It has long been known that diabetics eliminate sugar even after all administration of sugar is stopped. It has also been generally recognized that protein ingestion tends to increase the sugar output in the urine, while fat has no effect.

A large amount of information has been collected concerning the relation between the urinary nitrogen and sugar elimination in the fasting and meat-fed diabetic organism. The dextrose to nitrogen ratio (D : N) is a key to the problem of the quantity of sugar which can be derived from protein metabolism (p. 173).

Minkowski1 was the pioneer who discovered that depan-creatized dogs, whether fasting or fed with meat, showed a constant elimination of 2.8 grams of glucose for each gram of nitrogen in the urine. This ratio (D : N : : 2.8 : 1) was the average obtained from 7 dogs on twenty-two different days. The lowest ratio was 2.62 :1, the highest 3.05 : 1. Some other operators have been unable to obtain these ratios. Pfluger2 finds a variable and generally lower ratio, and his dogs all died of abscesses. Embden's3 ratios are all lower than Minkowski's, and are probably due to incomplete extirpation of the pancreas.

The accuracy of Minkowski's results is indicated by the fact that the ratio (D : N : : 2.8 : 1) may be easily established by the administration of phlorhizin to rabbits, goats, cats, and in certain dogs whose kidneys have been somewhat affected, as, for example, by giving camphor. Phlorhizin acts first to cause a sweeping out of the excess of sugar in the organism, with a subsequent establishment of the ratio. (See table, p. 463.) The ratios in different animals are given in the following table:

1 Minkowski: "Archiv fur exp. Path, und Pharm.," 1893, xxxi, pp. 85, 97.

2 Pfluger: "Das Glycogen," 1905, p. 491.

3Embden and Salomon: "Hofmeister's Beitrage," 1905, vi, 63.

Ratios In Diabetes Of D : N : : 2.8 : 1

The uniformity of the ratio as shown in different animals is very striking. One may calculate from these results that 45 per cent, of the protein molecule may be converted into dextrose in the course of metabolism.

This, however, does not complete the story of the D : N ratio, for a higher ratio, or 3.75 :1, was discovered by Reilly, Nolan, and Lusk6 in the urine of dogs with normal kidneys, after subcutaneous injections of phlorhizin. This ratio was subsequently revised by Stiles and Lusk7 and found to be 3.65 :1. The importance of this discovery was enhanced by the finding of Mandel and Lusk8 that the same ratio may exist in human diabetes when the patient is given a diet of meat and fat. The ratios found on successive days are thus comparable:

1 Minkowski: Loc. cil., p. 97.

2 Jackson: "American Journal of Physiology," 1902, viii, p. xxxii.

3 Arteaga: Ibid., 1901, vi, 175.

4 Lusk: "Zeitschrift fur Biologie," 1901, xlii, 43.

5 Reilly, Nolan, and Lusk: "American Journal of Physiology," 1898, i, 396.

6 Reilly, Nolan, and Lusk: Loc. cit.

7 Stiles and Lusk: "American Journal of Physiology," 1903, x, 67.

8 Mandel and Lusk: "Deutsches Archiv fur klin. Medizin," 1904, lxxxi, 479.

9 Stiles and Lusk: Loc. cit., p. 77. (Details, this book, p. 99).

10 Benedict, S. R., and Lewis, R. C: "Proceedings of the Society for Experimental Biology and Medicine," 1914, xi, 134. (Details unpublished).

11 Mandel and Lusk: Loc. cit., p. 479.

12 Greenwald: "Journal of Biological Chemistry," 1913-14, xvi, 375.

In another place (p. 174) it has been shown that the D : N ratio does not vary after the ingestion of sufficient meat to double the quantity of nitrogen in the urine; the sugar also doubles. The sugar production is therefore proportional to the protein metabolism, and, apparently, must be derived from protein.

Various objections have been raised to this statement. Other experiments, however, confirm the above proposition.

Luthje1 gave ".nutrose" to a depancreatizeddog. "Nutrose" contains casein, but no sugar. The-dog weighed 5.8 kilograms and eliminated 1176 grams of glucose during twenty-five days. The tissues of the dog could not possibly have contained over 232 grams of glycogen at the beginning of the experiment. The source of the sugar could not have been the aiiimars store of glycogen, but it must have arisen from either protein or fat.

The D : N ratio of 3.65 : 1 was accepted by Lusk as being true for the dog because the greater number of the higher ratios which were found were established at this level. Janney2 prefers to take the average of all determined D : N ratios and in this way arrives at a ratio of 3.43 :1. He argues that, since 4.7 per cent, of the urinary nitrogen is in the form of creatin and creatinin, which are not glucose formers, a correction would bring up the D : N ratio in the dog to 3.60. Although it is not clear why the urinary creatin and creatinin should be thus subtracted, for they are as truly metabolism products of protein as is urea, still Janney's3 experiments, which show the quantities of glucose produced from various forms of flesh, as determined through feeding experiments with the phlorhizinized dog, are of great interest and may thus be presented:

1 Luthje: "Pfluger's Archiv," 1905, cvi, 160.

2 Janney and Csonka: "Journal of Biological Chemistry," 1915, xxii, 203.

3 Janney and Blatherwick: Ibid.. 1915, xxiii, 77.

According to Janney,1 the percentage quantity of glucose derivable from the following proteins is: casein 48, ovalbumin 54, serum albumin 55, gelatin 65, fibrin 53, edestin 65, gliadin 80, and zein 53 per cent.

Pfluger2 would have it that fat metabolism is the principal source of sugar in diabetes.