CONSEQUENCES OF CROSSING-OVER 255 



Fig. 82. — Diagram to show the distinction between different genetical 

 interpretations of the same diplotene bivalent. This bivalent has four 

 chiasmata which have different structural relationships with one 

 another. Thus the second does not restore the association broken by the 

 first ; they are disparate ; the second and third, the third and fourth, on 

 the other hand, are comparate {v. Fig. 36, B and C). At anaphase there 

 will be interlocking of chromatids in the loop between the second and 

 third chiasmata {v. Fig. 36, C). A and B, alternative interpretations on 

 Janssens' hypothesis of " partial chiasmatypy " as simplified by Belling, 

 1928 c, and Darlington, 1930 b. Disparate chiasmata (i and 2) arise from 

 crossing-over of one chromatid at both chiasmata and two different 

 chromatids once at each chiasma. Comparate chiasmata arise either 

 from both affected chromatids, being the same (2 and 3 in A, and 

 3 and 4 in B, " reciprocal crossing-over "), or from both being different 

 (3 and 4 in A, 2 and 3 in B, " complementary cross-overs "). 



cross-overs. These various kinds of evidence will now be con- 

 sidered from a more general point of view. 



(ii) Prophase Interlocking. With few exceptions the chromosomes 

 are fortuitously distributed in the nucleus before they pair at the 

 prophase of meiosis, with regard to their homology. In the Diptera 

 {cf. Metz and Nonidez, 1924) the homologous partners probably lie 

 next to one another at the telophase of the preceding division owing 



