HEMOGLOBIN, CARBON MONOXIDE AND OXYGEN 167 



distance of any other point on the curve from the abscissa and 

 ordinate is also 310; we may express it thus for the whole curve 



[O,] [HbCO] ^ 

 [CO] [HbOg] 



This is conditioned by the fact that the curve is a hyperbola, and in 

 so far as it is not true of the actual figures given for the points 

 tabulated above, the error is due to the fact that they are read 

 from a diagram, i.e. to the fallibihty of human drawing. 

 Thirdly, if 



[O2] ^ [HbCO] 



it follows that 



[CO] ^ [HbOa] 



= 310, 



[HbCO] ^3^^ [CO] 



[HbO^] [O2] • 



Not merely one point on the curve — that at which the oxy- and 

 carboxyhsemoglobin are present in equal quantities — but every point 

 on the curve is defined by the number 310, because at every point the 

 ratio of carboxy- to oxyhaemoglobin is 310 times the ratio of carbon 

 monoxide to oxygen. 



The figure 310 is in fact the equihbrium constant of the reaction 

 and is usually denoted by the letter K. 



To retrace our steps. I said a while ago that Haldane and his 

 associates found a number of curves, with somewhat different values 

 of K, each of which appeared to represent the eqtiihbrium between 

 oxygen, carbon monoxide and haemoglobin. It was a matter of some 

 interest to discover why such differences should exist. At first the 

 presumption was in favour of some sort of experimental error, such 

 for instance as insufficient time given for the equiUbration of the blood. 

 The acceptance in 1909 of the general view that changes in hydrogen- 

 ion concentration, temperature and saUne content of haemoglobin 

 solutions, altered the dissociation curve of oxy haemoglobin, made 

 it more reasonable to believe that the various curves which had been 

 obtained for the reaction 



HbCO + O2 5^ CO + HbOg 



were not merely due to experimental error, but to more substantial 

 causes. To quote Douglas, Haldane and Haldane (2): 



The influence of CO2 and that of salts on the curve were also quite unsuspected. 

 The researches of Zuntz and Loewy, Bohr and Barcroft and their associates, have. 



