222 
Allen.— Nuclear Division in the 
just as was the case with the granddaughter chromosomes of the previous 
division, so with the same segments (now daughter chromosomes) in the 
present division — the majority are attached to the spindle at or close 
to their ends, resulting in giving them the shape of a straight or hooked 
rod, while a few (usually not more than one or two in a cell) are attached 
at some point between the two ends, resulting in the shape of a V, whose 
arms may or may not be of equal length. The same fact is very well 
illustrated in Fig. 103, which represents a polar view of the equatorial 
plates of two sister-cells. The chromosomes a and b in the upper cell, 
a' and h' in the lower, consist each of two V-shaped daughter segments, 
one superposed upon the other, while each of the other chromosomes 
in either cell consists of two rod-shaped segments similarly superposed one 
upon the other. From the fact that in each sister-cell appears the same 
number of chromosomes of each form, and the further fact that chromo- 
somes a and b in the upper cell correspond very closely in position with 
chromosomes d and b' in the lower, it seems quite certain that a and a', 
b and b' are respectively sister chromosomes of the previous division, 
comparable with those shown in Fig. 85 ; and the conclusion is strongly 
suggested that the method of attachment of each individual chromosome 
persists from one division to the next. All the evidence furnished by the 
shape and point of attachment of the chromosomes of the homoeotypic 
division, and by the general numerical proportion of the different forms 
at various stages, confirms this notion of a persistence in the point of 
attachment. Such a persistence is hardly conceivable, it seems to me, 
without supposing also that some of the kinoplasmic fibres have a con- 
tinuous existence ; and so the facts just described afford some confirmation 
for the doctrine of fibrillar persistence. 
The Metaphases and Anaphases. 
The separation of the more numerous rod-shaped daughter chromo- 
somes begins, as a result of the method of their attachment to the spindle, 
at the ends which have been so long in contact. As these ends are pulled 
apart (Fig. 107), the unattached, originally more or less divergent, ends are 
frequently brought into contact, and the final separation of the daughter 
chromosomes is then at the junction of these originally free ends in the 
equatorial plane. The ends which are now turned toward the poles are 
hooked (as was the case in the previous division), due to the lagging 
behind of some of the substance of the attached end of the chromosome, or 
to a bending at the point of attachment if this is not quite at the end, so that 
the daughter chromosome has the shape of a J. Some are not hooked, 
but appear as straight rods. 
The same variations in shape of the daughter chromosomes which 
