356 EDMUND B. WILSON 



to lie at right angles to the spindle. I am uncertain whether the 

 latter movement is a simple rotation of the group as a whole, or 

 whether the relative position of the individual chromosomes 

 changes more or less as they spread apart. It is certain, how- 

 ever, that when the metaphase has been attained (figs. 34-37, 

 photo. 4) the chromosomes have the same general grouping as in 

 the final anaphases of the first division or in the interkinesis, save 

 that they are less crowded. As before, the XF-bivalent lies 

 near the center, surrounded by the seven other chromosomes, 

 very often arranged in an irregular ring, though this is somewhat 

 variable. 



It seems probable from the facts just described that in these 

 animals the general grouping of the chromosomes is determined in 

 the prophases of the first spermatocyte-division. Already at 

 this time the X- and F-chromosomes are brought into position 

 for their ensuing conjugation; and their topographical relation 

 to the autosomes remains thenceforward unchanged until their 

 final delivery to the spermatid-nuclei. " In this respect these 

 species agree with such forms as Fitchia or Rocconota among the 

 reduvioids (Payne, '09) and difi'er from the coreids and other 

 forms in which a marked change of grouping occurs after the first 

 .division. I conclude, further, that neither the chromosomes nor 

 the centrioles lose their identity in the period between the first and 

 second divisions, and that a complete relation of continuity exists 

 between the two generations of spermatocytes in this respect. 



In side-views of the second metaphase the XF-bivalent is still 

 almost always distinguishable from the other chromosomes by 

 its deeply constricted dumb-bell shape (figs. 36, 37, 42, 43) ; and 

 in correlation with this, this element is apparently always the 

 first to divide, its two components having often completely sepa- 

 rated before the others have even become deeply constricted 

 (figs. 38 to 41, 44, 46, photo. 6). This precocious division of the 

 XF-bivalent is a very common phenomenon among the Hemip- 

 tera (as I have heretofore described). It is obviously due to the 

 comparatively loose uiiion of X and F after their conjugation, so 

 that they yield more readily to the poleward force (whatever it 

 may be) that operates during the division. 



