SUMMARYThe intrinsic innervation of the heart has been described anatomically but seldom, functionally. These experiments were designed to verify and to evaluate the influence of intrinsic neural activation on the contractility of the atria and ventricles. Nine mongrel dogs underwent total cardiac denerva-tion using the Geis two-stage technique. Following recovery, they were placed on total cardiopulmonary bypass. Balloons were inserted into each cardiac chamber, filled with saline, and used to record atrial and ventricular isovolumic pressures in response to intracoronary drug injections. The hearts were paced to avoid rate changes. Acetylcholine (ACh) (0.05–1.0 &mgr;g) was given to assess direct effects on the myocardium; nicotine (NIC) (25–200 &mgr;g) was used to produce indirect inotropic effects. d-Tubocurare (dTC) (1.0–2.5 mg) and tetrodotoxin (TTX) (10–30 &mgr;g) were used to differentiate the direct and the indirect effects. Both ACh and NIC produced qualitatively similar negative inotropic responses in both atria and ventricles. dTC usually produced a significantly greater blockade of responses to NIC than to ACh. TTX administered to 12 other denervated dogs to differentiate direct and neurally mediated responses, completely blocked NIC- induced negative inotropy but did not change responses to ACh. The data show that (1) the intrinsic cardiac innervation can exert significant negative inotropic effects on atria and ventricles (10 to >30%); (2) there is no functional evidence of participation by chromaffin cells in the inotropic responses to intrinsic nerve activation; (3) the negative inotropy produced by NIC is sensitive to dTC and, especially, TTX blockade. This is consistent with the mechanism of action being intrinsic neural stimulation.