KARYOKINESIS AND ITS RELATION TO FERTILIZATION. 175 
Rabl’s researches further show that the spindle figure 
changes its position during division, and, what is especially 
remarkable, that the loops of the chromatic figure follow this 
change of position. It has been known since the discovery of 
karyokinetic cell-division that the threads of the daughter- 
nuclei group themselves round the poles of the spindle. 
Flemming and E. van Beneden (Jl. cc.) have also taught that 
the daughter-loops move along the threads of the nuclear 
spindle in order to reach the poles. Rabl has shown that 
from the first appearance of the spindle figure it exerts an 
“ arranging ” influence, as it were, on the chromatic threads ; 
and this takes place during the phase of the ‘‘ segmented skein.” 
When the spindle figure first becomes visible it lies (in 
the Salamander), according to Rabl, in the neighbourhood of 
the polar area, and is so placed that its equator lies in the 
polar area, whilst its long axis is directed obliquely to the 
long axis of the nucleus. Later on it sinks deeper into the 
nuclear substance, and takes up a position such that its equator 
falls in that plane along which the division of the nucleus will 
occur (division-plane). The long axis of the spindle figure 
thus coincides with the “ division-axis”’ of the nucleus. We 
may here add that this division-axis of the nucleus is not 
always the same as its long axis or as the long axis of the 
cell; e.g. a cylindrical cell may divide, not only transversely, 
but also longitudinally, as, amongst others, Arthur Kollmann 
in his beautiful work on the tactile apparatus of the hand 
(110) has shown to be the case with the deep-lying cells of the 
rete Malpighii, and as Rabl describes for the Salamander. 
Oblique division, too, may occur. In many cases things go on 
in such a way that, according to Rabl and Strasburger, the 
axis of division passes through the polar and antipolar areas 
of the mother-nucleus ; in later stages therefore the polar area 
of one daughter-nucleus will coincide with the original polar 
area of the mother-nucleus, and that of the other daughter- 
nucleus with the original antipole (see fig. 12). This is not 
always the case, however, for Strasburger finds that in plants 
the division-axis more often lies parallel to the polar area. 
VOL. XXX, PART 2.—NEW SER. M 
