Hypothermia remains the primary adjunct employed to lower cellular metabolism during various cardiac procedures. In these experiments, left ventricular myocardial oxygen consumption (MVO2) and transmural blood flow (TBF) were measured during cardiopulmonary bypass with the range of temperatures used clinically. Determinations were made in empty beating normothermic hearts and after potassium cardioplegia at 37, 32, 28, 22, 18, and 15° (K+ = 15–37 meq/L: Hct 25 volumes %). Oxygen content of the total coronary sinus collection was compared with a large volume arterial sample using a Lex-O2-Con-TL analyzer (vs Van Slyke, R = 0.98). Transmural blood flow was measured at each temperature using microspheres (8μ), and perfusion was maintained at 80 mmHg. Asystole (37°) alone decreased MVO2 from 5.18 ± 0.55 to 1.85 ± 0.20 ml O2/min/100 g of left ventricle or approximately 65% (p < 0.001). With progressive cooling to 15° an additional 82% decrement in oxygen uptake occurred during asystole (p < 0.001). During asystole at 37° the decrease in MVO2 was reflected mainly by a large decrement (p < 0.01) in TBF (1.27 ± 0.19 to 0.74 ± 0.17 ml/min/g of mean left ventricular flow). However, with cooling below 32°, the arteriovenous oxygen difference narrowed progressively (p < 0.001) while TBF paradoxically returned to control levels. Endocardial/epicardial flow ratios were not altered by cooling. These data not only confirm earlier reports describing a sequential drop in MVO2 with incremental myocardial cooling, but also establish MVO2 levels for perfused hearts arrested by potassium at lower temperatures (18–15°). Moreover, as transmural blood flow becomes independent of metabolic necessity during hypothermia, coronary autoregulation appears to be impaired, possibly affecting detrimental tissue over perfusion.