Acute relationships between cerebrospinal fluid hydrostatic pressure and arterial pressure were quantified in conscious rats. The rats were catheterized with a left femoral artery catheter, and a double set of catheters was implanted into the third cerebral ventricle. In three groups of rats, artificial cerebrospinal fluid with various sodium concentrations (142, 152, and 162 mM) was infused into the third ventricle for 3 hours at 1.0 microliter/min. Mean arterial pressure (MAP) and third ventricular pressure were monitored simultaneously, and both increased progressively over the 3-hour infusion period; the rate of rise was significantly greater with infusion of the hypertonic solution. There were no significant differences between the rat groups infused with low, normal, or high artificial cerebrospinal fluid sodium in either the slope or the intercept of the regression equation relating cerebrospinal fluid pressure and MAP: a 1 cm H2O rise of cerebrospinal fluid pressure was always associated with nearly a 1 mm Hg rise in MAP. In other rats, changes in both cerebrospinal fluid pressure and MAP were shown to be highly dependent on the rate of ventricular infusion. We conclude that elevations of systemic arterial pressure are associated with only small elevations of cerebrospinal fluid pressure and that physiological changes of cerebrospinal sodium (+/- 10 mM) influence arterial pressure by altering intravascular hydrostatic pressure rather than sodium or osmosensitive receptors in the cerebral ventricles.