Reports regarding hemorrhage into the tissue with application of a tourniquet or under conditions similar to those which exist with the use of a tourniquet have appeared from time to time in the literature2,11,24. This project sought not only to quantitate the extent to which bleeding occurred but also to study those factors changes in which might render possible an explanation for the increased bleeding. Capillry permeability, blood-coagulation factors, and components of the fibrinolytic system offered likely avenues of inquiry. The clotting mechanism in the dog is more rapid in action than that in man; the Lee-White clotting time being one-half to two-thirds that of the human22 and the prothrombin time about half that of man20. While there are differences between the dog and the man, it is generally held that the similarities far outweigh the differences. For example, a strain of Factor VIII deficient dogs has been developed1. Most of the factors which have been found necessary for spontaneous coagulation in man have also been shown to be necessary for the coagulation mechanism of the dog. A lengthening of recalcification time and Lee-White clotting time suggested the development of an abnormality in blood coagulation. Blood drawn from the experimental animals after tourniquet removal showed an increase in euglobulin lytic activity and in plasma lytic activity. The moderate impairment of blood coagulation seen in the experimental animals may have been related to destruction by fibrinolysis of some fibrinogen and other coagulation factors4,5,10,24. Products of fibrinolytic activity may have been responsible for the appearance of an increase in antithrombin activity. It is well known that when fibrinogen is lysed by fibrinolysin, anticoagulant substances appear17,25,27. The increased fibrinolytic activity may have been caused by activators released from the hyposic area3. These substances could have been responsible for the transitory fibrinolytic activity observed as well as its sequelae, that is, increased antithrombin, and so forth. Some of the fibrinolytic activity may have been masked by the apparent increase in fibrinogen which took place. The studies of the tourniqueted limb indicated that capillary permeability was [See table in the PDF file] increased. There was no difference between the capillary permeability in this limb and that in other capillary beds in the same animal. A factor (probably a kinin)6 was released during the hypoxic phase which spread throughout the experimental animal when the tourniquet was removed16. Under these conditions, a large proportion of the water of the blood could have been lost to the interstitial space. This water would be unaccompanied by molecules as large as fibrinogen. Examination of the hematocrit showed a marked increase in every experimental animal (Table II). The decreased clot retraction closely paralleled the increased hematocrit. Hemoconcentration undoubtedly occurred. This renders more difficult the assignment of significance to the various changes in coagulation and fibrinolytic machinery. Suffice it to say changes in coagulation and fibrinolytic activity undoubtedly occurred and uniformly in a direction which would have favored an increased bleeding tendence. [See table in the PDF file] In an experiment to be published, in which the leg was cooled by 10 degrees centigrade before application of the tourniquet, a sharp diminution in the changes reported herein was observed. In preliminary experiments in man (four cases) in which the temperature of the limb to be operated was lowered by 10 degrees centigrade before application of the tourniquet, a constant decrease in postoperative complications was observed17. It is believed that if the temperature were lowered by as much as 20 degrees immediately before application of the tourniquet, the negative effects might be abolished and a valued surgical technique would be rendered entirely safe for the patient.