A protease has been found in the bile of 11 mammalian species investigated. The protease, given the tentative name of cholelysin, has been studied intensively in Abattoir ox bile. It makes up less than 10% of the ox bile protein, and is a potent fibrinolysin, as well as being active against the substrates α-casein, and the synthetic esters of tyrosin (ATEE) and arginine (BAEE). It is inactive against trypsinogen, chymotrypsinogen and plasminogen. Isolation and purification of this protease from ox bile proved complex, and was finally achieved by an 8 step procedure which yielded a dry white powder, stable for 45 months (to date) at 4°C. Quality control of this procedure was effected by means o f a fibrin plate assay, using chymotrypsin as a reference standard: the dose response curves of cholelysin and chymotrypsin were closely similar on the fibrin plate, enabling cholelysin (units/l) to be substituted for chymotrypsin (mg/l), in equivalent diameters of fibrinolysis. Gradient elution by tris/NaCl from Whatman DE 32 produced four areas, or Peaks, of fibrinolytic activity of cholelysin, each with some differing characteristics against the various substrates. These complexities were not studied in detail, and a simplification of the procedure was discovered, using batch-wise elution with tris buffer. Thereafter, only the first peak was studied (Peak I). Studies on the inhibition of cholelysin were done using many known inhibitors, including serum of man, and 4 laboratory animals; serum of two patients with homozygous deficiency of α1-antitrypsin; α2-macroglobulin, soy bean trypsin inhibitor, and aprotonin. The serum studies were done with heated and unheated material; platelet-rich and platelet-poor plasma were also studied. Serum was fractionated by paper electrophoresis in an attempt to discover the globulin fraction containing the inhibitor. No inhibition was found in the α-globulin fractions, and inhibition was maximal in the inter-α- globulin and α-globulin fractions. ATEE-esterase activity of cholelysin was inhibited by serum as strongly as fibrinolytic activity. A limited series of studies of coagulation was done; cholelysin was only found to influence fibrinogen, being quite strongly fibrinogenolytic in vitro. A slight effect on ADP-induced aggregation of platelets was found at a low ADP concentration. Using a rat model originally devised for the study of the anticoagulant effect of ancrod, cholelysin was found to be weakly anticoagulant. The dose was low by comparison with ancrod, and the result approached, but did not reach, statistical significance. The split products of plasmin and cholelysin digestion of stabilised fibrin were studied by polyacrylamide gel electrophoresis (PAGE), and these were found to be entirely different. The kinetics of the reaction between cholelysin and fibrin were studied by means of a new technique (the composite cuvette method) and by this method it was shown that cholelysin made a two-phase attack on the fibrin molecule. The attack was studied at 2, 10, and 30 minutes following commencement of fibrinolysis, and the biphasic nature of the attack was proved by a sharp increase in the number of reaction products present between the 2 and 10-minute samples. The molecular weight of fibrin split products by plasmin were shown to agree with those found in published work. Finally, the molecular weight of cholelysin was estimated by PAGE and by column chromatography with and without SDS. It seems probable that the basic molecular weight is ~7 000, with a dimer of ~13 000 and a tetramer of ~ 28 000.