326 THE METABOLISM OF THE CARBOHYDRATES 



causes a distinct increase in the sugar excretion, then we can state 

 that proteid is capable of producing carbohydrate in the organism. 

 This indirect method has within recent years been very thoroughly 

 investigated, and we shall discuss it in considerable detail later. 

 Meanwhile we will proceed with a consideration of the results 

 obtained by the direct method. 



It is evident that this direct method of ascertaining whether 

 or not a substance is convertible into glycogen in the organism 

 can be only an approximate one. We can ascribe a glycogen- 

 producing influence to a food-stuff only when in a large number of 

 experiments, an amount of glycogen is found to be deposited 

 which is constantly higher than that found during starvation ; and 

 we must further show that, meanwhile, insufficient tissue proteid 

 has been decomposed to account for the glycogen. This latter we 

 can do by estimating how much glycogen could have been derived 

 during the observation from proteid, and seeing whether this 

 amount is sufficient to account for the excess of glycogen deposited. 

 The amount of glycogen liberated from proteid can be calculated 

 from the nitrogen excretion, for each grm. of nitrogen excreted 

 in the urine will equal 2-500 grm. of carbon liberated in the 

 tissues, which is equivalent to 5-700 grm. of glycogen (i.e. 3*295 

 grm. carbon liberated in the tissues minus 0'795 grm. excreted 

 as urea, &c.). A method including this factor was adopted by 

 Ott ( 6 ) in a series of experiments on the glycogen-forming power 

 of the various carbohydrates. 



The food-stuffs which we would naturally expect to be marked 

 glycogen formers are the carbohydrates. Let us, therefore, con- 

 sider these first of all in this connection. 



For dextrose, the proof of its conversion into glycogen in the 

 body is, as might be expected, an easy matter. Dr. F. W. Pavy ( 10 ) 

 found that, whereas in eleven healthy dogs fed only on flesh the 

 average relation of the weight of the liver to that of the whole of 

 the body was 1 to 30, in five dogs fed on vegetable food it was 

 1 to 15. On animal food also, the average percentage of crude 

 glycogen in the liver was 7*19, whereas on vegetable food it was 

 17' 23. With such results, little doubt can exist that glycogen 

 is derived directly from dextrose. It is, however, always 

 possible, as indicated above, that its source might be the tissue 

 proteids, and that the large accumulation of glycogen in the liver 

 could be accounted for by the proteid not being required by the 



