BED CALORIMETER AND TENSION-EQUALIZER UNIT. 107 



to obtain an adequate idea of minute differences in the volumes of 

 respiration by means of the pneumograph and an error of 40 c.c. would 

 be equivalent to an error of 2 to 3 c.c. per minute in the measurement of 

 the oxygen consumption. 



If only one 15-minute experiment were considered, no definite con- 

 clusions could be drawn. If, however, three 15-minute experiments are 

 made consecutively and the results obtained agree closely, it may be 

 concluded that the error due to the variations in volume of air in the 

 lungs is extremely small or else it is a constant one, but from all of our 

 experience with respiration experiments, changes in the volume of air in 

 the lungs have rarely to be considered. 



If the particular form of respiration apparatus used caused any 

 change in the respiration, this would also lead to differences in the 

 determination of the respiratory exchange. For example, if the con- 

 ditions obtaining with this apparatus tended to make the subject 

 breathe deeper and more rapidly than under ordinary conditions, it 

 may be seen that the results would be abnormal because of the abnor- 

 mal character of the breathing. When the subject is in the bed calori- 

 meter, there is no mechanical reason for his breathing abnormally. 

 The carbon-dioxide content of the air in the chamber may, however, 

 be higher than normal and, indeed, may be as high as 1 per cent. 

 While this would lead to an increase in the volume of respiration, the 

 increase would not be so great as to alter the metabolism. With the 

 respiration apparatus, air free from carbon dioxide is supplied to the 

 subject, but there is a possibility that the mechanics of the apparatus 

 may produce abnormal ventilation of the lungs. During this series 

 of comparison experiments a test was made of this point in several of 

 the bed-calorimeter experiments 1 by having subjects breathe through 

 the nosepieces and three-way valve of the respiration apparatus, which 

 had been detached for the purpose and placed inside the bed calorimeter. 

 The results indicated that the use of the nosepieces and valve had 

 practically no influence upon the respiratory exchange. The only 

 effect noted was due to the fact that the subject was obliged to lie upon 

 his back during the whole period without changing his position; he was 

 therefore more weary at the end of the experiment than when he had 

 more freedom of movement. 



DIFFERENCES IN THE INDIVIDUAL COMPARISONS. 



Considering all of these possible sources of error in conducting experi- 

 ments with the two apparatus, it may be said that those with the bed 

 calorimeter are mainly of physical or technical origin and with the 

 respiration apparatus are principally of physiological origin. With 

 these possibilities in mind, the differences in the individual compari- 

 sons may be considered. These differences are shown in table 13, in 



l See p. 90. 



