490 



D. O. JORDAN 



whereas in the case of the former measurements it is doubtful whether vis- 

 cosities in very dilute solutions can be determined with sufficient accuracy 

 to decide whether the rjap./c — vs. — c curves obtained with and without 

 added electrolyte extrapolate to the same point. It is to be concluded there- 

 fore that the deoxypentose nucleate ion in solution is deformable, but to a 

 much lesser extent than a typical polyelectrolyte or uncharged polymer. 



One further fundamental problem should be mentioned. For a deoxy- 

 pentose nucleate ion having the structure suggested by Crick and Watson''^ 

 and a molecular weight of 8 X 10«, the dimensions would be 40,000 A. by 

 20 A. This length is too great by a factor of five or six and can clearly only 

 be reduced by increasing the smaller dimension either by coiling or by chain- 

 branching. It would therefore appear impossible for the experimental values 

 of 8 X 10« for the molecular weight and 6800 A. by 20 A. for the dimen- 

 sions to be correct for a particular set of conditions. The value for the 

 molecular weight and length were obtained from the same measurements 

 (light-scattering) but the width is much less certain and depends largely 

 on measurements on the dried material. It is unlikely that a particle of size 

 40,000 A. by 20 A. should remain as a rigid rod in solution, but would coil 

 to give a much less asymmetric particle as the light-scattering evidence 

 indicates, ^^-^i The suggestion ^^ that the symmetry is due entirely to chain- 

 branching would appear not to be in agreement with the streaming bire- 

 fringence results since deformation would not then be possible. It is clearly 

 evident that more reliable experimental data is required before a complete 

 picture of the size and shape of the nucleate ion in solution is obtained. 



c. The Influence of Changes of pH on the Size and Shape of the Deoxypentose 

 Nucleate Ion in Solution 



The comparison of the forward- with the back-titration curve of sodium 

 deoxypentose nucleate (p. 477, Fig. 15) shows that the action of both acid 

 and alkali produces a marked change in the properties of the nucleic acid. 

 This behavior has been ascribed by Gulland et al}^ to an irreversible break- 

 ing of the hydrogen bonds existing between the amino and — NH— CO — 

 groups in the nucleate ion. The breaking of the hydrogen bonds occurs at 

 pH values of 5.0 and 11.0, between these values no groups are titrated on 

 the addition of acid or alkali and no hydrogen bonds are broken. In a re- 

 lated study of the viscosity changes produced in a 0.24 % solution of sodium 

 deoxypentose nucleate, Creeth et al}^^ observed that the viscosity dropped 

 sharply in the region of the same critical pH values of 5.0 and 11.0 at which 

 the titration of hydrogen-bonded groups occurred. This behavior was in- 

 terpreted as indicating that the hydrogen bonds unite molecular units which 

 are more symmetrical and of lower molecular weight than the original 

 nucleate. These units become the disperse species in acid and alkaline solu- 



