THE ORIGIN OF THE SEED-PLANTS 59 
An important subject of inquiry is uggested by the evidence as to the 
relative antiquity of Ferns and Seed-plants. There seems now to be no 
proof of the existence of Ferns ‚in the ordinary sense, before the Upper 
Devonian. In the earlier Devonian floras, the only suggestion of Fern- 
fronds is in the form of a naked, branched rachis, with no lamina, a 
structure perhaps little removed from the undifferentiated thallus. of 
the simplest landplants. 
On the other hand, the Gymnosperms appear to have been well de- 
veloped in early times. The Upper Devonian genus Callixylon had a be- 
autifully-organised wood, even rivalling that of the highest Conifers of 
the present day. This was not a Pteridosperm, but already something 
more advanced. From the Middle Devonian we have Hugh Miller’s fa- 
mous, but somewhat neglected „ConiferousTree’ (now Paloeopitys Mil- 
ler), which, whatever it may have been, shows the structure of a well- 
developed Gymnosperm. If we may judge by anatomical characters, 
the Seed-plants seem to have been in advance of the Spore-plants from 
about the time of the earliest known Land Flora. 
The inference from all the facts at present available appears to be 
that the Seed-plants, of which the Pteridosperms are among the earlier 
representatives, constitute aa independent phylum, of equal antiquity 
with any of the recognised lines of Vascular Cryptogams. Some of the 
Seed-plants (not necessarily all, for there may have been distinct races 
within the main phylum) passed through a Fern-like phase, but we 
have no reason to believe that they were ever Ferns. What the origin 
of the Spermophyta may have been is still unknown, but they must 
have sprung from some very early race, at least as ancient, and per- 
haps as simple, as the Psilophytales of KIDSTON and LANG. That such 
a hypothetical race may have been the common source of the Ferns al- 
so is not impossible, though the Fern-like features of the Pteridosperms 
are no doubt due, for the most part, to parallel development, rather 
than to common descent. 
The question must suggest itself :if the Seed-plants have always been 
distinct from any known line of Vascular Cryptogams, how did the 
seed arise? Were there heterosporous forms among the ancient and un- 
known race from which we must suppose the Spermophyta to have 
been derived ? Does the current, Hofmeisterian, theory of the origin of 
the seed represent what really took place, or are the accepted homolo- 
gies merely analogies? These are speculative questions, on which it 
