METHODS OF MEASURING RESPIRATORY EXCHANGE 45 



and o have been stored in two metal boxes one of which is inserted into 

 the entrance tube of the gas meter at (P) and the other into the exit 

 tube. The air in these boxes communicates with the burette (E). The 

 enclosed volume of air will be affected by the temperature of the air en- 

 tering and leaving the meter and by the atmospheric pressure, and the 

 volume changes can be read off on the burette when the water in (G) and 

 (E) has been brought to the same level by moving (G). The burette is 

 so divided that, if a volume of say 107-4 ' ls rea< ^ off during an experi- 

 ment, the volume of the air which has passed through the meter can be 

 reduced to normal conditions (oand 760 mm. dry pressure) by multi- 

 plication with , This arrangement is certainly not more accurate 



107-4 



and scarcely more convenient than to reduce by means of a table after 

 reading the barometer and a thermometer placed in the exit tube of 

 the gas meter. 



In experiments involving exercise in the open a dry gas meter 

 was carried on the back of the person experimented on. 



In Bohr's laboratory the Zuntz method has been considerably 

 simplified. In the expiration pipe leading to the meter a mixing 

 vessel (usually placed in a water-bath at room temperature) was pro- 

 vided and samples were drawn from this, either by hand a few c.c. at 

 a time, or simply by letting the mercury run out slowly from a samp- 

 ling vessel during the experiment. These two improvements are 

 interdependent. The expired air moving along the pipe from the 

 valves to the meter is of very unequal composition, the first part being 

 almost pure atmospheric air from the dead space (mouth, trachea and 

 bronchi) and the last alveolar air. It is very incompletely mixed in a 

 tube, and when samples are taken from the tube it is absolutely neces- 

 sary that they are drawn only while the current is passing. Other- 

 wise too much alveolar air is sure to get into the sample, since it will 

 in most cases be chiefly alveolar air which is left in the tube after the 

 expiration. In a vessel which will hold at least the volume of two 

 expirations the different portions of expired air are mixed and an 

 average sample can easily be obtained. 



Higley and Bowen [1904] have constructed an apparatus by which 

 the carbon dioxide from the expired air is absorbed in a Haldane set 

 of vessels suspended from a balance which will record graphically the 

 rate at which CO 2 is being eliminated. Their method is certainly 

 useful in special conditions involving rapid changes in the activity 

 of the organism, 



