II. REPLICATION OF DNA IN CHROMOSOMES 101 



and following replication would form part of the linear axi.s of the two 

 sister chromatids. A new set would be formed on each chromatid at 

 prophase when it became double, i.e., when half-chromatid bridges can 

 be induced. In rapidly dividing cells these linkers might be closed most 

 of the time, but in cells where the chromosome acts as a single unit in 

 breakage and reunion these linkers would have to be assumed to be 

 open or rather labile. Since protein synthesis appears to play a role in the 

 reunion of chromosomes, the H linkers for the present will be assumed 

 to consist of polypeptides. 



As soon as the H linkers are stabilized the chromosome should be 

 capable of acting as a double structure in breakage and exchange. In 

 Vicia. this appears to occur for all chromosomes before DNA replication 

 begins or at the beginning of the S period (see Section VI). In other 

 species where there is a marked asynchrony of replication, the stabiliza- 

 tion would be expected to occur in some chromosomes or sectors of 

 chromosomes before its completion in others. 



Replication is assumed to begin at a pair of the 3' ends, perhaps by 

 the insertion of a 5' linker and the closure of the 3' ends by a new and 

 different 3' linker not susceptible to the enzyme which is operating to 

 open the regular 3' linkers. Since this 3' linker is necessary only during 

 the replication cycle, we will refer to it as the 3'R linker. It must be 

 changed to a regular 3' linker before another replication can occur. 

 Some evidence suggests that the change may occur during anaphase or 

 telophase (see below). To allow rotation and untwisting of the DNA 

 helices the 3' linkers should be open on the side opposite the insertion of 

 the new 5' linkers. If there w^ere an out-of-phase replication of the two 

 subunits of a chromatid, unwinding would appear to be facilitated. 

 However, in an organism where the rate of DNA replication may be a 

 factor in limiting growth, E. coli for example, the cell must have evolved 

 some control so that both subunits can be replicating simultaneously. 



When two adjacent chains finish replication we will assume that the 

 3' ends of the two new chains are now closed with a 3'R linker (Fig. 9). 

 The result would be a replicated package consisting of a pair of double 

 helices attached to one axis of the original chromatid. The comple- 

 mentary chains attached to the other axis would now replicate and 

 form other pairs of helices attached to a single axis. When both chroma- 

 tids w'ere complete, each would consist of a series of the four-stranded 

 packets of DNA attached to a linear axis with a length that might very 

 well be manipulated in the dimensions available in a cell (Fig. 10). For 

 example, in E. coli if the units of replication had a molecular weight of 

 2.4X10*"' (Cavalieri and Rosenberg, 1961a), and were attached 40 A 

 apart along the axis, the over-all length of the chromosome would be only 



