Since the microcirculation is impaired in diabetes and since burn injury is associated with microvascular thrombosis, the purpose of this study was to determine whether diabetes increases the severity of burns. Swiss white mice were made diabetic with alloxan, and control animals received saline. One week later, animals with over 2000 mg/dl urine sugar and the control animals were anesthetized and burned uniformly on the back with a steel disc at 100 C for 10 sec. At 24 hr, burn severity was evaluated on a scale of 0 to 4 using gross and microscopic criteria. The mean burn severity of the diabetic mice was 0.80 +/- .15 and that of the control mice was 3.22 +/- .09 at 24 hr (P less than 0.001.). At 5 days, the differences of the burn severity between the two groups were still significant, but 5/10 (50%) of the diabetic mice and none of the control mice died. In diabetic mice, through burn severity appears mild in the wound, the mortality is high. Therefore, the criteria for initial evaluation of the diabetic with burns need to be reassessed.