OBJECTIVE: To determine the familiarization period required to obtain consistent measurements of the angular movements of the lumbar spine and pelvis during treadmill walking. DESIGN: An in vivo study with repeated measures every 2 min over 10 min. BACKGROUND: Walking on a treadmill can initially be an unfamiliar experience. No data were available to indicate the length of time required for treadmill familiarization prior to taking measurements of the angular movements of the lumbar spine and pelvis. The familiarization period has implications for the use of this technique in clinical testing. METHODS: The angular movements of the lumbar spine and pelvis were examined by thePEAK 3D motion measurement system in 16 untrained, normal subjects walking for 10 min on a treadmill at either self-selected or 60% of self-selected speed. RESULTS: The reliability of the angular measurements of the spine and pelvis were all greater than 0.83 (ICC) after 4 min of treadmill walking and did not appear to increase after 4 min. No changes in the amplitudes of the angular movements of the lumbar spine and pelvis could be detected after 4 min of treadmill walking, or of the absolute difference scores beyond 2 min of treadmill walking, in either normal or slow-speed walking groups. CONCLUSIONS: Angular movements of the lumbar spine and pelvis were consistent after 4 min of treadmill walking by normal subjects, both at self-selected and slow walking speeds, indicating that reliable measurements of these parameters could be taken at that time.