The effect of continuously slowing respiration (from 0.46 to 0.05 Hz, eg, from 30 to 3 breaths per minute) on cardiovascular variables was studied in 14 hypertensive patients and 16 normotensive subjects. Beat-to-beat time-frequency (Wigner) distributions were used for dynamic analysis of RR interval and systolic and diastolic pressures. Dominant breathing frequency at rest did not differ in hypertensive patients from the control group (0.21 versus 0.19 Hz). However, in the hypertensive group it was disturbed 34.4% of the time by slow breathing and apneas, which evoked transient blood pressure instability and increased spectral powers at low frequencies (range, 0.01 to 0.1 Hz). The nonrespiratory fluctuations (NONRFs) and respiratory fluctuations (RFs) in RR interval and NONRFs in systolic pressure were smaller in hypertensive patients (/ > <.001). In both groups, slowing of respiratory frequency from 0.46 to 0.05 Hz entrained RFs in the RR interval and systolic and diastolic pressures. RFs in the RR interval remained diminished in hypertensive patients (P<.001), but RFs in systolic pressure increased higher at maximum, corresponding to breathing frequencies from 0.07 to 0.09 Hz (P< .001). A dynamic cardiorespiratory index (ratio of RFs in RR interval and systolic pressure) was smaller (P<.01) in hypertensive patients than in normotensive subjects. Irregular breathing at rest was found in hypertensive patients. The transfer from respiration into RR interval was diminished, suggesting an impaired parasympathetic responsiveness in mild hypertension. (Hypertension. 1994^3:104-113.) Key Words • spectrum analysis • cardiovascular system • respiration • hypertension, essential T he problem of hyperreactivity of the sympathetic nervous system in essential hypertension has been studied frequently in the last two decades. An exaggerated sympathetic tone in mild hypertension can result from an increased intrinsic activity of the brain stem vasomotor neurons or can be secondary to a reduction of the inhibitory potency of baroreceptors 12 and pulmonary volume receptors. An augmentation of sympathetic tone was considered a primary mechanism of hypertension in a subgroup of patients in which elevated levels of plasma norepinephrine and epineph-rine 34 were found. High RR interval fluctuations from 0.05 to 0.1 Hz were reported and attributed to the augmentation of sympathetic efferent activity. 5 ' 6 However , these oscillations are under combined parasympa-thetic and sympathetic influences. Only the parasympa-thetic muscarinic blockade consistently suppresses RR interval fluctuations over the range of 0.01 to 0.4 Hz, and the effect of sympathetic blockade is ambiguous. 7 ' 8 Moreover, direct recordings of sympathetic efferent activity did not show exaggerated resting levels of sympathetic tone. 9 Heart rate fluctuations at respiratory frequencies are recognized as a noninvasive measure of cardiovagal activity 7 ' 8 and reflect more modulation of the firing of barore-ceptors and of cardiac and pulmonary volume receptors than a...