BY DR. BURDON-S ANDERSON. 213 



31. Introduction of the Gas to be Analyzed. The 



measuring and laboratory tubes having been brought into con- 

 nection in the manner described above, and both filled with 

 mercury, the gas to be analyzed is introduced into the labo- 

 ratory tube from the test tube to which it has been discharged 

 by the Sprengel. It is then at once transferred to the mea- 

 suring tube by depressing t until the mercury rises in the 

 laboratory tube as far as the stop-cock g f . This done, the 

 stop-cock g is closed, and t raised or depressed till the column 

 stands at one of the marks of the graduation, in reference to 

 which the capacity of the tube has been determined. The 

 temperature is then observed, and the pressure determined by 

 adding the difference between the height of the column in the 

 measuring tube and that in the pressure tube, to the reading 

 of a barometer which stands by. A few drops of solution of 

 caustic potash having been introduced into the laboratory 

 tube, the gas is returned from the measuring tube. Absorption 

 takes place rapidly. It is accelerated by slightly agitating the 

 trough, and by allowing the mercury to stream into the labo- 

 ratory tube after the gas has passed. The measurement of 

 the gas after absorption is performed in the same manner us 

 before. About half a centimetre of strong solution of pyro- 

 gallic acid is then introduced in the same way as the potash, 

 and the gas again returned. After absorption of the oxygen, 

 what remains is nitrogen. In analysis of blood gases, the 

 proportion of nitrogen is nearl} r constant, viz., about 2.5 vol- 

 umes in 100 volumes of blood. If a larger quantity is obtained, 

 the fact indicates that air has entered. Whatever method of 

 anatysis is employed, the results must be reduced to tem- 

 perature and 760 millimetres pressure i. <?., they must be 

 expressed as if the measurements had been made under those 

 conditions. A further deduction must be made from each 

 measurement in respect of the aqueous vapor which the gas 

 contains (the measuring tube being always moist). This is 

 accomplished by the following well-known formula : 



V= Y/ H'-/ 



1 + t 0-00307 760 



V denotes the corrected volume ; V the volume read ; t the 

 temperature; H' the observed pressure ; and f the tension of 

 aqueous vapor at the temperature t. The values of 1 -f t 0.003G7 

 and / are always obtained from tables. For these, and many 

 other important practical details relating to the performance 

 of gas analysis, the reader is referred to Mr. Sutton's " Volu- 

 metrical Analysis," whom I have to thank for two of the 

 woodcuts with which this section is illustrated. To illustrate 

 the application of the method to the analysis of the gases of 

 the blood, I give the following example: 



