PRIMARY IRRITATION OF THE SKIN 22l used for testing. Neither the subjects nor the observers are aware of the identity of the test materials applied to particular areas of the skin. Patching procedure Using a disposable syringe of I ml capacity fitted with a size 0 hypo- dermic needle, 0.05 ml of test material is transferred to the lint square (Fig. 2). A maximum of six materials can be tested simultaneously on a patch 750 x 140 mm. The patch is placed in contact with the skin and remains there for 5 h. At the end of this time, the positioning of the corners of the patch is marked either with 0.5•o aqueous crystal violet or with dihydroxy- acetone to ensure precise relocation of the test site on subsequent days. The patch is then removed and the site is swabbed with warm water to remove any residual test material. 1 h later the test sites are assessed for irritation by a panel of three observers who record their scoring in- dependently. The interval between removal of patches and assessment helps to ensure that assessment will not be influenced by wrinkling or reddening due to the adhesive tape or to the process of removing it from the skin. This procedure is followed on 5 successive days. Assessment of irritation Assessment is carried out under standardized diffuse lighting conditions*, according to the following scoring system: 0 = No visible redness. 1 2 --• Slight redness. 3 4 = Distinct redness. 5 6 = Strong reaction (possibly some oedema). 7 8 := Very strong reaction (with blistering). The mean of the scores awarded by the three observers is taken as the daily score for a test material. If any material yields a score of 5 or more on a particular day of the test, patching of this material is not repeated on subsequent days. On the remaining 'no patch' days, the maximum score of 8 is awarded to the test material. *Koto-rite fluorescent tubes ex Mazda Lamps, England.

222 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS At the end of a test series, the daily scores are summed to give an overall score for each material on each subject. The mean overall score for the whole panel is then calculated. Comparisons between test materials may be made by subjecting the overall scores to Wilcoxon's Signed Ranks Matched Pairs Test (3). Interpretation Interpretation is based on the same numerical ratings as suggested for the mouse ear test, e.g. an overall score of 15 or more is likely to result in some irritation complaints from users under certain circumstances. When interpreting the results of 'closed' patch tests, the intended mode and frequency of use must be considered. Many products are used intermittently and are rinsed from the skin after brief contact, e.g. shampoos the level of complaints will probably not be unacceptable unless an irritation score of 25-30 is obtained. Products intended to have a long contact time on the skin, especially if used on infant skin, on the face or on areas normally occluded, will probably be unacceptable if the score exceeds 15-20. RESULTS Experiment I Agreement between observers in mouse ear test A range of test materials was applied in a 5 day test as described above, 10 I•l of each material being applied daily to one ear of each of five CF1 female mice. Two observers assessed the level of irritation independently, after a series of joint assessments to establish agreement on a scale of nu- merical ratings (Table I). These data indicate that the two observers gave the same rank order for the test materials and that their scoring for mouse ear irritation showed Table I Agreement between two observers in mouse ear test Total score throughout 5 day test Observer Shampoo S Shampoo C Ethyl oleate (undiluted) (undiluted) in ethanol (70: 30) 1 16.8 11.6 13.0 2 18.8 10.4 15.8