READJUSTMENTS OF REGULATION 53 
the gas simply pass through them by ordinary diffu- 
sion? This question has been debated ever since a 
suggestion that they may play some active part was 
made by Ludwig forty or fifty years ago. 
By means of an apparatus known as the aeroto- 
nometer, Pfliiger and his pupils compared the pressure 
of CO, in the blood with that in alveolar air, and 
found it to be about the same. The aerotonom- 
eter was then improved by Bohr of Copenhagen, 
an old pupil of Ludwig. His results seemed to show 
that sometimes there is a lower pressure of CO,, and 
a higher pressure of oxygen in the arterial blood than 
in the lung air, in which case an active secretion of 
oxygen inwards, and of CO, outwards, must be 
assumed. Fredericq then got results in favour of the 
simple diffusion theory. . Last of all Krogh of Copen- 
hagen improved the aerotonometer still further, and 
obtained results which again favoured the diffusion 
theory. 
Meanwhile Lorrain Smith and I attacked the prob- 
lem by a new method, which was suggested to me by 
the study of CO poisoning, and which eliminated cer- 
tain sources of very serious error in the aerotonometer 
method of measuring the arterial oxygen pressure. 
When blood is saturated with a mixture containing 
both oxygen and CO the haemoglobin combines partly 
with CO and partly with oxygen in perfectly definite 
proportions depending on the relative pressures of 
the two gases, although in consequence of the far 
greater affinity of CO for haemoglobin the pressure 
of oxygen must be about 300 times greater than the 
