356 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS pacity of a compound to adhere or combine with a keratinized substrate (2). In general it is accepted that a sunscreen preparation must have some degree of substantivity. Our literature search has failed to reveal studies oriented toward establishing standard methods of evaluating in the laboratory the ability of a U.V. absorber to resist being removed by water. Some studies of affinities of sunscreens for human skin in vivo and in vitro (3,4) or animal skin in vitro (5) have been reported. Nevertheless, very little is known about the significance of the in vitro methods and their capability to predict the behaviors of the compounds in human skin. An in vitro method that correlates with actual in vivo application on human skin is desirable to avoid the inconveniences and limitations of human experimentation. In our work we have studied the substantivities of p-amino benzoic acid (PABA) and two of its esters (Escalol 507 © and Amerscreen "P"©), using both an in vivo and in vitro method in order to find a correlation between the two methods. Healthy volunteers participated in the in vivo method, thus studying the substantivity of the sunscreen compounds under conditions which were near to actual use conditions. Human keratin in the form of excised human callus tissue was employed in the in vitro method. The results in terms of percentage of substantivity obtained by the two methods were compared and a very strong linear correlation was found. EXPERIMENTAL DESIGN MATERIALS The test materials were: p-amino benzoic acid, analytical reagent grade (USP) octyl dimethyl (Escalol 507 ©, Van Dyk & Co., Inc.) and ethyl dihydroxypropyl (Amerscreen "P"©, Amerchol Corporation) p-aminobenzoate and ethanol 95% and 2-propanol (E. Merck, Darmstadt, W. Germany). The human keratin was obtained from human foot callus tissue which had not been treated with chemical agents prior to removal. The callus tissue was pulvarized while suspended in acetone refrigerated below 0øC with solid CO2 using a Waring Blender ©. After the comminution procedure, the material was washed with ether and allowed to dry (6). After pulverization the keratin was sieved before use, utilizing the fraction that passed through a 70-mesh sieve but which was retained by a 100-mesh sieve. The mean particle size of the powder was 150 DETERMINATION OF SUBSTANTIVITY IN VIVO Eight healthy volunteers, one male and seven females, ranging in age from 20 to 30 yr (mean 24 yr) and in weight from 45 to 70 kg (mean = 60 kg), with no external signs of allergy or other skin disease, participated in this study. The method previously described by Cumpelik (4) was employed, utilizing 2% w/v, solutions of Escalol 507 © and Amerscreen "P"© in isopropanol (IPA). A 5% PABA solution in IPA was used for comparison. Hands and arms of all subjects were washed up to the elbows with soap and tapwater, dried, and then washed in IPA at 50øC repeatedly until 100% transmittance of the washings was obtained between 240 and 400 nm, using a Carl Zeiss spectrophotometer. Hands were air dried. Following the cleaning procedure, all the subjects had their right hand dipped into the PABA solution for 2 min. For comparison the left hand of each subject was dipped

SUNSCREEN SUBSTANTIVITY 357 into a comparison sunscreen solution. The amount of each sunscreen deposited on each subject's skin was obtained by weighting the solutions before and after the introduction of the hands. Hands were allowed to air dry for 35 min and then submerged in distilled water at 25øC for 30 min. Hands were again air dried and, finally, dipped into IPA at 50øC for 2 min. Aliquots of the IPA solution were read by spectrophotometer, thus obtaining the amount of screen remaining on the skin after the water treatment. The substantivity of the screen was estimated using the following equation: S%= amount of screen (mg) recovered x 100 amount of screen (mg) deposited initially on the skin In the spectrophotometric analysis PABA was read at 289 nm and both Escalol 507 © and Amerscreen "P"© at 310 nm using a Carl Zeiss M4 Q III spectrophotometer. DETERMINATION OF SUBSTANTIVITY IN VITRO Experiments were performed utilizing human keratin and ethanol as solvent because preliminary experiments proved that there were no differences in the results when using either IPA or ethanol. The concentration ranges for each sunscreen were established between those at which the affinity of the material for the skin remains constant, determining in each case the sorption isotherm. The values obtained for the three substances were: PABA, 80-115 mg% Escalol 507 ©, 80-120 mg% and Amerscreen "P"©, 80-120 mg%. Once the concentration range was selected, the substantivity was determined using the procedure described by Bottari et al. (1), modified by us as follows: A 100-mesh stainless steel basket containing 0.1 g of human keratin powder was placed in a flask with 20 ml of water and held 40 hr in a thermostatically controlled shaker bath, to allow hydration of the keratin. Hydration of the keratin previous to the treatment with the sunscreen was necessary. Experiments performed without previous hydration failed. The material was then allowed to dry at 20øC for 30 min. The following sunscreen solutions in ethanol were prepared: PABA, 85-90-105 mg% Escalol 507 ©, 100-110-120 mg% and Amerscreen "P"©, 80-110-120 mg%. Twenty ml of sunscreen solution were placed in a glass flask and the stainless steel basket containing the previously hydrated keratin was then added. The flask was stop- pered, sealed, and held for 72 hr in a thermostatically controlled shaker bath at 37øC. The solution was then filtered using a millipore H.A.-type filter and and assayed by spectrophotometer. The basket containing the keratin was then allowed to dry at 30øC for 30 min. After drying, the basket with the keratin was placed in a glass flask containing twenty ml of distilled water. The stoppered and sealed flask was held for 48 hr in a thermo- statically controlled (25 -+ iøC) bath. At the end of this time the basket was taken from the solution and allowed to dry for 60 min at 30øC. Finally, the basket containing the keratin was placed into a glass flask with twenty ml of ethanol. The flask was stoppered, sealed, and shaken in a constant temperature bath at 50 -+ iøC for 24 hr. The resultant solution was filtered and assayed by spectropho- tometer.