In three groups of rabbits, the rectus femoris muscle was subjected to 4 hours of total ischaemia. In Group 1 (normothermia, n = 5) the core temperature was maintained within the range 36-38°C for the duration of ischaemia. In Group 2 (total hypothermia, n = 5) the core temperature was allowed to fall to 31.5-33.5°C. In Group 3 (muscle only hypothermia, n = 5) core temperature was maintained as in Group 1 but the muscle temperature was allowed to fall to 29.5-31.5°C. After 24 hours of reperfusion the muscles were harvested and measurements made of muscle viability, oedema and myeloperoxidase content. The mean (s.e.m.) muscle viability of Group 1, 19.5 (3.8)%, was significantly less than that of both Group 2, 86.0 (2.0)%, and Group 3, 87 (4.1)%, (P < 0.001). Muscle oedema and myeloperoxidase levels were elevated in all experimental groups, but differences were not significant. These findings indicate that ischaemia-reperfusion injury in skeletal muscle in this model is highly temperature-sensitive, small reductions in muscle temperature during ischaemia providing significant protection against ischaemia-reperfusion injury.