Marage employed, in his experiments, the larynx of the dog. In order to spare the animal useless suffering, morphine was first administered hypodermically and, three hours later, the dog was put under the influence of chloroform, and the larynx, with five or six rings of the trachea, was excised. A rubber tube of the diameter of the trachea was then connected with the latter by means of a short tube of thing glass, so that a current of cold air could be forced through the extirpated larynx. The pressure of the air was measured with a very sensitive metallic manometer graduated in millimeters of water pressure. The compressed air was stored in a rubber bag similar to those which are employed for inhalations of oxygen, and was kept at the temperature of 98.6 deg. F. The muscles of the larynx were stimulated by the current of a small induction coil, which was energized by a storage battery, and the sounds emitted by the larynx were recorded by a phonograph. The following conclusions were reached:

When the larynx of a dog is removed during chloroform anesthesia, the laryngeal muscles retain their ability to contract for a short period, which varies from 3 to 10 minutes, but no contraction can be produced in the muscles of a dead larynx, even if it is removed immediately after the death of an animal, because the arterial blood has escapes.

In order to produce the vibrations, the current of air should be impelled by a pressure of from 6 to 8 inches of water, as it is in the normal production of the human voice. In these conditions the excised larynx of the dog barks and howls in every note of the canine register, from the deep baying of a mastiff to the shrill pipe of a terrier. These various notes are obtained at will by causing various muscles to contract.