1910] SMITH—ERIOCAULON 283 
About this time or a little later the nucellar tissue lateral to the 
megaspores begins to be broken down and absorbed by the growing 
embryo sac. A few of the apical cells of the nucellus, however, per- 
sist for a long time, and enlarging assume the appearance of a “tape- 
tum” (figs. 12-14, n). These too are ultimately absorbed, and then 
the embryo sac abuts directly upon the inner integument and micro- 
pyle. 
The development of the embryo sac is quite normal, so that little 
explanation need be added to the figures. The central vacuole first 
appears at the four-nucleate stage. During the three successive 
divisions there is little if any increase of nuclear material; probably 
the eight nuclei of jig. 16 have no greater aggregate volume than the 
single nucleus of the megaspore. 
In the succeeding pre-fertilization organization of the embryo sac 
some of these nuclei become greatly enlarged (cf. figs. 16 and 78, 
which are drawn to the same magnification). A nearly mature 
embryo sac is shown in fig.17. The antipodals are never conspicuous. | 
After the inception of endosperm formation they become difficult 
of recognition, but occasionally may be seen in a small pocket below 
the endosperm. The upper polar nucleus moves down to meet the 
lower one near the antipodals. From the frequency with which they 
are found in contact with each other, it may be inferred that their 
fusion is very gradual. It is attended with considerable growth of 
e fusing nuclei, and there is still further growth subsequent to 
their fusion (cf. figs. 17, 18, 22). Probably this enlargement of the 
definitive nucleus is correlated with the rapid production of endo- 
_ Sperm consequent on fertilization. There is much variation among 
plants as to the time when the polar nuclei fuse and the relation of 
this fusion to pollination and fertilization. Thus in Elodea WYLIE 
(15) found that the actual presence of the pollen tube in the embryo 
sac is necessary to bring about the fusion. In Eichhornia (SmrrH 14) 
the fusion occurs before pollination, and in Sagittaria (SCHAFFNER 13) 
before the entrance of the pollen tube. Further examples are given in 
CoutTer and CHAMBERLAIN’s M or phology of angiosperms (6, pp. 95, 
96). There can be no doubt that in Eriocaulon the fusion is inde- 
pendent of the stimulus of pollination, for it was found complete in 
numerous heads taken from beneath the surface of the water. On 
