456 ROLLIN D. HOTCHKISS 



(Chapter 13). Optimal bonding requires that adenine cross-Unk with 

 thymine, and guanine with the cytosines, therefore insuring that one hehx 

 will of necessity be the complement of the other. One can imagine the 

 double helix to be resynthesized upon a pattern furnished by either half. 

 These hypothetical relations are in agreement with the generalization that 

 the observed adenine-to-thymine and guanine-to-cytosine molar ratios of 

 all DNA's are near unity although, taken the other way, adenine and 

 guanine residues, for example, are often present in very different amounts. '^^ 

 This picture of DNA structure, then, in common with other less explicit 

 ones, would attribute the fundamental specificities to variation in the 

 sequences of the purine-pyrimidine pairs along the helical coils. As we have 

 seen, the analytical data offer some support for this possibility, and yet 

 offer no indications as to how many permutations in arrangement may 

 actually occur. 



h. Physical Evidence of Heterogeneity of DNA 



Bacterial transformation, especially with multiply marked DNA, has in- 

 dicated that different DNA's exist. Unfortunately, DNA does not prove to 

 be antigenic, so its heterogeneity has not been confirmed immunologically. 



Although most preparative methods tend to collect in one fraction all 

 of the DNA of a general polymer size class, there are a few recent indica- 

 tions that not all DNA material behaves alike. Fractions differing in meta- 

 bolic activity and salt solubility^^s g^j^j j,^ binding to basic protein^^^'^^'' 

 have been recognized. Separations of a single species DNA on the latter 

 principle have given fractions of somewhat different base composition. 



Chemical analyses in the years since the bacterial transforming agents 

 were recognized as DNA, have accordingly supported the view that DNA 

 molecules, once thought to be alike and of simple construction, are chemi- 

 cally as well as biologically diverse, in accordance with the important 

 genetic functions they have to fulfill. 



II. Deoxypentose Nucleic Acids in Growth and Metabolism 



Long before there was any convincing evidence that DNA's had a genetic 

 role, a number of workers were interested in investigating their distribution 

 and metabolism in tissues. Much of the findings can be summarized by 

 saying that DNA molecules are metabolically relatively stable in most 

 tissues and that they are produced in rather strict relation to growth. This 



1" E. Chargaff, Experientia 6, 201 (1950); Federation Proc. 10, 654 (1951). 



'" A. Bendich, Exptl. Cell. Research Suppl. 2, 181 (1952); A. Bendich, P. J. Russell, 



and G. B. Brown, J. Biol. Chem. 203, 305 (1953). 

 126 E. Chargaff, C. F. Crampton, and R. Lipshitz, Nature 172, 289 (1953). 

 "7 G. L. Brown and M. Watson, Nature 172, 339 (1953). 



