SECT. 8] 



CARBOHYDRATE METABOLISM 



1017 



purposes than combustion though there is, of course, the possibility 

 that the estimations of fat are wrong." If now we suppose that the 

 estimations were correct, and place beside the carbohydrate gained 

 the fat missing each day, as is done in Fig. 276, we find that a 

 correlation exists. Little account need be taken of the fact that the 

 maximum of the carbohydrate gained is reached before the maximum 

 of the fat lost, for the inferiority of methods, the comparative fewness 

 of estimations, and the varying conditions of individual workers must 

 all be borne in mind. Nor can stress be laid on the absolute magnitude 

 of each quota, for an exact balance would show more carbohydrate 

 gained than fat lost, since there 

 is more carbon in a gram of fat 

 than in a gram of carbohydrate. 

 But it is striking that nowhere 

 else in the development of the 

 hen's egg can two such dis- 

 crepancies be found, and that 

 the two are approximately of 

 the same order. 



If then we accept the view 

 that the increase in carbo- 



O Fab missing 



I.e,not found in chemical 

 analyses and not ac- 

 counted for by CO2 

 produced 



® Carbohydrate 

 formed 



Days 



Fig. 276. 



hydrate at this point is due to a transference from fat, there is one 

 interesting corollary. It is that a transfer of a less oxygenated to a more 

 oxygenated substance is taking place, and therefore that some oxygen 

 would tend to be retained in the egg, and therefore that the respira- 

 tory quotient would tend to be lower than it theoretically should. 

 If Fig. 1 44 be examined, it will be seen that there is a certain lowness 

 of the experimentally determined respiratory quotient points as com- 

 pared with those calculated from chemical analyses of food-stuff 

 burnt. It is true that the passage from fat to carbohydrate, if it is 

 of the order we are supposing, would not be large enough to make 

 much difference, but it might well be one of several factors. 



As is well known, the passage from fat to sugar is usually believed 

 to be impossible in vivo, and it is certainly not supported by any un- 

 assailable piece of evidence drawn from experiments on the adult 

 animal. In the starved dog the feeding of fat will not produce Hver- 

 glycogen, nor in the diabetic dog a rise in the G/N ratio, nor are 

 any of the balance-sheet experiments, in which glucose seems to 

 appear unaccountably and fat has to be postulated as its precursor, 



65-2 



