The Principles of Morphology. 
167 
sporangium always arises ontogenetically in the apparently axial 
position , some distance above the insertion of the sporophyll, so 
that it has actually been described as arising on the axis itself. But, 
during subsequent development, the sporangium becomes carried 
downwards on to the actual free base of the sporophyll, which 
seems to shew that it was from birth already situated on an upward 
basal extension of the sporophyll which is embedded in the axis. 
Now, suppose the ontogenetically primary position of the sporangium 
of Selaginella to become fixed and the subsequent transposition on 
to the free portion of the leaf to become completely eliminated (a 
state of things which might easily have come about), we should have 
a precisely parallel case to that of Calamostachys. 
For the existence of such an upward basal extension of the 
sporophyll as I have assumed above, there is further corroboration 
in the fact that it clearly obtains in the higher plants, e.g. in 
Primulaceae, where the ovules are situated in the same pseudo- 
axial position as the sporangiophores and sporangia of Calaniostachys 
and Selaginella. 
Turning now to the Ophioglossaceae, I must almost entirely 
agree with the conclusions of Celakovsky and of Bower in the view 
that, in order to explain the peculiar position of the “ fertile spike ” 
on the upper surface (in most forms) of the sporophyll, the theory 
of a gradual and, eventually, highly complex elaboration by means 
of progressive sterilisation, must be adopted. This complex 
structure must have been derived from the simple sporangium or 
sporangiophore of a Lycopod. We see an incipient stage of 
sterilization and differentiation in the case of Isoe tes. Imagine 
this proceeding further and in a different direction, and we might 
obtain at length the “ spike ” of Ophioglossum and, finally the very 
highly-differentiated “ spikes ” of Botrychium and Helminthostachys. 
The interesting parallelism in elaboration between the sterile and 
fertile portions of the sporophyll is a point to be noted. 
An important parallelism may also be observed if we compare 
together the cases of the simple Lycopodium-type and the complex 
type of the Ophioglossacete on the one hand, and, on the other, the 
type of the simpler Sphenophyllums, e.g. S', trichomatosum and 
Psilotaceie, and the complex type of Cheirostrobus or of S'. Ronieri. 
In Cheirostrobus we also see, as in Ophioglossaceae, a highly- 
differentiated sporangiophore with a correspondingly-developed sterile 
portion of the sporophyll. 
