358 BOTANICAL GAZETTE [May 
FERTILIZATION. 
Double fertilization has been found to occur in several Com- 
positae and, judging from some of my preparations, it occurs 
also in Aster; but my material has yielded so few stages just 
before and including the fusion of the sexual nucleithat I do not 
wish to speak unqualifiedly regarding this point at present. In 
fig. 9 we have an instance of the delayed fusion of the polar 
nuclei, and apparently an example of double fertilization in 4. 
undulatus. The polar nuclei are in an early stage of fusion, and 
the sperm nucleus is pressed against the lower polar nucleus. 
The fusion of the sperm with the egg nucleus is nearly com- 
pleted, so that the two appear almost as one nucleus with two 
nucleoli. 
A most interesting embryo sac is illustrated in fig. 18, a—0. 
There can be no doubt that we have in this instance the occur- 
rence of two oospheres and two endosperm nuclei in the same 
embryo sac. The upper part of the embryo sac resembles all 
normal embryo sacs at a similar stage of development; but in the 
lower portion, just above the first antipodal cell, there is a cell 
and a large nucleus which have the same structure and the same 
relative position as that of the egg and the endosperm nucleus at 
the micropylar end. These clearly represent, so far as position 
is concerned, an antipodal oosphere and a definitive nucleus. 
No positive statement regarding the origin of this extra egg 
and endosperm nucleus can be made, but several theories as to 
their probable derivation may be suggested. It is possible that 
the lower polar nucleus did not fuse with the upper polar nucleus, 
but that each has taken on the appearance of an endosperm 
nucleus, and that the lower oospere has arisen directly from an 
antipodal ; it may be that the normal egg has wandered to the 
antipodal region of the embryo sac, and that the upper egg has 
been derived from one of the synergids; or it is possible that the 
lower egg has developed from one of the synergids which had 
become displaced. In order to explain satisfactorily the struc- 
ture of this embryo sac on the basis that one of the eggs has 
been derived from a synergid, it is necessary to assume that at 
least three irregularities have occurred in its development. 
