206 Nichols, A morphological study of Juniperus communis var. depressa. 
pairing of the maternal and paternal elements of tlie chromatin, 
followed dnring synapsis by their apparent fusion. Miss Ferguson 
(1904) and Lewis (1908) are unable to find such procedure in the 
microspore mother cells of Pinus and Thuja bnt. maintain that 
there the reticulum of the nuclens ‘contracts directly into the 
synaptic condition. Lewis finds occasional paired threads but 
attaches no special significance to them. The writer’s observations, 
however, have led to the conclnsion that in the species under dis- 
cussion a presynaptic pairing does take place. 
The nnclei remain in s} T napsis for fiye or six days. Emerging 
from it they undergo the heterotjqnc and homotypic diyisions, and 
within a week the tetrads haye formed and haye giyen rise to the 
microspores. The ripening of the pollen takes about two weeks, 
and pollination occnrs abont May 25 th . 
Toward the close of synapsis the tightly interwoyen threads 
begin to loosen up, and the mass of chromatic material gradually 
resolyes itself into a slender, somewhat roughened spirem of uni¬ 
form thickness, which twists and coils throughout the nuclear cavity 
(figs. 19, 20). The nucleus at this period bears a striking re- 
semblance to the homologous stage asfigured in Drosera by Berghs 
(1905, fig. 8), and no trace of the bivalent nature of the spirem 
can be detected. As soon as the spirem has become completely 
disentangled it begins to shorten and thicken, and it can be seen 
in places that different portions of the thread are in contact 
(fig. 21). Miss Ferguson (1904) finds that in Pinus neighboring 
portions of the thread meet and fuse, and Noren (1907) States 
that in J. communis the conditions are similar. That parts ofthe 
spirem should touch one another is unavoidable, bnt a careful 
examination has led to the conclusion that there is no actnal 
coalescence at the points of contact and that the chromosomes 
arise, not from an “incompletely reticulated structure”, but from 
a continnons spirem. 
A longitudinal Splitting of the thick spirem soon becomes 
apparent, and there seems to be no reason for not assnming that 
the adjacent threads thns formed represent merely a reseparation 
of the chromatic elements which have been actually or seemingly 
fused dnring synapsis. Almost simifitaneously transverse breaks 
appear, so that the spirem becomes divided up into a number of 
double segments which are arranged end to end (fig. 22). Whether 
the number of segments formed at this time corresponds with the 
reduced number of chromosomes found dnring the later stages of 
the heterotypic division cannot be stated with certainty, but sub- 
sequent steps make this conclusion seem extremely probable. The 
end to end arrangement of the segments is of short duration, and 
the nucleus rapidly passes into the condition known as ‘diakinesis', 
the segments becoming variously oriented but for the most part 
lying in the peripheral region of the nucleus (fig. 23). These 
segments now represent the bivalent chromosomes of the hetero¬ 
typic division, and their double nature can readily be made out in 
well stained preparations. The two halves usually lie side by side 
