1931 



The secondary association of chromosomes 



357 



the components on either side of the equator of the spindle. When the 

 association is close it is impossible to say from a polar view whether 

 there is primary or secondary association. The criterion of secondary 

 association is therefore comparison with diakinesis. 



Analysis of complete side views is extremely difficult and can only 

 be made on the beginning of anaphase before the chiasmata have been 

 resolved by separation. Fig. 1 d shows 16 of the 18 bivalents from a 

 first metaphase. The bivalents have been set out in a line in the draw- 

 ing, care being taken not to separate those which were in immediate 

 proximity to each other. Eight bivalents are seen to be associated in 

 pairs and the two bivalents comprising a pair are similar (i) in size and 

 (ii) in configuration, (i.e. in having one or two terminal chiasmata). 

 Three of these pairs lie in a plane at right angles to the line of vision, 

 and in no case is there evidence of material connections. 



The bivalents marked ( x ) in Figs. Id, e have two terminal chias- 

 mata and are the last to divide at anaphase. The remaining bivalents 

 apparently have single terminal chiasmata. 



Polar views show a marked secondary association of the bivalents 

 into groups of two or three. Table 1 shows the number and kinds of 

 association in 12 metaphase plates. Groups of 3 bivalents are common- 

 ly found. The groups of 2 bivalents are tabulated in two columns ac- 

 cording to the degree of association. Those closely associated are 



Table 1 



Association of bivalents at MI in Dahlia Merckii. 



25* 



