1907] _ COKER—CEPHALOTAXUS FORTUNEI 
7 
the first time. Fig. 18 shows this stage just at the beginning of the 
cell wall formation. The cells are not arranged in regular tiers 
throughout, being most irregular in the central region. 
end are two fairly even tiers, and the tip is occu- 
pied nearly always by two superimposed cells. In 
archegonia with more rounded bases, which appear 
occasionally, there may be two cells side by side 
in the tip. 
After the formation of cell walls the sixteen cells 
divide ia to form thirty-two, and it seems prob- 
' able that the upper tier divides 
horizontally to form the rosette cells 
and suspensors. I have not found 
any case, however, where the rosette 
consists of free nuclei, as is so 
common in conifers. The cells of 
the more or less regular tier above 
cell-stage is represented, The upper 
two lower of one cell each. 
the suspensors are enclosed in the ,. 
cell walls, and occasionally one OF celled proembryo 
more of them may divide horizon- with cell walls be- 
tally. In jig. zg the thirty-two ae aoe 
At the upper 
IG. ¥8.—Sixteen 
tier (r), which answers to the rosette of five cells; 
the next tier (s), which forms the suspensors, also 
contains five cells; while the remaining cells are 
not in regular tiers except near the tip, where there 
are three distinct tiers, the upper of two cells, the 
The suspensors now begin to elongate and cell 
divisions occur in any of the cells of the middle 
Fic. r9.—Thirty- Fegion below the suspensors and above the two tip 
Beiteled proem- cells, These divisions are no longer simultaneous, 
ee aa =a but occur here and there as in ordinary growth. 
' 12) 
Suspensors. X 185. 
In fig. 20 is shown an embryo in which the 
suspensors have just begun to elongate. One cell 
of the upper tier has divided into two, and the number of cells in 
the middle region has increased somewhat. The two tip cells show 
