32 Chapter III 



The reader may say at this point that we have ourselves proved 

 that the molecule of haemoglobin contains but one atom of iron and 

 is therefore the simplest possible. That assumption underlies the 

 hyperbolic form of the curve which is based upon the equation 



For inasmuch as there are two atoms of oxygen to one of iron, 

 if the haemoglobin molecule contained two atoms of iron the equation 

 of the reaction would not be that just given but 



using Hb to denote the simplest possible formula for haemoglobin. 



Let us therefore inquire more particularly where we stand in this 

 matter. Let us see what we have assumed and what we have proved. 

 We have shown in Fig. 12 that the points obtained experimentally 

 from a dialysed haemoglobin solution are consistent with the view 

 that the dissociation curve is a hyperbola and, not as Bohr sup- 

 posed, inconsistent with that view ; moreover it must be remembered 

 that though the fluid in that experiment had a low electric conduc- 

 tivity it was not absolutely salt free. We have shown then that the 

 facts are consistent with the equation 



Hb + 3 ^: HbO 2 . 



But have we shown at that point that they are inconsistent with the 

 equation 



Hb, + 2O 2 = Hb 2 O 4 



or 



As a matter of fact the points as determined would agree nearly as 

 well with a curve which was drawn on the basis of the second of these 

 equations though not on one which was based on the third. Let it be 

 clear however that we are not giving up the position which we estab- 

 lished in Chapter II, namely the probability of the oxygen being 

 chemically united to the haemoglobin. 



I am indebted to Hill for calculating the curve (Fig. 15, Curve II) 

 which would be yielded by the equation 



Hb 2 + 2O 2 = Hb 2 4 . 



The points determined when compared with this curve appear to 

 be not altogether out of conformity with it and suggest that the 

 points are really between the two curves. And this is very likely 

 to be the case. For on our supposition it is clear that the actual 



