638 Davis.—Cytological Studies on Oenothera. II. 
of the chromosomes at the equatorial plate was very symmetrical because 
they remained joined in pairs in the form of seven rings throughout the pro¬ 
phases of the heterotypic mitosis (Davis ’ 09 , Figs. 31-36). The separated 
and scattered arrangement of the chromosomes in biennis allows far greater 
opportunity for numerical irregularities in their distribution during the 
heterotypic mitosis than is possible when they are closely associated in 
pairs. 
As the two sets of chromosomes (seven in each set) move away from 
one another towards the poles of the spindle the V-shaped form is especially 
evident (Fig. 23), but during anaphase of the mitosis their structure becomes 
complicated by a lengthwise fission of each chromosome in the plane of the 
page upon which the above letter (V) is printed. The seven chromosomes 
which leave the equatorial plate for each pole of the heterotypic spindle 
arrive as seven split chromosomes (Fig. 24), this division being a premature 
fission of each chromosome in preparation for the second, or homotypic, 
mitosis. The group of seven split chromosomes is best observed in polar 
views of late anaphase (Fig. 25), and at the time of the reconstruction of the 
daughter nuclei (Fig. 26). 
During the examination of a large amount of material in the stages 
described above (Figs. 25 and 26) two cases were observed that showed 
numerical irregularities in the distribution of the chromosomes by the 
heterotypic mitosis. In both examples eight split chromosomes were found 
at the poles of the heterotypic spindle where the normal number is seven. 
Fig. 27 illustrates this exceptional condition, and in this instance five 
chromosomes were present at the opposite pole of the spindle ; six would be 
the number expected, and it is quite probable that a chromosome had been 
carried away from the group by the microtome knife. The rarity of these 
exceptions would make it very difficult to determine whether or not such 
irregularities might continue throughout the homotypic mitosis and give 
rise to functional pollen grains with smaller and larger chromosome content 
than the normal. 
The history of the seven split chromosomes is readily followed through 
the interkinesis between the heterotypic and homotypic mitoses. After 
the daughter nuclei are organized the halves of each chromosome separate 
somewhat (Fig. 28), and the ends swing apart until the halves lie in approxi¬ 
mately the same plane and resemble somewhat two U’s joined together at 
the middle region (Figs. 29 and 30). It then becomes evident that there 
are seven pairs of chromosomes in each nucleus, a condition which is, as 
a rule, most clearly shown in polar views of the structure. The chromo¬ 
somes increase markedly in size, and remain in this expanded condition 
throughout the period of interkinesis. The free ends of the chromosomes 
sometimes exhibit a tendency to branch and become united to form a loose 
and imperfect network, but generally most of the chromosome pairs of 
