200 Proceedings of Royal Society of Edinburgh. [sess- 
ably it does, since it does not form any union with pure Hb0 2 im 
solution when the discs have been destroyed. But if it does not 
unite with Hb0 2 , and yet does unite with the physiologically intact 
disc, it can only unite with some constituent of the corpuscle other 
than the pigmentary. To this extent I would seem to agree with 
Bohr (10); though I could not say the union was with the globin 
any more than with the haematin, seeing that C0 2 effects no change 
upon haemoglobin in solution. In other words, C0 2 effects changes 
of reduction only upon blood. What is noteworthy is the ex- 
tremely unstable or loose character of this compound of C0 2 with 
the red disc, which can in an instant be reoxidised, i.e ., 0 can displace 
the C0 2 very readily. The artificial HbO, made by reducing Hb0 2 
with Am 2 S, is, in comparison, quite stable , for if reoxidised, it shows 
the two bands, and then they fade away to the one again. 
It is, in short, easy to preserve the Am 2 S-HbO intact for 
months, whereas the C0 2 -HbO is a most evanescent production. 
I am very much inclined to regard the union of C0 2 with the discs 
as not a chemical one, but much more of the nature of a physical, 
interstitial, or intermolecular one in the meshes of the protoplasm 
of the intact red corpuscle, — Hb0 2 having no chemical affinity for 
C0 2 , while it has a very strong one for 0. This state of matters 
has a most direct and important bearing upon the needs of the 
body as regards these gases, seeing that the red corpuscle of arterial 
blood must be able to carry its burden of respiratory oxygen to the 
remotest recesses of the tissues, and not till then give it up to the 
lymph, whereas the red corpuscle of venous blood must be ready 
to part with its dissociable C0 2 as soon as it has reached the pul- 
monary capillaries. 
XII .— On Carbon-Monoxide-Hcemoglobin. 
It is well known that chemists regard CO-HbO as “ a very stable 
pigment,” and one that “resists putrefaction for a long time.” I 
tried to answer the question ‘how long?’ and in doing so found 
a difference in stability between what one might call a large and a 
small quantity thus : — a large quantity (250 c.c.) was bottled upon 
30th January 1895, and showed the two bands upon 23rd of 
February, and up to 15th of March, whereas a small quantity put up 
in a test-tube 30th January had by 18th of February become HbO, 
