1200 BIOLOGICAL EFFECTS OF RADIATION 



The strong reduction of the frequency of crossing over in the neighborhood 

 of the breakage and attachment points (page 1198) is the consequence. 

 Short sections of chromosomes attached to long nonhomologous chromo- 

 somes are, all other conditions being equal, weak competitors. The 

 pairing of such sections with their homologues is most likely to be delayed 

 or not attained at all, hence the strong reduction of crossing over in these 

 sections, and little or no reduction in the chromosomes to which they are 

 attached (see page 1199). 



The validity of this interpretation of the reduction of crossing over 

 observed in individuals heterozygous for translocations and inversions 

 was tested in a variety of ways. A cytological investigation of the 

 chromosomes of Drosophila in the prophase stages of meiosis has so far 

 not been found possible on account of the extreme technical difficulties 

 presented by this object. Such an investigation, however, was carried 

 through in a plant object, namely, in Zea Mays, by McClintock (65, 66). 

 In maize, in contradistinction to Drosophila, little or no reduction of 

 crossing over is produced by most of the heterozygous chromosome 

 rearrangements. In a full agreement with this stands the fact, dis- 

 covered by McClintock, that in most heterozygous translocations in 

 maize the pairing of all the homologous sections of the chromosomes 

 involved is complete or nearly so. On the other hand, in some trans- 

 locations and inversions in maize some sections of the chromosomes fail 

 to pair with their homologues, and, instead, either remain unpaired, or 

 show a peculiar pairing with nonhomologous sections, which, as far as 

 the genetic consequences are concerned, is equivalent to a lack of pairing. 

 McClintock found furthermore that the failure of normal pairing is most 

 frequently observed in the parts of the chromosomes immediately adjacent 

 to the loci of the breakages and reattachments, that in case a translocation 

 involves an exchange of very unequal sections the short sections fail 

 to pair much more frequently than the long ones, that in inversions 

 involving short sections of chromosomes the inverted part fails to pair, 

 and in those involving very long sections the noninverted part is more 

 likely not to attain normal pairing. It is easy to see how nicely these 

 findings agree with the genetic facts discovered in Drosophila. 



A series of genetic tests of the hypothesis of competitive pairing was 

 devised. This hypothesis requires that if the attraction of one of the 

 competing sections toward its homologue is decreased by some factor, 

 the sections which are the competitors of the former should pair more 

 successfully, and, consequently, the frequency of crossing over in the 

 latter sections should rise, while the frequency of nondisjunction should 

 decrease. Dobzhansky and Sturtevant (133) and Dobzhansky (34) 

 actually found that if crossing over between some sections of the chromo- 

 somes involved in a translocation is prevented by introduction of inver- 

 sions, the frequency of crossing over in other sections increases sharply. 



