1922] BUCHHOLZ—VASCULAR PLANTS 267 
In eusporangiate ferns with well developed suspensors it is very 
much greater; hence the evidence to be found in the eliminated 
zygotes of this more active form of embryonic selection in ferns 
is not conspicuous. It occurs in the earliest stages, leaves only 
very small aborted embryos, and it is probably for this reason that 
embryonic selection has usually been overlooked as a normal 
process of the life cycle. 
While many of the living ferns probably do not possess embry- 
onic selection, at least as a very striking or prominent feature, 
practically all of them show good evidence of a derivation from 
forms possessing it; of having passed through this condition 
historically. During Paleozoic time, when pteridophytes consti- 
tuted the dominant vegetation, embryonic selection was probably 
the prevailing condition. Even the environmental forms of 
competition were much more keen, as our vast coal deposits would 
indicate. There is little doubt that the early seed plants which 
were derived from these ferns retained embryonic selection, as it 
is a feature which has persisted until today in, gymnosperms 
generally. The simple suicides of gymnosperms is therefore 
of fern origin. 
SUPPRESSION OF EMBRYONIC SELECTION IN FERNS.—There are 
some pteridophytes among which there is a more or less complete 
elimination of embryonic selection. Such a highly specialized © 
form as Marsilia presents a special variation in this direction, 
since only one archegonium is produced on the female gametophyte. 
Obviously there is no selection between two or more zygotes on the 
Same prothallium. It appears that in Marsilia and other pterido- 
phyte forms having only one archegonium, we have examples of 
the elimination of the embryonic selection, an advanced condition, 
doubtless the result of specialization. Lea tenpornnet ferns, whose 
are times very much reduced in size, may perhaps 
provide. additional id teresting examples of the complete elimination 
of embryonic selection. Should this selection occur between arche- 
gonia, or between the eggs of neighboring archegonia during their 
development, it could certainly not be classified as embryonic se- 
lection; it may perhaps be designated as gametic selection, or in 
some other category of developmental selection. 
