1. Chloroform anesthesia for two hours or more will cause more or less central liver necrosis in dogs, depending on the length of the anesthesia and the susceptibility of the animal.
2. If the fibrinogen of the blood of such an animal be estimated at intervals, it is found that this proteid shows a drop corresponding to the amount of liver necrosis.
3. By administering chloroform, the fibrinogen may be almost eliminated from the circulating blood, and the poisoned animal may bleed for hours from small skin pricks or cuts.
4. The liver can recover from a grave injury due to chloroform and return practically to a normal condition in about ten days.
5. The fibrinogen reappears in the blood as the liver effects its repair. It seems that the quantity of fibrinogen present is a good indicator of the liver efficiency and a fairly accurate index of the amount of liver injury.
6. Shortly after the recovery of the liver from an injury due to chloroform, one may find an excess of fibrinogen in the blood.
7. In severe cases of chloroform poisoning, the calcium of the blood was normal or slightly increased, and the thrombin was normal.
8. These experiments give no evidence that the formation of thrombin or prothrombin is dependent upon liver activity.
9. The hemorrhages of chloroform poisoning are due not to lack of blood clotting but to inefficient coagulation. The clot has not the body and toughness supplied by the fibrinogen, and is, therefore, unable to check even capillary hemorrhage.
10. Fibrinogen is either formed in the liver or is wholly dependent upon liver activity for its production.