518 OXYGEN CONTENT OF BLOOD IN PNEUMONIC INFLUENZA. 



The oxygen capacity and oxygen saturation of the samples of 

 blood were determined by Haldane's (1897) ferrieyanide method 

 in Barcroft's (1908) differential apparatus. The principle of 

 the method is as follows : Blood is laked with dilute ammonium 

 hydroxide. Its oxygen is then liberated by the addition of 

 potassium ferrieyanide. These operations are performed in 

 closed vessels connected with a clove oil manometer. The dif- 

 ference of pressure observed in the manometer multiplied by the 

 constant of the instrument gives the volume of oxygen liberated. 

 The volume is then corrected for temperature, barometric pres- 

 sure, and moisture, the final volume given being that of the dry 

 gas at normal temperature and pressure. In the case of un- 

 saturated blood, oxygen is absorbed during the process of 

 laking, and the negative pressure produced in the manometer is 

 similarly a measure of the volume of oxygen absorbed. 



For samples 1 — 6 Barcroft's small apparatus, which takes 

 . 1 cc. of blood was used ; for samples 7 — 11 the large apparatus, 

 which requires 1.0 cc. of blood was used. The small apparatus 

 was calibrated by liberating in it the oxygen from a known 

 volume of standard hydrogen peroxide (Barcxoft and Burn, 

 1913) . The large apparatus was calibrated by measuring the 

 diameters of the limbs of the manometer and the volumes of 

 the vessels in which the gas was liberated (Barcroft and Hig- 

 gins, 1911), the constants so found being increased by 2 % 

 to allow for the difference existing between this and the former 

 method of calibration. 



The sample of blood for the determination of oxygen capacity 

 was saturated with oxygen by spreading 2 cc. in a thin layer 

 over the inner surface of a 200 cc. flask. The flask was corked 

 to prevent evaporation of the blood, and was kept rotating so 

 as to expose the greatest surface to the air for about 15 minutes. 

 Such blood took up no more oxygen on being laked in the 

 differential apparatus. The samples for the determination of 

 oxygen content were cautiously introduced into the apparatus . 

 under a layer of the dilute ammonium hydroxide used for laking, 

 and did not come into contact with the air during the trans- 

 ference except at the narrow upper meniscus of the pipette, and 

 momentarily at its tip. As the pipette was not completely 

 emptied in delivering the sample, the upper portion of the 

 sample, which came into contact with the air was not used in 

 the analysis. The pipettes used were calibrated by weighing 



