The aim of this study was to investigate the reliability between different examiners when using the axis I of the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD). The hypothesis was that the standardized RDC/TMD examination protocol enables calibrated examiners to evaluate all examination items reliably. After calibration training by the RDC/TMD calibration team including the calibration of palpation pressure and the performance of the standardized examination protocol, four examiners, blinded to the patients' medical histories examined 24 subjects in a randomized sequence. One experienced examiner was the standard (hierarchical calibration). The recorded measurements strictly followed the RDC/TMD. Intraclass correlation coefficients (ICC), bias and precision were calculated to estimate interrater reliability. Acceptable (0.75 > or = CC > 0.4) to excellent (ICC > 0.75) reliability was found for 20 of the 23 (87%) examinations. Only sub-retromandibular muscle palpation and joint sound vibration recordings on lateral excursion showed poor-results (ICC < or = 0.4). The RDC/TMD examination protocol enables calibrated examiners to perform most (87%) examination items with satisfactory reliability. Therefore multi-site studies based on the RDC/TMD examination protocol may become feasible, keeping in mind the unsatisfactory reliability of 13% of the items (clicking during laterotrusion to the ipsilateral side, palpation of the posterior and submandibular region).