Both aprotinin and tranexamic acid are effective inhibitors of fibrinolysis in vitro and in vivo and both agents can act as plasmin inhibitors in purified systems, although there is some debate on their exact mechanism of action in vivo. The studies reported here using an in vitro clot lysis system designed to provide precise inhibition constants show that aprotinin remains a very potent inhibitor of plasmin even in the presence of fibrin with Ki = 2 nM. Plasmin-aprotinin interactions in solution are not affected by a number of kringle binding ligands, aminohexanoic acid, tranexamic acid or CNBr-fibrinogen fragments with Ki = 0.4 nM. The difference between these two Ki values is explained by competition for the plasmin active site between substrate (fibrin) and inhibitor (aprotinin). Inhibition of fibrinolysis by tranexamic acid is not readily analysed by a simple inhibition model which may be due to multiple overlapping ligand-kringle interactions or tranexamic-fibrin interactions. Experiments using combinations of aprotinin and tranexamic acid in the clot lysis system confirm the complementary nature of inhibitory mechanisms and suggest a slight synergism. These results support the idea that aprotinin inhibition of plasmin is a primary mode of action in vivo, and suggest that combination therapy of aprotinin with tranexamic acid might be more effective than either inhibitor alone.