206 Proceedings of Royal Society of Edinburgh. [sess. 
after death it would be difficult to say, but I find one experiment 
upon the point. 
A cat was killed by a blow on the bead, and the body lay for 
eighteen hours, when blood from the left ventricle gave the single 
band of HbO very easily oxidised ; blood from the right auricle 
gave HbO not so easily oxidised. 
On the Crystallisation of Hb0 2 of Rodents ’ Blood. 
For the last three years I have kept crystals of Hb0 2 under 
observation. Some were from the rat and were mounted both in 
water and in balsam (by Stein’s method, defibrinated blood mixed 
with balsam). It is interesting that Hb0 2 will not crystallise in 
plasma ; the blood needs to be defibrinated, and the red corpuscles 
must give up their Hb0 2 before crystallisation can be depended on. 
The relation of this fact to the needs of the circulation is too 
obvious to require mention. Whilst watching the familiar crystal- 
lisation in water of the Hb0 2 from the rat, I noticed that a con- 
siderable number of the spicules, whether singly or in stars, were 
colourless ; in every other respect they resembled by transmitted 
light their bright crimson neighbours. 
There were more colourless crystals in the water preparation 
than in the balsam — a fact evidently related to the higher solu- 
bility of Hb0 2 in water. It was also noticed that the water round 
about these colourless groups was stained deep red. What were 
these colourless crystals 1 It has been suggested they were crys- 
tallised proteid-globin ; but this involves the breaking up of 
Hb0 2 , and the pigmentary portion — the bsematin — is not soluble 
in water. If they were still Hb0 2 , colourless Hb0 2 is a substance 
unknown to physiological chemistry. At the same time, there is 
no other term by which they can be alluded to than “ colourless 
haemoglobin,” or the cbromogen of Hb0 2 . More recently, I have 
observed Hb0 2 crystals in balsam from the guinea-pig become 
within three weeks ‘bleached’ to this colourless condition; it 
certainly looks like a reduction of the Hb0 2 to colourless cbromo- 
gen, probably effected by changes in the defibrinated blood sur- 
rounding the crystals. 
On the other band, why should not all similarly formed crystals 
of Hb0 2 become similarly altered ? — the majority do not. I have 
