PHILIP GEORGE 



-2.9, +13.6, and +16.3 for the OH" F", Nj, and CNS" 

 complexes of ferrihemoglobin. These values, with those for 

 ferric iron complexes, are shown diagrammatically in Figures 2 

 and 3. Whereas the values for the triatomic ligands N^ and 

 CNS~ are appreciably higher than those for the monatomic F~ 

 and diatomic OH~, which could arise in part through the 

 additional vibration, the values as a whole are extremely nega- 

 tive. Values not very different from those for the cobaltic 

 complex ions might have been anticipated, say between +20 

 and +30 e.u., since, although the ligand is directly attached to 

 the metal in the ferrihemoprotein complexes, in both instances 

 the metal is already bound within a definite coordination 

 structure in the parent compounds. The very low values are 



e.u. 



20 



i 



HAEMOGLOBIN^ 

 — 'CNO 



=x:ns 



Fe; 



'Hb 



N3 

 -F 

 -OH 



Fe, 



Hb 



^-HbY 



Fig. 3. Relative values of S" for hemoglobin in the ferrous and ferric oxidation 

 states, together with values for some ferrihemoglobin complexes. 



again indicative of a loss in entropy, arising through some change 

 in solvation, or tightening of the prosthetic group and possibly 

 the protein fabric in its immediate vicinity, which is absent in 

 the other cases. A configurational change of this kind, lowering 

 AS^ for both ferro- and ferri-myoglobin reactions, would, 

 however, entail a favorable increment in Ai/o, heat being 

 liberated because of the stronger bonds formed as the structure 

 tightens, and so the more positive heats of formation (more 

 negative Ai/o values) for both sets of hemoprotein reactions are 

 consistent with these entropy effects. 



Finally a useful comparison can be made between ferri- 

 myoglobin and ferrihemoglobin. It has been suggested that 

 the equilibrium constants for complex formation are about 10 

 times greater for the latter (45). This may be true, however, 



356 



