186 IN THE SHOOT AS A WHOLE 
it does not follow from’ its acceptance in these cases that a theory of 
recapitulation can be applied consistently, or in detail, to all phases of 
development, or that evidence of it is to be found necessarily in the 
earliest steps of the embryogeny. It remains for the morphologist to draw 
for himself the reasonable limit of its application. If this be done, and 
especially if the variability which exists be duly appreciated, then the early 
stages of the initial embryogeny of the sporophyte will take their right 
place: and recapitulation will be traced as a limited phenomenon only, 
applicable, it is true, to the case of relatively recent adaptations, but not 
with equal certainty to the far-away facts of the past. For reasons such 
as are explained in this chapter, it will not be assumed that plants so 
diverse as are the main groups of Archegoniatae show in their early seg- 
mentation, or in the initial form of their embryos, any detailed reflection of 
ancestral characters. The facts observed should be used with the greatest 
caution, especially where the comparisons are made between representatives 
of phyla which must have diverged early from some primitive stock, if 
indeed they be related at all. 
Certain points touched in the above discussion will help towards an 
understanding of the relations of sporophylls and foliage leaves to the 
first leaves of the embryonic plant. In Chapter XIII. it was concluded 
that in certain cases at least foliage leaves are to be held phylogenetically 
as sterilised sporophylls: and the question remains whether or not all 
non-propagative leaves, including the cotyledons themselves, originated in 
this way. There seems to be a high probability that in the Pteridophytes 
they did. There is no reason to hold that their first leaves differ in any 
essential point from those which are formed later: frequently they resemble 
the later leaves closely in outline; but they are sometimes characterised by 
peculiarities of form, though® these are less marked than in the cotyledons 
of Phanerogams. Sometimes the first leaves in Pteridophytes arise laterally 
on an axis already defined (Zguise¢um); but in other cases, and especially 
in the megaphyllous forms, the first leaf or cotyledon may appear prior 
to any definite outgrowth of the axis itself. This fact may be held to be 
in itself an inherent objection to ranking the cotyledon as the equivalent 
of a foliage leaf which arises from the axis; but this objection is met by 
the fact that free-living leaves, apart from any obvious existent axis, do 
occur elsewhere in certain specialised cases: these may be interpreted as 
originating by an anticipatory development, though still in relation to an 
axis not yet defined by external growth. And so also the cotyledon in the 
Fern may be held to be essentially an appendage of the axis, the central 
point of which is already defined in close relation to the intersection of 
the octant walls of the epibasal segment, but not characterised as yet by 
external growth: the cotyledon, on the other hand, is hurried forward 
precociously in its development to meet the physiological need for nutrition, 
but maintains nevertheless its orientation relatively to the deferred axis. 
