10 Miyake, The development of the gametophytes etc. 
In the conjugating nnclei of Juniperus communis Noren 
(1907) observed the decrease of the chromatic substance and the 
corresponding increase of the nucleolus-like bodies the “Pseudo- 
nuclei” which he thinks contribute the greater part of the substance 
of the chromosomes. Although I do not have enongh evidence to 
speak for or against Noren’s view, some of my preparations like 
fig. 90 may suggest the possibility of such a process. In figs. 92 
and 114 are shown completely formed spindle, with chromosomes 
accumulating near the equator of the form er; the paternal "and 
maternal elements can no longer be distinguished. 
The result of the first division is shown in fig. 94; two 
danghter-nuclei are snrrounded by dense mass of starch-granules, 
and they travel toward the base of the egg. As they reach the 
base of the archegonium (figs. 95, 96), both nuclei diyide simul- 
taneously, and four free nuclei are formed as shown in fig. 97. 
The third division which now follows is also simnltaneons and 
results in the formation of eight nuclei, which are arranged in 
tiers as shown in fig. 98. Walls are then formed between the 
nuclei, the upper tier remaining open at the top. 
In Ot'yptomeria, Lawson (1904b) finds that the continuous 
fibrils of the spindle persists, and the first cell-membranes of the 
embryo are formed between the nnclei. According to Miss Fer¬ 
guson (1901, 1904), who has stndied the first wall-formation of 
the proembryo in Firnis , the spindle fibers of the third division 
seem to disappear before the walls are formed between the nnclei. 
Although I have not followed the process of the first wall-formation 
in detail, I am inclined to accept the view of the latter investigator 
on this point, The stage as shown in fig. 98, in which no trace 
of spindle fibers is visible between the eight free nnclei, can hardly 
be interpretted in accordance with Lawsons view. 
While in the Äbietineae , the eight nnclei of the proembryo, 
as a rule, arrange themselves in two tiers of four nuclei each, the 
same stage of Cunninghamia, Cryptomeria and Taxodium does not 
show such regularity in the arrangement of the nuclei. Coker 
(1903) States that in Taxodium two nuclei are situated at the base, 
and six nuclei lie above them in one plane. He adds: “while this 
is the nsual arrangement, it is not uncommon to find only one at 
the base, while the other seven are arranged above it. In a few 
cases there were three below and five above.” In Cunninghamia 
the number of cells in each of the two tiers is also not constant; 
it seems, however, that the upper tier generally contains five nuclei 
and the lower three, as shown in fig. 98. It was sometimes found 
that there are two below and six above. The other combinations 
of cell-arrangement are also possible. The lower group of com- 
pletely-walled cells may not lie in the same plane; they are some¬ 
times fonnd one above the other instead of side by side. 
The nnclei of the upper tier then divide simultaneously and 
the walls are formed between the daughter-nuclei (figs. 99—102). 
We have now in the proembryo two tiers of cells and one tier of 
the incompletely-walled cells (figs. 101, 102). Lawson (1904b) 
