1 Optimization of myocardial energy substrate metabolism improves the recovery of mechanical function of the post-ischaemic heart. This study investigated the role of K ATP -channels in the regulation of the metabolic and mechanical function of the aerobic and post-ischaemic heart by measuring the e ects of the selective K ATP -channel activator, cromakalim, and the e ects of the K ATP -channel antagonist, glibenclamide, in rat fatty acid perfused, working hearts in vitro. The role of K ATP channels in the cardioprotective actions of the adenosine A 1 -receptor agonist, N 6 -cyclohexyladenosine (CHA) was also investigated. 2 Myocardial glucose metabolism, mechanical function and e ciency were measured simultaneously in hearts perfused with modi®ed Krebs-Henseleit solution containing 2.5 mM Ca 2+ , 11 mM glucose, 1.2 mM palmitate and 100 mu l 71 insulin, and paced at 300 beats min 71 . Rates of glycolysis and glucose oxidation were measured from the quantitative production of 3 H 2 O and 14 CO 2 , respectively, from [5-3 H/ U-14 C]-glucose. 3 In hearts perfused under aerobic conditions, cromakalim (10 mM), CHA (0.5 mM) or glibenclamide (30 mM) had no e ect on mechanical function. Cromakalim did not a ect glycolysis or glucose oxidation, whereas glibenclamide signi®cantly increased rates of glycolysis and proton production. CHA signi®cantly reduced rates of glycolysis and proton production but had no e ect on glucose oxidation. Glibenclamide did not alter CHA-induced inhibition of glycolysis and proton production. 4 In hearts reperfused for 30 min following 30 min of ischaemia, left ventricular minute work (LV work) recovered to 24% of aerobic baseline values. Cromakalim (10 mM), administered 5 min before ischaemia, had no signi®cant e ect on mechanical recovery or glucose metabolism. CHA (0.5 mM) signi®cantly increased the recovery of LV work to 67% of aerobic baseline values and also signi®cantly inhibited rates of glycolysis and proton production. Glibenclamide (30 mM) signi®cantly depressed the recovery of mechanical function to 51% of aerobic baseline values and stimulated glycolysis and proton production. 5 Despite the deleterious actions of glibenclamide per se in post-ischaemic hearts, the bene®cial e ects of CHA (0.5 mM) on the recovery of mechanical function and proton production were not a ected by glibenclamide. 6 The data indicate that the cardioprotective mechanism of adenosine A 1 -receptor stimulation does not involve the activation of K ATP -channels. Furthermore, in rat fatty acid perfused, working hearts, stimulation of K ATP -channels is not cardioprotective and has no signi®cant e ects on myocardial glucose metabolism.