Electrical anesthesia in mammals, produced by means of an interrupted direct current, is reported by Leduc, Robinovitch, von Neergard, and Ivy and Barry, and denied by Sack and Koch. Von Haareveld and Kok obtained narcosis in the dog with a sinusoidal current and Scheminzky and others in the frog with a constant one. Conflicting results may have prevented the use of electro-narcosis in the clinic, yet it seems to offer much promise. This is a preliminary report of the successful production of anesthesia in rats by a constant direct current. Large dry cells (145 volts), a graphite rheostat, a milliammeter, and a reversing switch are connected in series with a rat through non-polarizable, zinc-zinc sulfate electrodes. The cathode is firmly pressed against the roof of the mouth and the anode is inserted about 1 cm into the rectum. The animal lies with his legs dependent through holes in a platform. The rat is usually anesthetized with ether prior to the application of current, to avoid struggling against confinement. Control animals without current, emerge from this anesthesia in 5 minutes (never more than 10 minutes). The experimental animals are subjected to a current of 10 mA (rheostat at 11,500 ohms) gradually increased from zero during a period of 3 minutes. As long as this current flows, the animal remains quiescent and gives no response to severe noxious stimuli, such as cutting, burning, or powerful tetanizing shocks. (Tests have been continued up to 4 hours. After a half-hour the anesthetizing current can be reduced to 8 mA.) To terminate anesthesia, the current is decreased to 4 mA during a minute, its direction reversed for 10 seconds, and then discontinued.