Meiotic Nuclear Divisions of Galtonia candicans . 743 
Chromosome Formation . 
In Galtonia the evolution of the eight bivalent chromosomes from the 
apparent chaos of the second contraction is very clearly demonstrated. 
Sometimes, as the second contraction loosens, the limits of the eight 
bivalent chromosomes can be identified each in the act of splitting into its 
two univalent chromosomes (Fig. 57). This split separates the univalent 
strands which became united during the preceding stages. It must not be 
confused with the true longitudinal fission in the substance of each univalent 
chromosome which is the homologue of the somatic fission. The real longi¬ 
tudinal fission is to be seen in the loops of Fig. 56, each side of which 
represents a univalent chromosome, the two chromosomes being independent 
of one another except at their extremity. In many cases the second con¬ 
traction mass of chromatin separates out into large oval or rounded blocks, 
often joined end to end like a string of beads (Fig. 58 a), in which the limits 
of the individual chromosomes are at first not discernible. Then a splitting 
of these bivalent homogeneous blocks takes place (Fig. 58 b). Whatever the 
shape and length of the segments, whether they are joined together by fine 
threads, or whether they are more isolated, a fission is to be seen gradually 
splitting them apart into somewhat flattened ribbon-like pieces with ragged 
edges (Fig. 58 B). These at first stain faintly, and are visibly of a different 
constitution to the unsplit segments. Thus by degrees the outline of each 
pair of univalent chromosomes is identified. Very quickly they become 
concentrated, assume an entire outline, and stain deeply and homogeneously. 
So soon does the concentration follow on the split, that the part that has 
already split is noticeably more deeply staining than the portions that are 
in the act of splitting (Figs. 59 A and 59 b). Where the segments are 
joined as a string of beads, they split independently of one another, and 
then the split portions of either side close up end to end (Figs. 57 and 59 B )> 
recalling the origin of the somatic chromosomes. 
All stages of this splitting apart can be seen, but apparently it is 
a phase that is quickly passed through. Whilst the nuclei of one anther 
lobe may show the actual splitting phenomena, the nuclei of the adjacent 
lobe may have fully formed, thick, contracted chromosomes. As the split 
proceeds, the sides divaricate (Figs. 60 and 61), and each represents a 
univalent somatic chromosome. 
Immediately on the separation of the two parts, representing univalent 
chromosomes, the original longitudinal fission may be recognized in the 
substance of each one (Figs. 59 B and 60). This fission was prepared for in 
the presynaptic stages by the condensation of parallel threads which repre¬ 
sented sister sides of a single somatic chromosome (PI. LX, Fig. 35, etc.). 
Though lost to sight during synapsis, the fission which reappeared in the 
lengths of univalent spireme as it emerged (PL LXI, Fig. 45) is believed to 
3 E % 
