9] X-RAY ANALYSIS OF HAEMOGLOBIN 137 



the first maps of the electron density distribution in the haemoglobin 

 molecule in projection along the dyad, or b axis.' Two maps were calcu- 

 lated; one showing a projected picture of the volume of hydrated protein 

 into which salt cannot penetrate, from which the external shape of the 

 molecule could be deduced. The other map showed a projected picture of 

 the electron density distribution within a row of haemoglobin molecules 

 suspended in water. Both projections were along the b axis of the crystal, 

 i.e. at right angles to a, the suspected chain direction along which Bragg, 

 Howells and myself had attempted to calculate the first projected picture 

 of the molecule. The b projection is centro-symmetric so that only the signs 

 of the X-ray reflexions have to be determined, whereas the a projection is 

 not and the phase angle may have any value. 



The maps give the external shape of the molecule as an ellipsoid of 

 55 X 55 X 70 Â, with a dimple at the centre, or possibly two dimples at the 

 centre on either side of the molecule. They also show that the four free 

 sulphydryl groups of native haemoglobin^ are arranged in two close pairs 

 spaced 30 Â apart and related by the dyad axis of symmetry. Unpublished 

 work on two projections of ox haemoglobin since done by D. W. Green and 

 A. C. T. North suggests that these SH-groups are situated at about 7 Â 

 from the surface of the molecule. Apart from the sulphydryl groups, which 

 can be located by the heavy atoms attached to them, the projection shows 

 a system of peaks and depressions which has so far defied interpretation 

 and is far too complex to fit into any simple structural theory. One obstacle 

 to interpretation is lack of resolution ; this is so low that single atoms are 

 smeared out to a diameter of 9 Â. The main obstacle, however, is the great 

 thickness of matter, corresponding to about 30 atoms, which is projected 

 on to one plane, thus obscuring the individual parts of the molecule by 

 superposition. 



At the next stage of the analysis Ann F. Cullis, H. M. Dintzis and myself 

 have now tried both to improve the resolution of the projection along the 

 b axis and to prepare some more heavy atom complexes in preparation for 

 a three-dimensional analysis of the structure. At the same time D. M. Blow 

 has made a renewed attempt to solve the projection along the a axis, this 

 time making no assumptions about the structure, but relying entirely on 

 isomorphous substitution to determine the phase angles. 



An additional pair of sites combining with heavy atoms was found by 

 blocking the sulphydryl groups with iodoacetamide and then allowing the 

 haemoglobin to react with two equivalents of mercuridiacetate. The posi- 

 tions of these new mercury-combining sites in relation to the sulphydryl 

 groups are shown in Fig. 1 . Their chemical character is still unknown. For 

 calculating the improved Fourier projection along the b axis we determined 

 the signs of 420 reflexions instead of the 94 used in the first round, and 

 thus attained a resolution of 2-8 Â instead of the original 6-5. The new map, 

 shown in Fig. 2, contains far more detail than the earlier one, but still fails 



