514 Influence of Inanition on Metabolism. 



errors involved in these assumptions on the total metabolism can hardly be 

 estimated with our present knowledge of physiological chemistry. Since for 

 the average of all the experiments, the discrepancy between the computed and 

 measured heat production is almost insignificant, it would appear that it is 

 reasonable to assume that in the few instances where noticeable discrepancies 

 appear there are specific errors other than those involved in the assumptions 

 above outlined, which have inevitably crept into the chemical and physical 

 measurements. Fortunately, the simultaneous determination of the oxygen 

 absorbed and carbon dioxide and heat production furnishes an unusual check 

 on the accuracy of the method in general and the computations in particular. 

 From the ratios between the oxygen and carbon dioxide and the heat, clues 

 regarding the causes of the few marked discrepancies observed in the com- 

 parisons may be obtained. It is necessary, therefore, to compare the data given 

 in table 244 with those in table 243. The most noticeable discrepancies 

 between the computed energy of body material and the total heat production 

 are to be observed on the last day of experiment No. 73, the third day of 

 experiment No. 77, and the last 2 days of experiment No. 74. Examining the 

 carbon dioxide and oxygen thermal quotients for these days, certain valuable 

 conclusions may be drawn. Thus, on the fifth day of experiment No. 73, the 

 oxygen thermal quotient is abnormally high while the carbon dioxide thermal 

 quotient is approximately normal for this particular experiment. Hence, on 

 this day, at least, the figures point strongly to the conclusion that there was 

 an error in the oxygen determination rather than in the heat determination, 

 and consequently the discrepancy of +4.3 p er cen + implies that the heat 

 production was accurately determined while the estimated energy of katabolized 

 body material was somewhat too high. Again, on the third day of experiment 

 No. 77, the oxygen thermal quotient is abnormally low while the corresponding 

 carbon dioxide thermal quotient is not abnormal. It is evident, then, that in 

 this case the error is again with the determination of oxygen, so that the total 

 heat production is probably not far from correct. Another factor enters into 

 food experiment No. 74. The oxygen-thermal quotient is high on both days 

 but the carbon dioxide thermal quotient is likewise abnormally high, hence it 

 would appear that on these 2 days the error is with the measurement of heat 

 rather than the computed products of katabolism. It is of course not impos- 

 sible that minor compensating errors may enter into the comparison of these 

 three factors, but certainly the simultaneous determination of the carbon diox- 

 ide elimination, oxygen consumption, and heat production is of the greatest 

 value in deducing the correct balance between the total heat production and 

 the estimated energy of katabolized body material. It is, furthermore, signifi- 

 cant that none of the ratios seem to point toward marked discrepancies in the 

 carbon dioxide determination, bearing out the results of check tests which show 

 tli at the carbon dioxide determinations with this apparatus are extremely 



