The respiratory and mean arterial pressure (MAP) responses to slow ramp pressure stimulation of carotid baroreceptors were compared in pentobarbital-anesthetized vagotomized dogs breathing 100% O2. Carotid sinus pressure (CSP) was raised from 50 (control) to 220 mmHg and then returned to control as linear ramps (+/- 1 mmHg/s) in isolated sinuses. MAP, heart rate (HR), ventilation (VE), frequency (f), and tidal volume (VT) were expressed as percent of control. The maximum difference between responses to positive and negative ramps at a given CSP (MAX) and the average difference (AVG) served as indicators of the hysteresis for each response. In 27 dogs MAP changed monotonically with varying CSP with insignificant (P = 0.27, MAX) or barely significant (P = 0.03, AVG) hysteresis, monotonic function being one that is continuously nondecreasing or continuously nonincreasing. Similar responses were obtained for HR. VE decreased as CSP increased, but the change was not monotonic. During negative ramp, VE increased back to control with an overshoot. Hysteresis for VE was pronounced (P less than 0.0001, both measures). The VE response was primarily determined by f; VT increased with CSP. To eliminate secondary respiratory effects due to alterations in MAP, in seven dogs similar experiments were performed after ganglionic blockade with hexamethonium. Hysteresis in VE and f persisted. To assess the role of changing arterial PCO2 (PaCO2) on VE, the CSP was held constant (after a ramp rise) at 140, 150, or 180 mmHg before reducing it at -1 mmHg/s to 50 mmHg; however, a significant hysteresis in VE was still observed. Further experiments, to eliminate secondary reflexes due to altered PaCO2, were performed in seven dogs after ganglionic blockade and paralysis with Flaxedil, with phrenic nerve activity as an indicator of ("neural") respiration. The hysteresis in VE and f were no longer significant. In summary, the results indicate that 1) slow ramp carotid baroreceptor stimulation elicits both VE and cardiovascular responses, the VE response showing a dramatically higher hysteresis than the cardiovascular responses; 2) the ventilatory hysteresis is partially explained by the secondary changes in PaCO2 and perhaps by cardiovascular variables; and 3) the central processing of the baroventilatory reflex appears to be rate sensitive at a slower rate of pressure change than that which causes rate sensitivity in the baropressure reflex.