639 
the Meiotic Divisions of Vicia Faba. 
may be laterally compressed so that one diameter of the equatorial plate 
is much longer than the other. The broad aspect of such a spindle is shown 
in Fig. 28. 
On the spindle the two arms of the V separate, so that the longitudinal 
fission, begun in the last premeiotic telophase, is here completed, and the 
daughter chromosomes pass as rods (Fig. 29) to the poles of the spindle 
(Fig. 30). Cross connexions appear between them, longitudinal fission in 
preparation for the first mitosis in the pollen-grain occurs (Fig. 31), and 
a spireme is produced (Fig. 32), the diamond-shaped mesh of which, as 
already shown (Fraser and Snell, Tl), is quite similar to that of the diploid 
nuclei. 
Conclusions. 
The points of special importance emerging from this series of events 
are the persistence in Vicia Faba of the longitudinal fission from the 
last sporophytic telophase to the metaphase of the homotype division, 
and the fact that, as far as any indication can be obtained, the association 
of the allelomorphs begins after synapsis, taking place in the course of the 
formation of the heterotype spireme, when the cross connexions of the 
reticulum break down. 
The relative unimportance of the position which the chromosomes of 
a pair assume in relation one to another has been recently emphasized by 
Farmer (1912), and attention may be called to the fact that in Vicia Faba 
lateral and end-to-end approximations take place in the same nucleus, and 
become recognizable at the same late stage of development, nor is either 
type related to the appearance of a double thread, either in the presynaptic 
prophases or in the premeiotic divisions. 
The chromosomes of the vegetative division are seen to be double struc- 
tures, not only in the premeiotic telophases, but in the postmeiotic also (Fraser 
and Snell, Ti), so that, unless a different interpretation be placed upon the 
duplication at different stages, it is clear that it cannot be ascribed in the 
former, since it is impossible to ascribe it in the latter, to an approximation of 
paternal and maternal segments. Rather it is due in both alike to longitu- 
dinal fission or vacuolization. As cross connexions appear, and as the 
nuclear area enlarges, the diamond-shaped mesh of the reticulum is formed 
by the pulling apart of the split portions of the chromosomes. In the 
meiotic prophase this mesh can be traced unbroken into full synapsis, and 
may be recognized in places at a considerably later stage. After the 
synaptic phase, as the cross connexions break down, the sides of the split 
strands tend to approximate, and the fission is only made clear here and 
there by fortunate accident. 
Having regard to the origin of the duplication, it is impossible to relate 
it to the formation of the gemini, and the confusion which has so frequently 
