196 SECTIONAL ADDREESES 



percentage of the measured heat was then plotted against the respiratory 

 quotient. The principle adopted is similar to that which was used in 

 figs. I and 2, where, however, the heat was calculated from the oxygen, 

 using the Zuntz-Schumburg relation. 



Here there is a tendency for the percentage differences to be positive 

 when the respiratory quotient exceeds 0-875, ^^^ the points do not He 

 so evenly distributed about the zero line as at intermediate quotients, 

 while below 0-775 they are on the average negative. Hence there is an 

 error in calculating the heat from the carbon dioxide output both at high 

 and low respiratory quotients, an error predicted from theory. This is 

 in contrast to the error, opposite in direction, which appeared when, in 

 figs. I and 2, the Zuntz-Schumburg figures were used, and for which there 

 was no explanation on the theory of the variable combustion ratio. 



Is it possible to prove quantitatively that conversion of carbohydrate 

 to fat and vice versa does take place at high and low quotients ? For this 

 purpose all the observations calculated for hourly periods of Du Bois and 

 his colleagues, apart from diabetes, have been divided into groups accord- 

 ing to the quotient, including febrile cases and experiments after food, 

 which fall into line with the figures for the basal metabolisms, though 

 the figures in febrile cases are not so regular as in non-febrile cases ; in 

 each group the CO 2 and the heat have been correlated and regression 

 equations obtained. 



Equations 6 and 7 are very similar to one another, and it would be 

 reasonable as an approximation to take the average — • 



Cals. = 3 -049CO2 — 2-75 . . . (9) 



and to call this the regression equation at the intermediate quotient o - 825 . 

 If we assume, as is compatible with fig. 4, that at this quotient a mixture 

 of carbohydrate and fat is burnt and there is no conversion, it will be 

 possible to calculate for the conversion process at high and low quotients 

 the calories absorbed or liberated per litre of oxygen liberated or absorbed 

 and to see how this agrees with the theoretical value from equation 2, viz. 

 3-93 Cals. per litre. The method of calculation is as follows: For 

 25 grms. CO2 (12 -7 litres) at R.Q. = 0-825, the O2 is 15 -4 litres, and the 

 heat from equation 9 is 73 -48 Cals. For 25 grms. COj at R.Q. = i, the 

 Og is 12-7 litres, and the heat from equation 3 is 62-88 Cals. Hence at 



