128 BOTANICAL GAZETTE [Aucust 
nuclei. In the upper or chalazal group two of them lie side by side 
without a separating wall. The upper two are in the normal position, 
but the usual separating wall is lacking. Fig. 34 seems to show an 
embryo sac of two nuclei derived from the chalazal group of two 
mother cells. The three disintegrating nuclei are probably mega- 
spore nuclei, and the micropylar mother cell has not divided. Fig. 
35 is taken from a much younger ovule, but a very broad one. It 
seems to be the product of two sporogenous cells, the chalazal mother 
cell having divided to form the two daughter nuclei, the micropylar 
mother cell being still in metaphase. It is possible that this is the 
result of the second division of the chalazal daughter cell without 
walls separating the two megaspore nuclei, the micropylar daughter 
cell being still in mitosis. But the chromosomes in this nucleus give 
every appearance of the heterotypic division, and the ovule is so un- 
usual in appearance that it seems to have had two mother cells. 
Synapsis continues for some time, if the extent of development of 
the integument be considered proof (figs. 4, 5). The division of 
the mother cell takes place in the usual way (figs. 5-10), giving the 
two daughter cells (jig. 17). The number of chromosomes is appar 
ently thirteen (fig. 6), although only a few counts were made, but 
this one seemed unusually distinct, so it is probably the correct num- 
ber. Several counts were attempted in a sporophytic area, giving 
approximately twenty-six. It is probable that sometimes the wall 
separating the daughter nuclei fails to appear. Fig. 9 would be 
expected to show some evidence of wall formation, if it were going 
to appear. Fig. 15 shows the division completed, with no suggestion 
of a wall. Fig. 16 gives no wall. Several examples of this failure 
of the wall were found, and in others it seemed very faint and probably 
disappeared. 
In many plants the chalazal daughter cell divides; but in Cale- 
pogon it seems quite common for both to divide, although the micro 
pylar cell is even then somewhat smaller than the chalazal cell (figs 
13, 14). These divisions may occur at the same time as in the above 
figures, or either may precede the other. Figs. 17 and 18 show the 
chalazal cell dividing first, the former showing the micropylat cell 
somewhat more advanced than the latter. In jig. 19 the micropylat 
daughter cell has almost completed the division, while the chal 
ST eS eer aa eT Ee eee a 
