Blood Oxygen Analyses. 



213 



the amount of oxygen physically held in the blood if the analysis 

 is intended to show the amount of oxygen loosely held in the 

 oxyhemoglobin. 



In the Henderson-Smith method a small correction should be 

 applied for water vapor pressure. This is however of theoretical 

 interest for it is only 0.1 to 0.3 volumes per cent, in amount. 

 Much more important is the fact that the oxygen content of the 

 air in the diffusion tube has been considered without regard to 

 the amount physically held in the 3 c.c. of blood-ferricyanide 

 mixture. The volume of air in the diffusion tube is constant at 

 about 10 c.c. When 0.5 c.c. potassium ferricyanide is injected 

 through the rubber stopper, this volume is reduced, and the pres- 

 sure within the tube correspondingly increased by about 5 per 

 cent. As a result of the chemical reaction oxygen is liberated, 

 and the pressure further increased by one per cent. Consequently, 

 the 3 c.c. of fluid within the tube will absorb oxygen in proportion 

 to its increased partial pressure. This physically held oxygen 

 should be added to that determined analytically to give the total 

 contained in the sample of blood. 



We have calculated the amount of oxygen that the blood- 

 ferricyanide mixture would hold on the basis of the gas laws 

 (assuming that the mixture has 90 per cent, the absorption capacity 

 of water), and it will be seen from Table II that the level of the 

 Henderson-Smith figures has been raised about 1.5 volumes per 

 cent. 



TABLE II. 



Sample. 



Van Slyk 



e Method. 



Henderson-Smith Method. 



Difference Be- 

 tween VS and 

 H-S Corrected 

 Values. 



Volumes 

 Old Value. 



Per Cent. 0 2 

 Corrected Value. 



Volumes 

 Old Value, 



Per Cent. 0 2 

 Corrected Value. 



18a. 



12.9 



IO.7 



IO.4 



11. 7 



— 1.0 



18&. 



18.5 



16.7 



12.8 



14.1 



2.6 



18c. 



20.1 



18.0 



I4.6 



16.0 



2.0 



i8d. 



20.0 



18.4 



15-5 



16.9 



I-S 



From Table II it is evident that our corrections have brought 

 the Van Slyke and Henderson-Smith values for blood oxygen 

 much closer together than they were originally. The figures are 

 however not identical, and we are now engaged in testing out 

 other factors that may be involved. 



