Studies were conducted on human cadaver thoracolumbar vertebrae, at the T₁₂−L₅ level, of five males and six females. Isolated vertebral bodies, free of posterior elements, were first scanned using dual photon absorptiometry and then underwent axial compressive loading. All of the vertebral bodies failed as a result of compressive fractures of the bone. Results indicated that the mechanical load-deflection response was non-linear and biphasic. The mean cross-sectional areas of the vertebral bodies progressively increased from L₁, to L₅. The maximum load carrying capacity was not dependent upon spinal level. The bone mineral content (BMC) obtained using dual photon absorptiometry in the lateral projected plane increased from L₁ to L₅. Male vertebral bodies consistently had higher BMC than female specimens. The cortical shell contributed 12.44% (mean) of the total cross-sectional area in the male, 17.56% in the female;​ 8.85% of the BMC in the male and 16.54% in the female. In contrast, it accounted for 43.8% (mean) of the total load in the male compared to 35.2% in the female specimens. Mean failure loads of decorticated vertebrae were significantly lower (p<0.001) when compared with that of the adjacent intact vertebral bodies. In one osteoporotic spine, the cortical shell accounted for 74% of the total strength. The anatomical placement of the thin shell which enables it to act as an encasing element to resist the collapse of the trabeculae under compression, and the difference in rigidity of the two structural components, and their differing sensitivity to metabolic influences, seem to explain this relatively high magnitude of load absorption in spite of its limited contribution to vertebral geometry.