A NEW HYPOTHESIS FOR 'CHROMATID' CHANGES 



secondly because an unexpectedly high proportion of exchanges seemed to 

 be homologous — that is to say, structurally similar to meiotic chiasmata. 

 For example, seventeen out of a series of forty X-ray interchanges between 

 the two long chromosomes in Vicia were found, so far as could be judged, 

 to be exactly homologous. This seemed to suggest that a close somatic 

 pairing must be involved in their formation. 



As is usual in this kind of material, it was also found that all exchanges, 

 whether homologous or not, were reciprocal. This last observation can by 

 itself be explained without difficulty by the orthodox theory, but, in con- 

 junction with the other evidence, raised doubts as to whether this theory 

 was after all entirely sound. 



The question of whether or not chromosomes might have to be function- 

 ally related in some way for exchange to occur has been the subject of much 

 experimentation and some controversy (see Muller*^) . ' Two opposite views 

 of the mechanism were conceivable : either that breakage of the chromo- 

 nemas occurred first — presumably independently in the different places — to 

 be followed by fusion as a later consequence, or that the process of breakage 

 was dependent on the prior approximation and perhaps fusion of the 

 chromonemas — the different breaks concerned being in this case results of 

 a common initiating process.' 



However, there is already an assumption implicit in this statement of the 

 problem : that a process of breakage is necessarily involved. This is also an 

 important premise involved in the actual scoring of visible changes at 

 metaphase. Described without prejudice these changes consist of discon- 

 tinuities (Loveless'^) and rearrangements. The classical theory assumes that 

 these discontinuities are breaks ; that these breaks are the primary aberra- 

 tions that were induced ; and that such breaks are the components — by their 

 reunion — of all the rearrangements. It is thus assumed that the changes 

 which can be seen at metaphase are of two kinds : the primary breaks and 

 the secondary rearrangements. 



Now the fact that breaks were found to rise about linearly and interchanges 

 to rise as the square of the X-ray dose (under certain conditions) often seems 

 to be taken as proof that each interchange involves two independent breaks. 

 Also, data from experiments where the X-ray dose was separated into two 

 halves by varying times, or where the dose intensity was varied, have been 

 used to estimate the mean time which is supposed to elapse between breakage 

 and reunion. But actually it is clear that this kind of evidence does not 

 itself indicate that two breaks are the two events induced because it cannot 

 by itself give any information about the nature of the changes in the two 

 chromosomes which cause them to exchange. 



However, the orthodox theory, as already pointed out, does assume that 

 ' chromatid breaks and chromatid exchanges represent different ultimate fates 

 of originally the same event, namely a chromatid break' (Catcheside^). 

 Thus the two independent events which are separable in this type of X-ray 

 experiment are identified as chromatid breaks. It does seem to the author 

 that the fact that this is an assumption is sometimes overlooked. Actually 

 there is no direct evidence that the aberrations interpreted as breaks at 

 metaphase were the potential components of the rearrangements, or that they 

 are breaks at all. 



244 



