714 CONCLUSION 
considerations support the conclusion that the Sporangtophoric Pteridophytes 
constitute a brush of naturally related phyletic lines. 
It has been argued at length above (Chapter XXXI.) that the Ophioglos- 
sales are an upgrade sequence, a view which accords with their homosporous 
state: also that ¢hezr spike illustrates various steps in the tncreasing complexity 
of a body of the nature of the sporangiophore. The elaboration of the 
subtending leaf runs parallel with it, while both leaf and spike show 
branchings and fissions comparable with those recognised in the sporangio- 
phoric Pteridophytes, but carried out here on a larger scale. On this view 
the whole unbranched shoot ts a simple strobilus bearing leaves, of which all 
are potentially fertile, and the great majority actually so. But the large 
size of the leaves, and their isolation in point of time (commonly only one 
being expanded at once), disguises the real nature of the strobilus. All 
the three genera have attained to great complexity, but in Ophzoglossum, 
and more clearly in Botrychium, the gradually increasing complexity of the 
leaf in the individual life indicates what has probably occurred also in the 
race. Along one line, that of Ophioglossum pendulum, intermedium, and 
simplex, it seems probable that reduction of the vegetative system has 
occurred ; but with this exception the Ophioglossaceae appear to have been 
an upgrade sequence, sprung from some sporangiophoric stock, and bearing 
no near relation to the large-leaved Ferns, Vhe anatomy here again points 
to an origin from a protostelic structure, while the single leaf-trace strand 
in all the simpler forms indicates a primitively simple structure of the leaf. 
The Filicales constitute a more isolated phylum than any of the smaller- 
leaved forms. Their general comparison among themselves has been fully 
discussed in Chapter XL., and the relations of their leading families graphically 
indicated on p. 653. It is now recognised that true Ferns were represented 
in the Primary Rocks by relatively few forms, while their derivative families 
increased in number and extent in later periods. The Leptosporangiate 
type is essentially modern: it is indeed doubtful whether any of the 
Palaeozoic Ferns had an annulus composed of a single row of cells: on the 
other hand, though Eusporangiate Ferns still survive, they were the leading 
type of the Palaeozoic Period. Accordingly, it is in the latter and not in 
the former that the features of interest for comparison with other phyla of 
Pteridophytes are to be found. 
It has been shown that the construction of the shoot of the primitive 
Eusporangiate Ferns is essentially strobiloid, maintaining constantly the same 
relations of axis and leaf as in smaller-leaved forms: the axis is in some of 
them permanently protostelic (Botryopterideae), while in the rest a protostelic 
structure figures in the early seedling of the forms still living. The leaf-trace 
is a single strand in primitive forms, though in the modern Marattiaceae it 
may be broken up into separate strands. In addition to this the outcome 
of anatomical comparison of the Ferns at large has been to show that the 
axial structure is constantly referable in origin to a primitive protostele, a 
construction which is held to be typical and primitive for strobiloid plants ; 
