CHROMOSOME ABERRATION 



conforms exactly with the ' contact ' hypothesis visuaKzed by earlier workers. It is 

 much more important, as well as more interesting, to try to find out what actually 

 happens. 



Again, it is certainly essential, as Gray says, to discover whether the independently 

 induced events, which are usually involved in each X-ray interchange, can occur 

 anywhere on the ' chromosome ' or only at loci in the defined associations. I think, 

 however, that it is also very important to recognize that this need not be just a 

 question of whether breakage is primary or secondary to exchange initiation, because 

 these events might not be breaks at all. The conclusion that breakage has to happen 

 at some stage in exchange induction only follows from the assumption that the 

 undoubtedly permanent genetic pattern is actually maintained as a permanent 

 structure. 



I also agree that the earlier stages in the actions of ionizing radiations and, say, 

 di-(2 : 3-epoxypropyl) ether must be different ; but the ' chromatid ' aberrations 

 which finally result from these two types of treatment are qualitatively indistinguish- 

 able, at least so far as I can see ; and therefore I prefer to think that the last stage 

 of both actions is probably the same. It is, of course, quite true that ' chromatid ' 

 changes produced in these two different ways can be distinguished statistically ; 

 the radiation changes are more nearly random in distribution and they show a 

 much higher proportion of 'open ends' (or 'imperfect exchanges') than do the 

 diepoxide changes ; but I do not think that these quantitative differences are by 

 themselves sufficient reason for supposing that the two agencies are inducing exactly 

 the same types of aberrations by two entirely different mechanisms. 



G. OsTERGREN : Our Studies on chromosome breakage in the onion by means of 

 coumarin suggest strongly that the figures which look like half-chromatid or part- 

 chromatid changes may in reality be ordinary full chromatid changes that are 

 masked by the presence of matrix. The reason for this opinion is the fact that in 

 the next division we get practically exclusively chromosome type changes. Further- 

 more, the presence in the Xg cells of induced constrictions or attached fragments, 

 occurring in both sister chromatids at the same locus, suggests to us that the treat- 

 ment has induced mutation to a labile state of some loci. These have reproduced 

 to give two labile sister loci which break during a critical period in the prophase 

 of the X2 division (Ostergren and Wakonig*). It seems to me that similar efTects 

 may occur in Revell's diepoxide treatments. The main effect of his treatment may 

 be mutations to a labile state occurring in early resting stage. After that follows 

 reproduction to give pairs of labile loci, one in each chromatid. Later on, during 

 a critical period in early prophase, these labile loci break and the breaks are recom- 

 bined. As breakage and reunion often occur in chromosomes having a matrix, 

 many of the changes are masked to subchromatid changes, although in reality they 

 may be of full chromatid type. It is conceivable that besides false sub-chromatid 

 changes there may also exist some ti'ue changes of this kind. Some of the labile 

 loci suggested by us may survive one or more mitoses and cause an effect of delayed 

 breakage (our observations on coumarin indicate that this may be the case). 



J. A. V. Butler : It would appear that the time for the explanation of these 

 chromosome breakage effects in terms of molecular structure has not yet arrived. 

 However, there are some simple facts which have to be allowed for. The chromo- 

 some is a large structure, e.g. the rat chromosome contains 10* particles of DNA. 

 According to the genetical results these must be united in a linear order (or multiples 

 of a linear order, as there may be multiple strands). How are they joined together ? 

 The chromosomes also contain proteins (histone) and the simplest possibility is that 

 the junctions between DNA particles are made of histone. This would perhaps give 

 a point of weakness at the junction of DNA and histone at which breaks would 

 most easily occur. The junction would presumably be a salt bond between NH3 "^ 

 and POi^. If by the action of radiation either of these terminal groups were removed 



280 



