54i 
in Pinus and Thuja. 
do not seem to occur so regularly in pairs as described for many other plants, 
but are scattered throughout the cavity. Orientation into the spindle plate 
soon follows (Fig. 28). The segments are generally placed tangential to 
the spindle and do not all occur in the same plane (Fig. 29). It is often 
difficult to determine the two segments which make up the respective 
chromosome. When the chromosomes finally become arranged in the plate 
of the spindle they are very much crowded together, and it is very difficult 
to follow all of the segments. It is quite certain, however, that from their 
manner of segmentation and orientation no such longitudinal fission occurs 
as that described by Miss Ferguson. The chromosomes often reach across 
the equatorial plate, so that it becomes difficult to determine toward which 
pole some of the segments will eventually pass. The fibres are attached to 
the ends in the great majority of cases, but in a few instances they may be 
attached near the middle. In a number of cases observed a few of the 
segments were seen to stand radially to the long axis of the spindle. The 
segments are entirely free from each other and each now constitutes a 
chromosome. At the beginning of metakinesis a much-tangled appearance 
is often presented and some of the segments seem to be standing radially 
to the axis of the spindle. This is due to the fact that some of the segments 
which pass to a given pole lie on the opposite side of the nuclear plate and 
are bent somewhat by the pull which seems to be exerted by the fibres. 
As the segments separate and pass to the poles, all of the forms observed 
in the prophase are met with. The segments seem to 'straighten as they 
move apart. This is always true of the forms which are attached to the 
ends. As the segments pass toward the poles the equatorial ends of 
neighbouring chromosomes diverge, while the polar ends come close together, 
thus forming the characteristic V. This figure is therefore due to two 
different chromosomes, which are so closely applied to each other as to 
appear fused. Some of the long rods which were attached to the spindle 
near the middle form the U shape, but the straight rods forming Vs are 
much the commoner form (Figs. 32,33). After the segments reach the pole 
they shorten and become closely applied to each other and their identity is 
soon lost. 
There is nothing, apart from purely theoretical reasons, in either of these 
genera, to indicate that the segments which occur in the homotype mitosis 
are the same as the ones which were present during the anaphase of the 
first. As already stated above, the chromosomes of the first mitosis lose 
their identity completely before the spirem of the second is formed. The 
loss of identity is as complete as is that which occurs in the resting nucleus 
preceding the heterotype division. What basis exists then for asserting that 
the homotype chromosomes are identical with the granddaughter segments 
of the first mitosis ? This centres again on the question of the individuality 
of the chromosomes. If the chromosomes are not capable of a complete 
