Oenothera Lamarckiana and O. gigas. 955 
such a close association as would result from the contraction of a complexly 
looped thread or system of threads anastomosing to form a reticulum. 
Following the period of mid-synapsis, when the contracted condition is 
most conspicuous, there begins a loosening of the synaptic knot, and it 
becomes at once evident that the threads have thickened and that the 
thread system has grown shorter (Fig. 55). Thus there emerges from 
synapsis a much more clearly defined system of loops than appeared in the 
involved tangle of threads that entered the synaptic knot (compare Fig.-55 
with Fig. 53). The thickened thread system then begins to assume the 
appearance of the spireme, which shortly after may be clearly recognized 
(Fig. 56). The entire history of the formation of the spireme from the stage 
of the nuclear reticulum preceding synapsis seems to be a process of 
contraction and thickening of the threads in the reticulum, the phenomena 
of synapsis being an expression of this process. 
The Formation of the Chromosomes. With the loosening of the 
synaptic knot following synapsis, as described above, it becomes at once 
apparent that the threads have grown markedly thicker and the thread 
system much shorter than at the beginning of the synaptic contraction 
(compare Fig. 55 with Figs. 51 and 53). Nevertheless, the thickened 
threads still present a very complicated coiled arrangement which, as in 
Lamarckiajia, cannot be followed throughout the nucleus. The structure 
is, however, as far as could be determined, that of a single thread or a group 
of threads intricately looped and occasionally anastomosing. The spireme 
(Figs. 56 and 57) is developed by the continued shortening or condensation 
of the thickened thread system until the band may be distinctly traced s 
throughout the greater number of its convolutions, although it is very 
difficult to determine whether the spireme is strictly continuous or composed 
of two or more portions. 
There then begins the process of segmentation (Fig. 58) that transforms 
the spireme into a chain of twenty-eight chromosomes (Fig. 59), which 
because of their large number can be counted only in very favourable 
preparations. The process is the same as in other Oenotheras, and consists 
in the constriction of the spireme and condensation of the chromatic material 
at twenty-eight centres. The result of the condensation is a further 
shortening of the spireme, so that much of the complicated looped arrange¬ 
ment becomes simplified, and the segments may be followed for long 
distances (Fig. 59), arranged end to end like links in a chain. The chain 
of chromosomes is frequently broken, perhaps because the spireme is made 
up of somewhat distinct portions, or because certain segments become 
displaced or separated, as might readily take place at points in the loops. 
The stage of second contraction is frequently very conspicuous in gigas, 
for the reason that so large a number of chromosomes is concerned with the 
phenomenon. It may appear as early as the segmentation of the spireme 
