COMPARATIVE DISCUSSION 485 
part. Such sterilisation, amounting even to the formation of permanent 
septa, has been shown to take place in other plants, and the theory is 
not open, therefore, to @ priorz objection (see Chapters VII. and X.). 
The frequent absence of sporangia from the tip of the spike is probably 
due to abortion: they cover the apex in some species, which also show 
abortion of sporangia lower down (e.g. O. pendulum). The presence of 
vascular strands in the septa themselves shows how the physiological supply 
followed the structural advance: on a theory of lateral fusion of sporangia 
such a position of the vascular strands would be, to say the least, 
improbable. Lastly, imperfect or irregular septa have sometimes been 
seen. Thus the structure, so far as it goes, readily coincides with a 
theory of extension, and progressive septation to produce the spike of 
Ophioglossum from some simpler beginning. 
The case of Botrychium is similar, though less obvious, owing to the 
isolation of the sporangia, while it is complicated by the fact that 
branching of the spike frequently accompanies septation. That a structure 
compatible with progressive septation exists is shown by Figs. 253, and 
its relation to the branching, which brings such conspicuous results in 
the spike of Botrychium, appears in its simplest form in Figs. 252. It 
only requires the repetition of the processes, which are thus illustrated 
in the individual, to lead from the simplest to the most complex spikes 
in the genus. 
Lastly, in Helminthostachys the ranks of sporangiophores correspond 
in position to the rows of sporangia in Op/ioglossum. An upgrowth from 
the sporangiogenic band, less regular, but of the same nature as that 
seen in the branching of the spike of Botrychium, would give the sporangio- 
phores of elminthostachys, while the individual development directly 
represents what this progressive theory demands. This, indeed, is the 
foundation upon which the present view of elaboration of the spike in 
the Ophioglossaceae is primarily based: without any preconceptions 
involving reduction or modification, the theory is founded directly upon 
the simple facts of individual development. 
The anatomical structure of the shoot in the Ophioglossaceae with 
its rare dichotomy, which compares rather with the microphyllous than 
with the megaphyllous Pteridophytes, may next be considered. It has 
been seen above (p. 464), that the facts observed are compatible with 
an origin of the axial system from a protostelic state. The stele of the 
seedling or adventitious bud, is either a protostele or slightly medullated 
monostele: passing upwards along the shoot there is an amiplification of 
the stele, with swelling of the central pith. In the lower region there is 
usually a well-marked endodermis: this may be continued throughout 
the length of the rhizome, but in some cases it fades out upwards, as 
the stele distends. The xylem in the upper region forms a hollow 
cylinder or funnel, more or less interrupted by leaf-gaps, where the 
single strands of the several leaf-traces pass off. The protoxylem is not 
