Hawkins and Whipple (I) recently described a number of abnormalities which occur commonly in dogs having long standing biliary fistulae. They observed a bleeding tendency which could be cured by feeding bile. If the animals were fed bile periodically in the earlier stages of the experiment the bleeding tendency did not develop. Subsequently, one of us in conjunction with Hawkins (2) showed that the bleeding was due to profound lowering of the plasma prothrombin level. When the bleeding was prevented by feeding bile, the prothrombin remained within normal limits.Several years ago a hemorrhagic deficiency disease of chicks was described in several laboratories. Dam (3) and Almquist and Stokstad (4) have identified the deficiency factor as a new fat-soluble vitamin, which has been designated the antibemorrhagic vitamin, or vitamin K. Both Schg~nheyder (5) and Quick (6) have concluded that the bleeding in this hemorrhagic chick disease is due to a low plasma prothrombin. These findings suggested that the biliary fistula dogs also suffer from vitamin K deficiency, a result of faulty absorption when bile is excluded from the gut. That this explanation is correct is indicated by a preliminary report of Greaves and Schmidt (7) on rats. Important confirmation is also supplied by our successful treatment of jaundiced human bleeders with vitamin K (8). A preliminary report of our work with patients was followed almost at once by a similar report from the Mayo Clinic (9).