492 D. O. JORDAN 



at pH 2.6. This is shown by the values of the root-mean-square end-to-end 

 distance, which assuming a monodisperse species are: at pH 6.5, 5030 A.; 

 pH 3.0, 4340 A.; pH 2.6, 2150 A. Furthermore this contraction is com- 

 pletely reversible, the value of the end-to-end distance being 5200 A. on 

 neutralizing the solution to pH 6.5 from pH 2.6. It is thus evident that, at 

 pH 2.6 in 0.2 M sodium chloride solution, degradation of the particle does 

 not occur, but only a marked contraction. It is also evident that the de- 

 formability of the molecule at pH 2.6 is greater than at pH 6. This will be 

 due to the removal of many of the hydrogen bonds, thus reducing the 

 rigidity of the molecule so that when the charged groups are neutralized by 

 combination with a proton, the molecule will assume the shape approaching 

 that of a random coil. The reversibility of the contraction (which is also 

 reflected in the viscosity measurements of Reichmann et al.^^^) is due to the 

 fact that, at pH 2.6 in 0.2 M sodium chloride, only about one-half of the 

 amino groups have been titrated and therefore a number of the hydrogen 

 bonds will remain unbroken and these are apparently sufficient to hold the 

 macro-ion together. At lower pH values, at this ionic strength, all the 

 hydrogen bonds will be broken and the molecular sub-units, capable of 

 independent existence, produced. Such a change will, in all probability, be 

 irreversible. 



The results of Reichmann et al}^^ are not in complete agreement with a 

 similar investigation carried out by Horn et al.^^^ The latter authors found 

 that an irreversible change in viscosity occurred when the pH of the solu- 

 tion was changed from pH 7.0 to 3.8 in 1 M sodium chloride. However, the 

 pH was changed by the addition of 0.01 M hydrochloric acid to the solution 

 and not by dialysis, which may account for the irreversible nature of the 

 change. The higher ionic strength used compared with that employed by 

 Reichmann et at. will make the degree of ionization at the different pH 

 values more comparable in view of the influence of ionic strength on the 

 dissociation constants of the amino groups. It would appear important that 

 much more precise information concerning the degree of ionization is neces- 

 sary for a true analysis of these studies. 



The properties of the products of acid treatment obtained by Horn et al}*^ 

 are interesting. Since the change was irreversible, it would appear that the 

 hydrogen-bonded structure had been destroyed and the sub-units, two or 

 more per molecule, liberated. The streaming birefringence results in solu- 

 tions of different viscosity (adjusted by the addition of glycerol), show that 

 molecular sub-units are much more deformable than the original nucleic 

 acid and appear to possess a much more typical polyelectrolyte behavior 

 as would be expected for a single, non-hydrogen-bonded polynucleotide 

 chain. 



