1. Respiratory modulation of cardiac parasympathetic activity and the relationship between respiratory sinus arrhythmia and parasympathetic control has been studied in twenty-nine conscious, healthy young adult subjects. 2. Changes in heart period in propranolol-treated subjects were taken as the measure of changes in cardiac parasympathetic activity; respiratory sinus arrhythmia was quantified as the difference between maximum and minimum heart periods in a given respiratory cycle; cardiac parasympathetic control was defined as the change in heart period after administration of a full dose of atropine. 3. During normal quiet breathing the inspiratory level of cardiac parasympathetic activity was not reduced to zero. The expiratory level was influenced by excitatory inputs whose activation was related to respiratory cycle length. 4. Slow breathing was associated with augmented sinus arrhythmia, but in different individuals the influence on minimum and maximum heart periods varied so that mean heart period was increased in some subjects but decreased in others. This occurred both in control conditions and after administration of a full dose of propranolol. 5. During normal breathing the correlation across subjects between respiratory sinus arrhythmia and parasympathetic control, although significant, was not close (r = 0.61). The relationship was not affected by beta-adrenergic blockade (r = 0.63). The strength of the correlation improved when multiple regression of respiratory sinus arrhythmia was performed on three variables: parasympathetic control, respiratory cycle length and tidal volume (R = 0.93). 6. It is concluded that in conscious human subjects the respiratory modulation of cardiac parasympathetic activity is different from that observed in the anaesthetized dog, and that variations in the amplitude of respiratory sinus arrhythmia do not necessarily reflect proportional changes in cardiac parasympathetic control.