488 CELL MECHANICS 



Change of direction might be expected on the other hand following 

 crossing-over between chromosomes with opposite coiling directions 

 if the direction were a specific property of the arms of the parent 

 chromosomes and were unaffected by crossing-over. Interpreta- 

 tions of three homologous chromatids having the same direction, 

 while the fourth at a corresponding point has the opposite direction, 

 contradict this conclusion but are perhaps mistaken. The constant 

 direction in arms of Rhceo and Tradescantia with terminal chiasmata 

 would require that chiasmata should have been formed originally 







ABC D E F 



Fig. 137. — Major spirals at first metaphase in Tradescantia revealed 

 by pre-treatment with hot water {cf. La Cour, 1935). A and B, 

 normal and mutant type of spiralisation in diploid species ; 

 C, D, E, F, tetraploid species showing different degrees of 

 chromatid association and contraction ; C, ring of four ; 

 E, change of direction at the centromere. X 2,000. 



very close to the end and that the amount of movement should not 

 have been proportionate to that found in the terminalisation of 

 chiasmata in small chromosomes. Such an abnormality of behavi- 

 our, a terminal localisation, has already been suggested on other 

 grounds (Ch. IV). 



In an organism with a considerable movement of chiasmata, a 

 change of direction at the point of crossing-over will mean that 

 chromatids with opposite directions of coiling will come to be 

 paired. Such an organism cannot therefore be supposed to have a 

 joint coiling of chromatids as in Tradescantia. It must have separate 

 coiling as in Fritillaria. 



The change of direction which results from crossing-over between 



