The interexaminer reliability of noninvasive methods of examining lumbar spinal segments is not well established. In this project, the interexaminer reliability of three experienced chiropractic examiners, who evaluated 21 symptomatic and 25 asymptomatic subjects, was explored. Eight noninvasive segmental examination strategies (dimensions) were employed at each spinal segment from T11/T12 through L5/S1. Marginal to good agreement beyond chance was noted for palpatory pain over osseous structures and in paraspinal soft-tissues. Weaker and less frequently, significant concordance between examiners was noted for detection of temperature differences (>/_1.5 degrees F) between adjacent segments and for visual inspection for segmental abnormality. Little significant agreement between examiners was found for active and passive motion palpation, muscle tension palpation and misalignment palpation. This study suggests that “subjective” findings (pain) may be among the most reliable of conservative spinal observation. Weak but significant correlations were found when positive findings for the eight dimensions at each lumbar segment were summed to form a composite joint abnormality index. When the multidimension index was developed using the four most reliable dimensions, slightly stronger correlations were found. The strongest agreement between examiners tended to be found in the lower lumbar spine.