ISOLATION AND COMPOSITION OF DEOXYPENTOSE NUCLEIC ACIDS 351 



oxypentose nucleic acids is compared; in Table XII, which will be dis- 

 cussed later, Nos. 1-4 are AT, Nos. 5-9 belong to the intermediate group, 

 Nos. 10-16 to the GC type. 



The subsequent account will demonstrate that different nucleic acid 

 specimens of the same origin show a remarkable identity in composition; 

 but this may be regarded, as was conjectured very early, "^ as "a statistical 

 expression of the unchanged state of the cell," for it is possible to obtain, 

 by the fractionation of nearly all deoxypentose nucleates, preparations of 

 a composition representative of all three types mentioned above. (See 

 Section VIII.) 



The comparison of analytical results obtained by different workers using 

 procedures that are almost never identical and often radically different 

 suffers from a great deal of uncertainty. If, moreover, a large number of 

 analyses of several well-characterized and purified preparations are re- 

 ported in one instance, and, in another, single analyses of ill-described 

 specimens, a dilemma between encyclopedic comprehensiveness and over- 

 judicious exclusiveness will occur. For analytical results to be acceptable 

 the following minimum requirements should be fulfilled. At least two com- 

 pletely characterized preparations isolated separately from two independ- 

 ently collected specimens of the starting material should be analyzed in at 

 least two different hydrolysis experiments, and a full balance of recovered 

 components on a molar basis should be drawn. It is surprising how seldom 

 these modest wishes are met. 



The compilation of data presented in Tables V to XIII comprises the 

 major part, if not all, of the analytical results with deoxypentose nucleic 

 acids. In general, only analyses performed on isolated nucleic acid speci- 

 mens have been included, and several results had to be omitted because of 

 lack of pertinent information. The total recovery has been indicated, when- 

 ever it was given in the literature; but, in order to facilitate comparison, all 

 molar proportions have been corrected for a 100% recovery in terms of 

 nucleic acid P. 



b. Composition Differences and Similarities 



After the existence of chemical differences between deoxypentose nucleic 

 acids of different origin was first suggested,*"^'"' -^'' agreement was not 

 general. This is not surprising, as in many cases very extensive work is re- 

 quired, in order to prove a divergent composition. But there is little doubt 

 at present that there exist many different deoxypentose nucleic acids and 

 that their number may be even much larger than would be revealed by 

 merely analytical differentiation. While, therefore, the question whether 

 there exists more than one deoxypentose nucleic acid can be answered by 



2" E. Chargaff, S. Zamenhof, and C. Green, Nature 165, 756 (1950). 



