Upper airway luminal patency is influenced by a number of factors including: intraluminal air pressure, upper airway dilator muscle activity, surrounding extraluminal tissue pressure and surface tension acting within the liquid layer lining the upper airway. In this study we examine the performance characteristics for the 'pull-off' force method for measuring the surface tension (gamma) of liquids. This method is then used to examine the gamma of the liquid lining the oro-pharynx in awake human subjects. The gamma of UAL samples (approximately 0.2 microL) obtained from the posterior pharyngeal wall was quantified using the pull-off force technique in which gamma is measured as the force required to separate two curved silica discs bridged by the liquid sample. Pull-off force measurement was not influenced by sample volumes or under different temperature or humidity conditions. The coefficients of variation for repeated measurements using the pull-off force technique for the three standard liquids ranged from 1.2% to 5.6%. The pull-off method tended to give slightly lower values than the Wihelmy balance method with the mean difference being 2.4 +/- 1.8 mN m(-1). For the group, the gamma of saliva was 59.2 +/- 0.6 mN m(-1) (CV 4.9 +/- 0.9%) for five measurements of a single sample. There was no significant relationship between the measured gamma values for five subjects measured on five separate days and the day of collection of the sample for any of the subjects or the group as a whole (both p > 0.6). There was no significant difference (p = 0.53, ANOVA) between the gamma values for samples obtained from under the tongue (61.6 +/- 1.7 mN m(-1)), at the oral surface of the soft palate (56.1 +/- 3.2 mN m(-1)) and from the posterior oro-pharyngeal wall (58.1 +/- 2.4 mN m(-1)). The pull-off force method provides adequate reproducibility to be able to measure the gamma of UAL. In healthy humans the gamma of UAL is similar to that of saliva.