Objective: Are substances released from rat coronary endothelial cells responsible for the increase in contractility and oxygen consumption (Gregg phenomenon) seen with an increase in cardiac perfusion? Methods: In an isovolumically contracting, Langendorff, crystalloid perfused rat heart ( n = 6) at 27 °C, coronary flow was changed (from 4.4 to 15.4 ml · min−1 · gWW−1) before and after the endothelium was made dysfunctional by Triton X-100. Vascular endothelium and smooth muscle function were tested with bradykinin (BK, 1 μM, an endothelium-dependent dilator) and papaverine (PAP, 1 μM, an endothelium-independent dilator) in a preconstricted vascular bed (vasopressin, VP, 3 nM). Results: Before Triton X-100, coronary resistance (at constant flow) decreased significantly in response to BK and to PAP. After Triton X-100 treatment the dilatory response to BK was abolished while the PAP response was still present, suggesting endothelial dysfunction with intact smooth muscle function. Due to Triton X-100 treatment, coronary resistance increased significantly. Therefore coronary flow changes were also applied during a similar increase in coronary resistance induced by VP infusion (3 nM) before Triton X-100 treatment. During control, developed left ventricular pressure (dev Plv) increased with 68 ± 21% and oxygen consumption (Vo2) increased with 122 ± 25% in response to the maximal increase in coronary flow. During increased coronary resistance with and without functional endothelium, dev Plv increased by 57 ± 16 and 64 ± 22%, respectively, and Vo2 increased by 126 ± 21 and 103 ± 20%, respectively, in response to the maximal increase in flow. These changes were not significantly different from control. Conclusion: The results suggest that the arterial endothelium is not involved in the Gregg phenomenon.