I 



RADIATION AND THE STUDY OF MUTATION IN ANIMALS 1247 



that at such stages the majority of the cells in the gonads, particularly 

 of the female, are immature and subsequently will divide. It may be, as 

 Harris (72) has indicated, that these divisions all concern a few apical 

 cells. Unless this can be demonstrated, however, Moore's adaptation of 

 Mailer's original assumption that fractionals are to be accounted for as 

 the result of a double chromosome cannot be accepted without emenda- 

 tion. If an apical-cell mechanism should be excluded, the choice is 

 between the assumption of an aftereffect or the assumption that the 

 chromosome is double, but segregation of sister chromonemata is random 

 at mitosis, which would account for Moore's results. It may be noted, 

 that on the latter assumption, complete mutations are really "double 

 mutations," two genes mutating at once — an apparent contradiction 

 to the evidence that only one of an allelomorphic pair mutates. This 

 has been considered by Moore and an explanation attempted on the basis 

 that the matrix within which the chromonemata are enclosed makes the 

 chromosome a unit. This, however, is not quite satisfactory, since, 

 according to Nebel (123), each chromonema has its own matrix. The 

 hypothesis that sister chromonemata segregate at random may be tested 

 as Sturtevant points out, by the data from somatic crossing over (Stern, 

 168). Until this is done, it is perhaps useless to speculate further in this 

 direction. 



Muller (113) recently has concluded that the induction of mutation 

 by X-rays may be an indirect effect. His reasoning is as follows: chromo- 

 some breakages tend to be associated, since most translocations are 

 mutual; this association precludes the possibility of a direct effect of 

 X-rays on chromosome breakage; since, however, there is a linear relation 

 between frequency of translocation and dosage, for translocation as for 

 mutation, if in the one case the effect is indirect, it may be so in the other 

 as well. The argument is far from convincing, although the conclusion 

 may be correct. One may assume a direct effect on attachment, with 

 the resultant double breakage a linear function of attachment; this would 

 give the apparent linear relation observed, and would depend on a direct 

 effect on the chromosome. But as has been pointed out, inferences 

 from the shape of the dosage-effect curve give no satisfying conclusion 

 as regards the mechanism of radiation effects. It is possible that the 

 further analysis of group effects will be profitable in this connection. 

 One may guess that both these and the translocat ion-group effects have 

 their origin in chromosome movements during the growth of the male 

 pronucleus. This is a time which has been untouched as yet by experi- 

 ment, which is in Drosophila not easy to manage in this regard. Yet, 

 so many things happen at this time that the lack of experiment is some- 

 what surprising, in view of the interest in the problem of direct effect. 



It is, of course, a question whether any changes at all occur in the 

 mature sperm, no matter how long it is aged. Timofeeff-Ressovsky 



