CLEANING HAIR 315 tion times, and raising the temperature to speed the elution led to loss of resolution among the other peaks. Since Thompson et al. (7) stated that the triglycerides are easily removed by the three surfactants they used, with no increased build-up at 10 or 20 cycles, we elected to concentrate on the other sebum components. ONE-CYCLE DATA The sebum component removal data for tresses were analyzed statistically for seven components, i.e., myristic (C14), palmitic (C16), stearic (C18), and unsaturated (oleic and linoleic) acids (C18:21), cholesterol (CHOL), paraffin waxes (11 fractions combined) (PW), and esters (from spermaceti wax five fractions combined) (EST). The total sebum removal data is shown in Table II. As previously stated, these figures correlate well with data acquired using a wool substrate (9) (Table II). Figure 2 shows results of component removal after one soil/wash cycle (0.01% deter- gent). The order of removal for the sebum components is similar: ester and paraffin wax removal is the most difficult, and cholesterol the easiest. The only difference is the magnitude of removal that is determined by the nature of the surfactant, i.e., whether it is a good or poor cleaner of lipid soils. As mentioned, the data show the relative total sebum removal of SLES-2, ALS, and SODS-1 to be similar from hair and wool surfaces, i.e., SLES-2 ) ALS ) SODS-1. This order confirms that predicted by surfactant theory for oily soil detergency (10). To determine if there is selective removal of components by a surfactant, one way ANOVA • C8,10-1EO • ALS • SLES-2 'o o lOO 8o 6o 4o 2o C14 C16 C18:21 C18 P.Wo CHOL. EST. Sebum Component Figure 2. Removal of sebum components by Cs,•o-iEO, ALS, and SLES-2 for one soil/wash cycle.

316 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS statistics have been performed on these data and removal of individual component groups compared. The following summarizes the statistical analyses of the sebum component removal by SODS-1, for one soil/wash cycle (95% confidence level). (Component removals are sig- nificantly different when components are not underlined by the same line): Least removed Most removed EST PW C 14 C 16 C 18:21 C 18 CHOL These analyses show that this surfactant is most effective in removing the cholesterol component from hair it is least effective in cleaning off the esters and paraffin waxes. For ALS, sebum component removal is as follows: Least removed Most removed EST PW C 14 C 16 C 18:21 C 18 CHOL Here, the order of removal is the same as SODS-1 but the data (Figure 2) show that ALS removes more of each component than SODS-! (95% confidence level). The esters and paraffin waxes are clearly more difficult for ALS to remove than the other components, with the exception of the C14 materials. On the other hand, when soiled hair is washed under similar conditions with SLES-2, there are no significant differences in removal among the sebum components (p - 0.05), and SLES-2 removes all components more effectively than either ALS or SODS-1. Consequently, after one soil/wash cycle (soiling level 0.04-0.055 g/g 0.01% surfac- tant), the removal of sebum components by each of the three surfactants tested can effectively be predicted by a value derived for the total sebum removal. SLES-2 is clearly the most effective against all groups of components and SODS-1 the least effective, a confirmation of surfactant theory (10) (Figure 2). Sebum removal data for hair soiled and washed under a second set of conditions, i.e., 0.03 g sebum/g hair and 0.1% surfactant solution, were also analyzed (9). The order of component removal for individual surfactants was found to be similar to the order under "A" soil/wash conditions. Additionally, under these conditions of lower soil loading, both ALS and SLES-2 remove all components at 94% levels, ap- proaching the limits of the experiment. Similar to "A" soil/wash conditions, the most difficult fractions to remove are the paraffin waxes and the esters. Clearly the one-cycle experiments indicate that some sebum components are more diffi- cult to remove, but the same pattern of removal exists for all three surfactants tested. Surfactant theory for oily soil detergency confirms this order (10). Thus, the one-cycle data show that a surfactant with good cleaning power removes all components well, a poor one less well. However, for all surfactants tested, the esters (from spermaceti wax) and the paraffin wax fractions are the most difficult materials to clean from the hair.