Techniques for organ preservation generally use hypothermia to retard metabolic requirements. However, excessive hypothermia may also produce injury. Using a canine left lung allotransplantation procedure, we compared two preservation temperatures (4° and 10° C) in terms of subsequent lung function measured by temporary occlusion of the right pulmonary artery after implantation of the preserved left donor lung. The lungs were flushed with low-potassium dextran electrolyte solution, inflated with 100 % oxygen, and preserved for 18 hours. To investigate possible changes of energy stores at different temperatures, we performed phosphorus 31-nuclear magnetic resonance analyses of lung samples. Sequential determinations of adenosine triphosphate levels in lung tissue preserved at 4°, 10°, and 22° C were studied. After transplantation, lungs preserved at 10° C (n = 6) provided significantly better arterial oxygen tension than those preserved at 4° C (n = 6), 451 ± 46 mm Hg versus 243 ± 86 mm Hg (p < 0.05), and lower pulmonary vascular resistance, 581 ± 68 dynes · sec · cm-5 versus 1006 ± 157 dynes · sec · cm-5 (p < 0.05). Adenosine triphosphate levels at 4° and 10° C were stable and did not differ from each other at the end of the 18-hour preservation period: 0.86 ± 0.04 μmol/gm wet weight for control versus 0.86 ± 0.07 μmol/gm wet weight for 4° C and 0.93 ± 0.06 μmol/gm wet weight for 10° C after 18 hours of preservation. Preservation at 22° C caused a 28% depression of adenosine triphosphate after 18 hours of preservation. These results lead us to conclude the following: (1) Optimal temperature for lung preservation is in the vicinity of 10° C, and (2) lung dysfunction caused by excessive hypothermia is not due to a failure to maintain adenosine triphosphate levels. We suspect that adenosine triphosphate is generated by oxidative phosphorylation during lung preservation.