Sphenophy llales. 99 
frond of Botrychium are not mere abnormal sports, they are 
ancestral structures. The spike of Botrychium is probably best 
regarded as a branch of the frond, to which the sporangia are now 
usually restricted ; but such dimorphism may well be secondary, 
and in the remote ancestors of Botrychium, both main branches of 
the frond may have been fertile,—a less specialized condition than 
dimorphism. In that case the sporangia on the usually sterile 
frond of Botrychium would not be fresh developments, as Dr. Scott 
supposes those borne by the dorsal sporangiophores of Spheno- 
phyllum fertile to be, but reminiscences of a primitive state. As 
already pointed out the want of differentiation between dorsal and 
ventral lobes is an important indication of primitiveness; other 
indications of this are the absence or small amount of union of the 
sporophylls at their base, and the fact that this theory of the 
primitiveness of the Sphenophyllum fertile type of cone enables us 
to interpret the cones of the allied Equisetales more satisfactorily, 
both as regards their relation to one another, and to the Spheno- 
phyllales. On the other hand the secondary growth of the axis of 
the cone of Sphenophyllum fertile seems to be an acquired character, 
and one that might well arise if, as Dr. Scott suggests, super¬ 
numerary sporangia were formed. Dr. Scott has ably defended his 
theory, grounding his defence chiefly on two facts. Firstly on the 
fact that Sphenophyllum fertile is isolated as regards the fertility 
of the dorsal segments, and that it would be rash to regard this 
exceptional character as primitive. Against this it may be urged 
that the detailed structure of the fructification is only known in a 
few species of the phylum, and the negative evidence drawn from 
the absence of fertile dorsal lobes in the Sphenophyllales is not 
therefore very strong. But Dr. Scott also relies on the dorsiventral 
lobing of the sporophyll as proving that the lower lobes originally 
had a protective function that had “ from some cause or other 
become superfluous, perhaps because the fertile laminae formed by 
themselves a sufficient envelope to the cone” (8). In our present 
ignorance of the conditions under which the Sphenophyllales lived 
we may as easily suppose that further protection became, for some 
unknown cause, advantageous and that this brought about the 
sterilization and modification into a bract-like structure of the lower 
sporangiophores. 
In Cheirostrobus each sporangiophore bore four sporangia ; in 
Sphenophyllum majus each sporangiophore (or perhaps sporophyll) 
bore four sporangia; in Bowmanites Romeri and Sphenophyllum 
