Psychophysical experiments show that the perception of posture is to a large degree affected by hitherto unknown graviceptors in the human trunk. By remote control subjects move themselves radially along their spinal axis over the horizontal platform of a rotating centrifuge until they feel horizontal. Normal subjects then set the centrifuge axis on average at 22-28 cm caudal of the meatus, neuromectomized subjects at 45-55 cm. Hence the mass centroid of these receptors should be situated near the last ribs. Evaluation of the residual faculties of paraplegic patients lead to the conclusion that somatic graviception is mediated by two distinctly localized inputs, the first entering the spinal cord at the 11th thoracic segment, and the second reaching the brain cranial of the 6th cervical segment, presumably via the N. phrenicus or the N. vagus. The effect of the first named input is abolished after bilateral nephrectomy. This proves that the kidneys affect gravity perception. But whether they function like statoliths or in another way cannot yet be decided. For the second input, however, the results show unequivocally that it yields gravity information through the inertia of a mass in the body. It is hypothesized that this mass may be that of the blood in the large vessels. This is corroborated by the effect of shifting blood craniad by means of positive pressure to the legs. It is inferred that the inertial forces are measured by mechanoreceptors in the structures that mechanically support the large vessels, rather than by baroreceptors.