EMOLLIENCY AND CHEMICAL STRUCTURE 651 Table XVI Lanolin and Derivatives Viscosity (Lotion)(cps) Lanolin (anhyd. USP) Liquid lanolin (oil) Lanolin alcohols Lanolin alcohols hydrocarbon extract Lanolin alcohols acetate Lanolin alcohols ricinoleate Isopropyl lanolate Initial Slip (Scale 1-5) Hydrocarbon extract 3.0 Alcohols 3.5 Acetate 3.8 Ricinoleate 4.5 Lanolin 4.5 Liquid lanolin 4.5 Isopropyl lanolate 5.0 7860 4680 39,500 10,720 13,560 8100 6400 End Feel (Scale 4-20) Liquid lanolin Alcohols Ricinoleate Lanolin Acetate Hydrocarbon extract Isopropyl lanolate 6.5 11.5 12.5 15.8 16.3 16.5 16.8 Table XVII Isopropyl Esters CHa O CHa--CH--O--C--R Approx Carbons on Viscosity Mol Wt Fatty Acid (Lotion)(cps) Branching Isopropyl myristate 271 14 2276 5.6 Isopropyl palmitate 299 16 5540 5.1 Isopropyl linoleate 320 18 1144 4.7 Isopropyl isostearate 321 18 5800 9.4 Isopropyl ricinoleate 337 18 4530 4.4 Isopropyl lanolate (10-32) 6400 P Diisopropyl adipate 228 6 800 13.1 Diisopropyl azelate 272 9 2040 11.0 Diisopropyl sebacate 286 10 6580 10.5 Initial Slip (Scale 1-5) Linoleate Sebacate Palmitate Myristate Isostearate Adipate Lanolate Azelate Ricinoleate 3.0 4.0 4.0 4.3 4.5 4.5 4.8 4.8 5.0 End Feel (Scale 4-20) Ricinoleate Linoleate Sebacate Azelate Isostearate Lanolate Myristate Palmitate Adipate 12 4 13 0 15 3 16 3 16 8 16 8 17.0 18.3 18.8

652 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS weight. On the other hand, the fact that both of these are diesters did not help them as much, apparently, as increasing the length of the fatty chain (isostearate, myristate, and palmkate isopropyl esters). Finally, note that the "best of the best" was diisopropyl adipate, with an End Feel score of 18.8. It is striking that the isopropyl ester giving the poorest End Feel (isopropyl ricinoleate) was the best in terms of Initial Slip, a pattern which was observed repeatedly throughout this study. SUMMARY AND CONCLUSIONS It is possible to quantitate skin feel of cosmetics, and to do so reason- ably reproducibly--whether the same investigator repeats his own work at various time intervals or the work of another investigator. From the work to date, we can draw a few "intuitive" conclusions, not yet backed by rigorous graphic or statistical analysis. Such analyses must await evaluation of further samples, not available at the time of this initial study. The intuitive conclusions include our belief that molecular weight per se is not relevant to final skin feel, but that "oiliness" (HLB?) is. This is evidenced by the fact that the presence of polar groups gen- erally gives very negative effects. Some degree of unsaturation and/or chain branching was usually helpful, perhaps due to the "liquefying ef- fect" such structural changes produced, lowering the melting point of the residual oil phase left on the skin. Viscosity of the final test product had to be at certain minimum levels, otherwise results were poor. Physical factors of this type may "mask" the more purely chemical effects, obtained by systematically varying the struc- ture of cosmetic emollients. Finally (and we are really not quite sure why), isopropyl alcohol esters of most fatty acids were outstanding as a group. They were invariably at or near the top of each series in which they were tested. Many of the trends demonstrated in this study have been known through practical experience in the past, or known "intuitively" by ex- perienced cosmetic formulators. Now, however, being able to assign numerical values to skin feel makes it possible to study the subject sys- tematically, and to estimate the additive skin feel effect of a series of ingredients proposed for a cosmetic formula. Also, we can now begin to choose emulsifiers and other cosmetic components on a known skin feel basis, considering their SFI index as well as HLB, solubility, etc. With this approach, it may be possible to custom-design molecules for certain