700 CONCLUSION 
been similar to, though phyletically quite independent of, that in Ophio- 
glossum; and the results are, in the former an elongated sorus attached 
to the leaf-surface, in the latter an elongated sporangiophore which is 
attached to the sporophyll only at its base. In many other Ferns there 
is evidence of amplification of the sori, whether by intercalary elongation 
of the receptacle and a basipetal succession of sporangia, as in the Gradatae, 
or by marginal extension, as in the Zéndsaya-Pteris series, or by superficial 
spread so as to produce the conditions seen in Gymnogramme, Acrostichum, 
or Flatycerium: associated with these is the profuse interpolation of new 
sporangia characteristic of the Mixtae. It is thus possible to picture how 
even the most complex and divergent types of spore-production in large- 
leaved forms may be referred back in their ultimate origin to elementary 
types, and to recognise how they conform to that general scheme of 
construction which obtains among the simpler strobiloid Pteridophytes. 
It remains to consider the distribution of the spore-producing members 
on the plant as a whole. We have recognised the shoot or primitive 
strobilus as composed of (i) axis, (ii) leaves or bracts, and (iii) spore- 
producing members. It has also been seen to be probable that originally 
all the leaves were sporophylls. The primitive shoot appears to have been 
a general-purposes shoot, in which vegetative and propagative regions were 
not segregated. But it is evident that two other conditions are possible, 
that is a shoot bearing (ii) alone, and one bearing (iii) alone; both of 
these states are known in living forms, and both may be held to be 
secondary and derivative. 
The former case, where leaves without spore-producing members are 
present, is by far the commoner condition of the two, and it appears 
in the early stage of the ontogeny in almost all Vascular Plants. But it 
also appears in successive intermediate zones higher up in various plants, 
and notably in Lycopodium »Selago, from which it is called the “Selago” 
condition (Frontispiece) (Chapter XIII.). It has been shown that this 
condition would result from abortion of the spore-producing members, and 
the fact that this has taken place is clearly indicated by the occurrence 
of imperfect sporangia about the limits of the region which has remained 
fertile (p. 162). The converse evidence, that in certain cases (Z. Seago, 
Botrychium, and Ophioglossum) the spore-producing members appear very 
early in the individual life, and that in Lygodium subalatum the very 
first leaf may be fertile, further strengthens the view that the whole plant 
was originally fertile (p. 186), and that the sterile regions, whether basal or 
intermediate, are so by abortion of the spore-producing members. 
The second case above mentioned, in which spore-producirig members 
are present but no leaves, is less common; it is seen in Archaeocalamites 
and in the modern £guzsetum. It has been argued at length above 
(pp. 382-4, and p. 429) that the leaves and sporangiophores in these 
plants are parts of distinct nature and origin, and that the condition of 
their strobili is due to abortion of the leaves, of which in Agudsetum the 
