The identification and synthesis of the hemorrhagic agent of hemorrhagic sweet clover disease, dicoumarin [3,3′-methylene-bis-(4-hydroxycoumarin)], by Link and his associates1have provided a pure substance which can be administered to animals and man and cause a considerable diminution of the prothrombin content of the blood. Their paper carried also the obvious suggestion that this compound might be of value in clinical conditions when interference with the coagulability and prothrombin content of blood was desirable. Since the most probable use of this substance in clinical medicine is in the prevention of thrombosis, its value in such conditions can be determined only by clinical trial. However, several questions should be answered by experiments with animals: First, does the lowering of the prothrombin content of the blood by this drug prevent or delay experimental thrombosis? Second, can the drug be given in amounts sufficient to produce the desired effects without producing untoward