BY H. S. MALCRO WARDLAW. 



791 



Periods of holding tlie breath, or, between expirations into a 

 bajr, are "iven in the Tables to the nearest second. 



The analyses of the samples of alveolar air were carried out in 

 a small Haldane-apparatus. About 20 cc. of alveolar air were 

 collected for each sample, about 9 cc. being used for an analysis. 

 Duplicate analyses were performed only in those cases in which 

 there was doubt as to the reliability of a result. The results of 

 duplicate analyses showed divergences ranging up to 2%. The 

 deviation from the mean is only half this figure. The results 

 given in the Tables, therefore, have an error of less than 2%; 

 that is to say, the percentages of carbon dioxide and of oxygen 

 are correct to less than one-tenth of one per cent. The amounts 

 of carbon dioxide and of oxygen in the alveolar air are expressed 

 as percentages by volume of tlie dry gas. The partial tensions 

 of carbon dioxide and of oxygen, in millimetres of mercury, are 

 also given. The tensions were calculated from the percentage 

 composition of the dry gas, the barometric pressure at the time 

 of the experiment, and the tension of aqueous vapour in the 

 lungs [43 mm. Hg, Loewy and Gerhartz(1913), Osborne (1913)]. 

 The tensions have an error of less than 1 mm.Hg. 



Effect of holding the Breath. 



In the following Table are shown the percentages and tensions 

 of carbon dioxide and of oxygen in alveolar air after holding the 

 breath for various pei'iods. In the last portion of the Table, the 

 averages of the values obtained in the individual experiments 

 are given. 



Tablk i. 

 Composition of alreolar air after lioldiiiy breath for rariouK jjeriods. 



