70 
“ The Origin of Gymnosperms ” 
were an old generalised or synthetic group, having much in common 
with archaic living ferns. Under other circumstances one might 
feel a trespasser, but Dr. Scott was like a kindly landowner who 
admitted others to the park which he had planted. He would draw 
attention to the Botryopterideae in one connexion only. The 
Pteridosperms included forms with a wide anatomical range. There 
were sojid monostelic forms ( Heterangium and Sutcliffia), medul- 
lated monostelic types ( Lyglnodendron) and complex polystelic forms 
(Meditllosa, Cladoxylon, etc.), in which each several stele is a centre 
of secondary thickening. Now it was a remarkable fact that among 
the Botryopterideffi there were represented vascular stem structures 
which might well serve as a starting point for the evolution of these 
different Pteridospermic types, as he would shew them by means of 
lantern slides. His view was that a great period of vegetative 
development was inaugurated among these archaic ferns in, perhaps, 
the Devonian epoch, and that this was correlated with a greatly 
increased demand for vascular tissue in the stem. This crisis had 
been met by an expansion of the primitive stele, sometimes limited, 
sometimes so violent that the verge of fragmentation was passed 
and the original stele split up into a number of parts. They would 
see by comparing the diagrams shewn that the Lyglnodendron- type, 
the simple Medullosa-type and the Cladoxylon- type could easily 
be derived from different forms of Botryopterid stele. The 
roots of Palms were instanced to illustrate the power of plants 
to fragment their steles under certain conditions. This suggestion, 
brought forward now in its crudest form, could not advance 
to the rank of a definite theory till the questions of leaf- 
traces and protoxylems had received a searching scrutiny. In con¬ 
clusion he would say that he was in sympathy with the general 
position laid down in Mr. Arber’s abstract. The Botryopterideae 
seemed to be the relics of a big generalised group of Cryptogamic 
plants, and contained the potentialities of evolution along many 
lines. In his opinion the fern-like plants remaining after all 
Pteridospermic deductions were adequate for the derivation of all 
the groups of seed plants, though these might have come off at 
very different epochs. There was only one more point he would 
raise. It was possible they were inclined to exaggerate the extent 
of the Pteridosperms. No doubt whole groups of Ferns had 
broken away from the Botryopterids, but might not others lie 
masked by the Pteridosperms. In the Lycopod phylum they knew 
they had Lepodendra with Lepidostrobus fructifications and others 
with the Lepidocarpon- type. The former were true Cryptogams, 
while the latter had several of the characters of seed-plants. To a 
certain extent the same thing was true of living Selaginellas. Was 
it not possible that, on the Filicinean side, our Heterangiums, 
Lyginodendra and Medullosae might be accompanied by purely 
cryptogamic species of like habit, but not necessarily heterosporous, 
whose real nature was masked by the recent discoveries of Pteri- 
dosperm seeds ? They would gather that he was inclined to a 
conservative view of the position, to depend still upon a crypto¬ 
gamic group for the origin of the Pteridosperms and eventually of 
the Cycads and Cordaiteae. And if of the Cordaiteae, then they 
were logically bound to include the Conifers as well, though the 
origin of these last was, of course, obscure, since their early forms 
had not been discovered in the petrified condition. 
