7 7 3 CHEM1STR Y OE RESPIRA TION. 
the tension of that gas varied between 09 and 57'8 mm. In the majority 
of the experiments the air of the trachea contained carbon dioxide with 
a higher tension than that of the gas in the blood. From these results 
Bohr concluded that the exchange of gases between the air of the alveoli 
and the blood in the lungs could not be accounted for by diffusion alone, 
and lie suggested that the tissues of the lungs played an active part in 
the absorption of oxygen and in the excretion of carbon dioxide. 
These results are so opposed to those obtained by Pnuger and his 
pupils, that they naturally are subject to considerable criticism. 1 In the 
first place, it is to be noted that the respiratory quotients obtained by 
Bohr during his experiments show values varying from 054 to l'Ol, 
results which suggest imperfect and irregular ventilation of the lungs. 
Hiifner 2 contests Bohr's results, and suggests that the irregularities in 
the results are due to a want of equilibrium in the tension of gases in 
the blood and in the air of the hsemataerometer. He finds that equilil >riuin 
only obtains after several minutes and vigorous shaking of the blood in 
the apparatus. Similar objections have been made by Fredericq, 3 who 
obtained, for the tension of oxygen in the peptonised arterial blood of the 
dog, results always lower than the partial pressure of oxygen in the air 
of the alveoli. Further, the results obtained by Fredericq for the carbon 
dioxide agree with those given by Pfluger and his pupils. 
The following values are given by Fredericq 4 for the tension of oxygen 
and of carbon dioxide in percentages of an atmosphere. 
Dog. 
External Air. Air of Alveoli. Arterial Blood. Tissues. 
Tension of oxygen . . 20'95 > 18 > 14 > 
External Air. Air of Alveoli. Venous Blood. Tissues. 
Tension of carbon dioxide . 0-03 < 2S < 3'81-5-4 < 5-9 
Quite recently Haldane and Lorrain Smith 5 have studied the tension of 
oxygen in the arterial blood of man by a new method, which, they maintain, 
avoids the probable sources of fallacy in the aerotonometer. In this new 
method the tension of oxygen in the arterial blood is calculated from the 
percentage of carbon monoxide breathed by the subject of the experiment, and 
from the final saturation of his blood with carbon monoxide. The results give, 
for the oxygen tension of human arterial blood, a value of 26 "2 per cent, of an 
atmosphere, or 200 mm. of mercury. This value is about twice as high as that 
of the oxygen in the pulmonary alveoli, and if it be correct, it follows that 
diffusion alone does not explain the absorption of oxygen by the blood in the 
lungs. Haldane and Lorrain Smith discuss some of the possible sources of 
error in their method, such as the estimation of the saturation of the blood 
with carbon monoxide, the dissociation of carboxyhoemoglobin, the effect of 
dilution of the haemoglobin, and the excretion or oxidation of carbon monoxide ; 
but the test experiments which they made confirm them in their opinion of its 
accuracy. 
Before, however, these results are accepted, further experiments are needed 
to test the method, for it is impossible with our present knowledge to judge 
1 See also criticism by Zuntz, Fortsclu. J. Med., Berlin, 1890, Bd. viii. S. 856. 
2 Arch. f. Physiol., Leipzig, 1890, S. 10. 
3 Centralbl.f. Physiol., Leipzig u. Wien, 1893, S. 33. 
4 Fredericq et Nuel, ''Elements de physiologic bumaine," Gand, 1893, pp. 156-158. 
8 Journ. Physiol., Cambridge and London, 1896, vol. xx. p. 497. 
