166 BOTANICAL GAZETTE | SEPTEMBER 
single megaspore and the persistent enclosure of the sporan- 
gium by the sporophyll, and the angiosperm peculiarities follow. 
The profound effect of these conditions upon the germination of 
the megaspore is so remarkable, and intergrading stages so com- 
pletely unknown, that there seems to be no clue to the sequence 
of changes. That an endosporic gametophyte might eliminate 
the archegonium seems evident, for the tendency is shown 
among gymnosperms by Gnetum, where oospheres are organized 
by free endosperm cells. That the reproductive region of the 
female gametophyte may be organized earlier than the nutritive 
region, when the gametophyte is supplied with outside nourish- 
ment by the retention of the megaspore, is hinted at among the 
heterosporous pteridophytes and gymnosperms. These tend- 
encies have found full expression in the angiosperms, where 
archegonia have disappeared and the reproductive tissue of the 
female gametophyte is persistently organized before the nutritive 
tissue. Evidenceas to the details of the evolution of this tend- 
ency is lacking, and may not be in existence, but the tendency 
has certainly reached a remarkably definite expression. The 
unvaried appearance and movement of eight free nuclei or cells, 
and the remarkable fusion of two of them, represent habits so 
fixed through such an enormous group that they baffle explana- 
tion, and argue both for the monophyletic origin of angiosperms, 
and against their derivation from so divergent a line as gymno- 
sperms. 
The earlier evolution of the gymnosperm line is probably to 
be explained by ecological conditions. The body as a rule is 
organized to endure extreme conditions. It is certainly not a 
mesophytic type, and its evolution was certainly not in response 
to prevailing mesophytic conditions. On the contrary, the 
angiosperm type is essentially a mesophytic one, with great foli- 
age display, and probably expanded in response to widely 
prevalent mesophytic conditions. This might explain the habit 
peculiarities of the two groups, but whether the more recondite 
morphological differences hold any relation to these or not is too 
obscure to permit even speculation. 
