Myofascial trigger points (TrPs) have been clinically described as discrete areas of muscle tenderness presenting in taut bands of skeletal muscle. Using well-defined clinical criteria, prior investigations have demonstrated interrater reliability in the diagnosis of TrPs within a given muscle. No reports exist, however, with respect to the precision with which experienced clinicians can determine the anatomic locations of TrPs within a muscle. This paper details a study wherein four trained clinicians achieved statistically significant reliability (see below) in estimating the precise locations of latent TrPs in the trapezius muscle of volunteer subjects (n=20). To do so, the clinicians trained extensively together prior to the study. The precise anatomic location of each subject's primary TrP was measured in a blinded fashion using a 3 dimensional (3-D) camera system. Use of this measurement system permitted the anatomic co-ordinates of each TrP to be located without providing feedback to subsequent clinicians. The clinicians each used a pressure algometer along with patient feedback to document the sensitivity of each suspected TrP site, however unlike routine clinical practice, the algometry was performed with a double-blinded approach hence the results were only examined post-hoc. At the time of data collection (algometry readings unknown), 16 of the 20 subjects were judged to present with a latent TrP. Subsequently, when subjected to a criterion pressure threshold value of <3.0 kg.cm(-2), 12 of these TrPs were classified as being clinically sensitive. To assess the 3-D measurement precision, and the reliability of the TrP estimates, statistical measures of the SEM and the Generalizability coefficient (G-coeff) were determined for all suspected TrP sites in the superior-inferior, medial-lateral and anterior-posterior directions. The best results were determined by pooling the measurements of all 4 clinicians, however, based upon exceeding a criterion reliability threshold of 80%, the use of just two testers was found to produce reliable results. The two-tester condition yielded a precision of 7.5, 7.6 and 6.5 mm (SEM) with reliability (G-coeff) of 0.92, 0.86 and 0.83, respectively. Given the double-blinded methodology, the use of pressure algometry was also found to demonstrate internal validity. The algometer responses associated with TrP estimates varied inversely with respect to the clinical group's reliability in identify the TrP locations. To summarize, for the trapezius muscle, this study demonstrates that two trained examiners can reliably localize latent TrPs with a precision that essentially approaches the physical dimensions of the clinician's own fingertips. Finally, it should be recognized that the ability to precisely document TrP location appears critical to the success of future studies that may be designed to investigate the etiology and pathogenesis of this commonly diagnosed clinical disorder.