734 
Dig by.—-The Somatic , Premeiotic , and 
At this stage there is still a definite concentration of the linin at the 
periphery (Fig. 18). As in the telophases, the prophases of the arche- 
sporium (PL LX, Figs. 23, 24, and 25) show a far more decided massing of 
the linin into solid blocks than do those of the roots (PL LIX, Figs. 9 and 
10). Concentration proceeds, and the parallel portions of linin, whether 
they be rows of granules or paired threads, gradually condense to form 
more homogeneous lengths of spireme (PL LX, Pigs. 26 and 27). The 
space which formerly separated the parallel rows, as the sides approached 
one another, perhaps becomes the longitudinal fission which will eventually 
split apart the sides of the chromosomes (PL LIX, Figs. 8 and 9) as suggested 
by Merriman ( 18 ). These stages in the formation of the somatic chromo¬ 
somes have been beautifully shown by Strasburger ( 32 ) (1907) in the root- 
tips of the Pea. 
It is not easy to form a mental picture of what is actually proceeding 
in the nucleus during chromosome formation. Two opposite processes are 
at work. There is the shrinkage of the chromosome upon itself, and at the 
same time the preparation for the split. 
From now onwards, there is a great increase in the stainable material, 
that is to say in the nuclein substances. 
Gregoire in his paper on the somatic divisions of Allium, &c. ( 9 ) (1906), 
has shown that the chromosomes may be formed in two ways. Either the 
chromosome band may keep its alveolar structure until it becomes a 
definite chromosome, and so retains a certain thickness throughout its 
formation ; or the chromosome band may uncurl into a thin filament which 
forms a zigzagging spireme owing to the uneven concentration of the 
alveolar structure. In Galto 7 iia these two types of chromosome formation 
are well seen. Strands of linin may be found whose granules, instead of 
lying in more or less parallel rows, are arranged in an irregular corkscrew¬ 
like way. As concentration proceeds a curling spireme results (P'ig. 11), 
the segments of which only straighten out during their final thickening 
prior to the evolution of the chromosomes. 
Chromosome Formation. 
No continuous spireme exists. As in the early prophases there is a 
great variety in the character of the linin framework leading to the for¬ 
mation of the chromosomes. In some nuclei the linin may be joined 
into lengths which are more or less polarized, whilst in others the linin 
is much fragmented (PL LX, Fig. 26), the separate portions being short and 
rounded, but of all manner of shapes and sizes. In either case the linin 
fragments, from being flat and ribbon-like, become thickened rods (Fig. 27). 
As their bulk increases, so does also their staining capacity. During this 
process of condensation the fragments gradually retreat from the nuclear 
periphery, and spread themselves across the nuclear cavity (Fig. 27). 
