II. REPLICATION OF DNA IN CHROMOSOMES 73 



The complementarity of the two chains and the specificity of the 

 hydrogen bonds suggested a mechanism for precise replication (Watson 

 and Crick, 1953b). If the chains came apart, each contained the hydrogen 

 bonding sites to line up the nucleotides of a complementary chain. This 

 type of replication would yield two identical molecules (Chapter I). 

 With this proposal of a precise model, the foundations were laid for 

 experimental tests. Techniques and suitable biological materials had to 

 be found. By taking the most optimistic view of all the findings, namely, 

 (1) that DNA was the genetic material of the cell, (2) that all of the 

 DNA of a chromosome was exactly reproduced once each division cycle, 



(3) that the proposed structure was correct for all of the DNA, and 



(4) that the Watson-Crick scheme for its replication was correct, one 

 could reach the conclusion that one chain of each original DNA molecule 

 in a chromosome or in a whole cell would be passed to each daughter 

 nucleus at mitosis. A new complementary chain would likewise be built 

 along each and passed on to the descendants. If a label (radioactive 

 isotope) were present only in the original DNA or available only for 

 the synthesis of the new chains, the distribution could be predicted. The 

 possibilities were formally stated by Delbriick and Stent (1957): (1) 

 if the original double helix remained intact and somehow produced a 

 completely new two-stranded molecule, the scheme would be conserva- 

 tive; {£) if the chains separated without breakage and each served as 

 a template for the formation of a complementary chain the scheme would 

 be semi-conservative; and (3) if the chains were broken or destroyed 

 in the process the scheme would be dispersive. 



Attempts to follow the distribution of carbon- 14-labeled DNA in 

 whole cells at anaphase by autoradiography might have provided a par- 

 tial answer. At least a clear result in a cell with a few chromosomes could 

 have distinguished between a conservative scheme versus a semi-con- 

 servative or dispersive replication. However, the published results 

 (Plaut and IMazia, 1956) did not provide a clear answer although they 

 were interpreted by the authors to favor a conservative scheme. 



To get a definitive answer a techniciue that could follow the distribu- 

 tion of label at the chromosomal level was required. This was provided 

 for large chromosomes by autoradiography with tritium-labeled thy- 

 midine, first prepared and used by Taylor et al. (1957). Because of its 

 low energy beta emission tritium gives better resolution for autoradi- 

 ography than any other isotope. In addition, compared to carbon-14 its 

 shorter half-life (12.3 years) allows one to prepare compounds with a 

 much higher specific activity. Since thymidine, which was labeled by an 

 exchange reaction (Taylor, 1960a), is also a selective label for DNA, 

 other components of the cells would not be labeled. Because the intra- 



