358 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS The K values for the sorption isotherms at 37øC were calculated according to Bottari (!) using the following equation: Ck K- Cs where Cs is the equilibrium concentration of the sunscreen in the solvent and Ck is the concentration of the sunscreen in the keratin substrate. The percentage of substantivity was calculated as indicated for the in vivo method. STATISTICAL ANALYSIS The differences in substantivity obtained in the in vivo method were examined using an analysis of variance (ANOVA) (7) and the test of Dunnett (8). RESULTS Table I shows the percentages of substantivity obtained for the three sunscreens in the in vivo studies. Tables II and Ill contain the ANOVA of the data and application of the Dunnett's significant differences test, respectively. Table IV shows the mean K values for different concentrations of PABA, Escalol 507 ©, and Amerscreen "P"©. In Table V a comparison of the solubilities in water and percentages of substantivity of PABA and its esters determined by the in vivo method can be seen. Table VI shows the percentages of substantivity obtained for the three sunscreens by the in vitro method. It can be seen that there are very significant differences between the substantivity found for Escalol 507 © and those obtained for the other two substances but that there are no significant differences between PABA and Amerscreen "P"©. DISCUSSION The mechanism of the interaction of the sunscreen substances and the skin have not been established clearly. It is known that there are several factors involved. Sayre eta/. Table I Substantivities of PABA and Its Esters Determined by the In Vivo Method Substantivity (%) Subject # PABA Escalol 507 © Amerscreen "P"© 1 0.29 57.5 0.58 2 0.28 58.2 0.39 3 0.32 58.5 0.46 4 0.30 57.4 0.41 5 0.27 57.5 0.56 6 0.32 57.2 0.46 7 0.36 58.6 0.46 8 0.29 58.8 0.37 Mean -+ SD 0.30 --+ 0.03 57.96 -+ 0.63 0.46 - 0.08

SUNSCREEN SUBSTANTIVITY 359 Table II Analysis of Variance (ANOVA) Determined From the % Substantivity of PABA and Its Esters in the I, Vivo Method IF = 5.78 (p 0.01)] Degrees of Sum of Squares Freedom Mean Squares F SS Explained k - 1 17682.48 2 SS Error N - k 2.84 21 SS Total N - 1 17685.32 23 8841.24 0.123 65490 Table III Experimental 't' Values From the Dunnett Test for the Substantivity of PABA and Its Esters in the In Vivo Method It = 0.573 (p 0.01)] PABA vs Escalol © PABA vs Amerscreen Escalol 507 © vs 507 "P"© Amerscreen "P"© t 57.66 0.16 57.5 Table IV Mean K Values for Different Concentrations of PABA, Escalol 507 ©, and Amerscreen "P"© in Ethanol PABA Escalol 507 © Amerscreen "P"© (rag%) K (mg%) K (mg%) 80 18.35 -+ 5.04 80 7.38 -+ 2.15 80 17.84 _+ 0.62 85 19.31 + 2.35 90 8.01 + 1.45 90 17.14 _+ 0.69 90 19.50 --- 2.76 100 7.01 --- 0.64 100 17.38 --- 0.72 105 19.50 - 2.59 110 7.09 -+ 0.55 110 18.08 _+ 1.41 115 18.10 +- 3.31 120 7.14 - 0.51 120 18.18 _ 1.67 Table V Comparison of the Solubilities in Water and Substantivities of PABA and Its Esters Solubility Substantivity Sunscreen (mg/ml) (%) PABA 4.7 0.30 Amerscreen "P"© 3.9 0.46 Escalol 507 © 1.6 X 10 -3 57.96 (9) have pointed out that both the chemical characteristics of the sunscreen and the vehicle are important. Other researchers have made clear the influence of the duration of the contact of the sunscreen and the skin before washing. It has been proposed that 15 min of contact allow the sunscreen to penetrate the horny layer, thus increasing the resistance to wash- off (10). It has also been suggested that some sunscreens need an optimum time to penetrate the stratum corneum (4). Whether substantivity confers detrimental or ben-