33. NUCLEIC ACIDS OF THE BACTERIAL VIRUSES 223 



P M -labeled parent, and one would expect that the actual labeled DNA chains of the 

 parent would be present in an appreciable fraction of such quadruply marked progeny. 

 If this were so, decay of these 1 >;)2 atoms upon storage of the over-all progeny would 

 specifically, and markedly, reduce the number of such quadruply marked progeny. 

 This result was not, however, obtained; if correct, these experiments would suggest 

 t hat these 4 loci are not located on one large DNA molecule which retains its integrity 

 (on the single-strand level). Stent's experiments have been criticized as to technical 

 details and Levinthal and Baylor made in 1956 154 a preliminary report of a contra- 

 dictory result in a 4 factor cross. 



While all of these experiments make it clear that a considerable amount 

 of parental DNA remains associated with the parental genotype, no quan- 

 titative explanation can be made at present. 



c. Chemical Mutagenesis 



The use of the technique of fine structure genetic mapping as developed 

 by Benzer 136 ' 137 permits detailed analysis of the effects of various chemical 

 mutagens upon the phage genome. These mutagens are believed to act by 

 causing mistakes during the replication of the DNA, thereby resulting in 

 changes in the nucleotide sequence. 



Among the most effective mutagens studied are: 5-bromouracil, 155 2- 

 aminopurine, 156 ' 157 nitrous acid, 158, 159 pH 5 at 45°, 159 and proflavine. 160 ' 161 



5-Bromouracil and 2-aminopurine are believed to act as tautomeric 

 mutagens 162 in the manner originally suggested by Watson and Crick. 163 

 Ordinarily, in its keto form, 5-bromouracil would be expected to replace 

 thymine and pair with adenine. However, in its enolic form, 5-bromouracil 

 will pair as does cytosine, so that it may enter a replicating DNA opposite 

 a guanine; at a later replication it may revert to its keto form and pair 

 with an adenine, thus, resulting in replacement of a guanine with an adenine 

 (mistake in incorporation). Or, if incorporated into DNA in place of thy- 

 mine (paired with adenine), a 5-bromouracil may during a later replica- 

 tion, enolize, and pair with a guanine, thus, causing a replacement of an 

 adenine with a guanine (mistake in replication). The presence of the bro- 



154 C. Levinthal and M. Baylor, in "The Chemical Basis of Heredity" (W. D. Mc- 

 Elroy and B. Glass, eds.), p. 745. Johns Hopkins, Baltimore. 1957. 



155 8. Benzer and E. Freese, Proc. Nail. Acad. Sci. U. S. 44, 112 (1958). 



156 E. Freese, Proc. Natl. Acad. Sci. U. S. 45, 622 (1959). 

 167 E. Freese, J. Mol. Biol. 1, 87 (1959). 



158 W. Vielmetter and C. M. Wieder, Z. Naturforsch. 14b, 312 (1959). 



159 E. Freese, Brookhaven Symposia in Biol. 12, 63 (1959). 



160 It. I. De Mars, Nature 172, 964 (1953). 



161 S. Brenner, S. Benzer, and L. Barnett, Nature 182, 983 (1958). 



162 2-Aminopurine could form a single hydrogen bond to 5-hydroxymethylcytosine 

 in its normal state without tautomerization. 167 



163 J. D. Watson and F. H. C. Crick, Cold Spring Harbor Symposia Quant. Biol. 18, 

 123 (1953). 



