CHROMOSOME BREAKAGE BY DIEPOXIDE AND BY X-RAYS 



One thing is clear, that in spite of the lack of demonstrable heterochro- 

 matin, di(2 : 3-epoxypropyl) ether is as effective in causing structural 

 change in the chromosomes of Tradescantia as it is in Vicia. The changes in 

 Vicia, though not confined to the demonstrable heterochromatic segments, 

 occur largely in them. It may be that changes outside these segments 

 occur in small undetectable segments of heterochromatin or at points which 

 have certain properties in common with heterochromatin. This of course 

 raises the whole question of the definition of heterochromatin. 



The chromosomes of Tradescantia are broken at many scattered locations. 

 Acentrics of certain lengths seem to predominate, but it has not yet been 

 possible to determine with certainty whether any particular points on the 

 chromosomes are sensitive to diepoxide. 



As Revell** has pointed out, there is considerable evidence that the hetero- 

 chromatic regions of Vicia associate with one another in the chromocentres 

 during the resting phase. These are just the regions in which, in the main, 

 breakage by diepoxide occurs. It is probable that no such association 

 between special regions sensitive to diepoxide occurs in Tradescantia where 

 there is no demonstrable heterochromatin and no obvious chromocentres. 

 Consequently it might be expected that interchange between chromosomes 

 (reunion) would occur in a more random manner. My observations suggest 

 that this is indeed the case. There is no such tendency towards homologous 

 chromatid interchange. 



RevelP- 'lays stress on the tendency for apparently homologous interchange 

 to occur in Vicia and compares this phenomenon with chiasma formation. 

 Clearly, reunion or interchange can take place only at points of association. 

 In meiosis, and to a certain extent in mitosis, in Vicia these points of associa- 

 tion are not random. Consequently, interchange will also tend to occur 

 between certain restricted regions of the chromosomes which are normally 

 associated rather than between points of chance association. 



This might be held to account for the lack of an exponential relation 

 between the frequency of interchanges and the dose. There are indications 

 in my data that there may be such an exponential relationship in Trades- 

 cantia. If both chromosome breaks (B") and interchanges show a linear 

 relationship with dose, their ratio should remain constant at all doses 

 provided that the threshold dose, the origin of the curves, is the same for 

 both (and there is no reason to believe that it is not). This is the case in 

 Vicia according to Revell*^- '. In Tradescantia the ratio rises from 0-5 at 20 

 minutes' exposure to approximately 1-4 at 30 minutes' exposure. 



The apparent great difference with a relatively small change of dose 

 may not truly reflect the relationship. Firstly, the samples are relatively 

 small and secondly, the actual dose may change to a much greater degree 

 than is suggested by the time values of the treatment. I have now reason 

 to believe that penetration into the root in Tradescantia may be much slower 

 than in Vicia. Certainly fixatives penetrate more slowly. It is there- 

 fore planned to carry out experiments with longer treatment at lower 

 concentrations. 



Revelli2 has interpreted all types of aberration including all the chromo- 

 some breaks with sister reunion (and presumably those without) as chro- 

 matid interchanges. Certainly there is considerable evidence that some of 



272 



