INGESTION OF PROTEIN DIETS. 303 



there was an increment of 10 calories in the heat production, but the 

 values for the carbon-dioxide production and the oxygen consumption 

 both present slight negative values. An examination of the details of 

 the experiment shows that positive increments occur in the first two 

 periods, these amounts being counterbalanced by negative values in 

 subsequent periods. In the first two experiments given in table 230 

 the food taken had practically the same nitrogen content and the 

 experiments may therefore be considered as duplicates, although with 

 different subjects; both indicate a considerable increase in all of the 

 three factors. In the next two experiments, which were made with 

 the same subjects, the food taken was approximately one-half that 

 ingested in the preceding experiments; the increments found were 

 proportionally smaller than those in the first two experiments. In 

 both pairs of experiments, the values for A. W. W. are lower than 

 those for A. H. M., especially for carbon-dioxide production and heat 

 production. From these four experiments, therefore, one may infer 

 that the influence of the ingestion of beefsteak is by no means the 

 same with different individuals. The values for oxygen consumption 

 for the two subjects are considerably at variance, as in the high-nitrogen 

 experiments the increments are alike, while with the low-nitrogen 

 intake the increment for A. H. M. was but half that with A. W. W., 

 thus showing the difficulties in comparing results by direct and indirect 

 calorimetry for experimental periods less than 24 hours. 



In comparing the data for exiperiments such as these, we should 

 expect to find that each gram of ncrement in carbon dioxide produced 

 would correspond to an increment in heat production of 1\ to 3 calories. 

 An examination of the results in table 230 shows that this ratio holds 

 true in but few instances. Thus, in the first four experiments the 

 amount is more nearly 2 calories per gram of carbon dioxide than 3 

 calories; this is true, also, for many other experiments. In the experi- 

 ment with V. G., January 21, 1911, we find that the ratio is 7 calories 

 per gram of carbon dioxide, while in the following experiment it is 

 only 1.2 calories, and in the two succeeding experiments the ratios 

 are less than 1 calorie. Such irregularities as these discredit the use of 

 direct calorimetry in short experiments. On the other hand, when the 

 computations of indirect calorimetry are based upon carbon dioxide 

 alone, they are open to the serious objection that the increase found 

 may be due to change in the character of the katabolism or to a forma- 

 tion of fat from carbohydrate, but when the measurements are made 

 by direct calorimetry it provides positive evidence that the increment 

 in heat production is due to the food alone. The data for heat produc- 

 tion in table 230 show that such an increment was found in every experi- 

 ment in which protein food was ingested, although in some cases the 

 increment was very small. 



