350 THE METABOLISM OF THE CARBOHYDRATES 



only substance, in fact, which does induce glycogen formation in 

 depancreated dogs is Icevulose. This interesting point we will 

 discuss later. The sugar must, therefore, first of all be derived 

 from glycogen. 



As we see, however, this soon disappears from the organism 

 after removal of the pancreas, and cannot be the only source 

 of the sugar. In a recently published experiment (by Ltithje 20 ), 

 a dog of 5'8 kg., after removal of the pancreas, excreted during 

 twenty-five days a total of 1176 grm. dextrose; during all this 

 time a carbohydrate-free diet was given. Assuming that on an 

 average a dog's body contains 40 grm. glycogen per kilo body 

 weight an average determined by Pfluger then 232 grm. glycogen 

 (5- 8 x 40) might be held accountable for the sugar; but this could 

 yield only 257 grm. dextrose, leaving 919 grm. to be otherwise 

 accounted for. 



Where does it come from, then, when all the available 

 glycogen has been used up ? The possible sources are pro- 

 teid and fat. The fact that the D : N ratio is the same during 

 starvation and proteid feeding would seem to point to the 

 derivation of both sugar and nitrogen from a common source, 

 i.e. from proteid. We have already seen how, chemically con- 

 sidered, such a derivation is possible. The extreme emaciation 

 of the dogs which supervenes, and the presence of /S-oxy- 

 butyric acid, &c., in the urine, is usually considered to support 

 this view. 



To explain the constant ratio of D:N, if proteid be the source 

 of the sugar, we must assume further that all the sugar liberated 

 in the tissues from proteid or a constant fraction of it reappears 

 as such in the urine. Supposing that all the carbon of proteid 

 were converted into sugar, then 100 grm. proteid could yield 

 113 grm. grape-sugar, and a D:N ratio of about 7 would be 

 obtained. We have seen, however, that it is only 3. Does this 

 mean that all the carbon of proteid is not converted into sugar, 

 or that all is thus converted, but that some of it is afterwards 

 destroyed ? To answer this question, let us see what proportion 

 of dextrose when fed to depancreated dogs reappears in the 

 urine : for if all, or nearly all, the administered sugar reappears, 

 then we may assume that the same will hold true for the 

 dextrose split off from proteid, and that the sugar which is 

 excreted in the urine on a carbohydrate-free diet represents all 



