208 
Allen . — Nuclear Division in the 
segment A comparison of Figs. 28 and 29 with Fig. 2 6 shows that for 
the most part the thickness is little greater at the time of segmentation 
than at that of longitudinal fission, although some portions of the thread 
have already thickened considerably. Fig. 33 shows a marked difference 
in thickness between different portions of the same chromosome. 
The chromosomes, formed as above described, now enter upon a period 
of shortening and thickening, two stages in which are shown in Figs. 30 
and 32. Quite early in this period, usually (Fig. 30), the chromosomes 
take up a position close to the nuclear membrane. This peripheral arrange- 
ment, however, may not be effected until a considerable thickening has 
occurred ; in such cases the massing at the centre persists, sometimes 
nearly to the stage of Fig. 32, with relatively short, thick chromosomes 
radiating toward the periphery. The various chromosomes do not shorten 
and thicken at the same rate; and it is common to find in the same 
nucleus long, slender chromosomes, and other much thicker ones, not more 
than half as long. Chromosomes a and b, Fig. 32, illustrate this difference, 
which is often much greater than in this instance. 
Figs. 33-54 show individual chromosomes at various stages, from 
segmentation to a time immediately before their arrangement on the 
equatorial plate. Such a series shows that the original longitudinal split, 
which is plainly apparent at the time of segmentation (Figs. 28, 29, 33), 
remains distinct until the chromosomes have reached their mature form. 
The chromosomes are always plainly double, and their double nature does 
not follow, as the figures show, from a folding over of a single chromosome 
and the approximation of its ends. The plane of separation between the 
daughter chromosomes on the equatorial plate is that of the original 
longitudinal split. The notion of the origin of the double appearance by 
a folding has found support in the shape of many of the chromosomes 
immediately after their formation, when, owing to the fact that their length 
is greater than the diameter of the nucleus, they are necessarily more or 
less curved, bent, or even looped. The chromosome shown at aa, Fig. 29, 
is bent into a loop at a point not very far from its centre ; and the 
chromosome at the same stage, shown in Fig. 33, is curved into a rather 
shallow U, the curve following in this case pretty closely the concavity 
of the nuclear membrane. As the chromosomes shorten they usually 
straighten out more or less (Figs. 34, 36, 37, 38) ; but in some cases 
(Figs. 40, 44, 53, 54) one or both daughter chromosomes remain curved 
down to a very late stage. 
Another point upon which stress has been laid as evidencing the 
formation of a double chromosome by the bending of a single one is the 
fact that the two parts are commonly continuous or fused at one end.. It 
is true that the daughter chromosomes often lie closely in contact at one 
or both ends ; and sometimes it is difficult to make out a line of demarca- 
