222 



WALDO E. COHN 



ionization of phosphate and of base groups as a function of pH is plotted in 

 Fig. 6. The order of elution predicted from these considerations is subject 

 to modification by the greater attraction of the exchanger for purines over 

 pyrimidines mentioned earlier. Thus, in order to superimpose the separa- 

 tions found (Fig. 7) on those predicted solely from the net-charge criterion 

 (Fig. 6), it must be assumed that the purine compounds are about three 

 times as strongly sorbed as the pyrimidines.^^ This factor remains fairly 

 constant over a wide range of pH (and hence of net charge), thus indicat- 

 ing that it probably has to do with intrinsic properties of the solutes — which 

 differ only in the base constituents — and not with ionic properties which 

 are pH dependent. 



The variation of elution position with pH (Fig. 7) is of some interest, 

 particularly in view of the increasing use of volatile salts for elution rather 

 than mineral acids. At neutral pH, similar purine nucleotides appear to- 

 gether and similar pyrimidine nucleotides appear together, the latter sep- 

 arated (by the factor of about 3) from the former. It may be concluded that 

 the separation of each pair at more acid pH is due to the development of a 

 cationic group (the ammonium group) , causing adenylic to precede guanylic 

 acid (pK 3.7 vs. 2.3) and C3d:idylic to precede uridylic acid (pK 4.2 vs. 



Fig. 6. Net charge per molecule of ribonucleotide as a function of pH, calculated" 

 from the pK values quoted in Levene and Bass."* 



