ORIGIN OF THE SUGAR. 409 



due to synthesis, where a disaccharide is formed. According to J. DE 

 MEYER * neither the pancreas nor the tissues as a whole contain any 

 glycolytic enzymes. According to him only the blood has a glycolytic 

 action, and this action is supported by a body acting as an amboceptor 

 and produced in the pancreas. Our knowledge as to the existence of the 

 glycolysis and the mode of action of the pancreas in the metabolism of 

 sugar in the animal body is very meager and incomplete. 



Where does the sugar eliminated in diabetes originate? Does it 

 depend entirely upon the carbohydrates of the food or the store of car- 

 bohydrates in the body, or has the body the power of producing sugar 

 from other material? To LUTHJE belongs the credit for positively 

 deciding this question. He has made experiments on dogs with pan- 

 creas diabetes, in which on a protein diet free from carbohydrates so much 

 sugar was eliminated that it could not possibly be accounted for by 

 the store of glycogen or other carbohydrate-containing substances in the 

 body. Similar experiments were also performed later by PFLUGER, 2 

 with the results that the power of the animal body to produce sugar from 

 non-carbohydrate material is now definitely proved. 



Is this sugar produced from protein or fat, or from both? This ques- 

 tion so far has not been answered, and it is the subject of continuous 

 dispute. It is not possible to enter into an exhaustive and detailed 

 discussion of the question in a text-book, and we will only mention, 

 briefly, certain of the most important observations and historical points. 



The largest amount of sugar which we can obtain theoretically from 

 protein is 8 grams of sugar from 1 gram of protein nitrogen, if we admit 

 that all the carbon of the protein, with the exception of that necessary 

 to form ammonium carbonate, is used for the formation of sugar. These 

 results are still somewhat too high for the average carbon and nitrogen 

 content of the proteins and the values D:N = 6.6 is probably more correct. 3 

 The actual relation between glucose and nitrogen in the urine, i.e., 

 the quotient D: N, has been repeatedly determined in various forms of 

 diabetes, and in depancreatized dogs it is generally 2.8 and in starving 

 dogs or dogs fed with protein and poisoned with phlorhizin it is equal to 

 3.65 (LUSK). It may undergo considerable variation, and in certain 

 cases it may indeed be lower than 1 as well as higher than 8, and high 

 results have been repeatedly obtained in cases of human diabetes. From 

 these quotients conclusions have been drawn as to the amount of sugar 



1 Cited from Centralbl. f. Physiol., 20 and 23. See also Lupine, Etat actuel de la 

 question de la Glycolyse, La semaine medicale, 1911. 



2 Lttthje, Deutsch. Arch. f. klin. Med., 79, and Pfliiger's Arch., 106; Pfliiger, Pflii- 

 ger's Arch., 108. 



3 See Falta, Zeitschr. f. klin. Med., 65; see also Gigon, Deutsch. Arch. f. klin. Med., 97. 



