470 D. O. JORDAN 



The complete three-dimensional Patterson function suggests that only 

 a part of the structure repeats itself in the plane at c = 3^ in the unit cell. 

 This is what would be expected for two coaxial chains related by a dyad 

 axis, as suggested by Watson and Crick. The phosphate groups repeat at 

 c = M, but since the two chains run in opposite directions, this will not be 

 true of the rest of the molecule. 



These important conclusions refer only to the structure of deoxypentose 

 nucleic acids and not of ribonucleic acids. The latter acids have yet to be 

 prepared in a form that will yield satisfactory X-ray diagrams and the 

 structural evidence based on chemical and enzymatic studies (see Chap- 

 ters 11 and 12) suggests that these acids have a branched-chain structure 

 in marked contrast to the unbranched chain of deoxypentose nucleic acids. 



2. Determination of Molecular Weight 



a. Deoxypentose Nucleic Acids 



The apparent molecular weight (particle weight) of the deoxypentose 

 nucleate ion in solution has been measured by a variety of methods, the 

 majority leading to a value of the order of 1 X 10^. The determination of 

 the molecular weight of the nucleate ion in the ultracentrifuge is compli- 

 cated by a marked variation of the sedimentation coefficient with concen- 

 tration of the nucleate ion. This difficulty, which is typical for poly electro- 

 lytes, may be reduced by increasing the ionic strength of the solution by 

 the addition of neutral salt, but even so, the results shown in Fig. 12 indi- 

 cate a marked variation of the sedimentation coefficient in a buffer of ionic 

 strength 3. This dependence upon concentration necessitates the extrapola- 

 tion of the sedimentation results to obtain the value of the sedimentation 

 coefficient at infinite dilution. Tennent and Vilbrandt*^ observed a par- 

 ticularly rapid increase of sedimentation constant at low concentrations 

 which did not permit extrapolation. This observation has not been con- 

 firmed by Cecil and Ogston,*^ Atlas and Stern, ^^ or by Kahler,®^ whose data 

 shown in Fig. 12 are in close agreement. Kahler^^ ^nd Ogston^^ extrapolated 

 to infinite dilution by making use of the fact that the reciprocal of the 

 sedimentation coefficient is a linear function of the concentration of the 

 nucleate ion at low concentrations (Fig. 13). The values obtained were 

 12.5 X 10-1^ c.g.s. (Kahler«2), 13.2 X 10-" c.g.s. (Ogston^^). These values 

 have been confirmed by the less detailed investigations of Conway et al.^* 

 and Krejci et al.^^ 



" H. G. Tennent and C. F. Vilbrandt, J. Am. Chem. Soc. 65, 424 (1943). 



•0 R. Cecil and A. G. Ogston, /. Chem. Soc. 1948, 1382. 



•1 S. M. Atlas and K. G. Stern, unpublished data. 



^^ H. Kahler, /. Phys. & Colloid Chem. 52, 676 (1948). 



"A. G. Ogston, Trans. Faraday Soc. 46, 791 (1950). 



'* B. E. Conway, L. Gilbert, and J. A. V. Butler, J. Chem. Soc, 1950, 3421. 



«5 L. E. Krejci, L. Sweeny, and J. Hambleton, J. Franklin Inst. 248, 177 (1949). 



