378 THE METABOLISM OF THE CARBOHYDRATES 



of bromine and sodium hydroxide, and that two molecules of 

 glycerose readily condense (aldol condensation) into a molecule of 

 laevulose. 



CH 2 OH CH 9 OH CH 2 OH 



I I 



CHOH CHOH CO 



I , I 



CH 2 OH CHO CH 2 OH 



(Glycerine) //>7 v 



v 9 ' (Glycerose) 



CH OH - CHOH - CHOH \. r 7 , 

 CHOH - CO - CH 2 OH f V*****) 



An index of whether the sugar in severe diabetes is derived 

 from glycerine would be the obtaining of a D : N ratio higher 

 than that which could exist were the sugar derived from proteid 

 alone, at a stage in the diabetes during which the glycogen in the 

 body had all disappeared. Von Mering calculates that 20 grm. of 

 proteid or 100 grm. of flesh (i.e. 3'4 grm. N), might yield 27 grm. 

 sugar, after allowing for the carbon of urea. This would give a 

 D : N quotient of 8:1. Other observers, including Minkowski, 

 consider that the D : N ratio obtained in depancreated dogs, 

 starved or fed on flesh alone (viz. 2*8: 1), represents the highest 

 one possible when proteid alone is the source of the sugar. 1 



Another way of answering the problem would be to see 

 whether the total sugar excreted by a diabetic animal during a 

 week or two could be accounted for by the glycogen previously 

 stored in the body (taking for the purpose of this calculation, 

 Pfltiger's average of 40 grm. of glycogen per kilo body weight), 2 



1 In using this quotient as a criterion of the source of dextrose in diabetes, 

 however, it must be borne in mind that the retention of nitrogen, as well as the 

 increased excretion of dextrose, might cause it to rise. Now several clinical 

 observers have shown that a considerable retention of nitrogen may occur in 

 Diabetes mellitus, so that, for the quotient to be of any value in indicating the 

 source of the sugar, the excretion of nitrogen should be at least as great as the 

 intake, and the observations should extend over several days. In other words, 

 the sugar and the nitrogenous excretory bodies formed in the organism, say 

 when egg albumin is given in the food, may not appear in the urine at the same 

 moment, the excretion of the sugar coming to an end before that of the nitro- 

 genous bodies. To obtain the correct quotient, therefore, the examination must 

 last as long as it is noticed that there is any effect on either the sugar or the 

 nitrogen excretions. 



2 When the observation is made at such a stage in diabetes during which 

 little glycogen could still remain in the body, a lower value than this would 

 have to be taken. 



