126 AN INTRODUCTION TO MODERN GENETICS 



somewhere near the centromere, and that subsequent chiasmata are 

 formed at distances which vary around a certain interval determined by 

 interference. If we plotted a curve giving the frequency of occurrence 

 of a chiasma at different places on the chromosome, we should obtain 

 a series of overlapping normal frequency curves. The peak of a curve 

 would correspond to a place of high chiasma frequency, i.e. of high 

 frequency of crossing- over. Thus two genes separated by a certain 

 length of chromosome would show crossing-over more frequently if 

 they lay in such a region than if they lay in the trough between the 

 peaks, where chiasmata are rarer. The peaks should thus correspond to 

 lengths of the cross-over map in which genes appear unduly separated, 

 while in the troughs we should expect to find them crowded together. 

 A preliminary analysis of Drosophila chromosomes can be made on 

 these lines. 



Chiasma interference and crossing-over interference has been found 

 between different chromosomes in a nucleus. If crossing-over is reduced 

 in one chromosome by the presence of an inversion, there is abnormally 

 frequent crossing-over in the other chromosomes.^ Similarly, there is a 

 negative correlation between chiasma frequencies in different chromo- 

 somes. ^ The mechanism of this is quite obscure. 



5. Other Theories of Crossing-Over and Chiasma Formation 



The alternative (so-called classical) theory^ of chiasma formation 

 explains a chiasma as resulting from the opening out of the bimdle of 

 four pachytene chromatids alternately along the plane separating the 

 two chromosomes (reductional plane) and the plane separating the two 

 daughter chromatids derived from each chromosome (equational plane). 

 Sax's* theory of crossing-over assumes that chiasma formation is of 

 the classical type, and that crossing-over takes place by a breakage and 

 rejoining of the two chromatids which cross one another at a chiasma. 

 This breakage at the same time destroys the chiasma. Two Unes of 

 argument have been advanced against this theory:^ {a) that chiasmata 

 are not actually of the classical type. We have seen above that at least 

 some chiasmata must be of the chiasmatype kind. Darlington also 

 points out that various configurations which would be expected in the 

 classical hypothesis actually do not occur. For instance, different 

 chromosomes sometimes become entangled during zygotene pairing, so 



1 Schultz and Redfield 1932, 1933- 



* Mather and Lamm 1935, Mather 1936a. 



3 Rev, McClung 1927. * Sax 1930, 1931. ^ Darlington 1932a. 



