Giving fat with meat to a diabetic will not ordinarily increase the sugar in the urine. The writer has never observed such an increase in any of the work of his laboratory. A large production of sugar from fat has been elsewhere reported,3 and Cremer4 finds that glycerin alone will increase the output of sugar in the urine. (See p. 262).

On giving meat in diabetes the fat metabolism is reduced as it would be in the normal organism, and yet there is no effect on the D : N ratio, and therefore the latter cannot be influenced by the quantity of fat burned. This is shown in a respiration experiment made by Mandel and Lusk5 on a dog with phlorhizin glycosuria whose metabolism starving and after meat ingestion was as follows:

The protein metabolism doubled when meat was ingested, the fat metabolism fell, but the D : N ratio remained constant.

It has also been demonstrated that neither exposure to cold nor mechanical exercise, both of which result in a largely increased metabolism of fat, has any effect on the sugar output in pancreas diabetes6 or in phlorhizin glycosuria.7 Freund and Marchand8 found that ten hours' exposure to the winter's cold reduced the blood-sugar of a phlorhizinized dog to zero. The writer found in a phlorhizinized dog which had been rid of glycogen by shivering and exercise that the composition of the urine was unchanged as the result of traveling 1500 meters in a revolving wheel, an effort which would have more than doubled the metabolism of fat during the hour when the exercise was taken. The analytic data for two-hour periods were the following:

1 Janney: "Journal of Biological Chemistry," 1915, xx, 321.

2 Pfluger: "Pfluger's Archiv," 1905, cviii, 115.

3 Hartogh and Schumm: "Archiv fur exp. Path, und Pharm.," 1901, xlv, 11.

4 Cremer: "Munchener med. Wochenschrift," 1902, xlix, 944.

5 Mandel and Lusk: "American Journal of Physiology," 1903, x, 54.

6 Allard: "Archiv fur exp. Path, und Pharm.," 1908, lix, III; Seo, Ibid., P. 341.

7 Lusk: "American Journal of Physiology," 1908, xxii, 163.

8 Freund and Marchand: "Archiv fur exp. Path, und Pharm.," 1913, lxxiii, 276.

In this experiment exercise was without influence on the excretion of nitrogen. If, however, the animal contains residues of glycogen which as a result of exercise are converted into sugar and eliminated, then there is also an increased nitrogen elimination as the result of work. This is suggestive of a chemical union between glycogen and nitrogenous substances.

The-theory of the origin of sugar from fat was supported by Falta,1 who found a largely increased sugar output after administering adrenalin to dogs with pancreas diabetes. Among the cases of high D : N in human diabetes reported from von Noorden's clinic that described by Bernstein, Bolaffio, and Westenrijk2 is the most remarkable. The ratio, after deducting the carbohydrates ingested in the food, often reached D : N : : 10 :1. The high ratios in diabetes are explained by Falta as being due to very great activity on the part of the adrenals which not only inhibits the internal secretion of the pancreas, but also causes a production of sugar from fat. However, Ringer,3 working in the author's laboratory, finds that if adrenalin be administered to a fasting phlorhizinized dog, although the first administration of the drug may bring about an elimination of "extra sugar" which may be discharged from the glycogen repositories of the body on account of the anemia of the tissues (see p. 447), a second injection of adrenalin may be entirely without influence on either the sugar or nitrogen elimination. This indicates that adrenalin does not cause a production of sugar from fat.

1 Eppinger, Falta, and Rudinger: "Zeitschrift fur klinische Medizin," 1908, lxvi, 1.

2 Bernstein, Bolaffio, and Westenrijk: Ibid., 1908, lxvi, 378.

3 Ringer: "Journal of Experimental Medicine," 1910, xii, 105.

The high D : N ratios reported above, as well as many similar observations described in the literature, are unquestionably due to the surreptitious ingestion of food containing carbohydrate.

Falta explains the results of many experiments by stating that while the secretory activities of thyroid and adrenals are each stimulated by the secretions of the other, the activity of the pancreas is in like manner inhibited by the secretions of the other two glands. Therefore supersecretion of adrenalin inhibits the secretory function of the pancreas so that the organism can no longer oxidize carbohydrates, and at the same time it stimulates the thyroid, causing increased protein metabolism. Furthermore, in exophthalmic goiter, where there is supersecretion in the thyroid gland, there is a tendency to glycosuria, and it is believed that true diabetes has been induced by this cause.1 Administration of thyroid extracts to dogs also produces glycosuria. Cecil,2 working under Opie's direction, finds lesions of the pancreas in cases of diabetes associated with exophthalmic goiter, and Forschbach and Severin,3 in Minkowski's clinic, believe that there is very likely a slight disturbance of the pancreas in some cases of hyperthyroidism.

1 Magnus-Levy: von Noorden's "Handbuch des Stoffwechsels," 1907, Bd. ii, p. 333.

2 Cecil: "Journal of Experimental Medicine," 1909, xi, 266.

3 Forschbach and Severin: "Archiv fur exp. Path, und Pharm.," 1914, lxxv, 168.