We validate a technique for measuring neural arch load-bearing in cadaveric spines, and use it to test the hypothesis that such load-bearing rises to high levels in old and degenerated spines. Fifty-nine cadaveric lumbar motion segments, aged 19–92 yr, were subjected to compressive creep loading to reduce intervertebral disc water content and height to in vivo levels. The distribution of compressive “stress” within the disc was then measured by pulling a miniature pressure transducer, side-mounted in a 1.3 mm-diameter needle, along its mid-sagittal diameter. During these measurements, the motion segment was subjected to a compressive load of 2 kN, and positioned in 2° of extension to simulate erect standing. Measurements of compressive “stress” were integrated over disc area, and this force subtracted from the applied 2 kN to give the force resisted by the neural arch. An empirical calibration factor was applied to normalise results from each disc to values obtained under conditions when all of the compressive force could be assumed to pass through the disc. Disc degeneration was graded macroscopically on a scale of 1–4. Validation tests showed that calculated values of disc loading were proportional to actual applied load (r²>0.96) and predicted it with errors of 2–8%. Neural arch load-bearing in non-degenerated specimens was generally less than 20%, but averaged 49% for specimens aged over 70 yr. Multiple regression showed that neural arch load bearing (%)=14.4×disc degeneration score+0.46×age−35. These results indicate a substantial shift in vertebral load-bearing with increasing age and degeneration.