IN a recent series of papers, Kroll [1933, a, b, c] has claimed that, by making an acetone extract of the fresh brains of animals killed during convulsions, material is obtained which produces convulsions when injected intravenously into other animals. The brains of sleeping animals, on the other hand, when similarly treated yield a sleep-producing substance, while the extract of the brains of normal animals has no effect.Kroll's technique is simple. Convulsions are produced by electrical stimulation of the cerebral cortex, by the use of convulsant drugs, or by causing insulin hypoglycaemia. Sleep is induced by hypnotic drugs, or else the brains of hibernating animals are used. The animal is killed, the brain excised, rapidly ground up with sand, and extracted with acetone. The acetone is removed by distillation in vacuo at a low temperature, and the resulting aqueous suspension diluted with normal saline, and injected intravenously into another animal. Kroll's papers should be consulted for further details: he used cats, rabbits, and, when hibernating animals were required, hamsters.We are already familiar with the part played by acetylcholine in the activity of the autonomic nerve cells. Sherrington has postulated the existence of " central excitatory " and " central inhibitory " states in the central nervous system, which may well be manifestations of the activity of a chemical agent. Many drugs are capable of provoking convulsions or sleep. Kroll's claims, therefore, startling as they appear, clearly merit serious investigation.EXPERIMENTAL. Rabbits were used in all the experiments described.A preliminary experiment was performed as follows. A large rabbit was ansesthetized with ether, and the cerebral cortex was exposed, and