546 EFFECT OF SUSPENDED EESFIIiATION OX COMPOSITION OF ALVEOLAR AIR, 



seemed to be approached in this case was 8.35 % or 60.5 mm. Hg. At the end 

 of the experiments, in about 40 seconds, the average alveolar content of carbon 

 dioxide had risen to 7.45 %, or 53.9 mm. Hg and was still rising. 



The alveolar oxygen content in each of the above two series of experiments 

 fell in a regular manner which showed only one phase, and could be expressed 

 by an exponential formula. In each case the oxygen content was still falling 

 rapidly at the end of an experiment. 



When the breath was held under atmospheric pressure the final alveolar 

 oxygen content which seemed to be approached was 7.62 %. or 55.0 mm. Hg. 

 The value actually reached at the end of the experiment was 11.21 % or 81.0 

 mm. Hg. 



When the breath was held under negative pressure or when the same air was 

 rebreathed, the alveolar oxygen content appeared to approach towards a final 

 value of zero. The average value actually reached at the end of the experiments 

 was 9.2 % or 66.5 mm. Hg. 



There is little doubt that the final values approached under conditions of 

 normal pressure represent the composition of alveolar air in eciuilibrium with 

 mixed venous blood. The figures obtained lie close to those found for the ten- 

 sions of the gases in venous blood by other workers using different methods, as 

 was stated before. The second rise in the alveolar carbon dioxide content is pro- 

 bably due to the return of a large body of blood to the lungs after leaving them 

 during an experiment in an incompletely arterialized condition. 



The question raised by these experiments, however, is this: What is the 

 significance of the final values which seemed to be approached under conditions 

 of negative pressure of rebreathing? Does the simple occurrence of negative 

 pressures in the lungs, or of respii'atory movements, alter the tensions of the 

 gases in the venous blood? 



In the work referred to, the final values of the alveolar carbon dioxide con- 

 tent were always 'approached from below, that is, the carbon dioxide content 

 always rose during an experiment. Tbe periods for which the experiments could 

 be continued were limited to 30-40 seconds, as the breath could not be held 

 longer without interfering with the success of the experiment. 



The experiments, therefore, had to be terminated before a definite con- 

 clusion could be reached as to whether the composition of the alveolar air actually 

 would reach the constant values which the form of the variation indicated. 



Methods. 



In the experiments described in the present paper, confirmation of the pre- 

 vious work has been sought by approaching the fhial carbon dioxide tensions 

 from above' as well as from below, and by adopting a method which enables the 

 experiment to be continued for much longer periods without inconvenience. 



The final alveolar carbon dioxide values have been approached from above 

 by breathing air containing a percentage of carbon dioxide gTeater than the 

 final value exijected. 



The periods of the experiments have been extended by using a method em- 

 ployed by Yandell Henderson and Prince (1917). Instead of holding the 

 breath for jDeriods increasing in length until no further increase is possible, the 

 subject holds .his breath for a series of short periods of only 5-10 seconds. 



He begins an experiment by expiring as deeply as possible into an empty 

 rubber-lined bag, and immediately closing the tvibe connecting . the mouthpiece 



