We have examined the effect of bradykinin on impulse traffic in sympathetic afferent fibres from the heart, great vessels and pleura, and have attempted to identify cardiac nociceptors that on the basis of their functional characteristics might have a role in the initiation of cardiac pain. 2. In anaesthetized cats, we recorded afferent impulses from 'single-fibre' slips of the left 2nd--5th thoracic rami communicantes and associated chain, and selected fibres arising from endings in the heart, great vessels, pericardium and pleura. We applied bradykinin solution (0 . 1--1 . 0 microgram/ml.) locally to the site of the ending; we also injected bradykinin (0 . 3--1 . 0 microgram/kg) into the left atrium. 3. Afferent endings excited by bradykinin (159 of 191 tested) were of two types. The larger group (140) were primarily mechanoreceptors with A delta of C fibres (mean conduction velocity, 7 . 5 +/- 0 . 6 m/sec). They were very sensitive to light touch. Those located in the heart, great vessels or overlying pleura had a cardiac rhythm of discharge and were stimulated by an increase in blood pressure or cardiac volume. 4. Bradykinin increased mechanoreceptor firing from 0 . 7 +/- to 5 . 0 +/- 0 . 3 (mean +/- S.E. of mean) impulses/sec. Some endings appeared to be stimulated directly by bradykinin, others sensitized by it so that they responded more vigorously to the pulsatile mechanical stimulation associated with the cardiac cycle. 5. The smaller group of eighteen endings, of which ten were in the left ventricle, were primarily chemosensitive. Most had C fibres, a few had A delta fibres (mean conduction velocity, 2 . 3 +/- 0 . 7 m/sec). They were insensitive to light touch. With one exception they never fired with a cardiac rhythm, and even large increases in aortic or left ventricular pressure had little effect on impulse frequency. 6. Chemosensitive endings were stimulated by bradykinin, impulse activity increasing from 0 . 6 to 15 . 6 +/- 1 . 3 impulses/sec and remaining above the control level for 1-3 min. The evoked discharge, which was either continuous or occurred in irregular bursts, was not secondary to mechanical changes in the heart and great vessels. 7. Tachyphylaxis occurred when the interval between successive applications of bradykinin was 20 min or less. It was a feature of the response of both mechanosensitive and chemosensitive endings. 8. Because of their responsiveness to changes in pressure and their sensitivity to light touch, the mechanosensitive endings appear to be unlikely to subserve a primarily nociceptive function, although they may be responsible for evoking some of the components of the pseudoaffective response. By contrast, the chemosensitive endings appear well fitted to act as cardiac nociceptors.
