74 M. M. RHOADES 



anaphase I configurations. Chromosomal fibers may arise from one or both 

 of the long arms of each dyad at late metaphase or early anaphase I. Al- 

 though it was not always possible to differentiate between long and short 

 arms, the neo-centric regions in general appear to be confined to the long 

 arm. When both long arms of the two chromatids of a dyad possessed a neo- 

 centric region, the chromosomal fibers arising from these centric regions were 

 usually directed toward the same pole. Occasionally they were oriented to 

 opposite poles thus causing a great attenuation. In such cases, however, 

 those chromosomal fibers nearest to one pole were powerful enough to over- 

 come the oppositely directed force of the second neo-centromere. Despite the 

 great complexity of configurations at anaphase I resulting from interacting 

 and conflicting half-spindle fibers arising from both the true and neo-centric 

 regions, the end of anaphase I usually finds ten dyads at each pole. Some- 

 times, however, greatly stretched chromosomes undergo breakage. This 

 breakage doubtless accounts for the higher pollen abortion (about 10 per 

 cent) found in homozygous abnormal 10 plants as contrasted to the lower 

 (0-5 per cent) pollen abortion of normal sibs. 



Even though one or two arms of some dyads are markedly stretched at 

 anaphase I, the ensuing telophase is normal. All four arms of each dyad con- 

 tract to form a spherical mass of chromatin which is loosely enveloped by 

 the lightly-staining matrical substance. The chromonemata uncoil during 

 interphase and early prophase II finds each daughter cell with ten, long X- 

 shaped dyads of typical appearance. The two chromatids comprising each 

 dyad are conjoined by the undivided primary centric region. There is no indi- 

 cation of neo-centric regions, although some of the long arms possessed chro- 

 mosomal fibers at the preceding anaphase. 



The onset of metaphase II sometimes occurs before the dyads have under- 

 gone their usual contraction. Occasionally chromosomal fibers arising from 

 neo-centric regions in the long arms are found at late prophase II. These 

 precociously acting fibers produce an extension of the long arms before any 

 spindle is visible. This observation is of singular importance. Some authori- 

 ties believe that the centromere region is attracted (whatever this term may 

 signify) to the spindle pole. Here we have a movement produced by the 

 chromosomal fibers of neo-centric regions in the absence of an organized 

 spindle. The way in which these neo-centric fibers act can only be conjec- 

 tured, but no interaction between centric regions and spindle pole is essential. 

 It is, indeed, probable that the only role of a bipolar spindle is to provide a 

 structural frame which channels the chromosomes to the spindle poles. 

 Clark's (1940) studies on divergent spindles are pertinent in this respect. 



The objection may be raised that the chromosomal fibers of neo-centric 

 regions are not comparable to those arising from the true centric region. I 

 do not believe this is a valid criticism. Not only are both kinds of fibers con- 

 cerned with chromosome movement, but, as will be shown in a later section, 



