Caroline Pellew and Eva Richardson Sansome 43 



interstitial chiasma on each side of the attachment constriction, and one 

 subterminal and one terminal chiasma (Fig. 2 b, second bivalent from 

 the left). 



Similar anaphase configurations have been described in Stenoboihrus 

 parallelus (Janssens, 1924) and in Fritillaria meleagris (Darhngton, 1929 a), 

 while Meurman (1929) shows chromatids separating from multiple 

 chiasmata in Aucuha. There is every reason to suppose that wherever 

 interstitial chiasmata exist at metaphase, the chromatids will pull apart 

 in the ways described. 



Janssens gives diagrams to show the probable arrangements of the 

 chromatids at metaphase to give the anaphase configurations observed 

 (Janssens, 1924, Planche schematique IV). In these diagrams he evidently 

 assumes that identical chromatids, that is, chromatids from the same 

 somatic chromosome, are associated, and that where a chiasma or change 

 of partners occurs, there has been a break leading to genetical "crossing- 

 over." Most of the diagrams referred to illustrate "partial chiasmatypy," 

 in which only two chromatids have been interchanged at one particular 

 point, but Janssens also assumed "total chiasmatypy" in which crossing- 

 over occurs between all four chromatids at once. 



Since Janssens' chiasma type affords a plausible interpretation of 

 the cytological features observed in this particular instance, and since 

 it is so important in accounting for genetical crossing-over, it seems 

 advisable to give a brief account of the theory as modified by Belling 

 and Darlington. 



These authors reject total chiasmatypy as being very rare, if it occurs 

 at all. Both assume that breakage occurs at the four-strand stage, and 

 that chiasmata are formed as a result of such breaks — the chiasma being 

 the point of breakage, the loops between the chiasmata being formed by 

 the tendency of identical chromatids to fall apart in pairs. Belling calls 

 such an interchange between chromatid parts "segmental interchange," 

 whereas DarUngton calls it " crossing over." Configurations in triploids and 

 tetraploids have been described, and Darlington (1930 b) shows that they 

 afford strong cytological evidence of "crossing-over" between the stages 

 of zygotene and diplotene. BelHng (1928) finds difiiculty in accounting 

 for the disappearance of chiasmata between diplotene and late diakinesis, 

 a phenomenon which is of widespread occurrence. Darlington explains 

 such a disappearance on the basis of terminalisation, by which chiasmata 

 move away from the attachment constriction towards the ends of the 

 chromosomes where they merge (Darlington, 1929 b). 



In those cases where terminalisation is complete it is presumed that 



