On autopsy of patients who have died of fevers, parenchymatous and fatty degenerations of the organs have been found. These changes have been ascribed to overheating of the cells.

Litten1 warmed guinea-pigs artificially and noted fatty but no parenchymatous degeneration of the tissues. The space in which the animals were kept was, however, insufficiently ventilated, and the fatty change might have been caused by dyspnea, as results in normal animals (p. 423).

Naunyn2 observed that rabbits might be artificially warmed for thirteen days, so that an average body temperature of 41.50 was maintained without any parenchymatous or fatty degeneration taking place. The animals were supplied with ample food, water, and a sufficient supply of air. Naunyn found that the red blood-cells of rabbits and dogs remained intact even at a body temperature of 420. Welch3 noticed fatty but no parenchymatous change in the tissues of rabbits after exposure to high temperature for at least a week. One rabbit which had been subjected to high temperature for four days was inoculated with the bacilli of the swine plague and died in thirty-six hours, showing extreme fatty changes in the heart and other organs.

Ziegler4 discovered degenerative changes, both parenchymatous and fatty, on artificially warming rabbits. The experiment was continued in 1 case for twenty-nine days. He found, however, a great reduction (30 per cent, and more) in the quantity of hemoglobin in his rabbits. It may well be a question whether the fatty change noticed in the liver and muscles was not due to anemia instead of to the hyperthermia. Since fatty infiltration is known to be caused by dyspnea, which frequently terminates life in fever, one might investigate this subject to see whether parenchymatous change in fever is not solely due to the toxins, and fatty change to the anaerobic cleavage of materials in the cells, which always induces fatty infiltration (p. 489).

1 Litten: "Virchow's Archiv," 1877, lxx, 10.

2 Naunyn: "Archiv fur ex. Path, und Pharm.," 1884, xviii, 49. 3 Welch: "Medical News," 1888, lii, 403.

4 Ziegler: "Kongress fur innere Medizin," 1895, xiii, 345.

Rosenthal1 states that if diphtheria toxin be administered to rabbits the liver is rendered incapable of retaining glycogen. There is hypoglycemia except following glucose administration, when a hyperglycemia greater than that possible in normal animals occurs.

Ever since the experiments of von Leyden2 a retention of water in fever has been assumed. It has also been shown that there is a retention of sodium chlorid within the body. The intimate relation between the retention of water and salt has been beautifully demonstrated by Sandelowsky3 in Luthje's clinic. Thus, during the period of high fever in pneumonia a gain in weight, a sodium chlorid retention, and a dilution of the organic contents of the blood usually went hand in hand. After the crisis, however, a loss in weight, a loss of chlorid, and a greater concentration of blood resulted. Similar conditions were found in scarlet fever.4 Sandelowsky observed that when sodium chlorid was given to a patient convalescent from pneumonia it was not so readily eliminated by the kidney as it would have been normally. He attributed this to a disturbed renal condition which was not wholly restored to the normal after the crisis. This brought about sodium chlorid retention, which in turn caused water retention, that the normal osmotic conditions might be preserved, thus accounting for the gain in body weight and the loss in the concentration of the blood in fever.

It has since been shown that failure to excrete chlorid during the acute stage of the disease is almost always associated with a concentration of sodium chlorid in the blood-plasma below 5.62 grams per liter, which is the normal thresh-hold value of excretion1 (see p. 167). Hence the retention of sodium chlorid is not due to kidney insufficiency.

1 Rosenthal, F.: "Archiv fur ex. Path, und Pharm.," 1914, lxxv, 99. 2 von Leyden: "Deutsches Archiv fur klin. Med.," 1869, v, 273. 3 Sandelowsky: Ibid., 1909, xcvi, 445. 4 Oppenheimer and Reiss: Ibid., p. 464.

As regards the etiology of fever, various attempts have been made to identify a single factor which would cause the high temperature.

Krehl and Matthes2 find that human urine during fever contains an increased quantity of albumoses which have been shown to possess a decidedly toxic action when introduced into ariimals. Klemperer3 denies that these albumoses have any toxic action, and asserts that the results were due to impurities in preparation. In other respects the urine has generally been found to be of normal character. Thus, Mohr4 finds that the relation C to N in the urine is unchanged from the normal, which indicates that there is no qualitative change in the character of the general protein metabolism.

However, there is a very noteworthy record made by A. R. Mandel5 that the rise of temperature in so-called aseptic or surgical fevers is accompanied by a large increase in the purin bases in the urine of patients fed with milk. The temperature rises and falls with the quantity of purin bases eliminated. The uric acid elimination is reduced. These relations are illustrated in Fig. 27 - a case of resection of the knee-joint for tubercular arthritis. The temperatures recorded represent the average of observations made every three hours during the day.

Fig. 27. - Resection of knee-joint for tubercular arthritis.

Another research available in this connection is that of von Jaksch,6 who noted that the purin bodies in the urine of tuberculous patients may increase from a normal equivalent of 4.4 per cent, of the total nitrogen excreted, to one representing 11.3, or even 17.39 per cent. Also Benjamin1 reports a case of typhoid where the urine contained the large quantity of 0.1 gram of purin bases with 0.54 gram of uric acid. Erben2 and Leathes3 report that the output of uric acid is always increased during high fever. Erben also finds that the content of the urine in xanthin bases and amino-acids is greatly augmen ted in measles and chicken-pox; and that the xanthin bases are also increased, though to a lesser extent, in scarlet fever and typhoid. Mandel4 has fed monkeys with bananas and xanthin and witnessed a rise in body temperature, and has noticed that if sodium salicylate be given at the same time no rise in temperature occurs. Ott1 reports that guanin, adenin, and hypoxanthin cause an elevation of temperature in rabbits, while uric acid does not.

1 Snapper: "Deutsches Archiv fur klin. Med.," 1913, cxi, 429; McLean, "Journal of Experimental Medicine," 1915, xxii, 366.

2 Krehl and Matthes: "Deutsches Archiv fur klinische Medizin," 1895, liv, 501.

3 Klemperer: "Naturforscherversammlung," 1903, 2, ii, 67.

4 Mohr: "Zeitschrift fur klinische Medizin," 1904, lii, 371.

5 Mandel: "American Journal of Physiology," 1904, x, 452. 6 von Jaksch: "Zeitschrift fur klinische Medizin," 1902, xlvii, 1.

1 Benjamin: "Salkowski's Festschrift," 1904, p. 61.

2 Erben: "Zeitschrift fur Heilkunde," 1904, xxv, 33.

3 Leathes: "Journal of Physiology," 1907, xxxv, 205.

4 Mandel: "American Journal of Physiology," 1907, xx, 439.

Mandel believes that the purin bases liberated through the toxic destruction of tissue may play a considerable part in producing the temperatures noted in fever. It is evident that the use of purin-free milk instead of purm-containing meat has its scientific justification.

It would indeed be a most striking fact if it should be found that the cause of the febrile temperature lies in the effect of purin bases on the heat-regulating apparatus of the mid-brain acting through the vasomotor system. Antipyretics do not lower body temperature in the normal organism in man. Is their action merely to nullify the action of purin bases upon the nerve-centers? Future research alone can decide this. Such conjectures indicate the extraordinary field which lies open to the investigator in clinical medicine.

1 Ott: "The Medical Bulletin" (Medico-Chir. College), October, 1907.