32 Over ton. —On the Organization of the Nuclei in the 
constant in number and bear no apparent relation to the number and 
arrangement of the chromosomes. The prochromosomes may be placed 
anywhere in the loops and are not at definite fixed points. The loops 
become the uniformly distributed spirem, which has much the same 
appearance as that of Thalictrum purpurascens , except that the bivalent 
chromosomes are smaller in proportion to the size of the nucleus. There is 
no continuous chromatic spirem present (Figs. 7 and 8, PI. II). It is a 
bivalent structure consisting of bivalent chromosomes and bivalent linin 
intervals. Segmentation of the spirem and segregation of the bivalent 
chromosomes take place in exactly the same way as has been described 
above for Thalictrum purpurascens , and need no further description here. 
Fig. 10, PI. II, represents a section of the nucleus in late diakinesis, in 
which the bivalent chromosomes are scattered freely in the nuclear cavity. 
The figures from this stage onward I was unable to obtain in my former 
study, and I have, therefore, drawn a complete series in order to trace the 
chromosomes throughout the various divisions. The forms of the diakinetic 
chromosomes are shown in Figs, na-iu, PI. II. The parallel double 
rods are the usual forms, but variations may occur, which are, however, 
exceptional. Fig. 11 a shows the two parts of a bivalent chromosome 
twisted closely about each other. In Fig. 11 b they form a figure 8, with 
the two portions united at the ends. In Fig. 11 c the two components are 
united at the ends and are wide apart in the middle, forming an open ring. 
The form shown in Fig. 11 d is that of the chromosomes in which the two 
univalent portions are united at one end to each other. Nearly all the 
chromosomes in Fig. 10, PL II, are of this type. Fig. lie shows the 
characteristic double rods, which are somewhat separated in the middle 
but loosely attached at the ends. It is interesting to observe that in this 
plant it is the chromatic aggregations of the bivalent spirem which form the 
same sort of figures as the entire segments of the bivalent spirems produce 
in forms like the lily. I have constantly observed at least one bivalent 
chromosome, which is apparently much longer than the others, and am of 
the opinion that it is this chromosome which forms the 8- and O-shaped 
figures described above, while the shorter chromosomes assume the shape 
of double rods with the portions either parallel or separated at one end. 
In Figs. 11 c and 11 d the longitudinal splitting of each univalent 
chromosome, which has been regarded as a second longitudinal fission, 
is visible, thus forming four-parted structures. That this split occurs by 
a longitudinal cleavage in each univalent component of the bivalent 
chromosomes, these figures show beyond a doubt. Since each univalent 
chromosome of the diakinetic pairs has been identified from the earlier 
stages, and since the line of separation of the univalent portions of each pair 
has been found to be the line of approximation, the appearance of a fission 
in each component of the pairs must certainly be a new splitting, and is not 
