and Fertilization in Oenothera. 
287 
in each of them, though the whole aspect cannot be included in this section. 
In Text-fig. Ill, 15, the lower one, whose polar is in the adjoining section, is 
gaining, and the filiform apparatus are developing in both of them. The 
dislocation of the megaspores often takes place in the course of their 
development (Text-fig. Ill, 4, 7), resulting in the formation of the condition 
delineated in Text-fig. Ill, 17, in which the smaller one appears as if partly 
embedded in the larger sac. If the above-stated dislocation proceeds farther 
and the inferior one remains, two embryo sacs attain a parallel position 
(Text-figs. Ill, 14, 16, and IV). Often the lower megaspore remains in uni- 
or binucleate state, as shown in Text-fig. Ill, 11, 12. A similar condition 
has been reported as occurring in a related species, Lopezia coronata , and 
in some others by Tackholm (68, 69 ). The writer has found the same 
phenomenon in Ludwigia prostrata (Text-fig. VIII, 10); such a case is, 
however, known to occur in a great many plants belonging to other families. 
8, 9, 10, Text-fig. Ill, show several younger stages which correspond to the 
cases as shown in Text-fig. Ill, 11, 13. Text-fig. Ill, 9, shows one of the inter¬ 
vening megaspores still surviving, and in Text-fig. Ill, 10, the upper one 
is in a tetranucleate stage, while the lower is binucleate, though the nuclei 
are not visible in this section. 
As to the fate of the excess embryo sac, nothing conclusive can be 
stated, owing to the lack of the material, except that it may disintegrate 
sooner or later when the ovule is entered by a single pollen-tube, or it may 
be visited by a pollen-tube (Text-fig. IV), when more than one pollen-tube 
penetrates the ovule. 
In Text-fig. IV, which has been reconstructed from three succeeding 
sections, ideally cut, the sac on the right-hand side is just before the fertili¬ 
zation, while the other is in contact with the pollen-tube. This figure 
suggests possibility of fertilization in both of the embryo sacs. Schwere 
( 59 ) working on Taraxacum officinale found a case of twin embryo sacs, 
each giving rise to a normal embryo ; such was also the case with 
Gastrodia elata studied by Kusano ( 36 ), in Lychnis alba x L . Flos-cuculi by 
Compton ( 9 ), and in Ulmus americana by Shattuck ( 61 ), though in the 
latter two cases it was not retraced farther up to the embryo sac formation. 
The persistency of more than one megaspore in Oenothera, which is 
situated in the higher rank among the Polypetalae, shows the atavistic 
inclination of the young gametophyte caused by an evolutional tendency. 
Male Gametophyte. 
A full description concerning the development of the pollen mother¬ 
cell of Oenothera has been given by Geerts, Gates, and Davis ; the further 
growth of the pollen, especially the development of the pollen-wall, has 
been faithfully followed by Beer ( 3 ). The ripe pollen grain, just before 
shedding, contains a vegetative nucleus (PL VII, Fig. 1, v) in the centre 
