CHEMICAL ANALYSIS OF BLOOD GASES. 131 



10 c.c. of blood. To make certain that all the oxygen has been 

 expelled, the bottle is again shaken. The two readings should agree. 

 Great care must be taken to keep the apparatus away from drafts 

 or other influences that might cause the temperature of the water 

 to change. If this occurs, the temperature in the bath and jacket 

 must be brought back to the original temperature before the final 

 reading is taken. For this purpose a small air thermometer T, in 

 the shape of a U-tube (containing coloured water) connected 

 with a test tube, (weighted with sand) is also placed in the bath. 

 If the meniscus of fluid in the thermometer changes during the 

 observation hot or cold water must be added to the bath and this 

 stirred until the original temperature is regained. 



The carbon dioxide is measured by a repetition of the same 

 technique, using tartaric acid in place of ferricyanide solution. 

 The steps are as follows: the stopcock B is opened, and the meniscus 

 of fluid in the burette brought back to zero by raising the levelling 

 tube, and removing the stopper of the bottle. The reagent tube is 

 withdrawn and washed into the bottle with as small a quantity of 

 CCVfree water as possible (water that has been boiled and cooled). 

 2.5 c.c. of the tartaric acid solution is placed in the tube, the stopper 

 reinserted with stopcock B opened, temperature adjustment made, 

 and the CC>2 displaced and measured by the same manipulations 

 as for oxygen. The bottle must be very thoroughly shaken since 

 the CO2 is difficult to dislodge from the viscid mixture of pre- 

 cipitated blood proteins now present in the bottle. 



When the estimations are completed, the flask, pipette, bottle, 

 reagent tube, etc., are washed thoroughly clean, and any fluid that 

 may have passed into the connecting tube, cleaned out thoroughly 

 with water and a pipe-cleaner. 



