The stimulus of the heart-beat probably starts at the juncture of the superior vena cava with the right auricle of the heart. Some biologists think that in this spot life takes its last stand before the final retreat, but that fact is disputed of late. In the hospital of the Rockefeller Institute for Medical Research in New York, Dr. G. Canby Robinson[60] made records from about eight patients before and during the actual stopping of the heart, using the electrocardiograph, which can be employed without disturbing the patient. He thus found--only in one case, however--that the heart may beat for a half hour after all vascular and circulatory sounds have ceased to be audible. In a letter to me Dr. Robinson said: "Undoubtedly the heart continues to show activity sufficient to be recorded by the string galvanometer very frequently after respiration has ceased, both in man and the lower animals; but this does not necessarily mean that it continues to be an efficient pump, maintaining the circulation. Undoubtedly also in other instances the cardiac activity ceases before the respiration, but I have never obtained electrocardiographic records of such cases." [60] _A Study with the Electrocardiograph of the Mode of Death of the Human Heart. Journal of Experimental Medicine_, 1912, xvi, 291. Crile's experiments upon dogs show that it is possible to resuscitate these animals after they have been apparently dead for periods of time up to seven and a half minutes. The cessation of the blood circulation causes degenerations in the nerve cells and fibres, and these lesions may last even if the animal has been resuscitated. Crile thinks the human respiratory centre may survive anemia from thirty to fifty minutes; the vasomotor and cardiac centres, about twenty to thirty minutes; the spinal cord, eight to ten minutes; the motor cortex, eight to ten minutes; the portion of the brain used in conscious activity as such, six to seven minutes. The higher neurons have been stimulated into reflex activity twenty-five minutes after complete clinical cardiac cessation of activity. In any attempt to resuscitate a person apparently dead the maintenance of the blood circulation is the chief end. If, however, the blood is not oxygenated the circulation will not go on automatically. Artificial respiration is used, and the active principle of the adrenal gland is injected to stimulate the heart. If the heart has stopped in diastole,--that is, when distended with blood,--this distention must be relieved by cardiac massage, commonly through an opening in the thoracic wall. Intratracheal insufflation of oxygen is also to be employed, as a rule.