664 CONCLUSION 
On the other hand, the examination of the embryos of various types of 
Pteridophytes has shown that the occurrence of a suspensor is variable 
within near phyletic limits, that the form of the embryo itself is in high 
degree plastic, and that a certain correspondence can be traced between 
biological requirement and the proportions, or even the actual position 
of the parts relatively to one another, and to the parent prothallus. Thus 
the haustorium or foot is found to be inconstant in position, and may 
be present or absent in plants of near affinity; the root may be entirely 
absent, or vary in its position; the cotyledons also may vary in number 
and in position as well as in form and dimensions. Such irregularities, 
together with a certain degree of physiological reasonableness which may 
often be seen to underlie them, led not unnaturally to the conclusion 
drawn by Goebel? that ‘‘root, shoot, and haustorium are laid down in the 
positions that are most beneficial for their function.” This implies that 
all parts are opportunist growths. To those who accept this thesis 
as true, embryology cannot form a secure basis for general comparisons. 
or for phylogenetic argument. For them comparative embryology would 
be little better than a study of the more or less immediate biological 
adaptations of the embryos themselves: there would be no common ground 
from which the comparison could start. 
But it may be questioned whether this extreme position is fully 
justified. The endeavour must be made to recognise and isolate those 
characters of the embryo which are variable, and to see whether there. is. 
not some element of constancy in shape or in construction which underlies. 
the fluctuating features, and runs through all the different forms. This. 
has been greatly facilitated by recent discoveries; for now the embryos. 
of all the leading types of living Pteridophytes are fairly well known, with 
the exception of the Psilotaceae—though possibly these are, for comparative 
purposes, the most important of them all. 
A revision of the embryology in the whole series of Pteridophytes. 
described above leads to the conclusion that the form is not so inchoate 
or immediately plastic as Goebel’s statement implies: comparison shows. 
that there is one point comparable in them all (where fully investigated) 
which does not appear susceptible of disturbance on a basis of opportunism,,. 
viz. the position of the apex of the axis relatively to the primary 
segmentation ; or, expressed in other words, the relation of the polarity of 
the embryo to its first cleavages. 
Of this primary segmentation there are two types, according as a 
suspensor is present or absent ; otherwise it shows that remarkable constancy 
of cleavage which led earlier writers to construct the theory of octants, 
now no longer to be upheld. It has been shown that these two types 
may appear in the same phylum (Lycopodiales, Ophioglossales) and even 
in the same genus (So¢rychium); and there is accordingly reason to- 
believe that, however important biologically, they do not mark such. 
1 Organography, ii. p 246. 
