Interpersonal differences can be observed in the human cerebrospinal fluid pressure (CSFP) in the cranium in an upright body position, varying from positive to subatmospheric values. So far, these changes have been explained by the Monroe-Kellie doctrine according to which CSFP should increase or decrease if a change in at least one of the three intracranial volumes (brain, blood, and CSF) occurs. According to our hypothesis, changes in intracranial CSFP can occur without a change in the volume of intracranial fluids. To test this hypothesis, we alternately added and removed 100 or 200 μl of fluid from the spinal CSF space of four anesthetized cats and from a phantom which, by its dimensions and biophysical characteristics, imitates the cat cerebrospinal system, subsequently comparing CSFP changes in the cranium and spinal space in both horizontal and vertical positions. The phantom was made from a rigid "cranial" part with unchangeable volume, while the "spinal" part was made of elastic material whose modulus of elasticity was in the same order of magnitude as those of spinal dura. When a fluid volume (CSF or artificial CSF) was removed from the spinal space, both lumbar and cranial CSFP pressures decreased by 2.0-2.5 cm H2O for every extracted 100 μL. On the other hand, adding fluid volume to spinal space causes an increase in both lumbar and cranial CSFP pressures of 2.6-3.0 cm H2O for every added 100 μL. Results observed in cats and phantoms did not differ significantly. The presented results on cats and a phantom suggest that changes in the spinal CSF volume significantly affect the intracranial CSFP, but regardless of whether we added or removed the CSF volume, the hydrostatic pressure difference between the measuring sites (lateral ventricle and lumbar subarachnoid space) was always constant. These results suggest that intracranial CSFP can be increased or decreased without significant changes in the volume of intracranial fluids and that intracranial CSFP changes in accordance with the law of fluid mechanics.