Currently, methods for the forensic analysis of human hair rely upon microscopic examination or DNA comparisons. Chemical techniques, on the other hand, have largely focussed on extracting and identifying drugs of abuse. Recent work in our laboratory has shown that organic material on the surface of hair is easily accessible using conventional extraction techniques and solvents, readily analyzed using gas chromatography/mass spectrometry (GC/MS), and provides qualitative and quantitative information that may be effective for forensic identifications or comparisons. In this work, various extraction methods such as ultrasound, Soxhlet, pressurized solvent, and supercritical fluid extraction have been evaluated for their efficiency and selectivity. Using these procedures, the organic fraction of the surface of hair was extracted by methanol, acetone, dichloromethane, hexane, and supercritical carbon dioxide. Yields from hair samples were 0.9% - 1.8% (mass fraction) for ultrasonic extraction, 1.1% - 4.1% for Soxhlet extraction, 2.5% - 3.6% for pressurized solvent extraction, and 2.4% for supercritical fluid extraction. Of the liquid solvent extraction techniques, pressurized fluid extraction was the most convenient and amenable to samples as small as 50 mg. In general, yields for all techniques decreased as solvent polarity decreased. Characterization of the hair extracts revealed a complex mixture of components ranging in carbon number from C14 to C33. This material consists of sebaceous excretions such as fatty alcohols, acids, esters, other lipids including squalene and cholesterol, and anthropogenic material from consumer products. Despite its lower extraction efficiency compared to the liquid solvents, supercritical fluid extraction has been the most successful method to date for very small hair samples. SFE has been applied as an on-line method using both static and dynamic modes. The pressure, temperature, and time of extraction have been optimized in order to maximize sensitivity and samples as small as 100 ug have been successfully analyzed. A restrictor is used to deposit the solutes directly onto the stationary phase of the chromatographic column. Subsequent chromatographic analysis time is on the order of 60 min. Analysis of head hair from 20 subjects varying in age and gender has shown that these chromatograms are suitably complex to provide some degree of individualization to a particular hair sample. Furthermore, there appears to be systematically higher levels of cholesterol in pre-pubescent versus post-pubescent individuals. Literature data suggest that further systematic differences in lipid composition according to gender and/or race may exist that would further increase the evidentiary value of chemical analysis. Finally, the chromatographic profiles obtained using this technique have proven to be stable under normal storage conditions and are reproducible for a single scalp location on an individual. Future work will include correlating the results with demographic data from subjects. An expanded sampling study is also needed to determine the amount of variation in the chromatographic profile of a single individual over time and scalp location. Lastly, chemometric pattern recognition algorithms will be used to quantitatively compare chromatographic profiles.