Computerized tomography ("tomodensitometry") should, in principle, be able to calculate the linear attenuation coefficient of a compact or spongy bone volume. This coefficient is related to the amount of mineralization, enabling the use of the Scanner X to measure the density of the bone. Thirty vertebrae from the vertebral columns of 14 subjects who died after acute infections were examined by C.T. and by standard methods, and then calcinified. The correlation between ashes weight-Houns-field number was positive but only moderate (r = 0.54), and lower than the result obtained (r = 0.84) with standard methods. Errors occur with presently used apparatuses which are related to: polychromatism, too high an energy of the beam, artefacts, and the effects of "partial volume". Futhermore, the mineral concentration by unit of vertebral volume is very low, and large variations must occur for it to be detectable. Finally, considerable errors arise from the intravertebral bone fat. Suggestions are made with the objective of improving the quality of C.T. as a measuring instrument.