It will be noticed that the nature of the action of caffein, whether it is a nerve or a muscle poison, formed the subject of several investigations. Binz(11) (1890) brought forward additional evidence in support of the view that caffein acts primarily on the ganglion cells, and not on the muscle directly. This he has shown by injecting 0.5 gram into each of two rabbits after cutting the sciatic nerve on one side; in one case he also resected the obdurator and crural nerves on the same side. Clonic spasms developed in both subjects soon after caffein was given, but in each rabbit the side operated upon remained paralyzed. Baldi(8) (1891) studied the action of caffein on _Rana esculenta_. After injecting from 4 to 20 mg tetanus, such as observed in strychnin poisoning, was noticed. Fröhner(26) (1892) made observations on the comparative toxicity of caffein in domesticated animals. After the administration of 5 grams of caffein sodium salicylate by mouth to a dog weighing 10 kilos, he noticed salivation, restlessness, vomiting, and convulsions as in strychnin poisoning. Death occurred three hours after the drug was given. On autopsy he noticed mild inflammation of the mucous membranes of the stomach and intestines and edema of the lungs; the heart was in diastole. A dose of 2 grams of caffein sodium salicylate given to the same animal subcutaneously two days previously provoked only very slight symptoms. The subcutaneous injection of 10 grams of the same preparation into a pig weighing 30 kilos caused death in two and a half hours, with the production of symptoms of disturbance of the nervous system and of gastrointestinal irritation. The same dose per kilo of body weight given to a goat likewise caused death in two and a half hours after its administration. Examination on autopsy revealed inflammation of the gastrointestinal tract. Similar lesions were found in a horse killed by 100 grams of caffein, in which he also noticed hemorrhage of the mucosa in the fundus of the stomach. Gourewitch(28) (1907) conducted experiments with caffein on rabbits, pigeons, and white rats. It appears from his protocol that single doses of about 0.2 to 0.25 gram caffein per kilo given subcutaneously proved to be fatal. He states, however, that the resistance to caffein was markedly diminished, when its administration was repeated daily, for much smaller amounts sufficed to cause death in these animals. A dose of 120 mg of caffein per kilo proved fatal after the third injection. When the dose was increased to 170 mg per kilo, the animal succumbed to the effects of caffein after the second injection. His experiments on the other animals do not indicate the degree of resistance to caffein, since the weights for some were not given while for the others no attempt was made to determine the minimum toxic or fatal dose.