110 EXPERIMENTAL PHYSIOLOGY 



at H and then allowing it slowly to fill the burette. The burette 

 is left filled with the weak acid. 



Analysis for CO 2 and O 2 . For this purpose, the absorp- 

 tion part of the apparatus is more complicated (see Fig. 32) 

 and is attached at G. 



The tube from the burette leads to a three-way stopcock J which accord- 

 ing to its position connects either with D or M; with the former for the ab- 

 sorption of CO2 as described above and with the latter for the absorption 

 of O2. This part of the apparatus consists of a spiral tube L ending below 

 in a bulb M which is connected with a reservoir N. M contains pyrogallic 

 acid dissolved in 60 per cent. KOH solution (10 grams per 100 c.c.) and N 

 contains mercury. These are introduced through the side tube O and the 

 levels are adjusted so that the jjieniscus of the pyro solution stands some- 

 where on the narrow tube e above the spiral. For convenience of filling and 

 cleaning there is a rubber joint at K surrounded by mercury so as to prevent 

 leaks. This is shown in detail in the small side sketch of Fig. 32. The spiral 

 is used to increase the surface of contact between gas and solution since 

 absorption of C>2 is much slower than that of CC>2. 



The analytical procedure is in principle the same for O 2 as 

 for CO 2 but it is necessary that the tubing and connections be 

 filled with nitrogen, instead of air, before starting the analysis. 

 This is done by passing some air back and forth between the 

 burette and the spiral, until, on bringing the pyrogallate 

 meniscus to the original mark on e, there is no further change 

 in the level of the mercury in A. The tap J is then turned so 

 that g and d communicate, the nitrogen passed into D and 

 then returned to A so as to displace the air entrapped in d. 

 The tap J is again turned so that g and e communicate and the 

 last traces of O 2 removed from the gas in the burette. 



This preliminary filling of the apparatus with N is necessary 

 only when it is used for the first time after assembling; other- 

 wise N remains over in the tubing from the previous analysis. 



In the actual analysis for CO 2 and O 2 , the percentage of 

 CO 2 is first of all determined by the procedure already described. 

 The tap J is then turned so that g and e communicate and the O 2 

 removed, great care being taken in moving the air between 

 A and L that the pyrogallate solution does not rise above the 

 mark on e. The small bulb on e serves to prevent this by 

 breaking up any bubbles that form. It will be noticed that 

 as the O 2 becomes removed the pyro solution adhering to the 



