Pollen Mother-cells of Lilinm canadense . 2 2 1 
been attached to them ; these fibres seem to be used in the formation 
of the spindle. Each daughter chromosome has now only a single bundle 
of fibres attached, in the majority of cases, at, or close to, its end. 
The Equatorial Plate . 
Fig. 1 01 shows the equatorial plate in lateral view ; in Figs. 104-106 
are represented individual chromosomes similarly viewed. The majority of 
the chromosomes are attached to the spindle at, or very close to, the end at 
which the daughter chromosomes are in contact with each other. To each 
daughter chromosome a bundle of fibres is attached, running toward the 
corresponding spindle-pole. In general the chromosomes lie in a direction 
which approximates more or less closely that of the long axis of the spindle. 
The two daughter chromosomes may be nearly parallel with each other, 
pointing toward the same pole (Fig. 104) ; or they may diverge toward 
opposite poles (Fig. 106). Less commonly, the chromosomes stand out 
nearly or quite at right angles to the axis of the spindle (Fig. 105). 
The arrangement of the chromosomes in the equatorial plate, as well 
as their number, are best determined in a polar view (Figs. 102, 103). 
Usually the chromosomes are attached at the periphery of the spindle ; it 
is not common to find any of them in the interior of the equatorial plate, as 
is regularly the case in the first division, although this occasionally happens, 
as appears in the lower of the two cells represented in Fig. 103. The 
general appearance of the chromosomes in polar view is that of V’s, usually 
considerably foreshortened on account of the orientation just described. 
The inner end of one arm of a V is directly superposed in such a view over 
the inner end of the other arm, as follows from the manner of attachment 
to the spindle shown in Figs. 104-106. The chromosome as a whole, then, 
lies in a plane nearly perpendicular to the equatorial plane. 
The plate shown in Fig. 102 contains fourteen V’s. instead of twelve, 
which, as has been said, is the number of chromosomes that, so far as I can 
determine, constantly occurs. The discrepancy in this case seems to be 
accounted for by the fact that in two instances (at a and b ) two V’s are in 
close contact with each other. Furthermore, as appears especially well 
at a, each V lies in the plane of the equatorial plate, instead of nearly 
at right angles to that plane as is ordinarily the case. The two V’s at a } 
Fig. 102, are, therefore, in all probability, to be thought of as corresponding 
respectively to the two arms of the V-shaped chromosome at c (same 
figure). One V-shaped daughter chromosome at a is superposed over 
its sister segment in exactly the same way that one of the rod-shaped 
daughter chromosomes at c is superposed over its sister. In other words, 
the varying methods of attachment of the chromosomes to the spindle 
fibres in the heterotypic division reappear in the homoeotypic division ; and, 
R 2 
