Baboons that were subjected to systemic hypothermia at 32 C had an arm skin temperature of 27.3 C and bleeding time of 5.8 minutes. With local warming of the arm skin to 34 C, the bleeding time was 2.4 minutes. In normothermic baboons with arm skin temperature of 34.6 C, the bleeding time was 3.1 minutes. Local cooling of the arm skin to 27.6 C produced a bleeding time of 6.9 minutes. Increasing the skin temperature of the arm in hypothermic baboons to 38.9 C and in normothermic baboons to 40.1 C reduced bleeding times to 2.1 and 2.3 minutes, respectively. In both hypothermic and normothermic baboons there was a negative and significant correlation between the bleeding time and the arm skin temperature and the thromboxane B2 level in the shed blood obtained at the template bleeding time site. There was a significant positive correlation between the thromboxane B2 level in the shed blood and the arm skin temperature. Both in-vivo and in-vitro studies have shown that the production of thromboxane B2 by platelets is temperature-dependent, and that a cooling of skin temperature produces a reversible platelet dysfunction. Data also suggest that when a hypothermic patient bleeds without surgical cause, skin and wound temperature should be restored to normal before the administration of blood products that are not only expensive but may also transmit disease.