Aims: To test in vitro and in vivo the reliability and accuracy of a new algometer, the pressure algometer for palpation (PAP), for measuring pressure pain thresholds (PPTs) and to compare its features with those of a commercially available pressure algometer. Methods: For in vitro accuracy testing, 6 repeated measurements were made at 8 defined test weights from 0.5 to 5 lb. In vivo validity testing compared the PAP to a standard instrument, the hand-held Somedic algometer, at 16 sites including the masticatory muscles, the temporomandibular joints, and the frontalis (as the control site) in 15 temporomandibular disorder (TMD) cases and 15 controls. Intraexaminer reliability was also assessed for both algometers. Results: In vitro reliability was high, with coefficients of variation of 5% and a single-measurement standard deviation of 2.1 kPa. Accuracy was also high, with PAP measurements correlating with test weights at r = .99 (P .001). Repeated measures reliability in vivo was high, with intraclass correlation estimates of 0.73 to 0.96 for the PAP and 0.78 to 0.99 for the Somedic algometer. PPT values correlated moderately between the 2 devices (r ranged from 0.38 to 0.66; P ¡Ü .05) and were consistently higher for the PAP at all sites (P .001). Differences between controls and TMD cases were also significant for both algometers (P .006). Conclusion: Both the PAP and the Somedic algometer showed high reliability. Concurrent validity was demonstrated by statistically significant correlations between the devices. Both showed equally high capacity for differentiating TMD cases from controls. The PAP yielded significantly higher PPTs than the Somedic algometer. Schlagwörter: algometry, pressure pain threshold, reliability, temporomandibular disorders, validity