Donald D. Van Slyke 129 
laked before addition of the ferricyanide, the first shaking prac- 
tically always removes all the oxygen from the blood solution. 
Even when laking has been incomplete, however, or the first 
extraction of the blood otherwise made incomplete, the deter- 
mination is not lost; for the right result will be obtained if the 
extraction is repeated until the reading becomes constant. 
After each analysis it is well to wash out the 50 cc. chamber 
of the apparatus with the dilute ammonia solution, as a black 
precipitate is formed by reaction of the reagents with the mer- 
’cury. Unless this precipitate is removed it tends to coagulate 
after a few analyses and interfere with further determinations. 
After the blood has been saturated with air the entire pro- 
cedure above outlined, including the final cleaning of the appa- 
ratus, is done in routine determinations in 7 or 8 minutes. 
In order to calculate the oxygen bound by the hemoglobin it 
is necessary to subtract from the gas measured the volume of air 
physically dissolved by the 2 cc. of blood at atmospheric pres- 
sure and the prevailing room temperature. The volume of gas 
thus corrected may be reduced to standard conditions by mul- 
tiplying by (0.999 — 0.0046 #) x vimnncier 
perature in degrees centigrade. The volumes of air dissolved by 
the blood at different temperatures are given in Table I. They 
are calculated in accordance with Bchr’s finding, that the solu- 
bility of gases in average whole blood is 90 per cent of their solu- 
bility in water. The table also gives factors by which one may 
transpose the readings directly into terms of volume per cent of 
chemically bound oxygen in the blood, or of per cent hemoglobin 
on the basis of Haldane’s normal average, 18.5 per cent of oxygen 
in the blood being taken as equivalent to 100 per cent hemoglobin.® 
Unless one is well experienced with the conditions used for 
saturating the blood with oxygen, it is advisable, after one por- 
tion of a blood sample has been analyzed, to aerate the remain- 
- der a second time and repeat the determination, in order to make 
certain that the hemoglobin of the first portion was completely 
saturated with oxygen. | 
, ¢ being the tem- 
3 Haldane, J., and Smith, J. L., J. Physiol., 1899-1900, xxv, 331. 
THE JOURNAL OF BIOLOGICAL CHEMISTRY, VOL. XXXIII, NO. 1 
