THE UNION OF OXYGEN WITH HEMOGLOBIN 131 



Hill's equation would stand 



^^3^,(1 + .12,)^ 



This equation corresponds well with the curve for blood. 



But there are difficulties : 



The first is that though 100 grams of blood contain about 

 14 grams of haemoglobin, it cannot be said to be a 14 per cent, solu- 

 tion of hsemoglobin. The corpuscle is about a 30 per cent, solution. 

 The calculation on that basis becomes much more difficult to accept. 



The second is that, according to the assumptions which we have 

 made, the dissociation curve of a dilute solution should be more 

 inflected than that of a strong solution. And the experimental 

 evidence is against that view. 



The bulk of evidence still seems to support the idea that the dis- 

 sociation curve of haemoglobin reduced to its simplest terms is a 

 rectangular hyperbola. The evidence for this is : 



(1) The work of Barcroft and Roberts (15), which gave a curve 

 approximating to that shape. 



(2) The work of Hartridge and Roughton(i6) which, at every point 

 at which the matter could be tested, indicated that in dilute solutions 

 the reaction between haemoglobin and oxygen was a bimolecular one. 



(3) That Hecht and Morgan (17) did on three occasions obtain a very 

 perfect rectangular hyperbola, while the curves which they obtained 

 on other occasions were intermediate in inflection between that of 

 blood and the hyperbola. 



Now the question arises : Can these difficulties be cleared away ? Is 

 the reaction 



Hb4 -f- 4O2 =:^ Hb^Os 



capable of being represented graphically by a rectangular hyperbola ? 

 The answer is in the affirmative, but certain assumptions must be 

 made which may be integrated in the statement that the reaction 

 must be regarded as four bimolecular reactions taking place inde- 

 pendently, thus: 



Hb^ +02^^ Hb^Oa 



Hb^Oa +02=:^ Hb404 



HbA+Oai^HbA 



Hb40e+ Oa^^HbA. 



If the reaction were of this character, and the molecule large enough 

 to allow of the supposition that each molecule of combined oxygen 



9-2 



