EMBRYOGENY OF THE PTERIDOPHYTES 675 



held to be the primitive axis, which in virtue of its constancy and of its 

 early development is regarded as the fundamental factor in the embryonic 

 shoot. The prior existence of the axis in the normally developing shoot, 

 and the origin of the leaf laterally upon it, have been held as the basis 

 of the enation theory of the leaf (Chapter XL) : since there is reason to 

 recognise the existence of that polarity of the embryo which defines the 

 axis, prior to the origin of the leaf in all the varying forms of the embryo, 

 the same arguments will apply even to the earliest phases of the ontogeny. 

 In fact, the embryo itself is from the first segmentation a shoot showing 

 polarity : the appendages appear later. Such results from the comparative 

 study of embryos greatly strengthen the strobiloid theory of the shoot, 

 as enunciated in Chapter XI. : at the same time they indicate that the 

 embryo is not a thing apart from the later developed shoot, but merely 

 its initial phase, modified in various ways to meet biological needs, 

 but preserving essentially the same relations of prior-existent axis and 

 of leaf produced in lateral relation to it. 



A question remains as to the relation of the embryos with suspensor to 

 those in which there is none. Is it possible to recognise either of these as 

 the prior state? The two types indicate different modes of prothallial nursing : 

 that with a suspensor is characteristic of stocks having relatively bulky 

 prothalli, often underground, and at the present time carrying on as a rule 

 a saprophytic nutrition : the type without suspensor is characteristic of stocks 

 with less bulky prothalli, usually above ground, and self-nourishing. The 

 determining factor would appear to have been the bulk of the nourishing 

 prothallus, rather than the exact way in which it obtained its nourishment. 

 The question will therefore be, whether the evidence points to a bulky 

 prothallus and embryo with suspensor as the prior condition, or a less 

 bulky prothallus and embryo without suspensor. There are two phyla from 

 which comparative evidence on this point may be drawn, viz. the Lycopo- 

 diales and the Ophioglossales. In the Lycopodiales, in view of the upward 

 curvature commonly seen in their embryos (Figs. 183, 186, 188, 190), 

 and the necessity of their bursting through the tissue of the prothallus 

 at some point apart from the archegonium to gain their freedom, the 

 complete inversion of the embryo, and its emergence in the neighbourhood 

 of the archegonial neck would be a simplification of an awkward and 

 inconvenient process. Such a simplification is found in Isoetes, which there 

 is good reason to look upon as a more specialised type of the Lycopodiales, 

 and in which the indeterminate position of the first segmentation of the 

 zygote suggests how the inversion may have come about. Moreover, 

 the condition with suspensor is found in its simplest form, and without 

 any tuberous complications, in such species as L. Selago and Sel. spinulosa, 

 both of which are believed to be relatively primitive forms. The facts 

 supply no proof, but they suggest a reasonable probability that within the 

 Lycopodiales there has been a progression from the state with suspensor, 

 and apex directed to the base of the archegonium, to the state without a 



