539 
in Pinus and Thuja. 
separated in this mitosis, but that the original longitudinal fission again 
becomes evident and the daughter segments separate along this line of 
fission, thus one half of each somatic passes to each daughter nucleus and 
this mitosis effects an equational division. After a careful study of such 
stages, as shown in Figs. 21, 22, 23, the opposite view has been adopted, 
namely, that the segments separate in such a way that one of the halves of 
the bivalent passes entire to each daughter nucleus, thus bringing about 
a reducing division. Fig. 21 shows the beginning of metakanesis ; the 
segments shown are not the longitudinal halves of the daughter segments, 
for, as will be shown in later stages, longitudinal fission has not yet taken 
place. The chromosome in the middle was a ring-shaped one, while those 
to the right and left were composed of two straight pieces which may have 
been oriented in any of the ways mentioned above. The ends are drawn 
together as they converge toward the poles and in many cases they are so 
close together that they may appear fused. This is, however, more appa- 
rent than real. The segments at this time are quite thick, but appear 
slightly more drawn out than when in the spindle plate before metakanesis 
begins. 
After the segments entirely separate from each other and move toward 
the poles, each is seen to split longitudinally. The appearance of this fission 
is often very much delayed, but it takes place quite generally, as shown in 
Fig. 22. It may be readily seen here that the segments begin their fission 
at the equatorial end and separate along the entire length. The equa- 
torial ends diverge somewhat and the characteristic V, U, or double V is 
formed. The second chromosome from the left represents a ring-shaped 
form in which longitudinal fission has taken place, giving rise to the double 
U, while that lying next to it may be due to a long rod-shaped one which 
has bent on itself at the place to which the spindle fibres were attached. 
Sometimes it happens that one of the granddaughter segments slips past its 
mate, the chromosome at the left, but they more than often remain together. 
Frequently their polar ends seem to be fused, but this does not appear to 
be the rule. In Fig. 23, the granddaughter chromosomes are shown just 
as they arrive at the poles. Longitudinal fission is now complete. The 
segments have not yet contracted and show the same long-drawn-out 
appearance which is typical of this stage of mitosis in Pinus. It is evident 
here that longitudinal fission has been complete, although the ends of some 
of the segments are very closely applied to each other and often seem fused. 
The granddaughter segments now approach very close to each other, and 
it becomes difficult to trace the individual members. A more or less com- 
plete spirem is formed, and about the same time the nuclear membrane is 
laid down. This spirem which has been formed by end-to-end fusion of the 
segments soon gives signs of dissolution, and, with the growth of the daughter 
nucleus to its full size, the spirem has entirely lost its identity and a com- 
