Analysis of nicotine and cotinine in human hair can provide information on nicotine intake and exposure to environmental tobacco smoke over a long period of time. Nonetheless, to better assess the usefulness of hair analysis to determine smoking habits or exposures, all procedures have to be standardized. Various solvents were tested as washing solvents to eliminate external contamination from nicotine. Dichloromethane was found effective when used for two washes prior to the extraction. Basic and acid digestion of hair followed by solid phase extraction with Extrelut-3 glass column using dichloromethane:isopropyl alcohol (9:1) as eluting mixture both gave good recoveries of nicotine and cotinine, when compared with extractions reported in the literature. The extraction method was free from substances, which could interfere in the chromatographic analysis. Furthermore, the addition of methanolic HCl to the eluting mixture prevented the loss of nicotine during the evaporation step before chromatography. Chromatography was performed using a reversed-phase column and a U.V. detection at 254 nm. Furthermore, hair treatments (dyes, permanent wave, hydrogen peroxide) caused a major decrease in the nicotine content in hair, and a smaller effect on cotinine levels. However, the effect of various treatments was not reproducible. Several attempts to produce reference materials were carried out. Nicotine and cotinine standard solutions at different concentrations were added to blank hair soaked in dimethylsulfoxide, methanol and water.