696 METABOLISM 



sufficient heat to maintain the body temperature. This method can be 

 rendered still more certain in effecting a removal of all carbohydrate 

 from the body by giving the animal phlorhizin every eight hours. Phlor- 

 hizin, as we shall see, renders the animal diabetic. 



After removing the glycogen, further deposition in the liver can be 

 readily shown to occur when any of the ordinary sugars or starches are 

 given as food. It does not occur, however, when chemical substances 

 closely related to ordinary sugar, such as the wood sugars (pentoses) 

 or the alcohols and acids corresponding to dextrose, are contained in the 

 diet. Nor does it occur with cellulose or with inulin, a polysaccharide 

 built up from pentose sugar. When proteins are fed the results are not 

 so definite, although many observers have claimed that glycogen is 

 formed. With fat, on the other hand, no glycogen formation can be 

 shown to occur, although we know that a trace of carbohydrate must be 

 formed out of the glycerine of the fat molecule. 



The results of. the direct method, even when the conditions are per- 

 fectly controlled, are very unreliable, especially when they are of a nega- 

 tive character, because any new sugar that may be produced by the in- 

 gested substance instead of being stored as glycogen is likely to be used 

 by the tissues as it is formed. Where only a slight degree of gluconeo- 

 genesis, as the process of sugar formation is called, is occurring, it is not 

 probable that any of the glucose will be retained in the body as glycogen. 



The methods employed for producing experimental diabetes in investi- 

 gation of these problems by the indirect method are (1) the entire removal 

 of the pancreas, and (2) the continuous administration of the drug 

 phlorhizin. The animal rendered diabetic by either of these methods is 

 first of all observed for several days to determine the normal daily ex- 

 cretion of sugar. At the same time the nitrogen excretion for the day 

 is determined, the ratio between the total nitrogen and the glucose 

 known as G to N ratio being about 3'.65 to 1 when complete diabetes 

 has become established. The foodstuff in question is then fed to the 

 animal, and the amount of extra glucose excreted thereby is taken to 

 represent that which has been derived from the ingested food. By this 

 method it has been possible to show that, not only the above mentioned 

 carbohydrates, but protein as well produce a very considerable quan- 

 tity of glucose in the animal body. Fats, however, yield only negative 

 results. 



The indirect method has another great advantage over the direct in 

 that the results are much more quantitative in character; for example, 

 Lusk and his pupils have been able to determine the amount of glucose 

 which can be produced by feeding certain of the building stones of the 

 protein molecule. The great practical importance of such results in 



