550 CELL MECHANICS 



chromosomes begin to repel one another. Secondly, it reduces 

 the longitudinal cohesion, for the new half-threads at the moment 

 of their origin must needs be weaker than the parent threads. 

 Now the first change will reduce the torsion of the threads by 

 uncoiling, but as observation has shown, uncoiling is very slow 

 (D., 1935 c). It lags behind the operation of the forces which 

 determine it. The second change on the other hand is immediate. 

 The weakened chromatids are therefore exposed to an unreduced 

 torsion and one of the four breaks under the strain. 



The break of one chromatid causes another upset of equilibrium 

 which is slight but sudden. The two broken ends will twist round 

 their unbroken sister chromatids, thus releasing the coiling of the 

 two which determined the breakage. But the relational coiling 

 of the chromosomes has been in equilibrium with that of the 

 chromatids. The release of that of one pair of chromatids therefore 

 imposes an extra strain on the unbroken pair. Since the first pair 

 yielded under the original strain the second pair is bound to yield 

 under the increased strain. One of its chromatids will break at a 

 point exactly opposite to the first break, since the strain will be 

 greatest at this point. 



The breakage of the chromatids will permit the uncoiling of 

 their relational coiling in both chromosomes. The broken chroma- 

 tids will reunite when, in the course of uncoiling, one of their ends 

 first meets another. This will always be the end of a chromatid 

 of a partner chromosome. Crossing-over will have occurred, and 

 when the lapse of attractions leads to separation, a chiasma will 

 appear. 



On this view crossing-over depends on the action of forces that 

 may be deduced from the behaviour of chromosomes at earlier 

 and later stages of their history. The reason for its occurrence at 

 meiosis is the division of the chromosomes at a time when they 

 are paired and coiled in such a way that when their equilibrium is 

 upset exceptional strains are imposed that can arise in no other way. 



The cross-over chromatids cannot be distinguished from the 

 non-cross-overs at a late diplotene stage because the chiasma 

 has by this time become symmetrical. But immediately after 

 the chiasma appears, the four chromatids can often be seen still 



