The content of "free-base nicotine" in cigarette smoke is a controversial subject, partly due to methodological issues. In this investigation, an improved method to measure volatile nicotine ("free-base nicotine") using headspace solid-phase micro-extraction (HS-SPME), combined with gas chromatography/mass spectrometry (GC/MS) was developed and validated for this purpose. Cigarette smoke particulate phase (PP) was collected onto a 44 mm glass fiber filter pad. The pad was then cut in half using one section for the determination of total nicotine and water contents. The other was transferred into a 20 mL amber glass vial for analysis by HS-SPME. The following factors were found to have a significant impact on the results: SPME fiber type, pre-equilibrium time before HS-SPME, extraction time and temperature, PP water content, and the solvent used for the preparation of standards. In this study, a PDMS/DVB fiber was found to provide sensitivity and peak shape superior to that obtained using a Carboxen/PDMS fiber. "Free-base nicotine" reached absorption equilibrium onto the fiber in less than 5 minutes, but more than 2 hours were required in order to obtain partitioning equilibrium between PP and the headspace. Control of the extraction temperature was critical for the reproducibility of analysis. Also, it was found that the partitioning into the headspace and extractability of "free-base nicotine" was dependent upon the PP water content. This necessitated the introduction of a moisture correction factor based on an experimentally determined reciprocal model. Under ISO smoking conditions method characteristics were: limit of quantification - 1.32 µg/cig, with a recovery from fortified solutions of 96.4%, and a coefficient of variation (CV) of 9.8%. Results for reference cigarettes (average, CV) were: Canadian flue-cured monitor - 34.6 µg/cig, 14% and Kentucky Reference Cigarette 2R4F - 36.1 µg/cig, 11%.