OXYGEN RESPIRATION 39 



at the microscope, and that the volume be recorded only if the 

 two readings coincide, since accidental adherence of the mercury 

 to the walls of the tube will lead to fluctuations in the volume 

 of the gas. This is an indication of the insufhcient purity or 

 dryness of^the tube or the mercury. 



The gas is now passed from the measuring tube into the 

 analysis tube filled with mercury, by raising the pear-shaped 

 bulb H and opening the stopcock a, which tube has been placed 

 so far unde^ mercury on the end of the measuring tube that the 

 end of the measuring tube touches the bend of the analysis 

 tube. The latter, with the gas contained within it, is then con- 

 veyed under mercury to the end of the absorption pipette D 

 and the gas is drawn into the pipette by lowering the bulb G 

 and opening the stopcock n. The gas is left about lo minutes 

 in contact with the potash. During this time the bulb 

 G is left lowered and the stopcock n opened so wide that mer- 

 cury drops continuously out of the bath into the pipette. After 

 the lapse of lo minutes the gas is forced back out of the pipette 

 into the analysis tube as follows: without changing the position 

 of the stopcock, the bulb G is raised and the greater part of 

 the gas is thereby quickly crowded out of the pipette into the 

 analysis tube. Then the stopcock is closed and the rest of the 

 gas is removed with the aid of the screw/, after which the level 

 of mercury in the bath C has been so lowered that the orifice 

 of the pipette with the analysis tube resting on it is visible. 

 In forcing gas out of the absorption pipette, care must be taken 

 that no trace of potash enters the analysis tube. If it does the 

 gas is replaced in the pipette and the analysis tube with the 

 impurities is replaced by another tube filled with pure, dry 

 mercury. If this precaution is not observed the lye might 

 easily enter the measuring tube and produce serious errors in 

 subsequent analyses. In order to prevent a gushing of the 

 KOH from the end of the pipette into the analysis tube, the 

 analysis tube must not be Hfted before the last trace of gas is 

 forced from the pipette. In addition, the level of mercury in 

 the bath must be so adjusted that the capillary tube of the 

 absorption pipette juts out only about i. 2 cm. after the displace- 

 ment of the gas. 



