II. REPLICATION OF DNA IN CHROxMOSOMES 91 



AVhilc both of these possibilities may be realized in some instances, 

 the reproduction of chromosomes determined by the types of aberrations 

 produced shows a good correlation with other events that would be 

 expected to occur during reproduction. Although both chromatids of a 

 chromosome may be broken simultaneously (isolocus breaks) the break- 

 age of one without the other being affected reveals a kind of reproduction 

 or change of state. Although broken ends may not undergo reunion 

 immediately, experimental evidence indicates that most breaks rejoin 

 or the ends exchange with other ends produced by breaks close by within 

 an hour or less in metabolically active cells (Wolff and Luippold, 1956a). 



Sax (1941) studied the change in state of chromosomes of Trades- 

 cantia microspores by analyzing the types of aberrations in cells fixed 

 at hourly intervals after X-radiation. The change from single chromo- 

 somes to double chromosomes occurred between 32 and 26 hours before 

 metaphase. When the time of DNA synthesis was determined in these 

 same cells by autoradiographic studies (Taylor, 1953) the two events 

 were shown to coincide. Later a study of DNA synthesis by autoradi- 

 ography and cytophotometric techniques at this stage and in the subse- 

 quent replication in the generative nucleus again showed a close 

 correlation of the change to doubleness to radiation and the increase 

 of DNA (Moses and Taylor, 1955). Thoday (1954) found a similar 

 correlation in root cells of Vicia. The change in state did not appear 

 to be abrupt but a gradual one spread over several hours. The experi- 

 ments were not designed in such a way that they could reveal whether 

 this difference was due to an asynchrony in the cell poinilation or a 

 gradual change of the chromosome complement of a single cell as its 

 DNA was replicated. Recent experiments indicate that the change may 

 be rather abrupt and may occur before DNA replication (Wolff, 1961 ; 

 J. H. Taylor, unpublished). However, the abruptness of the change may 

 vary with the cell type. In the cells studied DNA replication extends 

 over a period of several hours and is sometimes asynchronous with 

 respect to whole chromosomes or sectors of these. In Vicia and BeUevalia 

 root cells, all chromosomes are labeled along most of their length nearly 

 simultaneously, but in cells from other sources (Crepis roots, Chinese 

 hamster fibroblasts, human leucocytes, and grasshopper gonial cells) 

 individual chromosomes or sectors are asynchronous in their replication 

 as indicated by labeling with thymidine-H^ (Taylor, 1960c; Morishima 

 et (iL, 1962). A recent report by Chu and Giles (1961) shows that a 

 chromatid exchange and a chromosome exchange can occur in a single 

 cell of tlie human where the labeling mentioned above indicates 

 asynchrony. 



All of tlic information available indicates that the changed state, 



