other desperate cases. In one, hemorrhage of the newborn 30 c.c. were given, and six hours later 30 c.c., totaling for the first twenty-four hours 60 C.C. The second twenty-four hours the serum was given about the same way. After that the dose was diminished. The injections covered a period of five days, were twenty-three in number, and totaled 637 c.c. An extreme case of malnutrition of infancy was treated for twenty-one days, getting for the first week daily doses ranging from 20 c.c. to 78 c.c., the total for the twentyone days being 196 c.c.

Case "C. M.," a boy of ten years, mother hemorrhagic, and patient also bleeding excessively from slight wounds, received puncture wounds of mouth from his teeth. For five days he received calcium lactate internally, peroxid and adrenalin locally. Compresses of gelatin locally were changed q.2.h., and other recognized remedies were administered. On the fifth day, one ounce of serum was injected and the bleeding stopped within four hours,

never to return.

Case "I. G.," twenty-nine years of age. Epistaxis resisted all usual treat ment as just outlined and repeated packing of the nares. On the fourth day the red blood cells were down to 2,000,000 and the hemoglobin to 65%. Patient received one injection of serum. Bleeding stopped thirty-six hours later and did not return.

Such cases of bleeding are probably due to heredity plus delayed coagulation, the latter due in many cases to some poison either elaborated in the system or introduced from without. Bacteriemia or any grave sepsis or poisoning that effects the nutrition of the endothelial cells may produce this hemorrhagic diathesis. It is common

in post-operative cases with jaundice. Observers have claimed that the blood plates are diminished in hemophiliacs. This is difficult to decide as we have no very accurate way of estimating them. These blood plates are nuclei around which fibrin deposits, forming thrombi. These blood-plate-thrombi cause the pain of hemorrhagic points. As normal serum reduces the coagulation time of blood it might be supposed that such injections would increase the tendency of extravasated blood to clot in the tissue and to increase the attendant pain. This, however, is not so, for previous clotting in the tissues is absorbed after serum injections and no such clots are found. as a result of the injections.

Dr. Welch speculates that the beneficent action may be due to the nutritional effect of serum on the endothelium of the blood vessels. The serum supplies a prepared food, with molecules, receptors, etc. (according to Ehrlich's side-chain theory). The evidence is that the checking of hemorrhage in bacteriemias is not due to any punitive action toward the invading force, for there is no conclusive evidence that the serum has any effect in limiting the course of such disease, except, possibly, by its action on hemorrhage and nutrition. Cases of meningitis (without hemorrhage) due to the streptococcus and the pneumococcus have not responded to subcutaneous injections of serum.

Cases that succumb to infection may be placed in two groups: (1st) Those having in their blood sufficient antibodies but not sufficient complement; (2nd) those having sufficient complement but not sufficient antibodies. For the second group we could expect no benefit from normal human serum

tion than that mentioned, but some do, and additional stimulation may be provided by caffeine citrate in doses. of two or three grains every four hours, camphor in doses of two to five grains every four hours, aromatic spirits of ammonia in doses of onehalf to one dram every one or two hours, and digitalin in doses of one one-hundredth to one-fiftieth of a grain every four hours. Whisky in small or moderate amount should be given to patients who have been addicted to alcoholics.

In the beginning of the disease a small dose of calomel followed by a dose of jalap or castor oil is usually good treatment, and the bowels should be kept open during the disease, but at the time of the crisis the patient should not be disturbed even for bowel movements.

In the first twenty-four or thirtysix hours of the disease, if the patient can be seen so early, tincture of aconite in small doses may be useful; it is best given in half or quarter minim doses every hour.

To secure sleep, especially in the early days of the disease, morphine may be necessary.

The value of fresh air in the treatment of pneumonia has been abundantly proven, but not the value of exposure to excessive cold, especially at the time of and just after the crisis.

Hydrotherapeutic procedures to reduce the temperature are of questionable value in this disease.

In cases which show by a sudden fall in the systemic blood pressure, without notable signs of failure of the right heart, that vasomotor paralysis is impending, adrenalin and its congeners should be given to restore the tone of the dilated arteries; and normal sa

line solution should be given through the rectum and by hypodermoclysis.

THE TREATMENT OF INTERNAL

HEMORRHAGE

Only the pulmonary hemorrhage of phthisis and the intestinal hemorrhage occurring in typhoid fever are referred to here. In both these conditions quiet of both the nervous and muscular systems should be secured, which can be done most effectively by morphine. Calcium chloride, in ten or fifteen grain doses every four hours, can be given with a reasonable hope that it will increase the coagula

bility of the blood. External applications of cold over the region of the body which includes the bleeding part, may possibly help; in the case of intestinal hemorrhage they do so, if at all, probably more by quieting peristalsis than by reflexly inducing contraction of the bleeding artery; but their routine use is not to be recommended. In the case of intestinal hemorrhage all food should be stopped, or reduced to the blandest fluids, e.g., barley

water..

The arterio-constrictors, adrenalin, ergot, etc., are extensively employed in pulmonary and intestinal hemorrhage, but the reasons for their use in these conditions are not very good: even if the bleeding were from an artery instead of a vein, and these drugs could contract the bleeding vessel, they would also contract the arteries thruout the body and raise the general blood pressure, and thereby increase the hemorrhage.

The giving of astringents by mouth to check bleeding from a point in the intestines seems so irrational that I would not mention it even to condemn it, were it not that such treatment seems to have the sanction of usage by