282 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS immediately after wiping, giving the maximum water content or hygroscopicity. Then the measurement was repeated at intervals of 30 seconds for three minutes, measuring water-holding capacity (7). 2. Prehydration state, hygroscopicity, and water-holding capacity on moisturizer-applied skin (immediate and delayed observation). We selected two square areas, 5 x 5 cm 2 each, at the middle part of both forearms the upper line of the square was 5 cm below the ante- cubital fossa. Using the syringe, we applied 10% urea cream on the first square of the left forearm, 2 •l/cm 2, and spread it evenly by a glass rod. After waiting five minutes, we observed an immediate result by measuring the water content of the prehydration state (baseline). Then we applied water for ten seconds and removed it carefully, trying not to remove the agent. Then hygroscopicity and water-holding capacity were mea- sured in the same manner as in the first test procedure. The same method was used to measure other agents: cream base was applied on another square area of the left forearm, and 5% lactic acid cream and pH5-Eucerin were applied on square areas of the right forearm. Prehydration state, hygroscopicity, and water-holding capacity were measured in the same manner (Figure 2). After that, we observed the delayed result by measuring the prehydration state, hygroscopicity, and water-holding capacity of all four agents three hours after application. STATISTICS The Kruskal-Wallis test and least significant difference (LSD) analysis were used to compare the results and analyze data. RESULTS There were 109 volunteers, 24 males and 85 females, ages 20-44 (average age, 30.96), who completed the experiment. Two women were excluded from the test because they were pregnant and hormonal changes might have interfered with the result. WATER CONTENT OF NORMAL SKIN We found that the prehydration state, hygroscopicity, and water-holding capacity of the normal skin of both forearms were approximately the same and did not have any statistical difference. The water content of the skin increased sharply after water appli- RIGHT LEFT Lactic aci .--Urea cream pHS-Eucerin Cream base Figure 2. Site of application of testing agents.

WATER-HOLDING CAPACITY OF MOISTURIZERS 283 cation, showing high hygroscopicity. Most of this increase was lost within 30 seconds, followed by a gradual return to a prehydration level in 180 seconds (Figure 3). WATER CONTENT OF THE SKIN SURFACE AFTER APPLICATION OF MOISTURIZERS From statistical analysis, we found that the distribution pattern of the data was not in a normal distribution. Therefore, the analysis system that we used to analyze the data is a non-parametric test (Kruskal-Wallis test) to show the differences between the groups. From the Kruskal-Wallis test it was shown that the differences between the groups had a statistical significance of P 0.00001. Although the Kruskal-Wallis test can test the hypothesis that there is at least one group that differs from the others, it is unable to determine which specific group has a real difference in relation to another. Therefore, the least significant difference (LSD) analysis can be useful to determine comparatively how different each group is from another, or, in other words, we can rank the groups according to the value of interest by comparing the value of one group to that of another pair, by pairs, until every pair is analyzed. We then used a numeric number to replace the testing agents: 1 = normal skin as a control 2 = 10% urea cream 3 = cream base 4 = 5% lactic acid cream 5 = pH5-Eucerin Five minutes after the application of agents we used the least significant difference (LSD) ...w.s 1000 800 600 400 200 0 30 60 90 120 150 180 sec. i RIGHT FOREARM • LEFT FOREARM Figure 3. Hygroscopicity and water-holding capacity pattern of pretreated normal skin, right and left forearm.