The influence of myocardial oxygen consumption (MVO2) at the moment of coronary occlusion on the size of the ensuing infarct was investigated in two groups of anaesthetised dogs. In one group (n = 9) coronary occlusion was produced at a high MVO2, estimated to be 14.6 +/- 2.1 cm3 O2 X min-1 X 100g-1 which was changed midway during the occlusion period of 90 min to a low MVO2 estimated to be 6.7 +/- 1.6. In the second group (n = 9) the MVO2, was elevated after 45 min an estimate of 5.1 +/- 0.9 to one of 13.2 +/- 3.7 cm3 X min-1 X 100g-1. In this way the MVO2 averaged over the entire 90 min occlusion period was equal in both groups. Infarct size expressed as a percentage of perfusion area was 68 +/- 28% in group 1 and 32 +/- 30% in group 2 (P less than 0.025). The mass of the perfusion areas did not differ significantly between the groups. Haemodynamics changes were similar in both groups except for a higher rate in group 1 in the period with low MVO2 (109 vs 66 per min, P less than 0.001). Collateral flow was not different between the two groups 45 min after occlusion. It is concluded that the MOV2 at the moment of occlusion significantly influences infarct size. Drastic changes of MVO2 induced at 45 min (ie at half time of the occlusion period of 90 min) had no measurable effect on infarct size.