scales 4 = general, heavy powderiness and cracking, lifting scales 5 -- heavy cracking (possibly bleeding) and lifting scales 6 = severe scaling, bleeding cracks, sloughing of large scales Fractional increments were assigned to slight differences or intermediate responses when two test sites were compared at a single grading point. Statistical analyses of data. In the patch development phase studies, the Student's t-test was used to compare two treatments. When three or more treatments were being evaluated, an analysis of variance (ANOVA) was used. If the ANOVA indicated that the groups were significantly different, a Student's t-test between treatment pairs was conducted. In the first experiment in the FCAT modification phase, instrumental measurements were evaluated by a repeated ANOVA run on the values obtained from the treated sites