26 METABOLISM DURING WALKING. 



there collected; under these circumstances the change in weight of 

 these absorbers would not represent the amount of carbon dioxide 

 expired by the subject, but would give higher values. If, on the other 

 hand, the air-drier in the soda-lime train (i. e., the Williams bottle 

 following the two carbon-dioxide absorbers) were inefficient, so that 

 moisture carried over from the moist soda-lime was not wholly 

 absorbed by the sulphuric acid in the Williams bottle, the total increase 

 in weight would be less than the actual amount of carbon dioxide 

 expired by the subject. 



Efficiency of the water-absorbers. During rest experiments the usual 

 rate at which the absorbing system is run is not far from 30 to 35 liters 

 per minute. For testing the efficiency of the water-absorbers for the 

 special demands of this research, the absorbing system was run at 65 

 to 100 liters per minute without a subject. The gain in weight of the 

 water-absorbers and the loss in weight of the moistener were then com- 

 pared. If the water-absorbers were efficient, their gain in weight would 

 be equal to the loss in weight of the moistener. One such test showed 

 a difference after 1 hour of 0.19 gram. If, in 12 minutes (the usual 

 length of a period), such an amount of unabsorbed water were carried 

 to the soda-lime containers and there absorbed, the error in the deter- 

 mination of the carbon dioxide expired would amount to 1.5 c. c. per 

 minute. In another series of experiments with five periods, the first 

 Williams bottle gained 116 grams, while the second gained only 2 

 grams. In still another test an additional Williams bottle was used. 

 The first gained in weight 165.55 grams, the second 4.70 grams, while 

 the third lost 0.1 gram in a period of 3 hours. As it was the practice 

 to remove the first Williams bottle each day and advance the second 

 Williams bottle to first place, while a freshly filled bottle was used for 

 the second water-absorber, it may be safely assumed that none of the 

 moisture in the ventilating air-current reached the carbon-dioxide 

 absorbers. In the usual rest experiments, in which the ventilation is 

 about 30 liters per minute, the "error" would be negligible in all cases. 



Efficiency of the air-drier. The second point, viz, that the carbon- 

 dioxide absorbers lost no moisture that was not recovered by the air- 

 drier, was tested by passing the ventilating air-current through the 

 soda-lime containers, which were weighed separately. The moisture 

 taken up by the air in passing through the moist soda-lime should then 

 be entirely removed in the following Williams bottle or air-drier. The 

 largest gain found during any test was equivalent to an error of 3 c. c. 

 per minute in the carbon-dioxide determination, while the average 

 error in a series of tests was but a fraction of 1 c. c. per minute. It was 

 considered, therefore, that the absorbers as used in this research were 

 efficient, even at the high rate of ventilation of 100 liters per minute. 

 During the standing experiments the rate of ventilation was 65 liters 

 per minute, which allowed still greater efficiency in the absorbers. 



