JOURNAL OF COSMETIC SCIENCE 78 Table III HPLC Method Validation: DHA Recovery Sample DHA added DHA found % RSD % Recovery A 0.000 0.000 — — B 1.000 1.049 3.376 104.892 C 1.997 2.093 3.307 104.788 D 2.997 2.930 1.597 97.767 E 3.994 4.120 2.189 103.144 F 4.993 5.096 1.443 102.062 Figure 1. HPLC chromatograms. Table II Analytical Results Test article DHA (%) X 3.25 K 3.18 vitro data suggested that different in vivo performances are expected for X and K, espe- cially with respect to the yellow/red and lightness/chroma parameters of the skin color. IN VIVO EXPERIMENTAL DATA AND DISCUSSION The in vivo part of this study was conducted on fi ve panelists and focused on the applica- tion, results, and measurements related to test articles X and K. Four panelists were categorized based on their ITA° values (Table IV), according to Chardon et al. (11), as very

SUNLESS TANNING FORMULATIONS 79 light to light and one panelist was categorized as intermediate. In vivo experimental data are presented in Figures 5–7. Panelists noticed a slightly less intense color development after application of X within the fi rst 24 hours, but the color of the tan was more even throughout the application area compared to K. These visual observations were confi rmed instrumentally (Figure 5). The average standard deviation of ΔE*ab for all panelists at 24 hours was 4.2 times higher for K than for X. At 48 hours the average of ΔE*ab was now slightly higher for X than K, and it was still higher after 120 hours. Apparently the fading of the sunless tan occurred at a slower rate with X than with K. Color associated with X was still visible with the “naked eye” after 120 hours for four panelists and for one panelist with K. This Figure 3. Impact of the test articles on the color response in vitro at 72 hours. Figure 2. Impact of the test articles on ΔE*ab in vitro vs time.