This prospective and consecutive study was designed to evaluate the validity of different clinical tests, e.g. lumbar extension in lying and slump test for patients with suspected herniated nucleus pulposus, in comparison with findings on computed tomography (CT) and/or magnetic resonance imaging (MRI) scan. There were 105 patients who were seen and examined by the senior author (for the sake of the study) at the Orthopaedic Physiotherapy Department, on an average of 5.5 days (range 0-21 days) before CT and/or MRI examination were carried out. There were 36 women and 69 men with an average age of 42.7 +/- 9.8 (range 19-64) years. According to the radiological findings on CT and/or MRI, the patients were divided into three groups: 52 patients with disc hernia, 41 patients with bulging discs and 12 patients without positive findings. The mean values with standard deviations of 25 variables of three diagnostic groups were studied. Multiple comparison adjustment according to Bonferroni showed significant differences for three variables that were of diagnostic value (lumbar range of motion for forward flexion, left side-bending in standing, and pain distribution during extension in standing). The agreement between clinical and radiological findings for type and level of diagnosis of disc herniation was accurate in 72 patients (69%). The diagnostic sensitivity for disc herniation was 82.6% and the specificity 54.7%.