INDIVIDUALITY OF CHROMOSOMES. 



to lie in the region of the equatorial plate and neither pole was 

 able to attract it, and so being left in the region of the new cell 

 wall was constricted mechanically. I do not think that this 

 division is the same as that of normal cleavage of the accessory 

 as it occurs in the second spermatocyte division. 



The following fact in the behavior of the accessory has not 

 been recorded heretofore by any investigator as far as I know. 

 It throws some additional light on its own independence and 

 hence on the individuality of the chromosomes. 



As recorded by most writers on insect spermatogenesis there 

 is no resting stage or no true resting stage between the first and 

 the second spermatocyte divisions. But I have found in Gryllus 

 what I shall call a semiresting stage following Katharine Foot's 

 (i i) terminology. She has described a stage in the Allolobophora 

 egg to which my own findings correspond in many particulars. 

 At the close of the anaphase of the first division the chromosomes 

 are crowded around the pole. The centrosome divides about the 

 time or a little before the chromosomes reach the pole. The 

 two centrosomes begin to move apart with the radiating fibers of 

 the asters around them individually and the spindle fibers con- 

 necting them (Figs. 23 and 24). As the spindle is elongating 

 the chromosomes become somewhat vesicular (Figs. 25 and 26) 

 and have a nuclear wall formed around them, entirely around or 

 only part way around (Figs. 26 and 27). The diffusion of the 

 chromatin and the formation of the nuclear wall seems to go 

 farther in some cells than in others ; Fig. 27 shows as much dif- 

 fusion as any observed. The whole semiresting stage must be 

 very brief for a single cyst may show cells as far apart in devel- 

 opment as shown in Figs. 24 and 30. As the chromosomes 

 enter the equatorial plate of the second division they are usually 

 so crowded that it is impossible to count them or distinguish the 

 accessory from the others. 



After this brief description of the semiresting stage let us return 

 to the accessory chromosome. 



In the semiresting stage the accessory does not enter into the 

 nucleus but forms its own wall around itself. It becomes vesic- 

 ular, the chromatin becoming granular and showing vacuoles 

 (Figs. 24-27). Its position with relation to the nucleus varies. 



