1900 ] LIFE HISTORY OF SILPHIUM 109 
conduction are but two important phases. The view that the 
antipodals are homologous with the vegetative portion of a 
pteridophyte prothallus is strengthened, therefore, rather than 
overthrown by the discovery of the réle which they play in the 
nutrition of the embryo sac. 
I have reserved one feature in connection with the egg 
apparatus for a special consideration, namely, the peculiar cap 
which is usually to be found upon the micropylar end of the sac. 
Its ordinary appearance is as in fig. 38. It is frequently striated, 
and always shows a rift or superficial groove, which cannot be 
related, however, to the line of contact between the synergids. 
.The material of which the cap is composed stains deeply with 
all the stains used, but when lightly stained the outline of the 
tip of the sac can be seen within it, showing that the cap is not 
a part of the synergids. This fact precludes the possibility of 
its being a filiform apparatus, and I have been led to the conclu- 
sion that the cap represents the remnant of the nucellus, which 
may often be broken in pieces, as in fig. 37, but which, in most 
cases, is pushed forward by the sac and remains perched, calyptra- 
like, upon its apex. This explanation was first suggested by the 
form of the nucellus tip (see especially figs. 32, 35), and seems 
to be the only possible interpretation of the condition seen 
in fig. 43, where the cap has evidently continued its growth 
and become highly vacuolate. Finally, in fig. ¢4, from a 
Preparation poor in other respects, the cap consists of four 
cells in section, the walls and two of the nuclei still being 
visible. 
The above description of the embryo sac of Silphium integri- 
Solium will apply very well to S. ¢rifoliatum, S. terebinthinaceum, 
and S. laciniatum, the chief differences being in matters of size 
and proportions, and hardly deserving special mention. The 
peculiarities of the antipodal and micropylar regions are illus- 
trated from S. integrifollum, because its greater abundance, 
together with its being easier to kill and section, encouraged a 
more extended study. But abundant evidence was obtained 
Showing that the same peculiarities exist in the other species, 
