484 D. O. JORDAN 



crease of temperature and retarded by the presence of electrolyte.^^^ It 

 could also be caused by aggregation of the ions to form ionic micelles in the 

 more concentrated solutions. A further cause of the nonreproducibility is 

 that the properties vary considerably with the method of preparation of the 

 solution, and the most satisfactory method is to permit the fibrous nucleic 

 acid to dissolve slowly without stirring at ca. 0°, when solution is normally 

 complete within twelve hours. Solutions of sodium deoxypentose nucleate 

 in pure water as solvent and prepared in this way show a very high vis- 

 cosity, which is dependent to a marked degree on the rate of shear, and also 

 exhibit strong streaming birefringence. On the addition of electrolyte all 

 three effects are considerably reduced and the reproducibility of the meas- 

 urements greatly increased. 



The decrease of viscosity on the addition of electrolyte''^ ■"''•^2^'^2' was 

 first explained by Greenstein and Jenrette^^* as a reversible depolymeriza- 

 tion of the deoxypentose nucleate. However, comparison of the acid-base 

 properties in 1 .0 M potassium chloride solution with those in pure water as 

 solvent^^ shows that there is no increase in the titratable acidic and basic 

 groups when the deoxypentose nucleate is in the former solvent and further- 

 more indicates that the hydrogen-bonded structure is not affected by the 

 increase in ionic strength. More definite evidence that the process is not one 

 of depolymerization is afforded by the light-scattering measurements of 

 Reichmann et alJ^ which show that an increase of ionic strength does not 

 change the molecular weight, but only produces relatively small changes in 

 the shape of the ion. The similarity of the behavior of the deoxypentose 

 nucleate ion and of synthetic polyelectrolytes such as poly-A''-n-butyl-4- 

 vinylpyridonium bromide to changes of ionic strength, led Jordan^^" to sug- 

 gest that in water solution the nucleate ion was fully stretched by virtue of 

 the repulsion between the charged (-^P0~) groups. The fall in viscosity 

 on the addition of electrolyte would then be produced by a coiling of the 

 molecule permitted by a neutralization of these charged groups. This mech- 

 anism is identical with that suggested by Fuoss and Strauss^^' for the syn- 

 thetic polyelectrolytes. The early work of Bungenberg de Jong and Kwan'^- 

 had indicated that there was a similarity between the viscosimetric behavior 

 of nucleic acids and polyelectrolytes, and this has been confirmed by Basu.^^^ 

 In a more recent study, however, Pouyet^^^ has made measurements of the 



" T. Miyaji and V. E. Price, Proc. Soc. Exptl. Biol. Med. 75, 311 (1950). 



28 J. P. Greenstein and W. V. Jenrette, /. Natl. Cancer Inst. 1, 77 (1940). 



29 G. Vallet and H. Schwander, Helv. Chim. Acta 32, 2508 (1949). 

 '» D. O. Jordan, Trans. Faraday Soc. 46, 792 (1950). 



" R. M. Fuoss and U. P. Strauss, /. Polymer Sci. 3, 246; 602 (1948). 



32 H. G. Bungenberg de Jong and U. S. Kwan, Kolloidchem. Beih. 31, 89 (1930). 



" S. Basu, Nature 168, 341 (1951). 



3< J. Pouyet, Compt. rend. 234, 152 (1952). 



