206 BOTANICAL GAZETTE [SEPTEMBER 
located some distance below them, are fusing and migrating upward 
in the sac (fig. 60). The nucleus resulting from their union lies 
just below the egg and becomes the nucleus of the primordial cell 
of the proembryo (figs. 61, 62). Before division it enlarges greatly 
and is apparently cut off fromi the sac below by a transverse wall. 
At the same time two of the micropylar nuclei, the synergids, begin 
to degenerate. The other nucleus, which is homologous with the 
one in the sac of D. opuntioides from which the proembryo origi- 
nates, continues to enlarge for a short time as if preparing to divide, 
and then also begins to degenerate before division of the fusion 
nucleus occurs. The primordial cell of the proembryo divides by 
ordinary mitosis (figs. 62, 63). The two resulting cells (fig. 64) by 
further divisions form a proembryo, which is similar to that of 
D. opuntioides (fig. 65). The development which now follows is 
also the same as has been given above, hence further descriptions 
are unnecessary. The origin of the embryo as seen in D. gracile 
suggests the example described by FARMER and Dicpy (8) in Lastrea 
pseudo-mas var. polydactyla Wills, in which the embryo arises from 
a nucleus resulting from the union of two body cells of the prothallus 
which has come from a normally developed spore and has the 
haploid number of chromosomes. The difference in the mode of 
origin of the embryos of D. gracile and D. opuntioides also recalls 
that shown by the above mentioned authors between Lastrea 
pseudo-mas var. polydactyla Wills and the closely related Lastrea 
pseudo-mas var. polydactyla Dadds. In the latter the embryo 
arises from a projection or budding out of the prothallus which has 
arisen aposporously. In this same variety these authors also found 
nuclear fusions occurring in the prothalli which had come from 
normal spores, quite similar to those in the variety first mentioned. 
It is probable that a phenomenon similar to that just stated for 
Dadds’ fern occurs in D. opuntioides, though no evidence for such 
a supposition was seen. Owing to the lack of material, it was not 
possible to make a detailed study of the division of the megaspore 
mother cell. At the division of the microspore mother cell in 
D. gracile 9 chromosomes were found passing to each pole of the 
spindle; 18-20 chromosomes were estimated in the cells of the 
developing embryo. Apparently the same number is also present 
i. 

