Method.-Observations were made both on isolated mammalian preparations and on decapitated animals : 28 experiments were performed on 10 animals (two rats, six rabbits, and two cats).Rats and rabbits were used for the isolated preparations. The animal was killed in the usual way by a blow on the back of the neck, and the auricle or ventricle to be used was quickly removed and placed in oxygenated Ringer-Locke solution kept in a bath at 37° C. This was done by one of us while the other removed as quickly as possible one or more of the thin muscles from the forelimb and placed it in the same Ringer-Locke solution. The two tissues were then placed in two separate cylinders containing oxygenated Ringer-Locke solution and kept at 37° C. in a water bath. The arrangement was such that, when necessary, oxygen or nitrogen could be bubbled through the solu tions. The chambers could also be rapidly and simultaneously emptied and refilled with fresh Ringer-Locke solution. This was done when the gas was changed, so that there should be no delay in the response due to residual gas dissolved in the solution. The muscles were attached to light isotonic levers and were stimulated with break shocks through platinum electrodes at rates varying from 40 to 100 per minute.It was found that even with the greatest care the skeletal muscle preparation would not keep in good condition for more than a short time : consequently, the observations were repeated on decapitated cats and rabbits. Sherrington's method of decapitation was used (we are indebted to Dr. N. B. Dreyer for demonstrating the technique). Artificial respiration was instituted, the chest opened and the heart freed from its pericardium. The records of the ventricular contraction were made by attaching the apex to an ordinary lever. Cotton wool was loosely packed beneath the heart to prevent the shifting of its base with respiration and the consequent distortion of the records. The gastrocne mius muscle of one of the limbs was exposed and freed from its surroundings, care being taken not to damage the blood supply. The tendon was cut and fastened to a lever ; the knee and ankle joints were clamped to render the limb immobile. The heart and muscle were then stimulated as in the isolated preparations. The electrodes on the heart were placed on the auricles in the region of the sinus node, and so arranged that they moved with the heart and so offered no impediment to its contraction. In some experiments the heart was not stimulated but allowed to beat naturally. One or more electric
