322 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS The extraction efficiency of the solvents for lipids increased in the order: CHC13 CHCl,/H20 ether CHC13/MeOH CHCl•/MeOH/H20 ether/water, and in all cases previous swelling of the hair in water vapour leads to a further increase. Water-saturated ether was, therefore, chosen as extraction solvent for all future investigations and hair samples were conditioned for 2 h in a water-saturated atmosphere at room temperature prior to extraction. Optimization of extraction conditions: In a final set of trials the optimum extraction conditions regarding time and temperature were investigated. Preconditioned hair samples were extracted at room temperature and under reflux under conditions given in Table IV. In both cases, Table IV Selection of Extraction Conditions Hair Samples (batch C) Conditioned in Water-Saturated Atmosphere, Extracted with Water-Saturated Ether Extraction Conditions Amount Amount of of Extract Total Lipids (rag) (mg) 1 a) 10 min at room temp. 27.6 ] b) additional 20 min at room temp. 6.4 20.6 c) additional •0 min at room temp. 2.9 2 a) I0 min under reflux 27.6 ] b) additional 10 min under reflux 32.1 20.9 c) additional 40 min under reflux 2.8 roughly the same amounts of lipids with similar composition were extracted after 60 min. This indicates that only the external sebum was affected. At boiling temperature extraction is only faster. As it might be suspected that the small amount of lipids extracted in the last 40 min might already originate from diffusion of internal sebum, we finally decided to use 20 min at boiling temperature as standard conditions for all future extractions (15). ISOLATION OF INTERNAL LIPIDS AND SURFACE BOUND CA- and MG-CARBOXYLATES The existence of "internal lipids" has not yet been generally accepted. However, Curry and Golding have shown (8) that prolonged extraction (75 Soxhlet cycles) with CH2C12, ether, and ethanol does not remove all the lipids present in the hair. It is obvious that lipid substances left in the hair after this procedure bear no relation to the material usually recognized as sebum which is accessible to detergent solutions. Therefore, it may be supposed that it is part of the suspected "internal" lipids. First evidence for the origin of the "internal" lipids was provided by K.D. Bingham (personal communica- tion). He observed that solvent extraction over prolonged periods of time continuously yielded a lipid mixture consisting not only of structural lipids* but also components *I.e., lipids resulting from the keratinization process of hair, consisting mainly of phospholipids, steroIs, mono- and diglycerides and free fatty acids.

HAIR OILINESS 323 typical for surface lipids (squalene, wax esters (16)). Further evidence for the existence of an internal lipid fraction from external origin was provided by Sakamoto et al. (17) who detected wax esters after homogenization of hair samples which had previously been Soxhlet-extracted with CH2C12 for 100 h. As the extraction of internal lipids is so slow and its completeness cannot be secured even after much longer periods of extraction, we decided to digest the keratin fibres by treatment with papain/dithiothreitol (7) in order to make the internal lipids more easily accessible for organic solvents and extract them from the residual membranes. The lipid content and composition determined in the CHC13-extract of a hair sample previously extracted with water-saturated ether for 20 min and with CH2CI• for additional 16 h, and subsequently digested, is given in Table V. The figures show that Table V Composition of Lipids Detected in Hair After Preceding Soxhlet Extractions with Boiling Ether/H20 for 20 min and CH2CI• for 16 h and Subsequent Digestion Lipid Class Content in Hair [%] Squalene/paraffin traces Cholesterol ester/wax ester traces Triglyceride -- Free fatty acids 1.5 1,3-Diglyceride -- Cholesterol/1,2-diglyceride traces Monoglyceride 0.2 Total 1.7 only monoglycerides and free fatty acids are present in concentrations worth mentioning, although most other sebum components could be detected in trace amounts. Certainly this is a proof of the existence of "internal" lipids, but it is evident that the definition of "internal" lipids is conceived too narrowly because the term "external" lipid cannot include all lipids removable by such an intense extraction. The average commercial shampoo removes only 40-60% of those lipids which are extractable with water-saturated ether after 20 min of reflux (15), and even after repeated shampooing of the hair this amount does not exceed 70-90%. Thus, it may be assumed that this ether extraction under the conditions mentioned above is equivalent to the maximum possible shampooing effect with the best shampoo available. Therefore, all residual lipids left in the hair subsequent to extraction should be considered as internal with one exception: Curry and Golding detected the presence of a thin layer of fatty acids on the hair surface, possibly bound to free carboxylic groups of the protein by Mg 2+ or Ca 2+ bridges (8).* Due to hydrolysis these fatty acids are slowly released during extraction, and they are probably one of the reasons for the seemingly retarded extraction of internal lipids and also for the rather high concentra- tion of free fatty acids in the hair sample after 16 h of extraction with CH2C12. The Ca- and Mg-bridge-bound fatty acids cannot be considered as internal lipids, because they *In contrary to these authors who identified these fatty acids as stearic acid, we detected by GLC of the methyl esters all fatty acid species which are present in natural sebum, although their composition was somewhat different from that in the corresponding external sebum.