130 The Botanical Gazette. [March, 
the proembryo (fig. 21). At this stage the na Ps | 
one synergida still survives but it shows signs 0 an ‘a 
tion. It disappears after this stage and its ee: ye Bi. 
likely go to nourish the enlarging nucleus of t . fe oo 
cell. The next division is again in the lowest 0 oe mi 
cells, in a transverse direction (fig. 22). This ade olutel 
four cells of the proembryo. These divisions are a Ni ‘a 
certain, for they were traced out through stages in w a 
nuclei were in the close daughter-skein, which leaves no The 
as to the origin of the different nuclei in the series. a 
next division which occurs is in the lowest of the four “a 
This now divides in a longitudinal direction (fig. mie ye 
development was traced no farther. But the cou a 
scribed above was confirmed by numerous exam les. Bre 
the early development of the proembryo of — a 
presents variations, or the descriptions given by Hans be 
Famintzin are incorrect. At a later stage, three — a 
cells could still be seen (fig. 24), but whether these He oll 
Same as those seen in fig. 23, or represented cells whic 
nated by subsequent divisions, I did not determine. 
Summary, - 
Ol 
The results of the investigation may be summed up as 
lows: 
I. The development of the embryo-sac of ee ae 
represents nothing unusual, the fully matured sac hav 
usual eight nuclei. * the four 
2. During the conjugation of the two polar see vanatioa 
entrospheres conjugate by couples, resulting in the fo 
of two new ones for the definitive nucleus. - ely from 
The endosperm is not abundant and comes ee 
the division of the definitive nucleus, the antipodal ce 
ing no division or fra mentation. oc 
4. The division of ‘the generative nucleus of the pellen 
curs in the pollen grain. into 
5. Both of the et nuclei enter with the pollen Be act 
the embryo-sac, but only the lower one takes part in 
of fertilization, the oth : b 
- The nucleus of one of the synergidz is entirely 2 is late 
when in contact with the pollen tube. The other one's *, 
