Molecular weight of haemoglobin 



33 



hyperbola could only be reached if all the salts were got rid of, 

 that is to say that the hyperbola is a theoretical limit which cannot 

 actually be attained, and similarly the unimolecular is a theoretical 

 limit of the aggregation of the molecules. In actual fact it is not 

 unlikely that the great majority of the haemoglobin molecules are 

 single molecules and that these are mixed with a certain number of 

 others. Therefore there is the possibility of considering what curve 

 would be given if the average number was n = 1'5. This curve agrees 

 with the. determined points just about as well as the hyperbola and 

 it is therefore not unreasonable to suppose that the actual solution, 



30 



40 



50 60 70 80 90 100 



FIG. 15. Curves I, II and III represent dissociation curves in which the mean number of 

 molecules in each aggregate would be 1, 1-5 and 2 respectively. The points are those 

 experimentally determined. 



which was equivalent to a '004 normal salt solution and it is the 

 first traces of salt which would exercise the most profound influence 

 in causing the molecules to aggregate contained some aggregated 

 molecules. 



Such an hypothesis would at once hold out a hope of bringing our 

 own results into line with those of Roaf (5) whose general theme is 

 that the osmotic pressure of the protein, and therefore its molecular 

 weight, varies with the nature of the solvent in which the haemo- 

 globin is dissolved. As we shall have to discuss this matter more 

 particularly in the next chapter we shall leave it at this point. 



B. R. F. 



